Commit | Line | Data |
---|---|---|
6cbd5570 | 1 | /* |
d352ac68 | 2 | * Copyright (C) 2007,2008 Oracle. All rights reserved. |
6cbd5570 CM |
3 | * |
4 | * This program is free software; you can redistribute it and/or | |
5 | * modify it under the terms of the GNU General Public | |
6 | * License v2 as published by the Free Software Foundation. | |
7 | * | |
8 | * This program is distributed in the hope that it will be useful, | |
9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
11 | * General Public License for more details. | |
12 | * | |
13 | * You should have received a copy of the GNU General Public | |
14 | * License along with this program; if not, write to the | |
15 | * Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
16 | * Boston, MA 021110-1307, USA. | |
17 | */ | |
18 | ||
a6b6e75e | 19 | #include <linux/sched.h> |
5a0e3ad6 | 20 | #include <linux/slab.h> |
eb60ceac CM |
21 | #include "ctree.h" |
22 | #include "disk-io.h" | |
7f5c1516 | 23 | #include "transaction.h" |
5f39d397 | 24 | #include "print-tree.h" |
925baedd | 25 | #include "locking.h" |
9a8dd150 | 26 | |
e089f05c CM |
27 | static int split_node(struct btrfs_trans_handle *trans, struct btrfs_root |
28 | *root, struct btrfs_path *path, int level); | |
29 | static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root | |
d4dbff95 | 30 | *root, struct btrfs_key *ins_key, |
cc0c5538 | 31 | struct btrfs_path *path, int data_size, int extend); |
5f39d397 CM |
32 | static int push_node_left(struct btrfs_trans_handle *trans, |
33 | struct btrfs_root *root, struct extent_buffer *dst, | |
971a1f66 | 34 | struct extent_buffer *src, int empty); |
5f39d397 CM |
35 | static int balance_node_right(struct btrfs_trans_handle *trans, |
36 | struct btrfs_root *root, | |
37 | struct extent_buffer *dst_buf, | |
38 | struct extent_buffer *src_buf); | |
e089f05c CM |
39 | static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root, |
40 | struct btrfs_path *path, int level, int slot); | |
ad48fd75 YZ |
41 | static int setup_items_for_insert(struct btrfs_trans_handle *trans, |
42 | struct btrfs_root *root, struct btrfs_path *path, | |
43 | struct btrfs_key *cpu_key, u32 *data_size, | |
44 | u32 total_data, u32 total_size, int nr); | |
45 | ||
d97e63b6 | 46 | |
df24a2b9 | 47 | struct btrfs_path *btrfs_alloc_path(void) |
2c90e5d6 | 48 | { |
df24a2b9 | 49 | struct btrfs_path *path; |
e00f7308 JM |
50 | path = kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS); |
51 | if (path) | |
2cc58cf2 | 52 | path->reada = 1; |
df24a2b9 | 53 | return path; |
2c90e5d6 CM |
54 | } |
55 | ||
b4ce94de CM |
56 | /* |
57 | * set all locked nodes in the path to blocking locks. This should | |
58 | * be done before scheduling | |
59 | */ | |
60 | noinline void btrfs_set_path_blocking(struct btrfs_path *p) | |
61 | { | |
62 | int i; | |
63 | for (i = 0; i < BTRFS_MAX_LEVEL; i++) { | |
64 | if (p->nodes[i] && p->locks[i]) | |
65 | btrfs_set_lock_blocking(p->nodes[i]); | |
66 | } | |
67 | } | |
68 | ||
69 | /* | |
70 | * reset all the locked nodes in the patch to spinning locks. | |
4008c04a CM |
71 | * |
72 | * held is used to keep lockdep happy, when lockdep is enabled | |
73 | * we set held to a blocking lock before we go around and | |
74 | * retake all the spinlocks in the path. You can safely use NULL | |
75 | * for held | |
b4ce94de | 76 | */ |
4008c04a CM |
77 | noinline void btrfs_clear_path_blocking(struct btrfs_path *p, |
78 | struct extent_buffer *held) | |
b4ce94de CM |
79 | { |
80 | int i; | |
4008c04a CM |
81 | |
82 | #ifdef CONFIG_DEBUG_LOCK_ALLOC | |
83 | /* lockdep really cares that we take all of these spinlocks | |
84 | * in the right order. If any of the locks in the path are not | |
85 | * currently blocking, it is going to complain. So, make really | |
86 | * really sure by forcing the path to blocking before we clear | |
87 | * the path blocking. | |
88 | */ | |
89 | if (held) | |
90 | btrfs_set_lock_blocking(held); | |
91 | btrfs_set_path_blocking(p); | |
92 | #endif | |
93 | ||
94 | for (i = BTRFS_MAX_LEVEL - 1; i >= 0; i--) { | |
b4ce94de CM |
95 | if (p->nodes[i] && p->locks[i]) |
96 | btrfs_clear_lock_blocking(p->nodes[i]); | |
97 | } | |
4008c04a CM |
98 | |
99 | #ifdef CONFIG_DEBUG_LOCK_ALLOC | |
100 | if (held) | |
101 | btrfs_clear_lock_blocking(held); | |
102 | #endif | |
b4ce94de CM |
103 | } |
104 | ||
d352ac68 | 105 | /* this also releases the path */ |
df24a2b9 | 106 | void btrfs_free_path(struct btrfs_path *p) |
be0e5c09 | 107 | { |
df24a2b9 CM |
108 | btrfs_release_path(NULL, p); |
109 | kmem_cache_free(btrfs_path_cachep, p); | |
be0e5c09 CM |
110 | } |
111 | ||
d352ac68 CM |
112 | /* |
113 | * path release drops references on the extent buffers in the path | |
114 | * and it drops any locks held by this path | |
115 | * | |
116 | * It is safe to call this on paths that no locks or extent buffers held. | |
117 | */ | |
d397712b | 118 | noinline void btrfs_release_path(struct btrfs_root *root, struct btrfs_path *p) |
eb60ceac CM |
119 | { |
120 | int i; | |
a2135011 | 121 | |
234b63a0 | 122 | for (i = 0; i < BTRFS_MAX_LEVEL; i++) { |
3f157a2f | 123 | p->slots[i] = 0; |
eb60ceac | 124 | if (!p->nodes[i]) |
925baedd CM |
125 | continue; |
126 | if (p->locks[i]) { | |
127 | btrfs_tree_unlock(p->nodes[i]); | |
128 | p->locks[i] = 0; | |
129 | } | |
5f39d397 | 130 | free_extent_buffer(p->nodes[i]); |
3f157a2f | 131 | p->nodes[i] = NULL; |
eb60ceac CM |
132 | } |
133 | } | |
134 | ||
d352ac68 CM |
135 | /* |
136 | * safely gets a reference on the root node of a tree. A lock | |
137 | * is not taken, so a concurrent writer may put a different node | |
138 | * at the root of the tree. See btrfs_lock_root_node for the | |
139 | * looping required. | |
140 | * | |
141 | * The extent buffer returned by this has a reference taken, so | |
142 | * it won't disappear. It may stop being the root of the tree | |
143 | * at any time because there are no locks held. | |
144 | */ | |
925baedd CM |
145 | struct extent_buffer *btrfs_root_node(struct btrfs_root *root) |
146 | { | |
147 | struct extent_buffer *eb; | |
148 | spin_lock(&root->node_lock); | |
149 | eb = root->node; | |
150 | extent_buffer_get(eb); | |
151 | spin_unlock(&root->node_lock); | |
152 | return eb; | |
153 | } | |
154 | ||
d352ac68 CM |
155 | /* loop around taking references on and locking the root node of the |
156 | * tree until you end up with a lock on the root. A locked buffer | |
157 | * is returned, with a reference held. | |
158 | */ | |
925baedd CM |
159 | struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root) |
160 | { | |
161 | struct extent_buffer *eb; | |
162 | ||
d397712b | 163 | while (1) { |
925baedd CM |
164 | eb = btrfs_root_node(root); |
165 | btrfs_tree_lock(eb); | |
166 | ||
167 | spin_lock(&root->node_lock); | |
168 | if (eb == root->node) { | |
169 | spin_unlock(&root->node_lock); | |
170 | break; | |
171 | } | |
172 | spin_unlock(&root->node_lock); | |
173 | ||
174 | btrfs_tree_unlock(eb); | |
175 | free_extent_buffer(eb); | |
176 | } | |
177 | return eb; | |
178 | } | |
179 | ||
d352ac68 CM |
180 | /* cowonly root (everything not a reference counted cow subvolume), just get |
181 | * put onto a simple dirty list. transaction.c walks this to make sure they | |
182 | * get properly updated on disk. | |
183 | */ | |
0b86a832 CM |
184 | static void add_root_to_dirty_list(struct btrfs_root *root) |
185 | { | |
186 | if (root->track_dirty && list_empty(&root->dirty_list)) { | |
187 | list_add(&root->dirty_list, | |
188 | &root->fs_info->dirty_cowonly_roots); | |
189 | } | |
190 | } | |
191 | ||
d352ac68 CM |
192 | /* |
193 | * used by snapshot creation to make a copy of a root for a tree with | |
194 | * a given objectid. The buffer with the new root node is returned in | |
195 | * cow_ret, and this func returns zero on success or a negative error code. | |
196 | */ | |
be20aa9d CM |
197 | int btrfs_copy_root(struct btrfs_trans_handle *trans, |
198 | struct btrfs_root *root, | |
199 | struct extent_buffer *buf, | |
200 | struct extent_buffer **cow_ret, u64 new_root_objectid) | |
201 | { | |
202 | struct extent_buffer *cow; | |
be20aa9d CM |
203 | int ret = 0; |
204 | int level; | |
5d4f98a2 | 205 | struct btrfs_disk_key disk_key; |
be20aa9d CM |
206 | |
207 | WARN_ON(root->ref_cows && trans->transid != | |
208 | root->fs_info->running_transaction->transid); | |
209 | WARN_ON(root->ref_cows && trans->transid != root->last_trans); | |
210 | ||
211 | level = btrfs_header_level(buf); | |
5d4f98a2 YZ |
212 | if (level == 0) |
213 | btrfs_item_key(buf, &disk_key, 0); | |
214 | else | |
215 | btrfs_node_key(buf, &disk_key, 0); | |
31840ae1 | 216 | |
5d4f98a2 YZ |
217 | cow = btrfs_alloc_free_block(trans, root, buf->len, 0, |
218 | new_root_objectid, &disk_key, level, | |
219 | buf->start, 0); | |
220 | if (IS_ERR(cow)) | |
be20aa9d CM |
221 | return PTR_ERR(cow); |
222 | ||
223 | copy_extent_buffer(cow, buf, 0, 0, cow->len); | |
224 | btrfs_set_header_bytenr(cow, cow->start); | |
225 | btrfs_set_header_generation(cow, trans->transid); | |
5d4f98a2 YZ |
226 | btrfs_set_header_backref_rev(cow, BTRFS_MIXED_BACKREF_REV); |
227 | btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN | | |
228 | BTRFS_HEADER_FLAG_RELOC); | |
229 | if (new_root_objectid == BTRFS_TREE_RELOC_OBJECTID) | |
230 | btrfs_set_header_flag(cow, BTRFS_HEADER_FLAG_RELOC); | |
231 | else | |
232 | btrfs_set_header_owner(cow, new_root_objectid); | |
be20aa9d | 233 | |
2b82032c YZ |
234 | write_extent_buffer(cow, root->fs_info->fsid, |
235 | (unsigned long)btrfs_header_fsid(cow), | |
236 | BTRFS_FSID_SIZE); | |
237 | ||
be20aa9d | 238 | WARN_ON(btrfs_header_generation(buf) > trans->transid); |
5d4f98a2 YZ |
239 | if (new_root_objectid == BTRFS_TREE_RELOC_OBJECTID) |
240 | ret = btrfs_inc_ref(trans, root, cow, 1); | |
241 | else | |
242 | ret = btrfs_inc_ref(trans, root, cow, 0); | |
4aec2b52 | 243 | |
be20aa9d CM |
244 | if (ret) |
245 | return ret; | |
246 | ||
247 | btrfs_mark_buffer_dirty(cow); | |
248 | *cow_ret = cow; | |
249 | return 0; | |
250 | } | |
251 | ||
5d4f98a2 YZ |
252 | /* |
253 | * check if the tree block can be shared by multiple trees | |
254 | */ | |
255 | int btrfs_block_can_be_shared(struct btrfs_root *root, | |
256 | struct extent_buffer *buf) | |
257 | { | |
258 | /* | |
259 | * Tree blocks not in refernece counted trees and tree roots | |
260 | * are never shared. If a block was allocated after the last | |
261 | * snapshot and the block was not allocated by tree relocation, | |
262 | * we know the block is not shared. | |
263 | */ | |
264 | if (root->ref_cows && | |
265 | buf != root->node && buf != root->commit_root && | |
266 | (btrfs_header_generation(buf) <= | |
267 | btrfs_root_last_snapshot(&root->root_item) || | |
268 | btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC))) | |
269 | return 1; | |
270 | #ifdef BTRFS_COMPAT_EXTENT_TREE_V0 | |
271 | if (root->ref_cows && | |
272 | btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV) | |
273 | return 1; | |
274 | #endif | |
275 | return 0; | |
276 | } | |
277 | ||
278 | static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, | |
279 | struct btrfs_root *root, | |
280 | struct extent_buffer *buf, | |
f0486c68 YZ |
281 | struct extent_buffer *cow, |
282 | int *last_ref) | |
5d4f98a2 YZ |
283 | { |
284 | u64 refs; | |
285 | u64 owner; | |
286 | u64 flags; | |
287 | u64 new_flags = 0; | |
288 | int ret; | |
289 | ||
290 | /* | |
291 | * Backrefs update rules: | |
292 | * | |
293 | * Always use full backrefs for extent pointers in tree block | |
294 | * allocated by tree relocation. | |
295 | * | |
296 | * If a shared tree block is no longer referenced by its owner | |
297 | * tree (btrfs_header_owner(buf) == root->root_key.objectid), | |
298 | * use full backrefs for extent pointers in tree block. | |
299 | * | |
300 | * If a tree block is been relocating | |
301 | * (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID), | |
302 | * use full backrefs for extent pointers in tree block. | |
303 | * The reason for this is some operations (such as drop tree) | |
304 | * are only allowed for blocks use full backrefs. | |
305 | */ | |
306 | ||
307 | if (btrfs_block_can_be_shared(root, buf)) { | |
308 | ret = btrfs_lookup_extent_info(trans, root, buf->start, | |
309 | buf->len, &refs, &flags); | |
310 | BUG_ON(ret); | |
311 | BUG_ON(refs == 0); | |
312 | } else { | |
313 | refs = 1; | |
314 | if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID || | |
315 | btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV) | |
316 | flags = BTRFS_BLOCK_FLAG_FULL_BACKREF; | |
317 | else | |
318 | flags = 0; | |
319 | } | |
320 | ||
321 | owner = btrfs_header_owner(buf); | |
322 | BUG_ON(owner == BTRFS_TREE_RELOC_OBJECTID && | |
323 | !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)); | |
324 | ||
325 | if (refs > 1) { | |
326 | if ((owner == root->root_key.objectid || | |
327 | root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) && | |
328 | !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)) { | |
329 | ret = btrfs_inc_ref(trans, root, buf, 1); | |
330 | BUG_ON(ret); | |
331 | ||
332 | if (root->root_key.objectid == | |
333 | BTRFS_TREE_RELOC_OBJECTID) { | |
334 | ret = btrfs_dec_ref(trans, root, buf, 0); | |
335 | BUG_ON(ret); | |
336 | ret = btrfs_inc_ref(trans, root, cow, 1); | |
337 | BUG_ON(ret); | |
338 | } | |
339 | new_flags |= BTRFS_BLOCK_FLAG_FULL_BACKREF; | |
340 | } else { | |
341 | ||
342 | if (root->root_key.objectid == | |
343 | BTRFS_TREE_RELOC_OBJECTID) | |
344 | ret = btrfs_inc_ref(trans, root, cow, 1); | |
345 | else | |
346 | ret = btrfs_inc_ref(trans, root, cow, 0); | |
347 | BUG_ON(ret); | |
348 | } | |
349 | if (new_flags != 0) { | |
350 | ret = btrfs_set_disk_extent_flags(trans, root, | |
351 | buf->start, | |
352 | buf->len, | |
353 | new_flags, 0); | |
354 | BUG_ON(ret); | |
355 | } | |
356 | } else { | |
357 | if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF) { | |
358 | if (root->root_key.objectid == | |
359 | BTRFS_TREE_RELOC_OBJECTID) | |
360 | ret = btrfs_inc_ref(trans, root, cow, 1); | |
361 | else | |
362 | ret = btrfs_inc_ref(trans, root, cow, 0); | |
363 | BUG_ON(ret); | |
364 | ret = btrfs_dec_ref(trans, root, buf, 1); | |
365 | BUG_ON(ret); | |
366 | } | |
367 | clean_tree_block(trans, root, buf); | |
f0486c68 | 368 | *last_ref = 1; |
5d4f98a2 YZ |
369 | } |
370 | return 0; | |
371 | } | |
372 | ||
d352ac68 | 373 | /* |
d397712b CM |
374 | * does the dirty work in cow of a single block. The parent block (if |
375 | * supplied) is updated to point to the new cow copy. The new buffer is marked | |
376 | * dirty and returned locked. If you modify the block it needs to be marked | |
377 | * dirty again. | |
d352ac68 CM |
378 | * |
379 | * search_start -- an allocation hint for the new block | |
380 | * | |
d397712b CM |
381 | * empty_size -- a hint that you plan on doing more cow. This is the size in |
382 | * bytes the allocator should try to find free next to the block it returns. | |
383 | * This is just a hint and may be ignored by the allocator. | |
d352ac68 | 384 | */ |
d397712b | 385 | static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans, |
5f39d397 CM |
386 | struct btrfs_root *root, |
387 | struct extent_buffer *buf, | |
388 | struct extent_buffer *parent, int parent_slot, | |
389 | struct extent_buffer **cow_ret, | |
9fa8cfe7 | 390 | u64 search_start, u64 empty_size) |
02217ed2 | 391 | { |
5d4f98a2 | 392 | struct btrfs_disk_key disk_key; |
5f39d397 | 393 | struct extent_buffer *cow; |
7bb86316 | 394 | int level; |
f0486c68 | 395 | int last_ref = 0; |
925baedd | 396 | int unlock_orig = 0; |
5d4f98a2 | 397 | u64 parent_start; |
7bb86316 | 398 | |
925baedd CM |
399 | if (*cow_ret == buf) |
400 | unlock_orig = 1; | |
401 | ||
b9447ef8 | 402 | btrfs_assert_tree_locked(buf); |
925baedd | 403 | |
7bb86316 CM |
404 | WARN_ON(root->ref_cows && trans->transid != |
405 | root->fs_info->running_transaction->transid); | |
6702ed49 | 406 | WARN_ON(root->ref_cows && trans->transid != root->last_trans); |
5f39d397 | 407 | |
7bb86316 | 408 | level = btrfs_header_level(buf); |
31840ae1 | 409 | |
5d4f98a2 YZ |
410 | if (level == 0) |
411 | btrfs_item_key(buf, &disk_key, 0); | |
412 | else | |
413 | btrfs_node_key(buf, &disk_key, 0); | |
414 | ||
415 | if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) { | |
416 | if (parent) | |
417 | parent_start = parent->start; | |
418 | else | |
419 | parent_start = 0; | |
420 | } else | |
421 | parent_start = 0; | |
422 | ||
423 | cow = btrfs_alloc_free_block(trans, root, buf->len, parent_start, | |
424 | root->root_key.objectid, &disk_key, | |
425 | level, search_start, empty_size); | |
54aa1f4d CM |
426 | if (IS_ERR(cow)) |
427 | return PTR_ERR(cow); | |
6702ed49 | 428 | |
b4ce94de CM |
429 | /* cow is set to blocking by btrfs_init_new_buffer */ |
430 | ||
5f39d397 | 431 | copy_extent_buffer(cow, buf, 0, 0, cow->len); |
db94535d | 432 | btrfs_set_header_bytenr(cow, cow->start); |
5f39d397 | 433 | btrfs_set_header_generation(cow, trans->transid); |
5d4f98a2 YZ |
434 | btrfs_set_header_backref_rev(cow, BTRFS_MIXED_BACKREF_REV); |
435 | btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN | | |
436 | BTRFS_HEADER_FLAG_RELOC); | |
437 | if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) | |
438 | btrfs_set_header_flag(cow, BTRFS_HEADER_FLAG_RELOC); | |
439 | else | |
440 | btrfs_set_header_owner(cow, root->root_key.objectid); | |
6702ed49 | 441 | |
2b82032c YZ |
442 | write_extent_buffer(cow, root->fs_info->fsid, |
443 | (unsigned long)btrfs_header_fsid(cow), | |
444 | BTRFS_FSID_SIZE); | |
445 | ||
f0486c68 | 446 | update_ref_for_cow(trans, root, buf, cow, &last_ref); |
1a40e23b | 447 | |
3fd0a558 YZ |
448 | if (root->ref_cows) |
449 | btrfs_reloc_cow_block(trans, root, buf, cow); | |
450 | ||
02217ed2 | 451 | if (buf == root->node) { |
925baedd | 452 | WARN_ON(parent && parent != buf); |
5d4f98a2 YZ |
453 | if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID || |
454 | btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV) | |
455 | parent_start = buf->start; | |
456 | else | |
457 | parent_start = 0; | |
925baedd CM |
458 | |
459 | spin_lock(&root->node_lock); | |
02217ed2 | 460 | root->node = cow; |
5f39d397 | 461 | extent_buffer_get(cow); |
925baedd CM |
462 | spin_unlock(&root->node_lock); |
463 | ||
f0486c68 YZ |
464 | btrfs_free_tree_block(trans, root, buf, parent_start, |
465 | last_ref); | |
5f39d397 | 466 | free_extent_buffer(buf); |
0b86a832 | 467 | add_root_to_dirty_list(root); |
02217ed2 | 468 | } else { |
5d4f98a2 YZ |
469 | if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) |
470 | parent_start = parent->start; | |
471 | else | |
472 | parent_start = 0; | |
473 | ||
474 | WARN_ON(trans->transid != btrfs_header_generation(parent)); | |
5f39d397 | 475 | btrfs_set_node_blockptr(parent, parent_slot, |
db94535d | 476 | cow->start); |
74493f7a CM |
477 | btrfs_set_node_ptr_generation(parent, parent_slot, |
478 | trans->transid); | |
d6025579 | 479 | btrfs_mark_buffer_dirty(parent); |
f0486c68 YZ |
480 | btrfs_free_tree_block(trans, root, buf, parent_start, |
481 | last_ref); | |
02217ed2 | 482 | } |
925baedd CM |
483 | if (unlock_orig) |
484 | btrfs_tree_unlock(buf); | |
5f39d397 | 485 | free_extent_buffer(buf); |
ccd467d6 | 486 | btrfs_mark_buffer_dirty(cow); |
2c90e5d6 | 487 | *cow_ret = cow; |
02217ed2 CM |
488 | return 0; |
489 | } | |
490 | ||
5d4f98a2 YZ |
491 | static inline int should_cow_block(struct btrfs_trans_handle *trans, |
492 | struct btrfs_root *root, | |
493 | struct extent_buffer *buf) | |
494 | { | |
495 | if (btrfs_header_generation(buf) == trans->transid && | |
496 | !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN) && | |
497 | !(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID && | |
498 | btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC))) | |
499 | return 0; | |
500 | return 1; | |
501 | } | |
502 | ||
d352ac68 CM |
503 | /* |
504 | * cows a single block, see __btrfs_cow_block for the real work. | |
505 | * This version of it has extra checks so that a block isn't cow'd more than | |
506 | * once per transaction, as long as it hasn't been written yet | |
507 | */ | |
d397712b | 508 | noinline int btrfs_cow_block(struct btrfs_trans_handle *trans, |
5f39d397 CM |
509 | struct btrfs_root *root, struct extent_buffer *buf, |
510 | struct extent_buffer *parent, int parent_slot, | |
9fa8cfe7 | 511 | struct extent_buffer **cow_ret) |
6702ed49 CM |
512 | { |
513 | u64 search_start; | |
f510cfec | 514 | int ret; |
dc17ff8f | 515 | |
6702ed49 | 516 | if (trans->transaction != root->fs_info->running_transaction) { |
d397712b CM |
517 | printk(KERN_CRIT "trans %llu running %llu\n", |
518 | (unsigned long long)trans->transid, | |
519 | (unsigned long long) | |
6702ed49 CM |
520 | root->fs_info->running_transaction->transid); |
521 | WARN_ON(1); | |
522 | } | |
523 | if (trans->transid != root->fs_info->generation) { | |
d397712b CM |
524 | printk(KERN_CRIT "trans %llu running %llu\n", |
525 | (unsigned long long)trans->transid, | |
526 | (unsigned long long)root->fs_info->generation); | |
6702ed49 CM |
527 | WARN_ON(1); |
528 | } | |
dc17ff8f | 529 | |
5d4f98a2 | 530 | if (!should_cow_block(trans, root, buf)) { |
6702ed49 CM |
531 | *cow_ret = buf; |
532 | return 0; | |
533 | } | |
c487685d | 534 | |
0b86a832 | 535 | search_start = buf->start & ~((u64)(1024 * 1024 * 1024) - 1); |
b4ce94de CM |
536 | |
537 | if (parent) | |
538 | btrfs_set_lock_blocking(parent); | |
539 | btrfs_set_lock_blocking(buf); | |
540 | ||
f510cfec | 541 | ret = __btrfs_cow_block(trans, root, buf, parent, |
9fa8cfe7 | 542 | parent_slot, cow_ret, search_start, 0); |
f510cfec | 543 | return ret; |
6702ed49 CM |
544 | } |
545 | ||
d352ac68 CM |
546 | /* |
547 | * helper function for defrag to decide if two blocks pointed to by a | |
548 | * node are actually close by | |
549 | */ | |
6b80053d | 550 | static int close_blocks(u64 blocknr, u64 other, u32 blocksize) |
6702ed49 | 551 | { |
6b80053d | 552 | if (blocknr < other && other - (blocknr + blocksize) < 32768) |
6702ed49 | 553 | return 1; |
6b80053d | 554 | if (blocknr > other && blocknr - (other + blocksize) < 32768) |
6702ed49 CM |
555 | return 1; |
556 | return 0; | |
557 | } | |
558 | ||
081e9573 CM |
559 | /* |
560 | * compare two keys in a memcmp fashion | |
561 | */ | |
562 | static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2) | |
563 | { | |
564 | struct btrfs_key k1; | |
565 | ||
566 | btrfs_disk_key_to_cpu(&k1, disk); | |
567 | ||
20736aba | 568 | return btrfs_comp_cpu_keys(&k1, k2); |
081e9573 CM |
569 | } |
570 | ||
f3465ca4 JB |
571 | /* |
572 | * same as comp_keys only with two btrfs_key's | |
573 | */ | |
5d4f98a2 | 574 | int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2) |
f3465ca4 JB |
575 | { |
576 | if (k1->objectid > k2->objectid) | |
577 | return 1; | |
578 | if (k1->objectid < k2->objectid) | |
579 | return -1; | |
580 | if (k1->type > k2->type) | |
581 | return 1; | |
582 | if (k1->type < k2->type) | |
583 | return -1; | |
584 | if (k1->offset > k2->offset) | |
585 | return 1; | |
586 | if (k1->offset < k2->offset) | |
587 | return -1; | |
588 | return 0; | |
589 | } | |
081e9573 | 590 | |
d352ac68 CM |
591 | /* |
592 | * this is used by the defrag code to go through all the | |
593 | * leaves pointed to by a node and reallocate them so that | |
594 | * disk order is close to key order | |
595 | */ | |
6702ed49 | 596 | int btrfs_realloc_node(struct btrfs_trans_handle *trans, |
5f39d397 | 597 | struct btrfs_root *root, struct extent_buffer *parent, |
a6b6e75e CM |
598 | int start_slot, int cache_only, u64 *last_ret, |
599 | struct btrfs_key *progress) | |
6702ed49 | 600 | { |
6b80053d | 601 | struct extent_buffer *cur; |
6702ed49 | 602 | u64 blocknr; |
ca7a79ad | 603 | u64 gen; |
e9d0b13b CM |
604 | u64 search_start = *last_ret; |
605 | u64 last_block = 0; | |
6702ed49 CM |
606 | u64 other; |
607 | u32 parent_nritems; | |
6702ed49 CM |
608 | int end_slot; |
609 | int i; | |
610 | int err = 0; | |
f2183bde | 611 | int parent_level; |
6b80053d CM |
612 | int uptodate; |
613 | u32 blocksize; | |
081e9573 CM |
614 | int progress_passed = 0; |
615 | struct btrfs_disk_key disk_key; | |
6702ed49 | 616 | |
5708b959 CM |
617 | parent_level = btrfs_header_level(parent); |
618 | if (cache_only && parent_level != 1) | |
619 | return 0; | |
620 | ||
d397712b | 621 | if (trans->transaction != root->fs_info->running_transaction) |
6702ed49 | 622 | WARN_ON(1); |
d397712b | 623 | if (trans->transid != root->fs_info->generation) |
6702ed49 | 624 | WARN_ON(1); |
86479a04 | 625 | |
6b80053d | 626 | parent_nritems = btrfs_header_nritems(parent); |
6b80053d | 627 | blocksize = btrfs_level_size(root, parent_level - 1); |
6702ed49 CM |
628 | end_slot = parent_nritems; |
629 | ||
630 | if (parent_nritems == 1) | |
631 | return 0; | |
632 | ||
b4ce94de CM |
633 | btrfs_set_lock_blocking(parent); |
634 | ||
6702ed49 CM |
635 | for (i = start_slot; i < end_slot; i++) { |
636 | int close = 1; | |
a6b6e75e | 637 | |
5708b959 CM |
638 | if (!parent->map_token) { |
639 | map_extent_buffer(parent, | |
640 | btrfs_node_key_ptr_offset(i), | |
641 | sizeof(struct btrfs_key_ptr), | |
642 | &parent->map_token, &parent->kaddr, | |
643 | &parent->map_start, &parent->map_len, | |
644 | KM_USER1); | |
645 | } | |
081e9573 CM |
646 | btrfs_node_key(parent, &disk_key, i); |
647 | if (!progress_passed && comp_keys(&disk_key, progress) < 0) | |
648 | continue; | |
649 | ||
650 | progress_passed = 1; | |
6b80053d | 651 | blocknr = btrfs_node_blockptr(parent, i); |
ca7a79ad | 652 | gen = btrfs_node_ptr_generation(parent, i); |
e9d0b13b CM |
653 | if (last_block == 0) |
654 | last_block = blocknr; | |
5708b959 | 655 | |
6702ed49 | 656 | if (i > 0) { |
6b80053d CM |
657 | other = btrfs_node_blockptr(parent, i - 1); |
658 | close = close_blocks(blocknr, other, blocksize); | |
6702ed49 | 659 | } |
0ef3e66b | 660 | if (!close && i < end_slot - 2) { |
6b80053d CM |
661 | other = btrfs_node_blockptr(parent, i + 1); |
662 | close = close_blocks(blocknr, other, blocksize); | |
6702ed49 | 663 | } |
e9d0b13b CM |
664 | if (close) { |
665 | last_block = blocknr; | |
6702ed49 | 666 | continue; |
e9d0b13b | 667 | } |
5708b959 CM |
668 | if (parent->map_token) { |
669 | unmap_extent_buffer(parent, parent->map_token, | |
670 | KM_USER1); | |
671 | parent->map_token = NULL; | |
672 | } | |
6702ed49 | 673 | |
6b80053d CM |
674 | cur = btrfs_find_tree_block(root, blocknr, blocksize); |
675 | if (cur) | |
1259ab75 | 676 | uptodate = btrfs_buffer_uptodate(cur, gen); |
6b80053d CM |
677 | else |
678 | uptodate = 0; | |
5708b959 | 679 | if (!cur || !uptodate) { |
6702ed49 | 680 | if (cache_only) { |
6b80053d | 681 | free_extent_buffer(cur); |
6702ed49 CM |
682 | continue; |
683 | } | |
6b80053d CM |
684 | if (!cur) { |
685 | cur = read_tree_block(root, blocknr, | |
ca7a79ad | 686 | blocksize, gen); |
6b80053d | 687 | } else if (!uptodate) { |
ca7a79ad | 688 | btrfs_read_buffer(cur, gen); |
f2183bde | 689 | } |
6702ed49 | 690 | } |
e9d0b13b | 691 | if (search_start == 0) |
6b80053d | 692 | search_start = last_block; |
e9d0b13b | 693 | |
e7a84565 | 694 | btrfs_tree_lock(cur); |
b4ce94de | 695 | btrfs_set_lock_blocking(cur); |
6b80053d | 696 | err = __btrfs_cow_block(trans, root, cur, parent, i, |
e7a84565 | 697 | &cur, search_start, |
6b80053d | 698 | min(16 * blocksize, |
9fa8cfe7 | 699 | (end_slot - i) * blocksize)); |
252c38f0 | 700 | if (err) { |
e7a84565 | 701 | btrfs_tree_unlock(cur); |
6b80053d | 702 | free_extent_buffer(cur); |
6702ed49 | 703 | break; |
252c38f0 | 704 | } |
e7a84565 CM |
705 | search_start = cur->start; |
706 | last_block = cur->start; | |
f2183bde | 707 | *last_ret = search_start; |
e7a84565 CM |
708 | btrfs_tree_unlock(cur); |
709 | free_extent_buffer(cur); | |
6702ed49 | 710 | } |
5708b959 CM |
711 | if (parent->map_token) { |
712 | unmap_extent_buffer(parent, parent->map_token, | |
713 | KM_USER1); | |
714 | parent->map_token = NULL; | |
715 | } | |
6702ed49 CM |
716 | return err; |
717 | } | |
718 | ||
74123bd7 CM |
719 | /* |
720 | * The leaf data grows from end-to-front in the node. | |
721 | * this returns the address of the start of the last item, | |
722 | * which is the stop of the leaf data stack | |
723 | */ | |
123abc88 | 724 | static inline unsigned int leaf_data_end(struct btrfs_root *root, |
5f39d397 | 725 | struct extent_buffer *leaf) |
be0e5c09 | 726 | { |
5f39d397 | 727 | u32 nr = btrfs_header_nritems(leaf); |
be0e5c09 | 728 | if (nr == 0) |
123abc88 | 729 | return BTRFS_LEAF_DATA_SIZE(root); |
5f39d397 | 730 | return btrfs_item_offset_nr(leaf, nr - 1); |
be0e5c09 CM |
731 | } |
732 | ||
d352ac68 CM |
733 | /* |
734 | * extra debugging checks to make sure all the items in a key are | |
735 | * well formed and in the proper order | |
736 | */ | |
123abc88 CM |
737 | static int check_node(struct btrfs_root *root, struct btrfs_path *path, |
738 | int level) | |
aa5d6bed | 739 | { |
5f39d397 CM |
740 | struct extent_buffer *parent = NULL; |
741 | struct extent_buffer *node = path->nodes[level]; | |
742 | struct btrfs_disk_key parent_key; | |
743 | struct btrfs_disk_key node_key; | |
aa5d6bed | 744 | int parent_slot; |
8d7be552 CM |
745 | int slot; |
746 | struct btrfs_key cpukey; | |
5f39d397 | 747 | u32 nritems = btrfs_header_nritems(node); |
aa5d6bed CM |
748 | |
749 | if (path->nodes[level + 1]) | |
5f39d397 | 750 | parent = path->nodes[level + 1]; |
a1f39630 | 751 | |
8d7be552 | 752 | slot = path->slots[level]; |
7518a238 CM |
753 | BUG_ON(nritems == 0); |
754 | if (parent) { | |
a1f39630 | 755 | parent_slot = path->slots[level + 1]; |
5f39d397 CM |
756 | btrfs_node_key(parent, &parent_key, parent_slot); |
757 | btrfs_node_key(node, &node_key, 0); | |
758 | BUG_ON(memcmp(&parent_key, &node_key, | |
e2fa7227 | 759 | sizeof(struct btrfs_disk_key))); |
1d4f8a0c | 760 | BUG_ON(btrfs_node_blockptr(parent, parent_slot) != |
db94535d | 761 | btrfs_header_bytenr(node)); |
aa5d6bed | 762 | } |
123abc88 | 763 | BUG_ON(nritems > BTRFS_NODEPTRS_PER_BLOCK(root)); |
8d7be552 | 764 | if (slot != 0) { |
5f39d397 CM |
765 | btrfs_node_key_to_cpu(node, &cpukey, slot - 1); |
766 | btrfs_node_key(node, &node_key, slot); | |
767 | BUG_ON(comp_keys(&node_key, &cpukey) <= 0); | |
8d7be552 CM |
768 | } |
769 | if (slot < nritems - 1) { | |
5f39d397 CM |
770 | btrfs_node_key_to_cpu(node, &cpukey, slot + 1); |
771 | btrfs_node_key(node, &node_key, slot); | |
772 | BUG_ON(comp_keys(&node_key, &cpukey) >= 0); | |
aa5d6bed CM |
773 | } |
774 | return 0; | |
775 | } | |
776 | ||
d352ac68 CM |
777 | /* |
778 | * extra checking to make sure all the items in a leaf are | |
779 | * well formed and in the proper order | |
780 | */ | |
123abc88 CM |
781 | static int check_leaf(struct btrfs_root *root, struct btrfs_path *path, |
782 | int level) | |
aa5d6bed | 783 | { |
5f39d397 CM |
784 | struct extent_buffer *leaf = path->nodes[level]; |
785 | struct extent_buffer *parent = NULL; | |
aa5d6bed | 786 | int parent_slot; |
8d7be552 | 787 | struct btrfs_key cpukey; |
5f39d397 CM |
788 | struct btrfs_disk_key parent_key; |
789 | struct btrfs_disk_key leaf_key; | |
790 | int slot = path->slots[0]; | |
8d7be552 | 791 | |
5f39d397 | 792 | u32 nritems = btrfs_header_nritems(leaf); |
aa5d6bed CM |
793 | |
794 | if (path->nodes[level + 1]) | |
5f39d397 | 795 | parent = path->nodes[level + 1]; |
7518a238 CM |
796 | |
797 | if (nritems == 0) | |
798 | return 0; | |
799 | ||
800 | if (parent) { | |
a1f39630 | 801 | parent_slot = path->slots[level + 1]; |
5f39d397 CM |
802 | btrfs_node_key(parent, &parent_key, parent_slot); |
803 | btrfs_item_key(leaf, &leaf_key, 0); | |
6702ed49 | 804 | |
5f39d397 | 805 | BUG_ON(memcmp(&parent_key, &leaf_key, |
e2fa7227 | 806 | sizeof(struct btrfs_disk_key))); |
1d4f8a0c | 807 | BUG_ON(btrfs_node_blockptr(parent, parent_slot) != |
db94535d | 808 | btrfs_header_bytenr(leaf)); |
5f39d397 | 809 | } |
5f39d397 CM |
810 | if (slot != 0 && slot < nritems - 1) { |
811 | btrfs_item_key(leaf, &leaf_key, slot); | |
812 | btrfs_item_key_to_cpu(leaf, &cpukey, slot - 1); | |
813 | if (comp_keys(&leaf_key, &cpukey) <= 0) { | |
814 | btrfs_print_leaf(root, leaf); | |
d397712b | 815 | printk(KERN_CRIT "slot %d offset bad key\n", slot); |
5f39d397 CM |
816 | BUG_ON(1); |
817 | } | |
818 | if (btrfs_item_offset_nr(leaf, slot - 1) != | |
819 | btrfs_item_end_nr(leaf, slot)) { | |
820 | btrfs_print_leaf(root, leaf); | |
d397712b | 821 | printk(KERN_CRIT "slot %d offset bad\n", slot); |
5f39d397 CM |
822 | BUG_ON(1); |
823 | } | |
8d7be552 CM |
824 | } |
825 | if (slot < nritems - 1) { | |
5f39d397 CM |
826 | btrfs_item_key(leaf, &leaf_key, slot); |
827 | btrfs_item_key_to_cpu(leaf, &cpukey, slot + 1); | |
828 | BUG_ON(comp_keys(&leaf_key, &cpukey) >= 0); | |
829 | if (btrfs_item_offset_nr(leaf, slot) != | |
830 | btrfs_item_end_nr(leaf, slot + 1)) { | |
831 | btrfs_print_leaf(root, leaf); | |
d397712b | 832 | printk(KERN_CRIT "slot %d offset bad\n", slot); |
5f39d397 CM |
833 | BUG_ON(1); |
834 | } | |
aa5d6bed | 835 | } |
5f39d397 CM |
836 | BUG_ON(btrfs_item_offset_nr(leaf, 0) + |
837 | btrfs_item_size_nr(leaf, 0) != BTRFS_LEAF_DATA_SIZE(root)); | |
aa5d6bed CM |
838 | return 0; |
839 | } | |
840 | ||
d397712b | 841 | static noinline int check_block(struct btrfs_root *root, |
98ed5174 | 842 | struct btrfs_path *path, int level) |
aa5d6bed | 843 | { |
85d824c4 | 844 | return 0; |
aa5d6bed | 845 | if (level == 0) |
123abc88 CM |
846 | return check_leaf(root, path, level); |
847 | return check_node(root, path, level); | |
aa5d6bed CM |
848 | } |
849 | ||
74123bd7 | 850 | /* |
5f39d397 CM |
851 | * search for key in the extent_buffer. The items start at offset p, |
852 | * and they are item_size apart. There are 'max' items in p. | |
853 | * | |
74123bd7 CM |
854 | * the slot in the array is returned via slot, and it points to |
855 | * the place where you would insert key if it is not found in | |
856 | * the array. | |
857 | * | |
858 | * slot may point to max if the key is bigger than all of the keys | |
859 | */ | |
e02119d5 CM |
860 | static noinline int generic_bin_search(struct extent_buffer *eb, |
861 | unsigned long p, | |
862 | int item_size, struct btrfs_key *key, | |
863 | int max, int *slot) | |
be0e5c09 CM |
864 | { |
865 | int low = 0; | |
866 | int high = max; | |
867 | int mid; | |
868 | int ret; | |
479965d6 | 869 | struct btrfs_disk_key *tmp = NULL; |
5f39d397 CM |
870 | struct btrfs_disk_key unaligned; |
871 | unsigned long offset; | |
872 | char *map_token = NULL; | |
873 | char *kaddr = NULL; | |
874 | unsigned long map_start = 0; | |
875 | unsigned long map_len = 0; | |
479965d6 | 876 | int err; |
be0e5c09 | 877 | |
d397712b | 878 | while (low < high) { |
be0e5c09 | 879 | mid = (low + high) / 2; |
5f39d397 CM |
880 | offset = p + mid * item_size; |
881 | ||
882 | if (!map_token || offset < map_start || | |
883 | (offset + sizeof(struct btrfs_disk_key)) > | |
884 | map_start + map_len) { | |
479965d6 | 885 | if (map_token) { |
5f39d397 | 886 | unmap_extent_buffer(eb, map_token, KM_USER0); |
479965d6 CM |
887 | map_token = NULL; |
888 | } | |
934d375b CM |
889 | |
890 | err = map_private_extent_buffer(eb, offset, | |
479965d6 CM |
891 | sizeof(struct btrfs_disk_key), |
892 | &map_token, &kaddr, | |
893 | &map_start, &map_len, KM_USER0); | |
894 | ||
895 | if (!err) { | |
896 | tmp = (struct btrfs_disk_key *)(kaddr + offset - | |
897 | map_start); | |
898 | } else { | |
899 | read_extent_buffer(eb, &unaligned, | |
900 | offset, sizeof(unaligned)); | |
901 | tmp = &unaligned; | |
902 | } | |
5f39d397 | 903 | |
5f39d397 CM |
904 | } else { |
905 | tmp = (struct btrfs_disk_key *)(kaddr + offset - | |
906 | map_start); | |
907 | } | |
be0e5c09 CM |
908 | ret = comp_keys(tmp, key); |
909 | ||
910 | if (ret < 0) | |
911 | low = mid + 1; | |
912 | else if (ret > 0) | |
913 | high = mid; | |
914 | else { | |
915 | *slot = mid; | |
479965d6 CM |
916 | if (map_token) |
917 | unmap_extent_buffer(eb, map_token, KM_USER0); | |
be0e5c09 CM |
918 | return 0; |
919 | } | |
920 | } | |
921 | *slot = low; | |
5f39d397 CM |
922 | if (map_token) |
923 | unmap_extent_buffer(eb, map_token, KM_USER0); | |
be0e5c09 CM |
924 | return 1; |
925 | } | |
926 | ||
97571fd0 CM |
927 | /* |
928 | * simple bin_search frontend that does the right thing for | |
929 | * leaves vs nodes | |
930 | */ | |
5f39d397 CM |
931 | static int bin_search(struct extent_buffer *eb, struct btrfs_key *key, |
932 | int level, int *slot) | |
be0e5c09 | 933 | { |
5f39d397 CM |
934 | if (level == 0) { |
935 | return generic_bin_search(eb, | |
936 | offsetof(struct btrfs_leaf, items), | |
0783fcfc | 937 | sizeof(struct btrfs_item), |
5f39d397 | 938 | key, btrfs_header_nritems(eb), |
7518a238 | 939 | slot); |
be0e5c09 | 940 | } else { |
5f39d397 CM |
941 | return generic_bin_search(eb, |
942 | offsetof(struct btrfs_node, ptrs), | |
123abc88 | 943 | sizeof(struct btrfs_key_ptr), |
5f39d397 | 944 | key, btrfs_header_nritems(eb), |
7518a238 | 945 | slot); |
be0e5c09 CM |
946 | } |
947 | return -1; | |
948 | } | |
949 | ||
5d4f98a2 YZ |
950 | int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key, |
951 | int level, int *slot) | |
952 | { | |
953 | return bin_search(eb, key, level, slot); | |
954 | } | |
955 | ||
f0486c68 YZ |
956 | static void root_add_used(struct btrfs_root *root, u32 size) |
957 | { | |
958 | spin_lock(&root->accounting_lock); | |
959 | btrfs_set_root_used(&root->root_item, | |
960 | btrfs_root_used(&root->root_item) + size); | |
961 | spin_unlock(&root->accounting_lock); | |
962 | } | |
963 | ||
964 | static void root_sub_used(struct btrfs_root *root, u32 size) | |
965 | { | |
966 | spin_lock(&root->accounting_lock); | |
967 | btrfs_set_root_used(&root->root_item, | |
968 | btrfs_root_used(&root->root_item) - size); | |
969 | spin_unlock(&root->accounting_lock); | |
970 | } | |
971 | ||
d352ac68 CM |
972 | /* given a node and slot number, this reads the blocks it points to. The |
973 | * extent buffer is returned with a reference taken (but unlocked). | |
974 | * NULL is returned on error. | |
975 | */ | |
e02119d5 | 976 | static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root, |
5f39d397 | 977 | struct extent_buffer *parent, int slot) |
bb803951 | 978 | { |
ca7a79ad | 979 | int level = btrfs_header_level(parent); |
bb803951 CM |
980 | if (slot < 0) |
981 | return NULL; | |
5f39d397 | 982 | if (slot >= btrfs_header_nritems(parent)) |
bb803951 | 983 | return NULL; |
ca7a79ad CM |
984 | |
985 | BUG_ON(level == 0); | |
986 | ||
db94535d | 987 | return read_tree_block(root, btrfs_node_blockptr(parent, slot), |
ca7a79ad CM |
988 | btrfs_level_size(root, level - 1), |
989 | btrfs_node_ptr_generation(parent, slot)); | |
bb803951 CM |
990 | } |
991 | ||
d352ac68 CM |
992 | /* |
993 | * node level balancing, used to make sure nodes are in proper order for | |
994 | * item deletion. We balance from the top down, so we have to make sure | |
995 | * that a deletion won't leave an node completely empty later on. | |
996 | */ | |
e02119d5 | 997 | static noinline int balance_level(struct btrfs_trans_handle *trans, |
98ed5174 CM |
998 | struct btrfs_root *root, |
999 | struct btrfs_path *path, int level) | |
bb803951 | 1000 | { |
5f39d397 CM |
1001 | struct extent_buffer *right = NULL; |
1002 | struct extent_buffer *mid; | |
1003 | struct extent_buffer *left = NULL; | |
1004 | struct extent_buffer *parent = NULL; | |
bb803951 CM |
1005 | int ret = 0; |
1006 | int wret; | |
1007 | int pslot; | |
bb803951 | 1008 | int orig_slot = path->slots[level]; |
79f95c82 | 1009 | u64 orig_ptr; |
bb803951 CM |
1010 | |
1011 | if (level == 0) | |
1012 | return 0; | |
1013 | ||
5f39d397 | 1014 | mid = path->nodes[level]; |
b4ce94de | 1015 | |
925baedd | 1016 | WARN_ON(!path->locks[level]); |
7bb86316 CM |
1017 | WARN_ON(btrfs_header_generation(mid) != trans->transid); |
1018 | ||
1d4f8a0c | 1019 | orig_ptr = btrfs_node_blockptr(mid, orig_slot); |
79f95c82 | 1020 | |
234b63a0 | 1021 | if (level < BTRFS_MAX_LEVEL - 1) |
5f39d397 | 1022 | parent = path->nodes[level + 1]; |
bb803951 CM |
1023 | pslot = path->slots[level + 1]; |
1024 | ||
40689478 CM |
1025 | /* |
1026 | * deal with the case where there is only one pointer in the root | |
1027 | * by promoting the node below to a root | |
1028 | */ | |
5f39d397 CM |
1029 | if (!parent) { |
1030 | struct extent_buffer *child; | |
bb803951 | 1031 | |
5f39d397 | 1032 | if (btrfs_header_nritems(mid) != 1) |
bb803951 CM |
1033 | return 0; |
1034 | ||
1035 | /* promote the child to a root */ | |
5f39d397 | 1036 | child = read_node_slot(root, mid, 0); |
7951f3ce | 1037 | BUG_ON(!child); |
925baedd | 1038 | btrfs_tree_lock(child); |
b4ce94de | 1039 | btrfs_set_lock_blocking(child); |
9fa8cfe7 | 1040 | ret = btrfs_cow_block(trans, root, child, mid, 0, &child); |
f0486c68 YZ |
1041 | if (ret) { |
1042 | btrfs_tree_unlock(child); | |
1043 | free_extent_buffer(child); | |
1044 | goto enospc; | |
1045 | } | |
2f375ab9 | 1046 | |
925baedd | 1047 | spin_lock(&root->node_lock); |
bb803951 | 1048 | root->node = child; |
925baedd CM |
1049 | spin_unlock(&root->node_lock); |
1050 | ||
0b86a832 | 1051 | add_root_to_dirty_list(root); |
925baedd | 1052 | btrfs_tree_unlock(child); |
b4ce94de | 1053 | |
925baedd | 1054 | path->locks[level] = 0; |
bb803951 | 1055 | path->nodes[level] = NULL; |
5f39d397 | 1056 | clean_tree_block(trans, root, mid); |
925baedd | 1057 | btrfs_tree_unlock(mid); |
bb803951 | 1058 | /* once for the path */ |
5f39d397 | 1059 | free_extent_buffer(mid); |
f0486c68 YZ |
1060 | |
1061 | root_sub_used(root, mid->len); | |
1062 | btrfs_free_tree_block(trans, root, mid, 0, 1); | |
bb803951 | 1063 | /* once for the root ptr */ |
5f39d397 | 1064 | free_extent_buffer(mid); |
f0486c68 | 1065 | return 0; |
bb803951 | 1066 | } |
5f39d397 | 1067 | if (btrfs_header_nritems(mid) > |
123abc88 | 1068 | BTRFS_NODEPTRS_PER_BLOCK(root) / 4) |
bb803951 CM |
1069 | return 0; |
1070 | ||
559af821 | 1071 | btrfs_header_nritems(mid); |
54aa1f4d | 1072 | |
5f39d397 CM |
1073 | left = read_node_slot(root, parent, pslot - 1); |
1074 | if (left) { | |
925baedd | 1075 | btrfs_tree_lock(left); |
b4ce94de | 1076 | btrfs_set_lock_blocking(left); |
5f39d397 | 1077 | wret = btrfs_cow_block(trans, root, left, |
9fa8cfe7 | 1078 | parent, pslot - 1, &left); |
54aa1f4d CM |
1079 | if (wret) { |
1080 | ret = wret; | |
1081 | goto enospc; | |
1082 | } | |
2cc58cf2 | 1083 | } |
5f39d397 CM |
1084 | right = read_node_slot(root, parent, pslot + 1); |
1085 | if (right) { | |
925baedd | 1086 | btrfs_tree_lock(right); |
b4ce94de | 1087 | btrfs_set_lock_blocking(right); |
5f39d397 | 1088 | wret = btrfs_cow_block(trans, root, right, |
9fa8cfe7 | 1089 | parent, pslot + 1, &right); |
2cc58cf2 CM |
1090 | if (wret) { |
1091 | ret = wret; | |
1092 | goto enospc; | |
1093 | } | |
1094 | } | |
1095 | ||
1096 | /* first, try to make some room in the middle buffer */ | |
5f39d397 CM |
1097 | if (left) { |
1098 | orig_slot += btrfs_header_nritems(left); | |
bce4eae9 | 1099 | wret = push_node_left(trans, root, left, mid, 1); |
79f95c82 CM |
1100 | if (wret < 0) |
1101 | ret = wret; | |
559af821 | 1102 | btrfs_header_nritems(mid); |
bb803951 | 1103 | } |
79f95c82 CM |
1104 | |
1105 | /* | |
1106 | * then try to empty the right most buffer into the middle | |
1107 | */ | |
5f39d397 | 1108 | if (right) { |
971a1f66 | 1109 | wret = push_node_left(trans, root, mid, right, 1); |
54aa1f4d | 1110 | if (wret < 0 && wret != -ENOSPC) |
79f95c82 | 1111 | ret = wret; |
5f39d397 | 1112 | if (btrfs_header_nritems(right) == 0) { |
5f39d397 | 1113 | clean_tree_block(trans, root, right); |
925baedd | 1114 | btrfs_tree_unlock(right); |
e089f05c CM |
1115 | wret = del_ptr(trans, root, path, level + 1, pslot + |
1116 | 1); | |
bb803951 CM |
1117 | if (wret) |
1118 | ret = wret; | |
f0486c68 YZ |
1119 | root_sub_used(root, right->len); |
1120 | btrfs_free_tree_block(trans, root, right, 0, 1); | |
1121 | free_extent_buffer(right); | |
1122 | right = NULL; | |
bb803951 | 1123 | } else { |
5f39d397 CM |
1124 | struct btrfs_disk_key right_key; |
1125 | btrfs_node_key(right, &right_key, 0); | |
1126 | btrfs_set_node_key(parent, &right_key, pslot + 1); | |
1127 | btrfs_mark_buffer_dirty(parent); | |
bb803951 CM |
1128 | } |
1129 | } | |
5f39d397 | 1130 | if (btrfs_header_nritems(mid) == 1) { |
79f95c82 CM |
1131 | /* |
1132 | * we're not allowed to leave a node with one item in the | |
1133 | * tree during a delete. A deletion from lower in the tree | |
1134 | * could try to delete the only pointer in this node. | |
1135 | * So, pull some keys from the left. | |
1136 | * There has to be a left pointer at this point because | |
1137 | * otherwise we would have pulled some pointers from the | |
1138 | * right | |
1139 | */ | |
5f39d397 CM |
1140 | BUG_ON(!left); |
1141 | wret = balance_node_right(trans, root, mid, left); | |
54aa1f4d | 1142 | if (wret < 0) { |
79f95c82 | 1143 | ret = wret; |
54aa1f4d CM |
1144 | goto enospc; |
1145 | } | |
bce4eae9 CM |
1146 | if (wret == 1) { |
1147 | wret = push_node_left(trans, root, left, mid, 1); | |
1148 | if (wret < 0) | |
1149 | ret = wret; | |
1150 | } | |
79f95c82 CM |
1151 | BUG_ON(wret == 1); |
1152 | } | |
5f39d397 | 1153 | if (btrfs_header_nritems(mid) == 0) { |
5f39d397 | 1154 | clean_tree_block(trans, root, mid); |
925baedd | 1155 | btrfs_tree_unlock(mid); |
e089f05c | 1156 | wret = del_ptr(trans, root, path, level + 1, pslot); |
bb803951 CM |
1157 | if (wret) |
1158 | ret = wret; | |
f0486c68 YZ |
1159 | root_sub_used(root, mid->len); |
1160 | btrfs_free_tree_block(trans, root, mid, 0, 1); | |
1161 | free_extent_buffer(mid); | |
1162 | mid = NULL; | |
79f95c82 CM |
1163 | } else { |
1164 | /* update the parent key to reflect our changes */ | |
5f39d397 CM |
1165 | struct btrfs_disk_key mid_key; |
1166 | btrfs_node_key(mid, &mid_key, 0); | |
1167 | btrfs_set_node_key(parent, &mid_key, pslot); | |
1168 | btrfs_mark_buffer_dirty(parent); | |
79f95c82 | 1169 | } |
bb803951 | 1170 | |
79f95c82 | 1171 | /* update the path */ |
5f39d397 CM |
1172 | if (left) { |
1173 | if (btrfs_header_nritems(left) > orig_slot) { | |
1174 | extent_buffer_get(left); | |
925baedd | 1175 | /* left was locked after cow */ |
5f39d397 | 1176 | path->nodes[level] = left; |
bb803951 CM |
1177 | path->slots[level + 1] -= 1; |
1178 | path->slots[level] = orig_slot; | |
925baedd CM |
1179 | if (mid) { |
1180 | btrfs_tree_unlock(mid); | |
5f39d397 | 1181 | free_extent_buffer(mid); |
925baedd | 1182 | } |
bb803951 | 1183 | } else { |
5f39d397 | 1184 | orig_slot -= btrfs_header_nritems(left); |
bb803951 CM |
1185 | path->slots[level] = orig_slot; |
1186 | } | |
1187 | } | |
79f95c82 | 1188 | /* double check we haven't messed things up */ |
123abc88 | 1189 | check_block(root, path, level); |
e20d96d6 | 1190 | if (orig_ptr != |
5f39d397 | 1191 | btrfs_node_blockptr(path->nodes[level], path->slots[level])) |
79f95c82 | 1192 | BUG(); |
54aa1f4d | 1193 | enospc: |
925baedd CM |
1194 | if (right) { |
1195 | btrfs_tree_unlock(right); | |
5f39d397 | 1196 | free_extent_buffer(right); |
925baedd CM |
1197 | } |
1198 | if (left) { | |
1199 | if (path->nodes[level] != left) | |
1200 | btrfs_tree_unlock(left); | |
5f39d397 | 1201 | free_extent_buffer(left); |
925baedd | 1202 | } |
bb803951 CM |
1203 | return ret; |
1204 | } | |
1205 | ||
d352ac68 CM |
1206 | /* Node balancing for insertion. Here we only split or push nodes around |
1207 | * when they are completely full. This is also done top down, so we | |
1208 | * have to be pessimistic. | |
1209 | */ | |
d397712b | 1210 | static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans, |
98ed5174 CM |
1211 | struct btrfs_root *root, |
1212 | struct btrfs_path *path, int level) | |
e66f709b | 1213 | { |
5f39d397 CM |
1214 | struct extent_buffer *right = NULL; |
1215 | struct extent_buffer *mid; | |
1216 | struct extent_buffer *left = NULL; | |
1217 | struct extent_buffer *parent = NULL; | |
e66f709b CM |
1218 | int ret = 0; |
1219 | int wret; | |
1220 | int pslot; | |
1221 | int orig_slot = path->slots[level]; | |
e66f709b CM |
1222 | |
1223 | if (level == 0) | |
1224 | return 1; | |
1225 | ||
5f39d397 | 1226 | mid = path->nodes[level]; |
7bb86316 | 1227 | WARN_ON(btrfs_header_generation(mid) != trans->transid); |
e66f709b CM |
1228 | |
1229 | if (level < BTRFS_MAX_LEVEL - 1) | |
5f39d397 | 1230 | parent = path->nodes[level + 1]; |
e66f709b CM |
1231 | pslot = path->slots[level + 1]; |
1232 | ||
5f39d397 | 1233 | if (!parent) |
e66f709b | 1234 | return 1; |
e66f709b | 1235 | |
5f39d397 | 1236 | left = read_node_slot(root, parent, pslot - 1); |
e66f709b CM |
1237 | |
1238 | /* first, try to make some room in the middle buffer */ | |
5f39d397 | 1239 | if (left) { |
e66f709b | 1240 | u32 left_nr; |
925baedd CM |
1241 | |
1242 | btrfs_tree_lock(left); | |
b4ce94de CM |
1243 | btrfs_set_lock_blocking(left); |
1244 | ||
5f39d397 | 1245 | left_nr = btrfs_header_nritems(left); |
33ade1f8 CM |
1246 | if (left_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) { |
1247 | wret = 1; | |
1248 | } else { | |
5f39d397 | 1249 | ret = btrfs_cow_block(trans, root, left, parent, |
9fa8cfe7 | 1250 | pslot - 1, &left); |
54aa1f4d CM |
1251 | if (ret) |
1252 | wret = 1; | |
1253 | else { | |
54aa1f4d | 1254 | wret = push_node_left(trans, root, |
971a1f66 | 1255 | left, mid, 0); |
54aa1f4d | 1256 | } |
33ade1f8 | 1257 | } |
e66f709b CM |
1258 | if (wret < 0) |
1259 | ret = wret; | |
1260 | if (wret == 0) { | |
5f39d397 | 1261 | struct btrfs_disk_key disk_key; |
e66f709b | 1262 | orig_slot += left_nr; |
5f39d397 CM |
1263 | btrfs_node_key(mid, &disk_key, 0); |
1264 | btrfs_set_node_key(parent, &disk_key, pslot); | |
1265 | btrfs_mark_buffer_dirty(parent); | |
1266 | if (btrfs_header_nritems(left) > orig_slot) { | |
1267 | path->nodes[level] = left; | |
e66f709b CM |
1268 | path->slots[level + 1] -= 1; |
1269 | path->slots[level] = orig_slot; | |
925baedd | 1270 | btrfs_tree_unlock(mid); |
5f39d397 | 1271 | free_extent_buffer(mid); |
e66f709b CM |
1272 | } else { |
1273 | orig_slot -= | |
5f39d397 | 1274 | btrfs_header_nritems(left); |
e66f709b | 1275 | path->slots[level] = orig_slot; |
925baedd | 1276 | btrfs_tree_unlock(left); |
5f39d397 | 1277 | free_extent_buffer(left); |
e66f709b | 1278 | } |
e66f709b CM |
1279 | return 0; |
1280 | } | |
925baedd | 1281 | btrfs_tree_unlock(left); |
5f39d397 | 1282 | free_extent_buffer(left); |
e66f709b | 1283 | } |
925baedd | 1284 | right = read_node_slot(root, parent, pslot + 1); |
e66f709b CM |
1285 | |
1286 | /* | |
1287 | * then try to empty the right most buffer into the middle | |
1288 | */ | |
5f39d397 | 1289 | if (right) { |
33ade1f8 | 1290 | u32 right_nr; |
b4ce94de | 1291 | |
925baedd | 1292 | btrfs_tree_lock(right); |
b4ce94de CM |
1293 | btrfs_set_lock_blocking(right); |
1294 | ||
5f39d397 | 1295 | right_nr = btrfs_header_nritems(right); |
33ade1f8 CM |
1296 | if (right_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) { |
1297 | wret = 1; | |
1298 | } else { | |
5f39d397 CM |
1299 | ret = btrfs_cow_block(trans, root, right, |
1300 | parent, pslot + 1, | |
9fa8cfe7 | 1301 | &right); |
54aa1f4d CM |
1302 | if (ret) |
1303 | wret = 1; | |
1304 | else { | |
54aa1f4d | 1305 | wret = balance_node_right(trans, root, |
5f39d397 | 1306 | right, mid); |
54aa1f4d | 1307 | } |
33ade1f8 | 1308 | } |
e66f709b CM |
1309 | if (wret < 0) |
1310 | ret = wret; | |
1311 | if (wret == 0) { | |
5f39d397 CM |
1312 | struct btrfs_disk_key disk_key; |
1313 | ||
1314 | btrfs_node_key(right, &disk_key, 0); | |
1315 | btrfs_set_node_key(parent, &disk_key, pslot + 1); | |
1316 | btrfs_mark_buffer_dirty(parent); | |
1317 | ||
1318 | if (btrfs_header_nritems(mid) <= orig_slot) { | |
1319 | path->nodes[level] = right; | |
e66f709b CM |
1320 | path->slots[level + 1] += 1; |
1321 | path->slots[level] = orig_slot - | |
5f39d397 | 1322 | btrfs_header_nritems(mid); |
925baedd | 1323 | btrfs_tree_unlock(mid); |
5f39d397 | 1324 | free_extent_buffer(mid); |
e66f709b | 1325 | } else { |
925baedd | 1326 | btrfs_tree_unlock(right); |
5f39d397 | 1327 | free_extent_buffer(right); |
e66f709b | 1328 | } |
e66f709b CM |
1329 | return 0; |
1330 | } | |
925baedd | 1331 | btrfs_tree_unlock(right); |
5f39d397 | 1332 | free_extent_buffer(right); |
e66f709b | 1333 | } |
e66f709b CM |
1334 | return 1; |
1335 | } | |
1336 | ||
3c69faec | 1337 | /* |
d352ac68 CM |
1338 | * readahead one full node of leaves, finding things that are close |
1339 | * to the block in 'slot', and triggering ra on them. | |
3c69faec | 1340 | */ |
c8c42864 CM |
1341 | static void reada_for_search(struct btrfs_root *root, |
1342 | struct btrfs_path *path, | |
1343 | int level, int slot, u64 objectid) | |
3c69faec | 1344 | { |
5f39d397 | 1345 | struct extent_buffer *node; |
01f46658 | 1346 | struct btrfs_disk_key disk_key; |
3c69faec | 1347 | u32 nritems; |
3c69faec | 1348 | u64 search; |
a7175319 | 1349 | u64 target; |
6b80053d | 1350 | u64 nread = 0; |
3c69faec | 1351 | int direction = path->reada; |
5f39d397 | 1352 | struct extent_buffer *eb; |
6b80053d CM |
1353 | u32 nr; |
1354 | u32 blocksize; | |
1355 | u32 nscan = 0; | |
db94535d | 1356 | |
a6b6e75e | 1357 | if (level != 1) |
6702ed49 CM |
1358 | return; |
1359 | ||
1360 | if (!path->nodes[level]) | |
3c69faec CM |
1361 | return; |
1362 | ||
5f39d397 | 1363 | node = path->nodes[level]; |
925baedd | 1364 | |
3c69faec | 1365 | search = btrfs_node_blockptr(node, slot); |
6b80053d CM |
1366 | blocksize = btrfs_level_size(root, level - 1); |
1367 | eb = btrfs_find_tree_block(root, search, blocksize); | |
5f39d397 CM |
1368 | if (eb) { |
1369 | free_extent_buffer(eb); | |
3c69faec CM |
1370 | return; |
1371 | } | |
1372 | ||
a7175319 | 1373 | target = search; |
6b80053d | 1374 | |
5f39d397 | 1375 | nritems = btrfs_header_nritems(node); |
6b80053d | 1376 | nr = slot; |
d397712b | 1377 | while (1) { |
6b80053d CM |
1378 | if (direction < 0) { |
1379 | if (nr == 0) | |
1380 | break; | |
1381 | nr--; | |
1382 | } else if (direction > 0) { | |
1383 | nr++; | |
1384 | if (nr >= nritems) | |
1385 | break; | |
3c69faec | 1386 | } |
01f46658 CM |
1387 | if (path->reada < 0 && objectid) { |
1388 | btrfs_node_key(node, &disk_key, nr); | |
1389 | if (btrfs_disk_key_objectid(&disk_key) != objectid) | |
1390 | break; | |
1391 | } | |
6b80053d | 1392 | search = btrfs_node_blockptr(node, nr); |
a7175319 CM |
1393 | if ((search <= target && target - search <= 65536) || |
1394 | (search > target && search - target <= 65536)) { | |
ca7a79ad CM |
1395 | readahead_tree_block(root, search, blocksize, |
1396 | btrfs_node_ptr_generation(node, nr)); | |
6b80053d CM |
1397 | nread += blocksize; |
1398 | } | |
1399 | nscan++; | |
a7175319 | 1400 | if ((nread > 65536 || nscan > 32)) |
6b80053d | 1401 | break; |
3c69faec CM |
1402 | } |
1403 | } | |
925baedd | 1404 | |
b4ce94de CM |
1405 | /* |
1406 | * returns -EAGAIN if it had to drop the path, or zero if everything was in | |
1407 | * cache | |
1408 | */ | |
1409 | static noinline int reada_for_balance(struct btrfs_root *root, | |
1410 | struct btrfs_path *path, int level) | |
1411 | { | |
1412 | int slot; | |
1413 | int nritems; | |
1414 | struct extent_buffer *parent; | |
1415 | struct extent_buffer *eb; | |
1416 | u64 gen; | |
1417 | u64 block1 = 0; | |
1418 | u64 block2 = 0; | |
1419 | int ret = 0; | |
1420 | int blocksize; | |
1421 | ||
8c594ea8 | 1422 | parent = path->nodes[level + 1]; |
b4ce94de CM |
1423 | if (!parent) |
1424 | return 0; | |
1425 | ||
1426 | nritems = btrfs_header_nritems(parent); | |
8c594ea8 | 1427 | slot = path->slots[level + 1]; |
b4ce94de CM |
1428 | blocksize = btrfs_level_size(root, level); |
1429 | ||
1430 | if (slot > 0) { | |
1431 | block1 = btrfs_node_blockptr(parent, slot - 1); | |
1432 | gen = btrfs_node_ptr_generation(parent, slot - 1); | |
1433 | eb = btrfs_find_tree_block(root, block1, blocksize); | |
1434 | if (eb && btrfs_buffer_uptodate(eb, gen)) | |
1435 | block1 = 0; | |
1436 | free_extent_buffer(eb); | |
1437 | } | |
8c594ea8 | 1438 | if (slot + 1 < nritems) { |
b4ce94de CM |
1439 | block2 = btrfs_node_blockptr(parent, slot + 1); |
1440 | gen = btrfs_node_ptr_generation(parent, slot + 1); | |
1441 | eb = btrfs_find_tree_block(root, block2, blocksize); | |
1442 | if (eb && btrfs_buffer_uptodate(eb, gen)) | |
1443 | block2 = 0; | |
1444 | free_extent_buffer(eb); | |
1445 | } | |
1446 | if (block1 || block2) { | |
1447 | ret = -EAGAIN; | |
8c594ea8 CM |
1448 | |
1449 | /* release the whole path */ | |
b4ce94de | 1450 | btrfs_release_path(root, path); |
8c594ea8 CM |
1451 | |
1452 | /* read the blocks */ | |
b4ce94de CM |
1453 | if (block1) |
1454 | readahead_tree_block(root, block1, blocksize, 0); | |
1455 | if (block2) | |
1456 | readahead_tree_block(root, block2, blocksize, 0); | |
1457 | ||
1458 | if (block1) { | |
1459 | eb = read_tree_block(root, block1, blocksize, 0); | |
1460 | free_extent_buffer(eb); | |
1461 | } | |
8c594ea8 | 1462 | if (block2) { |
b4ce94de CM |
1463 | eb = read_tree_block(root, block2, blocksize, 0); |
1464 | free_extent_buffer(eb); | |
1465 | } | |
1466 | } | |
1467 | return ret; | |
1468 | } | |
1469 | ||
1470 | ||
d352ac68 | 1471 | /* |
d397712b CM |
1472 | * when we walk down the tree, it is usually safe to unlock the higher layers |
1473 | * in the tree. The exceptions are when our path goes through slot 0, because | |
1474 | * operations on the tree might require changing key pointers higher up in the | |
1475 | * tree. | |
d352ac68 | 1476 | * |
d397712b CM |
1477 | * callers might also have set path->keep_locks, which tells this code to keep |
1478 | * the lock if the path points to the last slot in the block. This is part of | |
1479 | * walking through the tree, and selecting the next slot in the higher block. | |
d352ac68 | 1480 | * |
d397712b CM |
1481 | * lowest_unlock sets the lowest level in the tree we're allowed to unlock. so |
1482 | * if lowest_unlock is 1, level 0 won't be unlocked | |
d352ac68 | 1483 | */ |
e02119d5 CM |
1484 | static noinline void unlock_up(struct btrfs_path *path, int level, |
1485 | int lowest_unlock) | |
925baedd CM |
1486 | { |
1487 | int i; | |
1488 | int skip_level = level; | |
051e1b9f | 1489 | int no_skips = 0; |
925baedd CM |
1490 | struct extent_buffer *t; |
1491 | ||
1492 | for (i = level; i < BTRFS_MAX_LEVEL; i++) { | |
1493 | if (!path->nodes[i]) | |
1494 | break; | |
1495 | if (!path->locks[i]) | |
1496 | break; | |
051e1b9f | 1497 | if (!no_skips && path->slots[i] == 0) { |
925baedd CM |
1498 | skip_level = i + 1; |
1499 | continue; | |
1500 | } | |
051e1b9f | 1501 | if (!no_skips && path->keep_locks) { |
925baedd CM |
1502 | u32 nritems; |
1503 | t = path->nodes[i]; | |
1504 | nritems = btrfs_header_nritems(t); | |
051e1b9f | 1505 | if (nritems < 1 || path->slots[i] >= nritems - 1) { |
925baedd CM |
1506 | skip_level = i + 1; |
1507 | continue; | |
1508 | } | |
1509 | } | |
051e1b9f CM |
1510 | if (skip_level < i && i >= lowest_unlock) |
1511 | no_skips = 1; | |
1512 | ||
925baedd CM |
1513 | t = path->nodes[i]; |
1514 | if (i >= lowest_unlock && i > skip_level && path->locks[i]) { | |
1515 | btrfs_tree_unlock(t); | |
1516 | path->locks[i] = 0; | |
1517 | } | |
1518 | } | |
1519 | } | |
1520 | ||
b4ce94de CM |
1521 | /* |
1522 | * This releases any locks held in the path starting at level and | |
1523 | * going all the way up to the root. | |
1524 | * | |
1525 | * btrfs_search_slot will keep the lock held on higher nodes in a few | |
1526 | * corner cases, such as COW of the block at slot zero in the node. This | |
1527 | * ignores those rules, and it should only be called when there are no | |
1528 | * more updates to be done higher up in the tree. | |
1529 | */ | |
1530 | noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level) | |
1531 | { | |
1532 | int i; | |
1533 | ||
5d4f98a2 | 1534 | if (path->keep_locks) |
b4ce94de CM |
1535 | return; |
1536 | ||
1537 | for (i = level; i < BTRFS_MAX_LEVEL; i++) { | |
1538 | if (!path->nodes[i]) | |
12f4dacc | 1539 | continue; |
b4ce94de | 1540 | if (!path->locks[i]) |
12f4dacc | 1541 | continue; |
b4ce94de CM |
1542 | btrfs_tree_unlock(path->nodes[i]); |
1543 | path->locks[i] = 0; | |
1544 | } | |
1545 | } | |
1546 | ||
c8c42864 CM |
1547 | /* |
1548 | * helper function for btrfs_search_slot. The goal is to find a block | |
1549 | * in cache without setting the path to blocking. If we find the block | |
1550 | * we return zero and the path is unchanged. | |
1551 | * | |
1552 | * If we can't find the block, we set the path blocking and do some | |
1553 | * reada. -EAGAIN is returned and the search must be repeated. | |
1554 | */ | |
1555 | static int | |
1556 | read_block_for_search(struct btrfs_trans_handle *trans, | |
1557 | struct btrfs_root *root, struct btrfs_path *p, | |
1558 | struct extent_buffer **eb_ret, int level, int slot, | |
1559 | struct btrfs_key *key) | |
1560 | { | |
1561 | u64 blocknr; | |
1562 | u64 gen; | |
1563 | u32 blocksize; | |
1564 | struct extent_buffer *b = *eb_ret; | |
1565 | struct extent_buffer *tmp; | |
76a05b35 | 1566 | int ret; |
c8c42864 CM |
1567 | |
1568 | blocknr = btrfs_node_blockptr(b, slot); | |
1569 | gen = btrfs_node_ptr_generation(b, slot); | |
1570 | blocksize = btrfs_level_size(root, level - 1); | |
1571 | ||
1572 | tmp = btrfs_find_tree_block(root, blocknr, blocksize); | |
cb44921a CM |
1573 | if (tmp) { |
1574 | if (btrfs_buffer_uptodate(tmp, 0)) { | |
1575 | if (btrfs_buffer_uptodate(tmp, gen)) { | |
1576 | /* | |
1577 | * we found an up to date block without | |
1578 | * sleeping, return | |
1579 | * right away | |
1580 | */ | |
1581 | *eb_ret = tmp; | |
1582 | return 0; | |
1583 | } | |
1584 | /* the pages were up to date, but we failed | |
1585 | * the generation number check. Do a full | |
1586 | * read for the generation number that is correct. | |
1587 | * We must do this without dropping locks so | |
1588 | * we can trust our generation number | |
1589 | */ | |
1590 | free_extent_buffer(tmp); | |
1591 | tmp = read_tree_block(root, blocknr, blocksize, gen); | |
1592 | if (tmp && btrfs_buffer_uptodate(tmp, gen)) { | |
1593 | *eb_ret = tmp; | |
1594 | return 0; | |
1595 | } | |
1596 | free_extent_buffer(tmp); | |
1597 | btrfs_release_path(NULL, p); | |
1598 | return -EIO; | |
1599 | } | |
c8c42864 CM |
1600 | } |
1601 | ||
1602 | /* | |
1603 | * reduce lock contention at high levels | |
1604 | * of the btree by dropping locks before | |
76a05b35 CM |
1605 | * we read. Don't release the lock on the current |
1606 | * level because we need to walk this node to figure | |
1607 | * out which blocks to read. | |
c8c42864 | 1608 | */ |
8c594ea8 CM |
1609 | btrfs_unlock_up_safe(p, level + 1); |
1610 | btrfs_set_path_blocking(p); | |
1611 | ||
cb44921a | 1612 | free_extent_buffer(tmp); |
c8c42864 CM |
1613 | if (p->reada) |
1614 | reada_for_search(root, p, level, slot, key->objectid); | |
1615 | ||
8c594ea8 | 1616 | btrfs_release_path(NULL, p); |
76a05b35 CM |
1617 | |
1618 | ret = -EAGAIN; | |
5bdd3536 | 1619 | tmp = read_tree_block(root, blocknr, blocksize, 0); |
76a05b35 CM |
1620 | if (tmp) { |
1621 | /* | |
1622 | * If the read above didn't mark this buffer up to date, | |
1623 | * it will never end up being up to date. Set ret to EIO now | |
1624 | * and give up so that our caller doesn't loop forever | |
1625 | * on our EAGAINs. | |
1626 | */ | |
1627 | if (!btrfs_buffer_uptodate(tmp, 0)) | |
1628 | ret = -EIO; | |
c8c42864 | 1629 | free_extent_buffer(tmp); |
76a05b35 CM |
1630 | } |
1631 | return ret; | |
c8c42864 CM |
1632 | } |
1633 | ||
1634 | /* | |
1635 | * helper function for btrfs_search_slot. This does all of the checks | |
1636 | * for node-level blocks and does any balancing required based on | |
1637 | * the ins_len. | |
1638 | * | |
1639 | * If no extra work was required, zero is returned. If we had to | |
1640 | * drop the path, -EAGAIN is returned and btrfs_search_slot must | |
1641 | * start over | |
1642 | */ | |
1643 | static int | |
1644 | setup_nodes_for_search(struct btrfs_trans_handle *trans, | |
1645 | struct btrfs_root *root, struct btrfs_path *p, | |
1646 | struct extent_buffer *b, int level, int ins_len) | |
1647 | { | |
1648 | int ret; | |
1649 | if ((p->search_for_split || ins_len > 0) && btrfs_header_nritems(b) >= | |
1650 | BTRFS_NODEPTRS_PER_BLOCK(root) - 3) { | |
1651 | int sret; | |
1652 | ||
1653 | sret = reada_for_balance(root, p, level); | |
1654 | if (sret) | |
1655 | goto again; | |
1656 | ||
1657 | btrfs_set_path_blocking(p); | |
1658 | sret = split_node(trans, root, p, level); | |
1659 | btrfs_clear_path_blocking(p, NULL); | |
1660 | ||
1661 | BUG_ON(sret > 0); | |
1662 | if (sret) { | |
1663 | ret = sret; | |
1664 | goto done; | |
1665 | } | |
1666 | b = p->nodes[level]; | |
1667 | } else if (ins_len < 0 && btrfs_header_nritems(b) < | |
cfbb9308 | 1668 | BTRFS_NODEPTRS_PER_BLOCK(root) / 2) { |
c8c42864 CM |
1669 | int sret; |
1670 | ||
1671 | sret = reada_for_balance(root, p, level); | |
1672 | if (sret) | |
1673 | goto again; | |
1674 | ||
1675 | btrfs_set_path_blocking(p); | |
1676 | sret = balance_level(trans, root, p, level); | |
1677 | btrfs_clear_path_blocking(p, NULL); | |
1678 | ||
1679 | if (sret) { | |
1680 | ret = sret; | |
1681 | goto done; | |
1682 | } | |
1683 | b = p->nodes[level]; | |
1684 | if (!b) { | |
1685 | btrfs_release_path(NULL, p); | |
1686 | goto again; | |
1687 | } | |
1688 | BUG_ON(btrfs_header_nritems(b) == 1); | |
1689 | } | |
1690 | return 0; | |
1691 | ||
1692 | again: | |
1693 | ret = -EAGAIN; | |
1694 | done: | |
1695 | return ret; | |
1696 | } | |
1697 | ||
74123bd7 CM |
1698 | /* |
1699 | * look for key in the tree. path is filled in with nodes along the way | |
1700 | * if key is found, we return zero and you can find the item in the leaf | |
1701 | * level of the path (level 0) | |
1702 | * | |
1703 | * If the key isn't found, the path points to the slot where it should | |
aa5d6bed CM |
1704 | * be inserted, and 1 is returned. If there are other errors during the |
1705 | * search a negative error number is returned. | |
97571fd0 CM |
1706 | * |
1707 | * if ins_len > 0, nodes and leaves will be split as we walk down the | |
1708 | * tree. if ins_len < 0, nodes will be merged as we walk down the tree (if | |
1709 | * possible) | |
74123bd7 | 1710 | */ |
e089f05c CM |
1711 | int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root |
1712 | *root, struct btrfs_key *key, struct btrfs_path *p, int | |
1713 | ins_len, int cow) | |
be0e5c09 | 1714 | { |
5f39d397 | 1715 | struct extent_buffer *b; |
be0e5c09 CM |
1716 | int slot; |
1717 | int ret; | |
33c66f43 | 1718 | int err; |
be0e5c09 | 1719 | int level; |
925baedd | 1720 | int lowest_unlock = 1; |
9f3a7427 CM |
1721 | u8 lowest_level = 0; |
1722 | ||
6702ed49 | 1723 | lowest_level = p->lowest_level; |
323ac95b | 1724 | WARN_ON(lowest_level && ins_len > 0); |
22b0ebda | 1725 | WARN_ON(p->nodes[0] != NULL); |
25179201 | 1726 | |
925baedd CM |
1727 | if (ins_len < 0) |
1728 | lowest_unlock = 2; | |
65b51a00 | 1729 | |
bb803951 | 1730 | again: |
5d4f98a2 YZ |
1731 | if (p->search_commit_root) { |
1732 | b = root->commit_root; | |
1733 | extent_buffer_get(b); | |
1734 | if (!p->skip_locking) | |
1735 | btrfs_tree_lock(b); | |
1736 | } else { | |
1737 | if (p->skip_locking) | |
1738 | b = btrfs_root_node(root); | |
1739 | else | |
1740 | b = btrfs_lock_root_node(root); | |
1741 | } | |
925baedd | 1742 | |
eb60ceac | 1743 | while (b) { |
5f39d397 | 1744 | level = btrfs_header_level(b); |
65b51a00 CM |
1745 | |
1746 | /* | |
1747 | * setup the path here so we can release it under lock | |
1748 | * contention with the cow code | |
1749 | */ | |
1750 | p->nodes[level] = b; | |
1751 | if (!p->skip_locking) | |
1752 | p->locks[level] = 1; | |
1753 | ||
02217ed2 | 1754 | if (cow) { |
c8c42864 CM |
1755 | /* |
1756 | * if we don't really need to cow this block | |
1757 | * then we don't want to set the path blocking, | |
1758 | * so we test it here | |
1759 | */ | |
5d4f98a2 | 1760 | if (!should_cow_block(trans, root, b)) |
65b51a00 | 1761 | goto cow_done; |
5d4f98a2 | 1762 | |
b4ce94de CM |
1763 | btrfs_set_path_blocking(p); |
1764 | ||
33c66f43 YZ |
1765 | err = btrfs_cow_block(trans, root, b, |
1766 | p->nodes[level + 1], | |
1767 | p->slots[level + 1], &b); | |
1768 | if (err) { | |
33c66f43 | 1769 | ret = err; |
65b51a00 | 1770 | goto done; |
54aa1f4d | 1771 | } |
02217ed2 | 1772 | } |
65b51a00 | 1773 | cow_done: |
02217ed2 | 1774 | BUG_ON(!cow && ins_len); |
5f39d397 | 1775 | if (level != btrfs_header_level(b)) |
2c90e5d6 | 1776 | WARN_ON(1); |
5f39d397 | 1777 | level = btrfs_header_level(b); |
65b51a00 | 1778 | |
eb60ceac | 1779 | p->nodes[level] = b; |
5cd57b2c CM |
1780 | if (!p->skip_locking) |
1781 | p->locks[level] = 1; | |
65b51a00 | 1782 | |
4008c04a | 1783 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de CM |
1784 | |
1785 | /* | |
1786 | * we have a lock on b and as long as we aren't changing | |
1787 | * the tree, there is no way to for the items in b to change. | |
1788 | * It is safe to drop the lock on our parent before we | |
1789 | * go through the expensive btree search on b. | |
1790 | * | |
1791 | * If cow is true, then we might be changing slot zero, | |
1792 | * which may require changing the parent. So, we can't | |
1793 | * drop the lock until after we know which slot we're | |
1794 | * operating on. | |
1795 | */ | |
1796 | if (!cow) | |
1797 | btrfs_unlock_up_safe(p, level + 1); | |
1798 | ||
123abc88 | 1799 | ret = check_block(root, p, level); |
65b51a00 CM |
1800 | if (ret) { |
1801 | ret = -1; | |
1802 | goto done; | |
1803 | } | |
925baedd | 1804 | |
5f39d397 | 1805 | ret = bin_search(b, key, level, &slot); |
b4ce94de | 1806 | |
5f39d397 | 1807 | if (level != 0) { |
33c66f43 YZ |
1808 | int dec = 0; |
1809 | if (ret && slot > 0) { | |
1810 | dec = 1; | |
be0e5c09 | 1811 | slot -= 1; |
33c66f43 | 1812 | } |
be0e5c09 | 1813 | p->slots[level] = slot; |
33c66f43 | 1814 | err = setup_nodes_for_search(trans, root, p, b, level, |
c8c42864 | 1815 | ins_len); |
33c66f43 | 1816 | if (err == -EAGAIN) |
c8c42864 | 1817 | goto again; |
33c66f43 YZ |
1818 | if (err) { |
1819 | ret = err; | |
c8c42864 | 1820 | goto done; |
33c66f43 | 1821 | } |
c8c42864 CM |
1822 | b = p->nodes[level]; |
1823 | slot = p->slots[level]; | |
b4ce94de | 1824 | |
f9efa9c7 CM |
1825 | unlock_up(p, level, lowest_unlock); |
1826 | ||
925baedd | 1827 | if (level == lowest_level) { |
33c66f43 YZ |
1828 | if (dec) |
1829 | p->slots[level]++; | |
5b21f2ed | 1830 | goto done; |
925baedd | 1831 | } |
ca7a79ad | 1832 | |
33c66f43 | 1833 | err = read_block_for_search(trans, root, p, |
c8c42864 | 1834 | &b, level, slot, key); |
33c66f43 | 1835 | if (err == -EAGAIN) |
c8c42864 | 1836 | goto again; |
33c66f43 YZ |
1837 | if (err) { |
1838 | ret = err; | |
76a05b35 | 1839 | goto done; |
33c66f43 | 1840 | } |
76a05b35 | 1841 | |
b4ce94de | 1842 | if (!p->skip_locking) { |
4008c04a | 1843 | btrfs_clear_path_blocking(p, NULL); |
33c66f43 | 1844 | err = btrfs_try_spin_lock(b); |
b4ce94de | 1845 | |
33c66f43 | 1846 | if (!err) { |
b4ce94de CM |
1847 | btrfs_set_path_blocking(p); |
1848 | btrfs_tree_lock(b); | |
4008c04a | 1849 | btrfs_clear_path_blocking(p, b); |
b4ce94de CM |
1850 | } |
1851 | } | |
be0e5c09 CM |
1852 | } else { |
1853 | p->slots[level] = slot; | |
87b29b20 YZ |
1854 | if (ins_len > 0 && |
1855 | btrfs_leaf_free_space(root, b) < ins_len) { | |
b4ce94de | 1856 | btrfs_set_path_blocking(p); |
33c66f43 YZ |
1857 | err = split_leaf(trans, root, key, |
1858 | p, ins_len, ret == 0); | |
4008c04a | 1859 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de | 1860 | |
33c66f43 YZ |
1861 | BUG_ON(err > 0); |
1862 | if (err) { | |
1863 | ret = err; | |
65b51a00 CM |
1864 | goto done; |
1865 | } | |
5c680ed6 | 1866 | } |
459931ec CM |
1867 | if (!p->search_for_split) |
1868 | unlock_up(p, level, lowest_unlock); | |
65b51a00 | 1869 | goto done; |
be0e5c09 CM |
1870 | } |
1871 | } | |
65b51a00 CM |
1872 | ret = 1; |
1873 | done: | |
b4ce94de CM |
1874 | /* |
1875 | * we don't really know what they plan on doing with the path | |
1876 | * from here on, so for now just mark it as blocking | |
1877 | */ | |
b9473439 CM |
1878 | if (!p->leave_spinning) |
1879 | btrfs_set_path_blocking(p); | |
76a05b35 CM |
1880 | if (ret < 0) |
1881 | btrfs_release_path(root, p); | |
65b51a00 | 1882 | return ret; |
be0e5c09 CM |
1883 | } |
1884 | ||
74123bd7 CM |
1885 | /* |
1886 | * adjust the pointers going up the tree, starting at level | |
1887 | * making sure the right key of each node is points to 'key'. | |
1888 | * This is used after shifting pointers to the left, so it stops | |
1889 | * fixing up pointers when a given leaf/node is not in slot 0 of the | |
1890 | * higher levels | |
aa5d6bed CM |
1891 | * |
1892 | * If this fails to write a tree block, it returns -1, but continues | |
1893 | * fixing up the blocks in ram so the tree is consistent. | |
74123bd7 | 1894 | */ |
5f39d397 CM |
1895 | static int fixup_low_keys(struct btrfs_trans_handle *trans, |
1896 | struct btrfs_root *root, struct btrfs_path *path, | |
1897 | struct btrfs_disk_key *key, int level) | |
be0e5c09 CM |
1898 | { |
1899 | int i; | |
aa5d6bed | 1900 | int ret = 0; |
5f39d397 CM |
1901 | struct extent_buffer *t; |
1902 | ||
234b63a0 | 1903 | for (i = level; i < BTRFS_MAX_LEVEL; i++) { |
be0e5c09 | 1904 | int tslot = path->slots[i]; |
eb60ceac | 1905 | if (!path->nodes[i]) |
be0e5c09 | 1906 | break; |
5f39d397 CM |
1907 | t = path->nodes[i]; |
1908 | btrfs_set_node_key(t, key, tslot); | |
d6025579 | 1909 | btrfs_mark_buffer_dirty(path->nodes[i]); |
be0e5c09 CM |
1910 | if (tslot != 0) |
1911 | break; | |
1912 | } | |
aa5d6bed | 1913 | return ret; |
be0e5c09 CM |
1914 | } |
1915 | ||
31840ae1 ZY |
1916 | /* |
1917 | * update item key. | |
1918 | * | |
1919 | * This function isn't completely safe. It's the caller's responsibility | |
1920 | * that the new key won't break the order | |
1921 | */ | |
1922 | int btrfs_set_item_key_safe(struct btrfs_trans_handle *trans, | |
1923 | struct btrfs_root *root, struct btrfs_path *path, | |
1924 | struct btrfs_key *new_key) | |
1925 | { | |
1926 | struct btrfs_disk_key disk_key; | |
1927 | struct extent_buffer *eb; | |
1928 | int slot; | |
1929 | ||
1930 | eb = path->nodes[0]; | |
1931 | slot = path->slots[0]; | |
1932 | if (slot > 0) { | |
1933 | btrfs_item_key(eb, &disk_key, slot - 1); | |
1934 | if (comp_keys(&disk_key, new_key) >= 0) | |
1935 | return -1; | |
1936 | } | |
1937 | if (slot < btrfs_header_nritems(eb) - 1) { | |
1938 | btrfs_item_key(eb, &disk_key, slot + 1); | |
1939 | if (comp_keys(&disk_key, new_key) <= 0) | |
1940 | return -1; | |
1941 | } | |
1942 | ||
1943 | btrfs_cpu_key_to_disk(&disk_key, new_key); | |
1944 | btrfs_set_item_key(eb, &disk_key, slot); | |
1945 | btrfs_mark_buffer_dirty(eb); | |
1946 | if (slot == 0) | |
1947 | fixup_low_keys(trans, root, path, &disk_key, 1); | |
1948 | return 0; | |
1949 | } | |
1950 | ||
74123bd7 CM |
1951 | /* |
1952 | * try to push data from one node into the next node left in the | |
79f95c82 | 1953 | * tree. |
aa5d6bed CM |
1954 | * |
1955 | * returns 0 if some ptrs were pushed left, < 0 if there was some horrible | |
1956 | * error, and > 0 if there was no room in the left hand block. | |
74123bd7 | 1957 | */ |
98ed5174 CM |
1958 | static int push_node_left(struct btrfs_trans_handle *trans, |
1959 | struct btrfs_root *root, struct extent_buffer *dst, | |
971a1f66 | 1960 | struct extent_buffer *src, int empty) |
be0e5c09 | 1961 | { |
be0e5c09 | 1962 | int push_items = 0; |
bb803951 CM |
1963 | int src_nritems; |
1964 | int dst_nritems; | |
aa5d6bed | 1965 | int ret = 0; |
be0e5c09 | 1966 | |
5f39d397 CM |
1967 | src_nritems = btrfs_header_nritems(src); |
1968 | dst_nritems = btrfs_header_nritems(dst); | |
123abc88 | 1969 | push_items = BTRFS_NODEPTRS_PER_BLOCK(root) - dst_nritems; |
7bb86316 CM |
1970 | WARN_ON(btrfs_header_generation(src) != trans->transid); |
1971 | WARN_ON(btrfs_header_generation(dst) != trans->transid); | |
54aa1f4d | 1972 | |
bce4eae9 | 1973 | if (!empty && src_nritems <= 8) |
971a1f66 CM |
1974 | return 1; |
1975 | ||
d397712b | 1976 | if (push_items <= 0) |
be0e5c09 CM |
1977 | return 1; |
1978 | ||
bce4eae9 | 1979 | if (empty) { |
971a1f66 | 1980 | push_items = min(src_nritems, push_items); |
bce4eae9 CM |
1981 | if (push_items < src_nritems) { |
1982 | /* leave at least 8 pointers in the node if | |
1983 | * we aren't going to empty it | |
1984 | */ | |
1985 | if (src_nritems - push_items < 8) { | |
1986 | if (push_items <= 8) | |
1987 | return 1; | |
1988 | push_items -= 8; | |
1989 | } | |
1990 | } | |
1991 | } else | |
1992 | push_items = min(src_nritems - 8, push_items); | |
79f95c82 | 1993 | |
5f39d397 CM |
1994 | copy_extent_buffer(dst, src, |
1995 | btrfs_node_key_ptr_offset(dst_nritems), | |
1996 | btrfs_node_key_ptr_offset(0), | |
d397712b | 1997 | push_items * sizeof(struct btrfs_key_ptr)); |
5f39d397 | 1998 | |
bb803951 | 1999 | if (push_items < src_nritems) { |
5f39d397 CM |
2000 | memmove_extent_buffer(src, btrfs_node_key_ptr_offset(0), |
2001 | btrfs_node_key_ptr_offset(push_items), | |
2002 | (src_nritems - push_items) * | |
2003 | sizeof(struct btrfs_key_ptr)); | |
2004 | } | |
2005 | btrfs_set_header_nritems(src, src_nritems - push_items); | |
2006 | btrfs_set_header_nritems(dst, dst_nritems + push_items); | |
2007 | btrfs_mark_buffer_dirty(src); | |
2008 | btrfs_mark_buffer_dirty(dst); | |
31840ae1 | 2009 | |
79f95c82 CM |
2010 | return ret; |
2011 | } | |
2012 | ||
2013 | /* | |
2014 | * try to push data from one node into the next node right in the | |
2015 | * tree. | |
2016 | * | |
2017 | * returns 0 if some ptrs were pushed, < 0 if there was some horrible | |
2018 | * error, and > 0 if there was no room in the right hand block. | |
2019 | * | |
2020 | * this will only push up to 1/2 the contents of the left node over | |
2021 | */ | |
5f39d397 CM |
2022 | static int balance_node_right(struct btrfs_trans_handle *trans, |
2023 | struct btrfs_root *root, | |
2024 | struct extent_buffer *dst, | |
2025 | struct extent_buffer *src) | |
79f95c82 | 2026 | { |
79f95c82 CM |
2027 | int push_items = 0; |
2028 | int max_push; | |
2029 | int src_nritems; | |
2030 | int dst_nritems; | |
2031 | int ret = 0; | |
79f95c82 | 2032 | |
7bb86316 CM |
2033 | WARN_ON(btrfs_header_generation(src) != trans->transid); |
2034 | WARN_ON(btrfs_header_generation(dst) != trans->transid); | |
2035 | ||
5f39d397 CM |
2036 | src_nritems = btrfs_header_nritems(src); |
2037 | dst_nritems = btrfs_header_nritems(dst); | |
123abc88 | 2038 | push_items = BTRFS_NODEPTRS_PER_BLOCK(root) - dst_nritems; |
d397712b | 2039 | if (push_items <= 0) |
79f95c82 | 2040 | return 1; |
bce4eae9 | 2041 | |
d397712b | 2042 | if (src_nritems < 4) |
bce4eae9 | 2043 | return 1; |
79f95c82 CM |
2044 | |
2045 | max_push = src_nritems / 2 + 1; | |
2046 | /* don't try to empty the node */ | |
d397712b | 2047 | if (max_push >= src_nritems) |
79f95c82 | 2048 | return 1; |
252c38f0 | 2049 | |
79f95c82 CM |
2050 | if (max_push < push_items) |
2051 | push_items = max_push; | |
2052 | ||
5f39d397 CM |
2053 | memmove_extent_buffer(dst, btrfs_node_key_ptr_offset(push_items), |
2054 | btrfs_node_key_ptr_offset(0), | |
2055 | (dst_nritems) * | |
2056 | sizeof(struct btrfs_key_ptr)); | |
d6025579 | 2057 | |
5f39d397 CM |
2058 | copy_extent_buffer(dst, src, |
2059 | btrfs_node_key_ptr_offset(0), | |
2060 | btrfs_node_key_ptr_offset(src_nritems - push_items), | |
d397712b | 2061 | push_items * sizeof(struct btrfs_key_ptr)); |
79f95c82 | 2062 | |
5f39d397 CM |
2063 | btrfs_set_header_nritems(src, src_nritems - push_items); |
2064 | btrfs_set_header_nritems(dst, dst_nritems + push_items); | |
79f95c82 | 2065 | |
5f39d397 CM |
2066 | btrfs_mark_buffer_dirty(src); |
2067 | btrfs_mark_buffer_dirty(dst); | |
31840ae1 | 2068 | |
aa5d6bed | 2069 | return ret; |
be0e5c09 CM |
2070 | } |
2071 | ||
97571fd0 CM |
2072 | /* |
2073 | * helper function to insert a new root level in the tree. | |
2074 | * A new node is allocated, and a single item is inserted to | |
2075 | * point to the existing root | |
aa5d6bed CM |
2076 | * |
2077 | * returns zero on success or < 0 on failure. | |
97571fd0 | 2078 | */ |
d397712b | 2079 | static noinline int insert_new_root(struct btrfs_trans_handle *trans, |
5f39d397 CM |
2080 | struct btrfs_root *root, |
2081 | struct btrfs_path *path, int level) | |
5c680ed6 | 2082 | { |
7bb86316 | 2083 | u64 lower_gen; |
5f39d397 CM |
2084 | struct extent_buffer *lower; |
2085 | struct extent_buffer *c; | |
925baedd | 2086 | struct extent_buffer *old; |
5f39d397 | 2087 | struct btrfs_disk_key lower_key; |
5c680ed6 CM |
2088 | |
2089 | BUG_ON(path->nodes[level]); | |
2090 | BUG_ON(path->nodes[level-1] != root->node); | |
2091 | ||
7bb86316 CM |
2092 | lower = path->nodes[level-1]; |
2093 | if (level == 1) | |
2094 | btrfs_item_key(lower, &lower_key, 0); | |
2095 | else | |
2096 | btrfs_node_key(lower, &lower_key, 0); | |
2097 | ||
31840ae1 | 2098 | c = btrfs_alloc_free_block(trans, root, root->nodesize, 0, |
5d4f98a2 | 2099 | root->root_key.objectid, &lower_key, |
ad3d81ba | 2100 | level, root->node->start, 0); |
5f39d397 CM |
2101 | if (IS_ERR(c)) |
2102 | return PTR_ERR(c); | |
925baedd | 2103 | |
f0486c68 YZ |
2104 | root_add_used(root, root->nodesize); |
2105 | ||
5d4f98a2 | 2106 | memset_extent_buffer(c, 0, 0, sizeof(struct btrfs_header)); |
5f39d397 CM |
2107 | btrfs_set_header_nritems(c, 1); |
2108 | btrfs_set_header_level(c, level); | |
db94535d | 2109 | btrfs_set_header_bytenr(c, c->start); |
5f39d397 | 2110 | btrfs_set_header_generation(c, trans->transid); |
5d4f98a2 | 2111 | btrfs_set_header_backref_rev(c, BTRFS_MIXED_BACKREF_REV); |
5f39d397 | 2112 | btrfs_set_header_owner(c, root->root_key.objectid); |
5f39d397 CM |
2113 | |
2114 | write_extent_buffer(c, root->fs_info->fsid, | |
2115 | (unsigned long)btrfs_header_fsid(c), | |
2116 | BTRFS_FSID_SIZE); | |
e17cade2 CM |
2117 | |
2118 | write_extent_buffer(c, root->fs_info->chunk_tree_uuid, | |
2119 | (unsigned long)btrfs_header_chunk_tree_uuid(c), | |
2120 | BTRFS_UUID_SIZE); | |
2121 | ||
5f39d397 | 2122 | btrfs_set_node_key(c, &lower_key, 0); |
db94535d | 2123 | btrfs_set_node_blockptr(c, 0, lower->start); |
7bb86316 | 2124 | lower_gen = btrfs_header_generation(lower); |
31840ae1 | 2125 | WARN_ON(lower_gen != trans->transid); |
7bb86316 CM |
2126 | |
2127 | btrfs_set_node_ptr_generation(c, 0, lower_gen); | |
d5719762 | 2128 | |
5f39d397 | 2129 | btrfs_mark_buffer_dirty(c); |
d5719762 | 2130 | |
925baedd CM |
2131 | spin_lock(&root->node_lock); |
2132 | old = root->node; | |
5f39d397 | 2133 | root->node = c; |
925baedd CM |
2134 | spin_unlock(&root->node_lock); |
2135 | ||
2136 | /* the super has an extra ref to root->node */ | |
2137 | free_extent_buffer(old); | |
2138 | ||
0b86a832 | 2139 | add_root_to_dirty_list(root); |
5f39d397 CM |
2140 | extent_buffer_get(c); |
2141 | path->nodes[level] = c; | |
925baedd | 2142 | path->locks[level] = 1; |
5c680ed6 CM |
2143 | path->slots[level] = 0; |
2144 | return 0; | |
2145 | } | |
2146 | ||
74123bd7 CM |
2147 | /* |
2148 | * worker function to insert a single pointer in a node. | |
2149 | * the node should have enough room for the pointer already | |
97571fd0 | 2150 | * |
74123bd7 CM |
2151 | * slot and level indicate where you want the key to go, and |
2152 | * blocknr is the block the key points to. | |
aa5d6bed CM |
2153 | * |
2154 | * returns zero on success and < 0 on any error | |
74123bd7 | 2155 | */ |
e089f05c CM |
2156 | static int insert_ptr(struct btrfs_trans_handle *trans, struct btrfs_root |
2157 | *root, struct btrfs_path *path, struct btrfs_disk_key | |
db94535d | 2158 | *key, u64 bytenr, int slot, int level) |
74123bd7 | 2159 | { |
5f39d397 | 2160 | struct extent_buffer *lower; |
74123bd7 | 2161 | int nritems; |
5c680ed6 CM |
2162 | |
2163 | BUG_ON(!path->nodes[level]); | |
f0486c68 | 2164 | btrfs_assert_tree_locked(path->nodes[level]); |
5f39d397 CM |
2165 | lower = path->nodes[level]; |
2166 | nritems = btrfs_header_nritems(lower); | |
c293498b | 2167 | BUG_ON(slot > nritems); |
123abc88 | 2168 | if (nritems == BTRFS_NODEPTRS_PER_BLOCK(root)) |
74123bd7 CM |
2169 | BUG(); |
2170 | if (slot != nritems) { | |
5f39d397 CM |
2171 | memmove_extent_buffer(lower, |
2172 | btrfs_node_key_ptr_offset(slot + 1), | |
2173 | btrfs_node_key_ptr_offset(slot), | |
d6025579 | 2174 | (nritems - slot) * sizeof(struct btrfs_key_ptr)); |
74123bd7 | 2175 | } |
5f39d397 | 2176 | btrfs_set_node_key(lower, key, slot); |
db94535d | 2177 | btrfs_set_node_blockptr(lower, slot, bytenr); |
74493f7a CM |
2178 | WARN_ON(trans->transid == 0); |
2179 | btrfs_set_node_ptr_generation(lower, slot, trans->transid); | |
5f39d397 CM |
2180 | btrfs_set_header_nritems(lower, nritems + 1); |
2181 | btrfs_mark_buffer_dirty(lower); | |
74123bd7 CM |
2182 | return 0; |
2183 | } | |
2184 | ||
97571fd0 CM |
2185 | /* |
2186 | * split the node at the specified level in path in two. | |
2187 | * The path is corrected to point to the appropriate node after the split | |
2188 | * | |
2189 | * Before splitting this tries to make some room in the node by pushing | |
2190 | * left and right, if either one works, it returns right away. | |
aa5d6bed CM |
2191 | * |
2192 | * returns 0 on success and < 0 on failure | |
97571fd0 | 2193 | */ |
e02119d5 CM |
2194 | static noinline int split_node(struct btrfs_trans_handle *trans, |
2195 | struct btrfs_root *root, | |
2196 | struct btrfs_path *path, int level) | |
be0e5c09 | 2197 | { |
5f39d397 CM |
2198 | struct extent_buffer *c; |
2199 | struct extent_buffer *split; | |
2200 | struct btrfs_disk_key disk_key; | |
be0e5c09 | 2201 | int mid; |
5c680ed6 | 2202 | int ret; |
aa5d6bed | 2203 | int wret; |
7518a238 | 2204 | u32 c_nritems; |
eb60ceac | 2205 | |
5f39d397 | 2206 | c = path->nodes[level]; |
7bb86316 | 2207 | WARN_ON(btrfs_header_generation(c) != trans->transid); |
5f39d397 | 2208 | if (c == root->node) { |
5c680ed6 | 2209 | /* trying to split the root, lets make a new one */ |
e089f05c | 2210 | ret = insert_new_root(trans, root, path, level + 1); |
5c680ed6 CM |
2211 | if (ret) |
2212 | return ret; | |
b3612421 | 2213 | } else { |
e66f709b | 2214 | ret = push_nodes_for_insert(trans, root, path, level); |
5f39d397 CM |
2215 | c = path->nodes[level]; |
2216 | if (!ret && btrfs_header_nritems(c) < | |
c448acf0 | 2217 | BTRFS_NODEPTRS_PER_BLOCK(root) - 3) |
e66f709b | 2218 | return 0; |
54aa1f4d CM |
2219 | if (ret < 0) |
2220 | return ret; | |
be0e5c09 | 2221 | } |
e66f709b | 2222 | |
5f39d397 | 2223 | c_nritems = btrfs_header_nritems(c); |
5d4f98a2 YZ |
2224 | mid = (c_nritems + 1) / 2; |
2225 | btrfs_node_key(c, &disk_key, mid); | |
7bb86316 | 2226 | |
5d4f98a2 | 2227 | split = btrfs_alloc_free_block(trans, root, root->nodesize, 0, |
31840ae1 | 2228 | root->root_key.objectid, |
5d4f98a2 | 2229 | &disk_key, level, c->start, 0); |
5f39d397 CM |
2230 | if (IS_ERR(split)) |
2231 | return PTR_ERR(split); | |
2232 | ||
f0486c68 YZ |
2233 | root_add_used(root, root->nodesize); |
2234 | ||
5d4f98a2 | 2235 | memset_extent_buffer(split, 0, 0, sizeof(struct btrfs_header)); |
5f39d397 | 2236 | btrfs_set_header_level(split, btrfs_header_level(c)); |
db94535d | 2237 | btrfs_set_header_bytenr(split, split->start); |
5f39d397 | 2238 | btrfs_set_header_generation(split, trans->transid); |
5d4f98a2 | 2239 | btrfs_set_header_backref_rev(split, BTRFS_MIXED_BACKREF_REV); |
5f39d397 CM |
2240 | btrfs_set_header_owner(split, root->root_key.objectid); |
2241 | write_extent_buffer(split, root->fs_info->fsid, | |
2242 | (unsigned long)btrfs_header_fsid(split), | |
2243 | BTRFS_FSID_SIZE); | |
e17cade2 CM |
2244 | write_extent_buffer(split, root->fs_info->chunk_tree_uuid, |
2245 | (unsigned long)btrfs_header_chunk_tree_uuid(split), | |
2246 | BTRFS_UUID_SIZE); | |
54aa1f4d | 2247 | |
5f39d397 CM |
2248 | |
2249 | copy_extent_buffer(split, c, | |
2250 | btrfs_node_key_ptr_offset(0), | |
2251 | btrfs_node_key_ptr_offset(mid), | |
2252 | (c_nritems - mid) * sizeof(struct btrfs_key_ptr)); | |
2253 | btrfs_set_header_nritems(split, c_nritems - mid); | |
2254 | btrfs_set_header_nritems(c, mid); | |
aa5d6bed CM |
2255 | ret = 0; |
2256 | ||
5f39d397 CM |
2257 | btrfs_mark_buffer_dirty(c); |
2258 | btrfs_mark_buffer_dirty(split); | |
2259 | ||
db94535d | 2260 | wret = insert_ptr(trans, root, path, &disk_key, split->start, |
5f39d397 | 2261 | path->slots[level + 1] + 1, |
123abc88 | 2262 | level + 1); |
aa5d6bed CM |
2263 | if (wret) |
2264 | ret = wret; | |
2265 | ||
5de08d7d | 2266 | if (path->slots[level] >= mid) { |
5c680ed6 | 2267 | path->slots[level] -= mid; |
925baedd | 2268 | btrfs_tree_unlock(c); |
5f39d397 CM |
2269 | free_extent_buffer(c); |
2270 | path->nodes[level] = split; | |
5c680ed6 CM |
2271 | path->slots[level + 1] += 1; |
2272 | } else { | |
925baedd | 2273 | btrfs_tree_unlock(split); |
5f39d397 | 2274 | free_extent_buffer(split); |
be0e5c09 | 2275 | } |
aa5d6bed | 2276 | return ret; |
be0e5c09 CM |
2277 | } |
2278 | ||
74123bd7 CM |
2279 | /* |
2280 | * how many bytes are required to store the items in a leaf. start | |
2281 | * and nr indicate which items in the leaf to check. This totals up the | |
2282 | * space used both by the item structs and the item data | |
2283 | */ | |
5f39d397 | 2284 | static int leaf_space_used(struct extent_buffer *l, int start, int nr) |
be0e5c09 CM |
2285 | { |
2286 | int data_len; | |
5f39d397 | 2287 | int nritems = btrfs_header_nritems(l); |
d4dbff95 | 2288 | int end = min(nritems, start + nr) - 1; |
be0e5c09 CM |
2289 | |
2290 | if (!nr) | |
2291 | return 0; | |
5f39d397 CM |
2292 | data_len = btrfs_item_end_nr(l, start); |
2293 | data_len = data_len - btrfs_item_offset_nr(l, end); | |
0783fcfc | 2294 | data_len += sizeof(struct btrfs_item) * nr; |
d4dbff95 | 2295 | WARN_ON(data_len < 0); |
be0e5c09 CM |
2296 | return data_len; |
2297 | } | |
2298 | ||
d4dbff95 CM |
2299 | /* |
2300 | * The space between the end of the leaf items and | |
2301 | * the start of the leaf data. IOW, how much room | |
2302 | * the leaf has left for both items and data | |
2303 | */ | |
d397712b | 2304 | noinline int btrfs_leaf_free_space(struct btrfs_root *root, |
e02119d5 | 2305 | struct extent_buffer *leaf) |
d4dbff95 | 2306 | { |
5f39d397 CM |
2307 | int nritems = btrfs_header_nritems(leaf); |
2308 | int ret; | |
2309 | ret = BTRFS_LEAF_DATA_SIZE(root) - leaf_space_used(leaf, 0, nritems); | |
2310 | if (ret < 0) { | |
d397712b CM |
2311 | printk(KERN_CRIT "leaf free space ret %d, leaf data size %lu, " |
2312 | "used %d nritems %d\n", | |
ae2f5411 | 2313 | ret, (unsigned long) BTRFS_LEAF_DATA_SIZE(root), |
5f39d397 CM |
2314 | leaf_space_used(leaf, 0, nritems), nritems); |
2315 | } | |
2316 | return ret; | |
d4dbff95 CM |
2317 | } |
2318 | ||
99d8f83c CM |
2319 | /* |
2320 | * min slot controls the lowest index we're willing to push to the | |
2321 | * right. We'll push up to and including min_slot, but no lower | |
2322 | */ | |
44871b1b CM |
2323 | static noinline int __push_leaf_right(struct btrfs_trans_handle *trans, |
2324 | struct btrfs_root *root, | |
2325 | struct btrfs_path *path, | |
2326 | int data_size, int empty, | |
2327 | struct extent_buffer *right, | |
99d8f83c CM |
2328 | int free_space, u32 left_nritems, |
2329 | u32 min_slot) | |
00ec4c51 | 2330 | { |
5f39d397 | 2331 | struct extent_buffer *left = path->nodes[0]; |
44871b1b | 2332 | struct extent_buffer *upper = path->nodes[1]; |
5f39d397 | 2333 | struct btrfs_disk_key disk_key; |
00ec4c51 | 2334 | int slot; |
34a38218 | 2335 | u32 i; |
00ec4c51 CM |
2336 | int push_space = 0; |
2337 | int push_items = 0; | |
0783fcfc | 2338 | struct btrfs_item *item; |
34a38218 | 2339 | u32 nr; |
7518a238 | 2340 | u32 right_nritems; |
5f39d397 | 2341 | u32 data_end; |
db94535d | 2342 | u32 this_item_size; |
00ec4c51 | 2343 | |
34a38218 CM |
2344 | if (empty) |
2345 | nr = 0; | |
2346 | else | |
99d8f83c | 2347 | nr = max_t(u32, 1, min_slot); |
34a38218 | 2348 | |
31840ae1 | 2349 | if (path->slots[0] >= left_nritems) |
87b29b20 | 2350 | push_space += data_size; |
31840ae1 | 2351 | |
44871b1b | 2352 | slot = path->slots[1]; |
34a38218 CM |
2353 | i = left_nritems - 1; |
2354 | while (i >= nr) { | |
5f39d397 | 2355 | item = btrfs_item_nr(left, i); |
db94535d | 2356 | |
31840ae1 ZY |
2357 | if (!empty && push_items > 0) { |
2358 | if (path->slots[0] > i) | |
2359 | break; | |
2360 | if (path->slots[0] == i) { | |
2361 | int space = btrfs_leaf_free_space(root, left); | |
2362 | if (space + push_space * 2 > free_space) | |
2363 | break; | |
2364 | } | |
2365 | } | |
2366 | ||
00ec4c51 | 2367 | if (path->slots[0] == i) |
87b29b20 | 2368 | push_space += data_size; |
db94535d CM |
2369 | |
2370 | if (!left->map_token) { | |
2371 | map_extent_buffer(left, (unsigned long)item, | |
2372 | sizeof(struct btrfs_item), | |
2373 | &left->map_token, &left->kaddr, | |
2374 | &left->map_start, &left->map_len, | |
2375 | KM_USER1); | |
2376 | } | |
2377 | ||
2378 | this_item_size = btrfs_item_size(left, item); | |
2379 | if (this_item_size + sizeof(*item) + push_space > free_space) | |
00ec4c51 | 2380 | break; |
31840ae1 | 2381 | |
00ec4c51 | 2382 | push_items++; |
db94535d | 2383 | push_space += this_item_size + sizeof(*item); |
34a38218 CM |
2384 | if (i == 0) |
2385 | break; | |
2386 | i--; | |
db94535d CM |
2387 | } |
2388 | if (left->map_token) { | |
2389 | unmap_extent_buffer(left, left->map_token, KM_USER1); | |
2390 | left->map_token = NULL; | |
00ec4c51 | 2391 | } |
5f39d397 | 2392 | |
925baedd CM |
2393 | if (push_items == 0) |
2394 | goto out_unlock; | |
5f39d397 | 2395 | |
34a38218 | 2396 | if (!empty && push_items == left_nritems) |
a429e513 | 2397 | WARN_ON(1); |
5f39d397 | 2398 | |
00ec4c51 | 2399 | /* push left to right */ |
5f39d397 | 2400 | right_nritems = btrfs_header_nritems(right); |
34a38218 | 2401 | |
5f39d397 | 2402 | push_space = btrfs_item_end_nr(left, left_nritems - push_items); |
123abc88 | 2403 | push_space -= leaf_data_end(root, left); |
5f39d397 | 2404 | |
00ec4c51 | 2405 | /* make room in the right data area */ |
5f39d397 CM |
2406 | data_end = leaf_data_end(root, right); |
2407 | memmove_extent_buffer(right, | |
2408 | btrfs_leaf_data(right) + data_end - push_space, | |
2409 | btrfs_leaf_data(right) + data_end, | |
2410 | BTRFS_LEAF_DATA_SIZE(root) - data_end); | |
2411 | ||
00ec4c51 | 2412 | /* copy from the left data area */ |
5f39d397 | 2413 | copy_extent_buffer(right, left, btrfs_leaf_data(right) + |
d6025579 CM |
2414 | BTRFS_LEAF_DATA_SIZE(root) - push_space, |
2415 | btrfs_leaf_data(left) + leaf_data_end(root, left), | |
2416 | push_space); | |
5f39d397 CM |
2417 | |
2418 | memmove_extent_buffer(right, btrfs_item_nr_offset(push_items), | |
2419 | btrfs_item_nr_offset(0), | |
2420 | right_nritems * sizeof(struct btrfs_item)); | |
2421 | ||
00ec4c51 | 2422 | /* copy the items from left to right */ |
5f39d397 CM |
2423 | copy_extent_buffer(right, left, btrfs_item_nr_offset(0), |
2424 | btrfs_item_nr_offset(left_nritems - push_items), | |
2425 | push_items * sizeof(struct btrfs_item)); | |
00ec4c51 CM |
2426 | |
2427 | /* update the item pointers */ | |
7518a238 | 2428 | right_nritems += push_items; |
5f39d397 | 2429 | btrfs_set_header_nritems(right, right_nritems); |
123abc88 | 2430 | push_space = BTRFS_LEAF_DATA_SIZE(root); |
7518a238 | 2431 | for (i = 0; i < right_nritems; i++) { |
5f39d397 | 2432 | item = btrfs_item_nr(right, i); |
db94535d CM |
2433 | if (!right->map_token) { |
2434 | map_extent_buffer(right, (unsigned long)item, | |
2435 | sizeof(struct btrfs_item), | |
2436 | &right->map_token, &right->kaddr, | |
2437 | &right->map_start, &right->map_len, | |
2438 | KM_USER1); | |
2439 | } | |
2440 | push_space -= btrfs_item_size(right, item); | |
2441 | btrfs_set_item_offset(right, item, push_space); | |
2442 | } | |
2443 | ||
2444 | if (right->map_token) { | |
2445 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2446 | right->map_token = NULL; | |
00ec4c51 | 2447 | } |
7518a238 | 2448 | left_nritems -= push_items; |
5f39d397 | 2449 | btrfs_set_header_nritems(left, left_nritems); |
00ec4c51 | 2450 | |
34a38218 CM |
2451 | if (left_nritems) |
2452 | btrfs_mark_buffer_dirty(left); | |
f0486c68 YZ |
2453 | else |
2454 | clean_tree_block(trans, root, left); | |
2455 | ||
5f39d397 | 2456 | btrfs_mark_buffer_dirty(right); |
a429e513 | 2457 | |
5f39d397 CM |
2458 | btrfs_item_key(right, &disk_key, 0); |
2459 | btrfs_set_node_key(upper, &disk_key, slot + 1); | |
d6025579 | 2460 | btrfs_mark_buffer_dirty(upper); |
02217ed2 | 2461 | |
00ec4c51 | 2462 | /* then fixup the leaf pointer in the path */ |
7518a238 CM |
2463 | if (path->slots[0] >= left_nritems) { |
2464 | path->slots[0] -= left_nritems; | |
925baedd CM |
2465 | if (btrfs_header_nritems(path->nodes[0]) == 0) |
2466 | clean_tree_block(trans, root, path->nodes[0]); | |
2467 | btrfs_tree_unlock(path->nodes[0]); | |
5f39d397 CM |
2468 | free_extent_buffer(path->nodes[0]); |
2469 | path->nodes[0] = right; | |
00ec4c51 CM |
2470 | path->slots[1] += 1; |
2471 | } else { | |
925baedd | 2472 | btrfs_tree_unlock(right); |
5f39d397 | 2473 | free_extent_buffer(right); |
00ec4c51 CM |
2474 | } |
2475 | return 0; | |
925baedd CM |
2476 | |
2477 | out_unlock: | |
2478 | btrfs_tree_unlock(right); | |
2479 | free_extent_buffer(right); | |
2480 | return 1; | |
00ec4c51 | 2481 | } |
925baedd | 2482 | |
44871b1b CM |
2483 | /* |
2484 | * push some data in the path leaf to the right, trying to free up at | |
2485 | * least data_size bytes. returns zero if the push worked, nonzero otherwise | |
2486 | * | |
2487 | * returns 1 if the push failed because the other node didn't have enough | |
2488 | * room, 0 if everything worked out and < 0 if there were major errors. | |
99d8f83c CM |
2489 | * |
2490 | * this will push starting from min_slot to the end of the leaf. It won't | |
2491 | * push any slot lower than min_slot | |
44871b1b CM |
2492 | */ |
2493 | static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root | |
99d8f83c CM |
2494 | *root, struct btrfs_path *path, |
2495 | int min_data_size, int data_size, | |
2496 | int empty, u32 min_slot) | |
44871b1b CM |
2497 | { |
2498 | struct extent_buffer *left = path->nodes[0]; | |
2499 | struct extent_buffer *right; | |
2500 | struct extent_buffer *upper; | |
2501 | int slot; | |
2502 | int free_space; | |
2503 | u32 left_nritems; | |
2504 | int ret; | |
2505 | ||
2506 | if (!path->nodes[1]) | |
2507 | return 1; | |
2508 | ||
2509 | slot = path->slots[1]; | |
2510 | upper = path->nodes[1]; | |
2511 | if (slot >= btrfs_header_nritems(upper) - 1) | |
2512 | return 1; | |
2513 | ||
2514 | btrfs_assert_tree_locked(path->nodes[1]); | |
2515 | ||
2516 | right = read_node_slot(root, upper, slot + 1); | |
2517 | btrfs_tree_lock(right); | |
2518 | btrfs_set_lock_blocking(right); | |
2519 | ||
2520 | free_space = btrfs_leaf_free_space(root, right); | |
2521 | if (free_space < data_size) | |
2522 | goto out_unlock; | |
2523 | ||
2524 | /* cow and double check */ | |
2525 | ret = btrfs_cow_block(trans, root, right, upper, | |
2526 | slot + 1, &right); | |
2527 | if (ret) | |
2528 | goto out_unlock; | |
2529 | ||
2530 | free_space = btrfs_leaf_free_space(root, right); | |
2531 | if (free_space < data_size) | |
2532 | goto out_unlock; | |
2533 | ||
2534 | left_nritems = btrfs_header_nritems(left); | |
2535 | if (left_nritems == 0) | |
2536 | goto out_unlock; | |
2537 | ||
99d8f83c CM |
2538 | return __push_leaf_right(trans, root, path, min_data_size, empty, |
2539 | right, free_space, left_nritems, min_slot); | |
44871b1b CM |
2540 | out_unlock: |
2541 | btrfs_tree_unlock(right); | |
2542 | free_extent_buffer(right); | |
2543 | return 1; | |
2544 | } | |
2545 | ||
74123bd7 CM |
2546 | /* |
2547 | * push some data in the path leaf to the left, trying to free up at | |
2548 | * least data_size bytes. returns zero if the push worked, nonzero otherwise | |
99d8f83c CM |
2549 | * |
2550 | * max_slot can put a limit on how far into the leaf we'll push items. The | |
2551 | * item at 'max_slot' won't be touched. Use (u32)-1 to make us do all the | |
2552 | * items | |
74123bd7 | 2553 | */ |
44871b1b CM |
2554 | static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, |
2555 | struct btrfs_root *root, | |
2556 | struct btrfs_path *path, int data_size, | |
2557 | int empty, struct extent_buffer *left, | |
99d8f83c CM |
2558 | int free_space, u32 right_nritems, |
2559 | u32 max_slot) | |
be0e5c09 | 2560 | { |
5f39d397 CM |
2561 | struct btrfs_disk_key disk_key; |
2562 | struct extent_buffer *right = path->nodes[0]; | |
be0e5c09 | 2563 | int i; |
be0e5c09 CM |
2564 | int push_space = 0; |
2565 | int push_items = 0; | |
0783fcfc | 2566 | struct btrfs_item *item; |
7518a238 | 2567 | u32 old_left_nritems; |
34a38218 | 2568 | u32 nr; |
aa5d6bed CM |
2569 | int ret = 0; |
2570 | int wret; | |
db94535d CM |
2571 | u32 this_item_size; |
2572 | u32 old_left_item_size; | |
be0e5c09 | 2573 | |
34a38218 | 2574 | if (empty) |
99d8f83c | 2575 | nr = min(right_nritems, max_slot); |
34a38218 | 2576 | else |
99d8f83c | 2577 | nr = min(right_nritems - 1, max_slot); |
34a38218 CM |
2578 | |
2579 | for (i = 0; i < nr; i++) { | |
5f39d397 | 2580 | item = btrfs_item_nr(right, i); |
db94535d CM |
2581 | if (!right->map_token) { |
2582 | map_extent_buffer(right, (unsigned long)item, | |
2583 | sizeof(struct btrfs_item), | |
2584 | &right->map_token, &right->kaddr, | |
2585 | &right->map_start, &right->map_len, | |
2586 | KM_USER1); | |
2587 | } | |
2588 | ||
31840ae1 ZY |
2589 | if (!empty && push_items > 0) { |
2590 | if (path->slots[0] < i) | |
2591 | break; | |
2592 | if (path->slots[0] == i) { | |
2593 | int space = btrfs_leaf_free_space(root, right); | |
2594 | if (space + push_space * 2 > free_space) | |
2595 | break; | |
2596 | } | |
2597 | } | |
2598 | ||
be0e5c09 | 2599 | if (path->slots[0] == i) |
87b29b20 | 2600 | push_space += data_size; |
db94535d CM |
2601 | |
2602 | this_item_size = btrfs_item_size(right, item); | |
2603 | if (this_item_size + sizeof(*item) + push_space > free_space) | |
be0e5c09 | 2604 | break; |
db94535d | 2605 | |
be0e5c09 | 2606 | push_items++; |
db94535d CM |
2607 | push_space += this_item_size + sizeof(*item); |
2608 | } | |
2609 | ||
2610 | if (right->map_token) { | |
2611 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2612 | right->map_token = NULL; | |
be0e5c09 | 2613 | } |
db94535d | 2614 | |
be0e5c09 | 2615 | if (push_items == 0) { |
925baedd CM |
2616 | ret = 1; |
2617 | goto out; | |
be0e5c09 | 2618 | } |
34a38218 | 2619 | if (!empty && push_items == btrfs_header_nritems(right)) |
a429e513 | 2620 | WARN_ON(1); |
5f39d397 | 2621 | |
be0e5c09 | 2622 | /* push data from right to left */ |
5f39d397 CM |
2623 | copy_extent_buffer(left, right, |
2624 | btrfs_item_nr_offset(btrfs_header_nritems(left)), | |
2625 | btrfs_item_nr_offset(0), | |
2626 | push_items * sizeof(struct btrfs_item)); | |
2627 | ||
123abc88 | 2628 | push_space = BTRFS_LEAF_DATA_SIZE(root) - |
d397712b | 2629 | btrfs_item_offset_nr(right, push_items - 1); |
5f39d397 CM |
2630 | |
2631 | copy_extent_buffer(left, right, btrfs_leaf_data(left) + | |
d6025579 CM |
2632 | leaf_data_end(root, left) - push_space, |
2633 | btrfs_leaf_data(right) + | |
5f39d397 | 2634 | btrfs_item_offset_nr(right, push_items - 1), |
d6025579 | 2635 | push_space); |
5f39d397 | 2636 | old_left_nritems = btrfs_header_nritems(left); |
87b29b20 | 2637 | BUG_ON(old_left_nritems <= 0); |
eb60ceac | 2638 | |
db94535d | 2639 | old_left_item_size = btrfs_item_offset_nr(left, old_left_nritems - 1); |
0783fcfc | 2640 | for (i = old_left_nritems; i < old_left_nritems + push_items; i++) { |
5f39d397 | 2641 | u32 ioff; |
db94535d | 2642 | |
5f39d397 | 2643 | item = btrfs_item_nr(left, i); |
db94535d CM |
2644 | if (!left->map_token) { |
2645 | map_extent_buffer(left, (unsigned long)item, | |
2646 | sizeof(struct btrfs_item), | |
2647 | &left->map_token, &left->kaddr, | |
2648 | &left->map_start, &left->map_len, | |
2649 | KM_USER1); | |
2650 | } | |
2651 | ||
5f39d397 CM |
2652 | ioff = btrfs_item_offset(left, item); |
2653 | btrfs_set_item_offset(left, item, | |
db94535d | 2654 | ioff - (BTRFS_LEAF_DATA_SIZE(root) - old_left_item_size)); |
be0e5c09 | 2655 | } |
5f39d397 | 2656 | btrfs_set_header_nritems(left, old_left_nritems + push_items); |
db94535d CM |
2657 | if (left->map_token) { |
2658 | unmap_extent_buffer(left, left->map_token, KM_USER1); | |
2659 | left->map_token = NULL; | |
2660 | } | |
be0e5c09 CM |
2661 | |
2662 | /* fixup right node */ | |
34a38218 | 2663 | if (push_items > right_nritems) { |
d397712b CM |
2664 | printk(KERN_CRIT "push items %d nr %u\n", push_items, |
2665 | right_nritems); | |
34a38218 CM |
2666 | WARN_ON(1); |
2667 | } | |
2668 | ||
2669 | if (push_items < right_nritems) { | |
2670 | push_space = btrfs_item_offset_nr(right, push_items - 1) - | |
2671 | leaf_data_end(root, right); | |
2672 | memmove_extent_buffer(right, btrfs_leaf_data(right) + | |
2673 | BTRFS_LEAF_DATA_SIZE(root) - push_space, | |
2674 | btrfs_leaf_data(right) + | |
2675 | leaf_data_end(root, right), push_space); | |
2676 | ||
2677 | memmove_extent_buffer(right, btrfs_item_nr_offset(0), | |
5f39d397 CM |
2678 | btrfs_item_nr_offset(push_items), |
2679 | (btrfs_header_nritems(right) - push_items) * | |
2680 | sizeof(struct btrfs_item)); | |
34a38218 | 2681 | } |
eef1c494 Y |
2682 | right_nritems -= push_items; |
2683 | btrfs_set_header_nritems(right, right_nritems); | |
123abc88 | 2684 | push_space = BTRFS_LEAF_DATA_SIZE(root); |
5f39d397 CM |
2685 | for (i = 0; i < right_nritems; i++) { |
2686 | item = btrfs_item_nr(right, i); | |
db94535d CM |
2687 | |
2688 | if (!right->map_token) { | |
2689 | map_extent_buffer(right, (unsigned long)item, | |
2690 | sizeof(struct btrfs_item), | |
2691 | &right->map_token, &right->kaddr, | |
2692 | &right->map_start, &right->map_len, | |
2693 | KM_USER1); | |
2694 | } | |
2695 | ||
2696 | push_space = push_space - btrfs_item_size(right, item); | |
2697 | btrfs_set_item_offset(right, item, push_space); | |
2698 | } | |
2699 | if (right->map_token) { | |
2700 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2701 | right->map_token = NULL; | |
be0e5c09 | 2702 | } |
eb60ceac | 2703 | |
5f39d397 | 2704 | btrfs_mark_buffer_dirty(left); |
34a38218 CM |
2705 | if (right_nritems) |
2706 | btrfs_mark_buffer_dirty(right); | |
f0486c68 YZ |
2707 | else |
2708 | clean_tree_block(trans, root, right); | |
098f59c2 | 2709 | |
5f39d397 CM |
2710 | btrfs_item_key(right, &disk_key, 0); |
2711 | wret = fixup_low_keys(trans, root, path, &disk_key, 1); | |
aa5d6bed CM |
2712 | if (wret) |
2713 | ret = wret; | |
be0e5c09 CM |
2714 | |
2715 | /* then fixup the leaf pointer in the path */ | |
2716 | if (path->slots[0] < push_items) { | |
2717 | path->slots[0] += old_left_nritems; | |
925baedd | 2718 | btrfs_tree_unlock(path->nodes[0]); |
5f39d397 CM |
2719 | free_extent_buffer(path->nodes[0]); |
2720 | path->nodes[0] = left; | |
be0e5c09 CM |
2721 | path->slots[1] -= 1; |
2722 | } else { | |
925baedd | 2723 | btrfs_tree_unlock(left); |
5f39d397 | 2724 | free_extent_buffer(left); |
be0e5c09 CM |
2725 | path->slots[0] -= push_items; |
2726 | } | |
eb60ceac | 2727 | BUG_ON(path->slots[0] < 0); |
aa5d6bed | 2728 | return ret; |
925baedd CM |
2729 | out: |
2730 | btrfs_tree_unlock(left); | |
2731 | free_extent_buffer(left); | |
2732 | return ret; | |
be0e5c09 CM |
2733 | } |
2734 | ||
44871b1b CM |
2735 | /* |
2736 | * push some data in the path leaf to the left, trying to free up at | |
2737 | * least data_size bytes. returns zero if the push worked, nonzero otherwise | |
99d8f83c CM |
2738 | * |
2739 | * max_slot can put a limit on how far into the leaf we'll push items. The | |
2740 | * item at 'max_slot' won't be touched. Use (u32)-1 to make us push all the | |
2741 | * items | |
44871b1b CM |
2742 | */ |
2743 | static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root | |
99d8f83c CM |
2744 | *root, struct btrfs_path *path, int min_data_size, |
2745 | int data_size, int empty, u32 max_slot) | |
44871b1b CM |
2746 | { |
2747 | struct extent_buffer *right = path->nodes[0]; | |
2748 | struct extent_buffer *left; | |
2749 | int slot; | |
2750 | int free_space; | |
2751 | u32 right_nritems; | |
2752 | int ret = 0; | |
2753 | ||
2754 | slot = path->slots[1]; | |
2755 | if (slot == 0) | |
2756 | return 1; | |
2757 | if (!path->nodes[1]) | |
2758 | return 1; | |
2759 | ||
2760 | right_nritems = btrfs_header_nritems(right); | |
2761 | if (right_nritems == 0) | |
2762 | return 1; | |
2763 | ||
2764 | btrfs_assert_tree_locked(path->nodes[1]); | |
2765 | ||
2766 | left = read_node_slot(root, path->nodes[1], slot - 1); | |
2767 | btrfs_tree_lock(left); | |
2768 | btrfs_set_lock_blocking(left); | |
2769 | ||
2770 | free_space = btrfs_leaf_free_space(root, left); | |
2771 | if (free_space < data_size) { | |
2772 | ret = 1; | |
2773 | goto out; | |
2774 | } | |
2775 | ||
2776 | /* cow and double check */ | |
2777 | ret = btrfs_cow_block(trans, root, left, | |
2778 | path->nodes[1], slot - 1, &left); | |
2779 | if (ret) { | |
2780 | /* we hit -ENOSPC, but it isn't fatal here */ | |
2781 | ret = 1; | |
2782 | goto out; | |
2783 | } | |
2784 | ||
2785 | free_space = btrfs_leaf_free_space(root, left); | |
2786 | if (free_space < data_size) { | |
2787 | ret = 1; | |
2788 | goto out; | |
2789 | } | |
2790 | ||
99d8f83c CM |
2791 | return __push_leaf_left(trans, root, path, min_data_size, |
2792 | empty, left, free_space, right_nritems, | |
2793 | max_slot); | |
44871b1b CM |
2794 | out: |
2795 | btrfs_tree_unlock(left); | |
2796 | free_extent_buffer(left); | |
2797 | return ret; | |
2798 | } | |
2799 | ||
2800 | /* | |
2801 | * split the path's leaf in two, making sure there is at least data_size | |
2802 | * available for the resulting leaf level of the path. | |
2803 | * | |
2804 | * returns 0 if all went well and < 0 on failure. | |
2805 | */ | |
2806 | static noinline int copy_for_split(struct btrfs_trans_handle *trans, | |
2807 | struct btrfs_root *root, | |
2808 | struct btrfs_path *path, | |
2809 | struct extent_buffer *l, | |
2810 | struct extent_buffer *right, | |
2811 | int slot, int mid, int nritems) | |
2812 | { | |
2813 | int data_copy_size; | |
2814 | int rt_data_off; | |
2815 | int i; | |
2816 | int ret = 0; | |
2817 | int wret; | |
2818 | struct btrfs_disk_key disk_key; | |
2819 | ||
2820 | nritems = nritems - mid; | |
2821 | btrfs_set_header_nritems(right, nritems); | |
2822 | data_copy_size = btrfs_item_end_nr(l, mid) - leaf_data_end(root, l); | |
2823 | ||
2824 | copy_extent_buffer(right, l, btrfs_item_nr_offset(0), | |
2825 | btrfs_item_nr_offset(mid), | |
2826 | nritems * sizeof(struct btrfs_item)); | |
2827 | ||
2828 | copy_extent_buffer(right, l, | |
2829 | btrfs_leaf_data(right) + BTRFS_LEAF_DATA_SIZE(root) - | |
2830 | data_copy_size, btrfs_leaf_data(l) + | |
2831 | leaf_data_end(root, l), data_copy_size); | |
2832 | ||
2833 | rt_data_off = BTRFS_LEAF_DATA_SIZE(root) - | |
2834 | btrfs_item_end_nr(l, mid); | |
2835 | ||
2836 | for (i = 0; i < nritems; i++) { | |
2837 | struct btrfs_item *item = btrfs_item_nr(right, i); | |
2838 | u32 ioff; | |
2839 | ||
2840 | if (!right->map_token) { | |
2841 | map_extent_buffer(right, (unsigned long)item, | |
2842 | sizeof(struct btrfs_item), | |
2843 | &right->map_token, &right->kaddr, | |
2844 | &right->map_start, &right->map_len, | |
2845 | KM_USER1); | |
2846 | } | |
2847 | ||
2848 | ioff = btrfs_item_offset(right, item); | |
2849 | btrfs_set_item_offset(right, item, ioff + rt_data_off); | |
2850 | } | |
2851 | ||
2852 | if (right->map_token) { | |
2853 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2854 | right->map_token = NULL; | |
2855 | } | |
2856 | ||
2857 | btrfs_set_header_nritems(l, mid); | |
2858 | ret = 0; | |
2859 | btrfs_item_key(right, &disk_key, 0); | |
2860 | wret = insert_ptr(trans, root, path, &disk_key, right->start, | |
2861 | path->slots[1] + 1, 1); | |
2862 | if (wret) | |
2863 | ret = wret; | |
2864 | ||
2865 | btrfs_mark_buffer_dirty(right); | |
2866 | btrfs_mark_buffer_dirty(l); | |
2867 | BUG_ON(path->slots[0] != slot); | |
2868 | ||
44871b1b CM |
2869 | if (mid <= slot) { |
2870 | btrfs_tree_unlock(path->nodes[0]); | |
2871 | free_extent_buffer(path->nodes[0]); | |
2872 | path->nodes[0] = right; | |
2873 | path->slots[0] -= mid; | |
2874 | path->slots[1] += 1; | |
2875 | } else { | |
2876 | btrfs_tree_unlock(right); | |
2877 | free_extent_buffer(right); | |
2878 | } | |
2879 | ||
2880 | BUG_ON(path->slots[0] < 0); | |
2881 | ||
2882 | return ret; | |
2883 | } | |
2884 | ||
99d8f83c CM |
2885 | /* |
2886 | * double splits happen when we need to insert a big item in the middle | |
2887 | * of a leaf. A double split can leave us with 3 mostly empty leaves: | |
2888 | * leaf: [ slots 0 - N] [ our target ] [ N + 1 - total in leaf ] | |
2889 | * A B C | |
2890 | * | |
2891 | * We avoid this by trying to push the items on either side of our target | |
2892 | * into the adjacent leaves. If all goes well we can avoid the double split | |
2893 | * completely. | |
2894 | */ | |
2895 | static noinline int push_for_double_split(struct btrfs_trans_handle *trans, | |
2896 | struct btrfs_root *root, | |
2897 | struct btrfs_path *path, | |
2898 | int data_size) | |
2899 | { | |
2900 | int ret; | |
2901 | int progress = 0; | |
2902 | int slot; | |
2903 | u32 nritems; | |
2904 | ||
2905 | slot = path->slots[0]; | |
2906 | ||
2907 | /* | |
2908 | * try to push all the items after our slot into the | |
2909 | * right leaf | |
2910 | */ | |
2911 | ret = push_leaf_right(trans, root, path, 1, data_size, 0, slot); | |
2912 | if (ret < 0) | |
2913 | return ret; | |
2914 | ||
2915 | if (ret == 0) | |
2916 | progress++; | |
2917 | ||
2918 | nritems = btrfs_header_nritems(path->nodes[0]); | |
2919 | /* | |
2920 | * our goal is to get our slot at the start or end of a leaf. If | |
2921 | * we've done so we're done | |
2922 | */ | |
2923 | if (path->slots[0] == 0 || path->slots[0] == nritems) | |
2924 | return 0; | |
2925 | ||
2926 | if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size) | |
2927 | return 0; | |
2928 | ||
2929 | /* try to push all the items before our slot into the next leaf */ | |
2930 | slot = path->slots[0]; | |
2931 | ret = push_leaf_left(trans, root, path, 1, data_size, 0, slot); | |
2932 | if (ret < 0) | |
2933 | return ret; | |
2934 | ||
2935 | if (ret == 0) | |
2936 | progress++; | |
2937 | ||
2938 | if (progress) | |
2939 | return 0; | |
2940 | return 1; | |
2941 | } | |
2942 | ||
74123bd7 CM |
2943 | /* |
2944 | * split the path's leaf in two, making sure there is at least data_size | |
2945 | * available for the resulting leaf level of the path. | |
aa5d6bed CM |
2946 | * |
2947 | * returns 0 if all went well and < 0 on failure. | |
74123bd7 | 2948 | */ |
e02119d5 CM |
2949 | static noinline int split_leaf(struct btrfs_trans_handle *trans, |
2950 | struct btrfs_root *root, | |
2951 | struct btrfs_key *ins_key, | |
2952 | struct btrfs_path *path, int data_size, | |
2953 | int extend) | |
be0e5c09 | 2954 | { |
5d4f98a2 | 2955 | struct btrfs_disk_key disk_key; |
5f39d397 | 2956 | struct extent_buffer *l; |
7518a238 | 2957 | u32 nritems; |
eb60ceac CM |
2958 | int mid; |
2959 | int slot; | |
5f39d397 | 2960 | struct extent_buffer *right; |
d4dbff95 | 2961 | int ret = 0; |
aa5d6bed | 2962 | int wret; |
5d4f98a2 | 2963 | int split; |
cc0c5538 | 2964 | int num_doubles = 0; |
99d8f83c | 2965 | int tried_avoid_double = 0; |
aa5d6bed | 2966 | |
a5719521 YZ |
2967 | l = path->nodes[0]; |
2968 | slot = path->slots[0]; | |
2969 | if (extend && data_size + btrfs_item_size_nr(l, slot) + | |
2970 | sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root)) | |
2971 | return -EOVERFLOW; | |
2972 | ||
40689478 | 2973 | /* first try to make some room by pushing left and right */ |
99d8f83c CM |
2974 | if (data_size) { |
2975 | wret = push_leaf_right(trans, root, path, data_size, | |
2976 | data_size, 0, 0); | |
d397712b | 2977 | if (wret < 0) |
eaee50e8 | 2978 | return wret; |
3685f791 | 2979 | if (wret) { |
99d8f83c CM |
2980 | wret = push_leaf_left(trans, root, path, data_size, |
2981 | data_size, 0, (u32)-1); | |
3685f791 CM |
2982 | if (wret < 0) |
2983 | return wret; | |
2984 | } | |
2985 | l = path->nodes[0]; | |
aa5d6bed | 2986 | |
3685f791 | 2987 | /* did the pushes work? */ |
87b29b20 | 2988 | if (btrfs_leaf_free_space(root, l) >= data_size) |
3685f791 | 2989 | return 0; |
3326d1b0 | 2990 | } |
aa5d6bed | 2991 | |
5c680ed6 | 2992 | if (!path->nodes[1]) { |
e089f05c | 2993 | ret = insert_new_root(trans, root, path, 1); |
5c680ed6 CM |
2994 | if (ret) |
2995 | return ret; | |
2996 | } | |
cc0c5538 | 2997 | again: |
5d4f98a2 | 2998 | split = 1; |
cc0c5538 | 2999 | l = path->nodes[0]; |
eb60ceac | 3000 | slot = path->slots[0]; |
5f39d397 | 3001 | nritems = btrfs_header_nritems(l); |
d397712b | 3002 | mid = (nritems + 1) / 2; |
54aa1f4d | 3003 | |
5d4f98a2 YZ |
3004 | if (mid <= slot) { |
3005 | if (nritems == 1 || | |
3006 | leaf_space_used(l, mid, nritems - mid) + data_size > | |
3007 | BTRFS_LEAF_DATA_SIZE(root)) { | |
3008 | if (slot >= nritems) { | |
3009 | split = 0; | |
3010 | } else { | |
3011 | mid = slot; | |
3012 | if (mid != nritems && | |
3013 | leaf_space_used(l, mid, nritems - mid) + | |
3014 | data_size > BTRFS_LEAF_DATA_SIZE(root)) { | |
99d8f83c CM |
3015 | if (data_size && !tried_avoid_double) |
3016 | goto push_for_double; | |
5d4f98a2 YZ |
3017 | split = 2; |
3018 | } | |
3019 | } | |
3020 | } | |
3021 | } else { | |
3022 | if (leaf_space_used(l, 0, mid) + data_size > | |
3023 | BTRFS_LEAF_DATA_SIZE(root)) { | |
3024 | if (!extend && data_size && slot == 0) { | |
3025 | split = 0; | |
3026 | } else if ((extend || !data_size) && slot == 0) { | |
3027 | mid = 1; | |
3028 | } else { | |
3029 | mid = slot; | |
3030 | if (mid != nritems && | |
3031 | leaf_space_used(l, mid, nritems - mid) + | |
3032 | data_size > BTRFS_LEAF_DATA_SIZE(root)) { | |
99d8f83c CM |
3033 | if (data_size && !tried_avoid_double) |
3034 | goto push_for_double; | |
5d4f98a2 YZ |
3035 | split = 2 ; |
3036 | } | |
3037 | } | |
3038 | } | |
3039 | } | |
3040 | ||
3041 | if (split == 0) | |
3042 | btrfs_cpu_key_to_disk(&disk_key, ins_key); | |
3043 | else | |
3044 | btrfs_item_key(l, &disk_key, mid); | |
3045 | ||
3046 | right = btrfs_alloc_free_block(trans, root, root->leafsize, 0, | |
31840ae1 | 3047 | root->root_key.objectid, |
5d4f98a2 | 3048 | &disk_key, 0, l->start, 0); |
f0486c68 | 3049 | if (IS_ERR(right)) |
5f39d397 | 3050 | return PTR_ERR(right); |
f0486c68 YZ |
3051 | |
3052 | root_add_used(root, root->leafsize); | |
5f39d397 CM |
3053 | |
3054 | memset_extent_buffer(right, 0, 0, sizeof(struct btrfs_header)); | |
db94535d | 3055 | btrfs_set_header_bytenr(right, right->start); |
5f39d397 | 3056 | btrfs_set_header_generation(right, trans->transid); |
5d4f98a2 | 3057 | btrfs_set_header_backref_rev(right, BTRFS_MIXED_BACKREF_REV); |
5f39d397 CM |
3058 | btrfs_set_header_owner(right, root->root_key.objectid); |
3059 | btrfs_set_header_level(right, 0); | |
3060 | write_extent_buffer(right, root->fs_info->fsid, | |
3061 | (unsigned long)btrfs_header_fsid(right), | |
3062 | BTRFS_FSID_SIZE); | |
e17cade2 CM |
3063 | |
3064 | write_extent_buffer(right, root->fs_info->chunk_tree_uuid, | |
3065 | (unsigned long)btrfs_header_chunk_tree_uuid(right), | |
3066 | BTRFS_UUID_SIZE); | |
44871b1b | 3067 | |
5d4f98a2 YZ |
3068 | if (split == 0) { |
3069 | if (mid <= slot) { | |
3070 | btrfs_set_header_nritems(right, 0); | |
3071 | wret = insert_ptr(trans, root, path, | |
3072 | &disk_key, right->start, | |
3073 | path->slots[1] + 1, 1); | |
3074 | if (wret) | |
3075 | ret = wret; | |
925baedd | 3076 | |
5d4f98a2 YZ |
3077 | btrfs_tree_unlock(path->nodes[0]); |
3078 | free_extent_buffer(path->nodes[0]); | |
3079 | path->nodes[0] = right; | |
3080 | path->slots[0] = 0; | |
3081 | path->slots[1] += 1; | |
3082 | } else { | |
3083 | btrfs_set_header_nritems(right, 0); | |
3084 | wret = insert_ptr(trans, root, path, | |
3085 | &disk_key, | |
3086 | right->start, | |
3087 | path->slots[1], 1); | |
3088 | if (wret) | |
3089 | ret = wret; | |
3090 | btrfs_tree_unlock(path->nodes[0]); | |
3091 | free_extent_buffer(path->nodes[0]); | |
3092 | path->nodes[0] = right; | |
3093 | path->slots[0] = 0; | |
3094 | if (path->slots[1] == 0) { | |
3095 | wret = fixup_low_keys(trans, root, | |
3096 | path, &disk_key, 1); | |
d4dbff95 CM |
3097 | if (wret) |
3098 | ret = wret; | |
5ee78ac7 | 3099 | } |
d4dbff95 | 3100 | } |
5d4f98a2 YZ |
3101 | btrfs_mark_buffer_dirty(right); |
3102 | return ret; | |
d4dbff95 | 3103 | } |
74123bd7 | 3104 | |
44871b1b | 3105 | ret = copy_for_split(trans, root, path, l, right, slot, mid, nritems); |
31840ae1 ZY |
3106 | BUG_ON(ret); |
3107 | ||
5d4f98a2 | 3108 | if (split == 2) { |
cc0c5538 CM |
3109 | BUG_ON(num_doubles != 0); |
3110 | num_doubles++; | |
3111 | goto again; | |
a429e513 | 3112 | } |
44871b1b | 3113 | |
be0e5c09 | 3114 | return ret; |
99d8f83c CM |
3115 | |
3116 | push_for_double: | |
3117 | push_for_double_split(trans, root, path, data_size); | |
3118 | tried_avoid_double = 1; | |
3119 | if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size) | |
3120 | return 0; | |
3121 | goto again; | |
be0e5c09 CM |
3122 | } |
3123 | ||
ad48fd75 YZ |
3124 | static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans, |
3125 | struct btrfs_root *root, | |
3126 | struct btrfs_path *path, int ins_len) | |
459931ec | 3127 | { |
ad48fd75 | 3128 | struct btrfs_key key; |
459931ec | 3129 | struct extent_buffer *leaf; |
ad48fd75 YZ |
3130 | struct btrfs_file_extent_item *fi; |
3131 | u64 extent_len = 0; | |
3132 | u32 item_size; | |
3133 | int ret; | |
459931ec CM |
3134 | |
3135 | leaf = path->nodes[0]; | |
ad48fd75 YZ |
3136 | btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); |
3137 | ||
3138 | BUG_ON(key.type != BTRFS_EXTENT_DATA_KEY && | |
3139 | key.type != BTRFS_EXTENT_CSUM_KEY); | |
3140 | ||
3141 | if (btrfs_leaf_free_space(root, leaf) >= ins_len) | |
3142 | return 0; | |
459931ec CM |
3143 | |
3144 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
ad48fd75 YZ |
3145 | if (key.type == BTRFS_EXTENT_DATA_KEY) { |
3146 | fi = btrfs_item_ptr(leaf, path->slots[0], | |
3147 | struct btrfs_file_extent_item); | |
3148 | extent_len = btrfs_file_extent_num_bytes(leaf, fi); | |
3149 | } | |
459931ec CM |
3150 | btrfs_release_path(root, path); |
3151 | ||
459931ec | 3152 | path->keep_locks = 1; |
ad48fd75 YZ |
3153 | path->search_for_split = 1; |
3154 | ret = btrfs_search_slot(trans, root, &key, path, 0, 1); | |
459931ec | 3155 | path->search_for_split = 0; |
ad48fd75 YZ |
3156 | if (ret < 0) |
3157 | goto err; | |
459931ec | 3158 | |
ad48fd75 YZ |
3159 | ret = -EAGAIN; |
3160 | leaf = path->nodes[0]; | |
459931ec | 3161 | /* if our item isn't there or got smaller, return now */ |
ad48fd75 YZ |
3162 | if (ret > 0 || item_size != btrfs_item_size_nr(leaf, path->slots[0])) |
3163 | goto err; | |
3164 | ||
109f6aef CM |
3165 | /* the leaf has changed, it now has room. return now */ |
3166 | if (btrfs_leaf_free_space(root, path->nodes[0]) >= ins_len) | |
3167 | goto err; | |
3168 | ||
ad48fd75 YZ |
3169 | if (key.type == BTRFS_EXTENT_DATA_KEY) { |
3170 | fi = btrfs_item_ptr(leaf, path->slots[0], | |
3171 | struct btrfs_file_extent_item); | |
3172 | if (extent_len != btrfs_file_extent_num_bytes(leaf, fi)) | |
3173 | goto err; | |
459931ec CM |
3174 | } |
3175 | ||
b9473439 | 3176 | btrfs_set_path_blocking(path); |
ad48fd75 | 3177 | ret = split_leaf(trans, root, &key, path, ins_len, 1); |
f0486c68 YZ |
3178 | if (ret) |
3179 | goto err; | |
459931ec | 3180 | |
ad48fd75 | 3181 | path->keep_locks = 0; |
b9473439 | 3182 | btrfs_unlock_up_safe(path, 1); |
ad48fd75 YZ |
3183 | return 0; |
3184 | err: | |
3185 | path->keep_locks = 0; | |
3186 | return ret; | |
3187 | } | |
3188 | ||
3189 | static noinline int split_item(struct btrfs_trans_handle *trans, | |
3190 | struct btrfs_root *root, | |
3191 | struct btrfs_path *path, | |
3192 | struct btrfs_key *new_key, | |
3193 | unsigned long split_offset) | |
3194 | { | |
3195 | struct extent_buffer *leaf; | |
3196 | struct btrfs_item *item; | |
3197 | struct btrfs_item *new_item; | |
3198 | int slot; | |
3199 | char *buf; | |
3200 | u32 nritems; | |
3201 | u32 item_size; | |
3202 | u32 orig_offset; | |
3203 | struct btrfs_disk_key disk_key; | |
3204 | ||
b9473439 CM |
3205 | leaf = path->nodes[0]; |
3206 | BUG_ON(btrfs_leaf_free_space(root, leaf) < sizeof(struct btrfs_item)); | |
3207 | ||
b4ce94de CM |
3208 | btrfs_set_path_blocking(path); |
3209 | ||
459931ec CM |
3210 | item = btrfs_item_nr(leaf, path->slots[0]); |
3211 | orig_offset = btrfs_item_offset(leaf, item); | |
3212 | item_size = btrfs_item_size(leaf, item); | |
3213 | ||
459931ec | 3214 | buf = kmalloc(item_size, GFP_NOFS); |
ad48fd75 YZ |
3215 | if (!buf) |
3216 | return -ENOMEM; | |
3217 | ||
459931ec CM |
3218 | read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, |
3219 | path->slots[0]), item_size); | |
459931ec | 3220 | |
ad48fd75 | 3221 | slot = path->slots[0] + 1; |
459931ec | 3222 | nritems = btrfs_header_nritems(leaf); |
459931ec CM |
3223 | if (slot != nritems) { |
3224 | /* shift the items */ | |
3225 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + 1), | |
ad48fd75 YZ |
3226 | btrfs_item_nr_offset(slot), |
3227 | (nritems - slot) * sizeof(struct btrfs_item)); | |
459931ec CM |
3228 | } |
3229 | ||
3230 | btrfs_cpu_key_to_disk(&disk_key, new_key); | |
3231 | btrfs_set_item_key(leaf, &disk_key, slot); | |
3232 | ||
3233 | new_item = btrfs_item_nr(leaf, slot); | |
3234 | ||
3235 | btrfs_set_item_offset(leaf, new_item, orig_offset); | |
3236 | btrfs_set_item_size(leaf, new_item, item_size - split_offset); | |
3237 | ||
3238 | btrfs_set_item_offset(leaf, item, | |
3239 | orig_offset + item_size - split_offset); | |
3240 | btrfs_set_item_size(leaf, item, split_offset); | |
3241 | ||
3242 | btrfs_set_header_nritems(leaf, nritems + 1); | |
3243 | ||
3244 | /* write the data for the start of the original item */ | |
3245 | write_extent_buffer(leaf, buf, | |
3246 | btrfs_item_ptr_offset(leaf, path->slots[0]), | |
3247 | split_offset); | |
3248 | ||
3249 | /* write the data for the new item */ | |
3250 | write_extent_buffer(leaf, buf + split_offset, | |
3251 | btrfs_item_ptr_offset(leaf, slot), | |
3252 | item_size - split_offset); | |
3253 | btrfs_mark_buffer_dirty(leaf); | |
3254 | ||
ad48fd75 | 3255 | BUG_ON(btrfs_leaf_free_space(root, leaf) < 0); |
459931ec | 3256 | kfree(buf); |
ad48fd75 YZ |
3257 | return 0; |
3258 | } | |
3259 | ||
3260 | /* | |
3261 | * This function splits a single item into two items, | |
3262 | * giving 'new_key' to the new item and splitting the | |
3263 | * old one at split_offset (from the start of the item). | |
3264 | * | |
3265 | * The path may be released by this operation. After | |
3266 | * the split, the path is pointing to the old item. The | |
3267 | * new item is going to be in the same node as the old one. | |
3268 | * | |
3269 | * Note, the item being split must be smaller enough to live alone on | |
3270 | * a tree block with room for one extra struct btrfs_item | |
3271 | * | |
3272 | * This allows us to split the item in place, keeping a lock on the | |
3273 | * leaf the entire time. | |
3274 | */ | |
3275 | int btrfs_split_item(struct btrfs_trans_handle *trans, | |
3276 | struct btrfs_root *root, | |
3277 | struct btrfs_path *path, | |
3278 | struct btrfs_key *new_key, | |
3279 | unsigned long split_offset) | |
3280 | { | |
3281 | int ret; | |
3282 | ret = setup_leaf_for_split(trans, root, path, | |
3283 | sizeof(struct btrfs_item)); | |
3284 | if (ret) | |
3285 | return ret; | |
3286 | ||
3287 | ret = split_item(trans, root, path, new_key, split_offset); | |
459931ec CM |
3288 | return ret; |
3289 | } | |
3290 | ||
ad48fd75 YZ |
3291 | /* |
3292 | * This function duplicate a item, giving 'new_key' to the new item. | |
3293 | * It guarantees both items live in the same tree leaf and the new item | |
3294 | * is contiguous with the original item. | |
3295 | * | |
3296 | * This allows us to split file extent in place, keeping a lock on the | |
3297 | * leaf the entire time. | |
3298 | */ | |
3299 | int btrfs_duplicate_item(struct btrfs_trans_handle *trans, | |
3300 | struct btrfs_root *root, | |
3301 | struct btrfs_path *path, | |
3302 | struct btrfs_key *new_key) | |
3303 | { | |
3304 | struct extent_buffer *leaf; | |
3305 | int ret; | |
3306 | u32 item_size; | |
3307 | ||
3308 | leaf = path->nodes[0]; | |
3309 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
3310 | ret = setup_leaf_for_split(trans, root, path, | |
3311 | item_size + sizeof(struct btrfs_item)); | |
3312 | if (ret) | |
3313 | return ret; | |
3314 | ||
3315 | path->slots[0]++; | |
3316 | ret = setup_items_for_insert(trans, root, path, new_key, &item_size, | |
3317 | item_size, item_size + | |
3318 | sizeof(struct btrfs_item), 1); | |
3319 | BUG_ON(ret); | |
3320 | ||
3321 | leaf = path->nodes[0]; | |
3322 | memcpy_extent_buffer(leaf, | |
3323 | btrfs_item_ptr_offset(leaf, path->slots[0]), | |
3324 | btrfs_item_ptr_offset(leaf, path->slots[0] - 1), | |
3325 | item_size); | |
3326 | return 0; | |
3327 | } | |
3328 | ||
d352ac68 CM |
3329 | /* |
3330 | * make the item pointed to by the path smaller. new_size indicates | |
3331 | * how small to make it, and from_end tells us if we just chop bytes | |
3332 | * off the end of the item or if we shift the item to chop bytes off | |
3333 | * the front. | |
3334 | */ | |
b18c6685 CM |
3335 | int btrfs_truncate_item(struct btrfs_trans_handle *trans, |
3336 | struct btrfs_root *root, | |
3337 | struct btrfs_path *path, | |
179e29e4 | 3338 | u32 new_size, int from_end) |
b18c6685 CM |
3339 | { |
3340 | int ret = 0; | |
3341 | int slot; | |
5f39d397 CM |
3342 | struct extent_buffer *leaf; |
3343 | struct btrfs_item *item; | |
b18c6685 CM |
3344 | u32 nritems; |
3345 | unsigned int data_end; | |
3346 | unsigned int old_data_start; | |
3347 | unsigned int old_size; | |
3348 | unsigned int size_diff; | |
3349 | int i; | |
3350 | ||
5f39d397 | 3351 | leaf = path->nodes[0]; |
179e29e4 CM |
3352 | slot = path->slots[0]; |
3353 | ||
3354 | old_size = btrfs_item_size_nr(leaf, slot); | |
3355 | if (old_size == new_size) | |
3356 | return 0; | |
b18c6685 | 3357 | |
5f39d397 | 3358 | nritems = btrfs_header_nritems(leaf); |
b18c6685 CM |
3359 | data_end = leaf_data_end(root, leaf); |
3360 | ||
5f39d397 | 3361 | old_data_start = btrfs_item_offset_nr(leaf, slot); |
179e29e4 | 3362 | |
b18c6685 CM |
3363 | size_diff = old_size - new_size; |
3364 | ||
3365 | BUG_ON(slot < 0); | |
3366 | BUG_ON(slot >= nritems); | |
3367 | ||
3368 | /* | |
3369 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3370 | */ | |
3371 | /* first correct the data pointers */ | |
3372 | for (i = slot; i < nritems; i++) { | |
5f39d397 CM |
3373 | u32 ioff; |
3374 | item = btrfs_item_nr(leaf, i); | |
db94535d CM |
3375 | |
3376 | if (!leaf->map_token) { | |
3377 | map_extent_buffer(leaf, (unsigned long)item, | |
3378 | sizeof(struct btrfs_item), | |
3379 | &leaf->map_token, &leaf->kaddr, | |
3380 | &leaf->map_start, &leaf->map_len, | |
3381 | KM_USER1); | |
3382 | } | |
3383 | ||
5f39d397 CM |
3384 | ioff = btrfs_item_offset(leaf, item); |
3385 | btrfs_set_item_offset(leaf, item, ioff + size_diff); | |
b18c6685 | 3386 | } |
db94535d CM |
3387 | |
3388 | if (leaf->map_token) { | |
3389 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3390 | leaf->map_token = NULL; | |
3391 | } | |
3392 | ||
b18c6685 | 3393 | /* shift the data */ |
179e29e4 CM |
3394 | if (from_end) { |
3395 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
3396 | data_end + size_diff, btrfs_leaf_data(leaf) + | |
3397 | data_end, old_data_start + new_size - data_end); | |
3398 | } else { | |
3399 | struct btrfs_disk_key disk_key; | |
3400 | u64 offset; | |
3401 | ||
3402 | btrfs_item_key(leaf, &disk_key, slot); | |
3403 | ||
3404 | if (btrfs_disk_key_type(&disk_key) == BTRFS_EXTENT_DATA_KEY) { | |
3405 | unsigned long ptr; | |
3406 | struct btrfs_file_extent_item *fi; | |
3407 | ||
3408 | fi = btrfs_item_ptr(leaf, slot, | |
3409 | struct btrfs_file_extent_item); | |
3410 | fi = (struct btrfs_file_extent_item *)( | |
3411 | (unsigned long)fi - size_diff); | |
3412 | ||
3413 | if (btrfs_file_extent_type(leaf, fi) == | |
3414 | BTRFS_FILE_EXTENT_INLINE) { | |
3415 | ptr = btrfs_item_ptr_offset(leaf, slot); | |
3416 | memmove_extent_buffer(leaf, ptr, | |
d397712b CM |
3417 | (unsigned long)fi, |
3418 | offsetof(struct btrfs_file_extent_item, | |
179e29e4 CM |
3419 | disk_bytenr)); |
3420 | } | |
3421 | } | |
3422 | ||
3423 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
3424 | data_end + size_diff, btrfs_leaf_data(leaf) + | |
3425 | data_end, old_data_start - data_end); | |
3426 | ||
3427 | offset = btrfs_disk_key_offset(&disk_key); | |
3428 | btrfs_set_disk_key_offset(&disk_key, offset + size_diff); | |
3429 | btrfs_set_item_key(leaf, &disk_key, slot); | |
3430 | if (slot == 0) | |
3431 | fixup_low_keys(trans, root, path, &disk_key, 1); | |
3432 | } | |
5f39d397 CM |
3433 | |
3434 | item = btrfs_item_nr(leaf, slot); | |
3435 | btrfs_set_item_size(leaf, item, new_size); | |
3436 | btrfs_mark_buffer_dirty(leaf); | |
b18c6685 CM |
3437 | |
3438 | ret = 0; | |
5f39d397 CM |
3439 | if (btrfs_leaf_free_space(root, leaf) < 0) { |
3440 | btrfs_print_leaf(root, leaf); | |
b18c6685 | 3441 | BUG(); |
5f39d397 | 3442 | } |
b18c6685 CM |
3443 | return ret; |
3444 | } | |
3445 | ||
d352ac68 CM |
3446 | /* |
3447 | * make the item pointed to by the path bigger, data_size is the new size. | |
3448 | */ | |
5f39d397 CM |
3449 | int btrfs_extend_item(struct btrfs_trans_handle *trans, |
3450 | struct btrfs_root *root, struct btrfs_path *path, | |
3451 | u32 data_size) | |
6567e837 CM |
3452 | { |
3453 | int ret = 0; | |
3454 | int slot; | |
5f39d397 CM |
3455 | struct extent_buffer *leaf; |
3456 | struct btrfs_item *item; | |
6567e837 CM |
3457 | u32 nritems; |
3458 | unsigned int data_end; | |
3459 | unsigned int old_data; | |
3460 | unsigned int old_size; | |
3461 | int i; | |
3462 | ||
5f39d397 | 3463 | leaf = path->nodes[0]; |
6567e837 | 3464 | |
5f39d397 | 3465 | nritems = btrfs_header_nritems(leaf); |
6567e837 CM |
3466 | data_end = leaf_data_end(root, leaf); |
3467 | ||
5f39d397 CM |
3468 | if (btrfs_leaf_free_space(root, leaf) < data_size) { |
3469 | btrfs_print_leaf(root, leaf); | |
6567e837 | 3470 | BUG(); |
5f39d397 | 3471 | } |
6567e837 | 3472 | slot = path->slots[0]; |
5f39d397 | 3473 | old_data = btrfs_item_end_nr(leaf, slot); |
6567e837 CM |
3474 | |
3475 | BUG_ON(slot < 0); | |
3326d1b0 CM |
3476 | if (slot >= nritems) { |
3477 | btrfs_print_leaf(root, leaf); | |
d397712b CM |
3478 | printk(KERN_CRIT "slot %d too large, nritems %d\n", |
3479 | slot, nritems); | |
3326d1b0 CM |
3480 | BUG_ON(1); |
3481 | } | |
6567e837 CM |
3482 | |
3483 | /* | |
3484 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3485 | */ | |
3486 | /* first correct the data pointers */ | |
3487 | for (i = slot; i < nritems; i++) { | |
5f39d397 CM |
3488 | u32 ioff; |
3489 | item = btrfs_item_nr(leaf, i); | |
db94535d CM |
3490 | |
3491 | if (!leaf->map_token) { | |
3492 | map_extent_buffer(leaf, (unsigned long)item, | |
3493 | sizeof(struct btrfs_item), | |
3494 | &leaf->map_token, &leaf->kaddr, | |
3495 | &leaf->map_start, &leaf->map_len, | |
3496 | KM_USER1); | |
3497 | } | |
5f39d397 CM |
3498 | ioff = btrfs_item_offset(leaf, item); |
3499 | btrfs_set_item_offset(leaf, item, ioff - data_size); | |
6567e837 | 3500 | } |
5f39d397 | 3501 | |
db94535d CM |
3502 | if (leaf->map_token) { |
3503 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3504 | leaf->map_token = NULL; | |
3505 | } | |
3506 | ||
6567e837 | 3507 | /* shift the data */ |
5f39d397 | 3508 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + |
6567e837 CM |
3509 | data_end - data_size, btrfs_leaf_data(leaf) + |
3510 | data_end, old_data - data_end); | |
5f39d397 | 3511 | |
6567e837 | 3512 | data_end = old_data; |
5f39d397 CM |
3513 | old_size = btrfs_item_size_nr(leaf, slot); |
3514 | item = btrfs_item_nr(leaf, slot); | |
3515 | btrfs_set_item_size(leaf, item, old_size + data_size); | |
3516 | btrfs_mark_buffer_dirty(leaf); | |
6567e837 CM |
3517 | |
3518 | ret = 0; | |
5f39d397 CM |
3519 | if (btrfs_leaf_free_space(root, leaf) < 0) { |
3520 | btrfs_print_leaf(root, leaf); | |
6567e837 | 3521 | BUG(); |
5f39d397 | 3522 | } |
6567e837 CM |
3523 | return ret; |
3524 | } | |
3525 | ||
f3465ca4 JB |
3526 | /* |
3527 | * Given a key and some data, insert items into the tree. | |
3528 | * This does all the path init required, making room in the tree if needed. | |
3529 | * Returns the number of keys that were inserted. | |
3530 | */ | |
3531 | int btrfs_insert_some_items(struct btrfs_trans_handle *trans, | |
3532 | struct btrfs_root *root, | |
3533 | struct btrfs_path *path, | |
3534 | struct btrfs_key *cpu_key, u32 *data_size, | |
3535 | int nr) | |
3536 | { | |
3537 | struct extent_buffer *leaf; | |
3538 | struct btrfs_item *item; | |
3539 | int ret = 0; | |
3540 | int slot; | |
f3465ca4 JB |
3541 | int i; |
3542 | u32 nritems; | |
3543 | u32 total_data = 0; | |
3544 | u32 total_size = 0; | |
3545 | unsigned int data_end; | |
3546 | struct btrfs_disk_key disk_key; | |
3547 | struct btrfs_key found_key; | |
3548 | ||
87b29b20 YZ |
3549 | for (i = 0; i < nr; i++) { |
3550 | if (total_size + data_size[i] + sizeof(struct btrfs_item) > | |
3551 | BTRFS_LEAF_DATA_SIZE(root)) { | |
3552 | break; | |
3553 | nr = i; | |
3554 | } | |
f3465ca4 | 3555 | total_data += data_size[i]; |
87b29b20 YZ |
3556 | total_size += data_size[i] + sizeof(struct btrfs_item); |
3557 | } | |
3558 | BUG_ON(nr == 0); | |
f3465ca4 | 3559 | |
f3465ca4 JB |
3560 | ret = btrfs_search_slot(trans, root, cpu_key, path, total_size, 1); |
3561 | if (ret == 0) | |
3562 | return -EEXIST; | |
3563 | if (ret < 0) | |
3564 | goto out; | |
3565 | ||
f3465ca4 JB |
3566 | leaf = path->nodes[0]; |
3567 | ||
3568 | nritems = btrfs_header_nritems(leaf); | |
3569 | data_end = leaf_data_end(root, leaf); | |
3570 | ||
3571 | if (btrfs_leaf_free_space(root, leaf) < total_size) { | |
3572 | for (i = nr; i >= 0; i--) { | |
3573 | total_data -= data_size[i]; | |
3574 | total_size -= data_size[i] + sizeof(struct btrfs_item); | |
3575 | if (total_size < btrfs_leaf_free_space(root, leaf)) | |
3576 | break; | |
3577 | } | |
3578 | nr = i; | |
3579 | } | |
3580 | ||
3581 | slot = path->slots[0]; | |
3582 | BUG_ON(slot < 0); | |
3583 | ||
3584 | if (slot != nritems) { | |
3585 | unsigned int old_data = btrfs_item_end_nr(leaf, slot); | |
3586 | ||
3587 | item = btrfs_item_nr(leaf, slot); | |
3588 | btrfs_item_key_to_cpu(leaf, &found_key, slot); | |
3589 | ||
3590 | /* figure out how many keys we can insert in here */ | |
3591 | total_data = data_size[0]; | |
3592 | for (i = 1; i < nr; i++) { | |
5d4f98a2 | 3593 | if (btrfs_comp_cpu_keys(&found_key, cpu_key + i) <= 0) |
f3465ca4 JB |
3594 | break; |
3595 | total_data += data_size[i]; | |
3596 | } | |
3597 | nr = i; | |
3598 | ||
3599 | if (old_data < data_end) { | |
3600 | btrfs_print_leaf(root, leaf); | |
d397712b | 3601 | printk(KERN_CRIT "slot %d old_data %d data_end %d\n", |
f3465ca4 JB |
3602 | slot, old_data, data_end); |
3603 | BUG_ON(1); | |
3604 | } | |
3605 | /* | |
3606 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3607 | */ | |
3608 | /* first correct the data pointers */ | |
3609 | WARN_ON(leaf->map_token); | |
3610 | for (i = slot; i < nritems; i++) { | |
3611 | u32 ioff; | |
3612 | ||
3613 | item = btrfs_item_nr(leaf, i); | |
3614 | if (!leaf->map_token) { | |
3615 | map_extent_buffer(leaf, (unsigned long)item, | |
3616 | sizeof(struct btrfs_item), | |
3617 | &leaf->map_token, &leaf->kaddr, | |
3618 | &leaf->map_start, &leaf->map_len, | |
3619 | KM_USER1); | |
3620 | } | |
3621 | ||
3622 | ioff = btrfs_item_offset(leaf, item); | |
3623 | btrfs_set_item_offset(leaf, item, ioff - total_data); | |
3624 | } | |
3625 | if (leaf->map_token) { | |
3626 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3627 | leaf->map_token = NULL; | |
3628 | } | |
3629 | ||
3630 | /* shift the items */ | |
3631 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), | |
3632 | btrfs_item_nr_offset(slot), | |
3633 | (nritems - slot) * sizeof(struct btrfs_item)); | |
3634 | ||
3635 | /* shift the data */ | |
3636 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
3637 | data_end - total_data, btrfs_leaf_data(leaf) + | |
3638 | data_end, old_data - data_end); | |
3639 | data_end = old_data; | |
3640 | } else { | |
3641 | /* | |
3642 | * this sucks but it has to be done, if we are inserting at | |
3643 | * the end of the leaf only insert 1 of the items, since we | |
3644 | * have no way of knowing whats on the next leaf and we'd have | |
3645 | * to drop our current locks to figure it out | |
3646 | */ | |
3647 | nr = 1; | |
3648 | } | |
3649 | ||
3650 | /* setup the item for the new data */ | |
3651 | for (i = 0; i < nr; i++) { | |
3652 | btrfs_cpu_key_to_disk(&disk_key, cpu_key + i); | |
3653 | btrfs_set_item_key(leaf, &disk_key, slot + i); | |
3654 | item = btrfs_item_nr(leaf, slot + i); | |
3655 | btrfs_set_item_offset(leaf, item, data_end - data_size[i]); | |
3656 | data_end -= data_size[i]; | |
3657 | btrfs_set_item_size(leaf, item, data_size[i]); | |
3658 | } | |
3659 | btrfs_set_header_nritems(leaf, nritems + nr); | |
3660 | btrfs_mark_buffer_dirty(leaf); | |
3661 | ||
3662 | ret = 0; | |
3663 | if (slot == 0) { | |
3664 | btrfs_cpu_key_to_disk(&disk_key, cpu_key); | |
3665 | ret = fixup_low_keys(trans, root, path, &disk_key, 1); | |
3666 | } | |
3667 | ||
3668 | if (btrfs_leaf_free_space(root, leaf) < 0) { | |
3669 | btrfs_print_leaf(root, leaf); | |
3670 | BUG(); | |
3671 | } | |
3672 | out: | |
3673 | if (!ret) | |
3674 | ret = nr; | |
3675 | return ret; | |
3676 | } | |
3677 | ||
74123bd7 | 3678 | /* |
44871b1b CM |
3679 | * this is a helper for btrfs_insert_empty_items, the main goal here is |
3680 | * to save stack depth by doing the bulk of the work in a function | |
3681 | * that doesn't call btrfs_search_slot | |
74123bd7 | 3682 | */ |
44871b1b CM |
3683 | static noinline_for_stack int |
3684 | setup_items_for_insert(struct btrfs_trans_handle *trans, | |
3685 | struct btrfs_root *root, struct btrfs_path *path, | |
3686 | struct btrfs_key *cpu_key, u32 *data_size, | |
3687 | u32 total_data, u32 total_size, int nr) | |
be0e5c09 | 3688 | { |
5f39d397 | 3689 | struct btrfs_item *item; |
9c58309d | 3690 | int i; |
7518a238 | 3691 | u32 nritems; |
be0e5c09 | 3692 | unsigned int data_end; |
e2fa7227 | 3693 | struct btrfs_disk_key disk_key; |
44871b1b CM |
3694 | int ret; |
3695 | struct extent_buffer *leaf; | |
3696 | int slot; | |
e2fa7227 | 3697 | |
5f39d397 | 3698 | leaf = path->nodes[0]; |
44871b1b | 3699 | slot = path->slots[0]; |
74123bd7 | 3700 | |
5f39d397 | 3701 | nritems = btrfs_header_nritems(leaf); |
123abc88 | 3702 | data_end = leaf_data_end(root, leaf); |
eb60ceac | 3703 | |
f25956cc | 3704 | if (btrfs_leaf_free_space(root, leaf) < total_size) { |
3326d1b0 | 3705 | btrfs_print_leaf(root, leaf); |
d397712b | 3706 | printk(KERN_CRIT "not enough freespace need %u have %d\n", |
9c58309d | 3707 | total_size, btrfs_leaf_free_space(root, leaf)); |
be0e5c09 | 3708 | BUG(); |
d4dbff95 | 3709 | } |
5f39d397 | 3710 | |
be0e5c09 | 3711 | if (slot != nritems) { |
5f39d397 | 3712 | unsigned int old_data = btrfs_item_end_nr(leaf, slot); |
be0e5c09 | 3713 | |
5f39d397 CM |
3714 | if (old_data < data_end) { |
3715 | btrfs_print_leaf(root, leaf); | |
d397712b | 3716 | printk(KERN_CRIT "slot %d old_data %d data_end %d\n", |
5f39d397 CM |
3717 | slot, old_data, data_end); |
3718 | BUG_ON(1); | |
3719 | } | |
be0e5c09 CM |
3720 | /* |
3721 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3722 | */ | |
3723 | /* first correct the data pointers */ | |
db94535d | 3724 | WARN_ON(leaf->map_token); |
0783fcfc | 3725 | for (i = slot; i < nritems; i++) { |
5f39d397 | 3726 | u32 ioff; |
db94535d | 3727 | |
5f39d397 | 3728 | item = btrfs_item_nr(leaf, i); |
db94535d CM |
3729 | if (!leaf->map_token) { |
3730 | map_extent_buffer(leaf, (unsigned long)item, | |
3731 | sizeof(struct btrfs_item), | |
3732 | &leaf->map_token, &leaf->kaddr, | |
3733 | &leaf->map_start, &leaf->map_len, | |
3734 | KM_USER1); | |
3735 | } | |
3736 | ||
5f39d397 | 3737 | ioff = btrfs_item_offset(leaf, item); |
9c58309d | 3738 | btrfs_set_item_offset(leaf, item, ioff - total_data); |
0783fcfc | 3739 | } |
db94535d CM |
3740 | if (leaf->map_token) { |
3741 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3742 | leaf->map_token = NULL; | |
3743 | } | |
be0e5c09 CM |
3744 | |
3745 | /* shift the items */ | |
9c58309d | 3746 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), |
5f39d397 | 3747 | btrfs_item_nr_offset(slot), |
d6025579 | 3748 | (nritems - slot) * sizeof(struct btrfs_item)); |
be0e5c09 CM |
3749 | |
3750 | /* shift the data */ | |
5f39d397 | 3751 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + |
9c58309d | 3752 | data_end - total_data, btrfs_leaf_data(leaf) + |
d6025579 | 3753 | data_end, old_data - data_end); |
be0e5c09 CM |
3754 | data_end = old_data; |
3755 | } | |
5f39d397 | 3756 | |
62e2749e | 3757 | /* setup the item for the new data */ |
9c58309d CM |
3758 | for (i = 0; i < nr; i++) { |
3759 | btrfs_cpu_key_to_disk(&disk_key, cpu_key + i); | |
3760 | btrfs_set_item_key(leaf, &disk_key, slot + i); | |
3761 | item = btrfs_item_nr(leaf, slot + i); | |
3762 | btrfs_set_item_offset(leaf, item, data_end - data_size[i]); | |
3763 | data_end -= data_size[i]; | |
3764 | btrfs_set_item_size(leaf, item, data_size[i]); | |
3765 | } | |
44871b1b | 3766 | |
9c58309d | 3767 | btrfs_set_header_nritems(leaf, nritems + nr); |
aa5d6bed CM |
3768 | |
3769 | ret = 0; | |
5a01a2e3 | 3770 | if (slot == 0) { |
44871b1b | 3771 | struct btrfs_disk_key disk_key; |
5a01a2e3 | 3772 | btrfs_cpu_key_to_disk(&disk_key, cpu_key); |
e089f05c | 3773 | ret = fixup_low_keys(trans, root, path, &disk_key, 1); |
5a01a2e3 | 3774 | } |
b9473439 CM |
3775 | btrfs_unlock_up_safe(path, 1); |
3776 | btrfs_mark_buffer_dirty(leaf); | |
aa5d6bed | 3777 | |
5f39d397 CM |
3778 | if (btrfs_leaf_free_space(root, leaf) < 0) { |
3779 | btrfs_print_leaf(root, leaf); | |
be0e5c09 | 3780 | BUG(); |
5f39d397 | 3781 | } |
44871b1b CM |
3782 | return ret; |
3783 | } | |
3784 | ||
3785 | /* | |
3786 | * Given a key and some data, insert items into the tree. | |
3787 | * This does all the path init required, making room in the tree if needed. | |
3788 | */ | |
3789 | int btrfs_insert_empty_items(struct btrfs_trans_handle *trans, | |
3790 | struct btrfs_root *root, | |
3791 | struct btrfs_path *path, | |
3792 | struct btrfs_key *cpu_key, u32 *data_size, | |
3793 | int nr) | |
3794 | { | |
44871b1b CM |
3795 | int ret = 0; |
3796 | int slot; | |
3797 | int i; | |
3798 | u32 total_size = 0; | |
3799 | u32 total_data = 0; | |
3800 | ||
3801 | for (i = 0; i < nr; i++) | |
3802 | total_data += data_size[i]; | |
3803 | ||
3804 | total_size = total_data + (nr * sizeof(struct btrfs_item)); | |
3805 | ret = btrfs_search_slot(trans, root, cpu_key, path, total_size, 1); | |
3806 | if (ret == 0) | |
3807 | return -EEXIST; | |
3808 | if (ret < 0) | |
3809 | goto out; | |
3810 | ||
44871b1b CM |
3811 | slot = path->slots[0]; |
3812 | BUG_ON(slot < 0); | |
3813 | ||
3814 | ret = setup_items_for_insert(trans, root, path, cpu_key, data_size, | |
3815 | total_data, total_size, nr); | |
3816 | ||
ed2ff2cb | 3817 | out: |
62e2749e CM |
3818 | return ret; |
3819 | } | |
3820 | ||
3821 | /* | |
3822 | * Given a key and some data, insert an item into the tree. | |
3823 | * This does all the path init required, making room in the tree if needed. | |
3824 | */ | |
e089f05c CM |
3825 | int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root |
3826 | *root, struct btrfs_key *cpu_key, void *data, u32 | |
3827 | data_size) | |
62e2749e CM |
3828 | { |
3829 | int ret = 0; | |
2c90e5d6 | 3830 | struct btrfs_path *path; |
5f39d397 CM |
3831 | struct extent_buffer *leaf; |
3832 | unsigned long ptr; | |
62e2749e | 3833 | |
2c90e5d6 CM |
3834 | path = btrfs_alloc_path(); |
3835 | BUG_ON(!path); | |
2c90e5d6 | 3836 | ret = btrfs_insert_empty_item(trans, root, path, cpu_key, data_size); |
62e2749e | 3837 | if (!ret) { |
5f39d397 CM |
3838 | leaf = path->nodes[0]; |
3839 | ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
3840 | write_extent_buffer(leaf, data, ptr, data_size); | |
3841 | btrfs_mark_buffer_dirty(leaf); | |
62e2749e | 3842 | } |
2c90e5d6 | 3843 | btrfs_free_path(path); |
aa5d6bed | 3844 | return ret; |
be0e5c09 CM |
3845 | } |
3846 | ||
74123bd7 | 3847 | /* |
5de08d7d | 3848 | * delete the pointer from a given node. |
74123bd7 | 3849 | * |
d352ac68 CM |
3850 | * the tree should have been previously balanced so the deletion does not |
3851 | * empty a node. | |
74123bd7 | 3852 | */ |
e089f05c CM |
3853 | static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root, |
3854 | struct btrfs_path *path, int level, int slot) | |
be0e5c09 | 3855 | { |
5f39d397 | 3856 | struct extent_buffer *parent = path->nodes[level]; |
7518a238 | 3857 | u32 nritems; |
aa5d6bed | 3858 | int ret = 0; |
bb803951 | 3859 | int wret; |
be0e5c09 | 3860 | |
5f39d397 | 3861 | nritems = btrfs_header_nritems(parent); |
d397712b | 3862 | if (slot != nritems - 1) { |
5f39d397 CM |
3863 | memmove_extent_buffer(parent, |
3864 | btrfs_node_key_ptr_offset(slot), | |
3865 | btrfs_node_key_ptr_offset(slot + 1), | |
d6025579 CM |
3866 | sizeof(struct btrfs_key_ptr) * |
3867 | (nritems - slot - 1)); | |
bb803951 | 3868 | } |
7518a238 | 3869 | nritems--; |
5f39d397 | 3870 | btrfs_set_header_nritems(parent, nritems); |
7518a238 | 3871 | if (nritems == 0 && parent == root->node) { |
5f39d397 | 3872 | BUG_ON(btrfs_header_level(root->node) != 1); |
bb803951 | 3873 | /* just turn the root into a leaf and break */ |
5f39d397 | 3874 | btrfs_set_header_level(root->node, 0); |
bb803951 | 3875 | } else if (slot == 0) { |
5f39d397 CM |
3876 | struct btrfs_disk_key disk_key; |
3877 | ||
3878 | btrfs_node_key(parent, &disk_key, 0); | |
3879 | wret = fixup_low_keys(trans, root, path, &disk_key, level + 1); | |
0f70abe2 CM |
3880 | if (wret) |
3881 | ret = wret; | |
be0e5c09 | 3882 | } |
d6025579 | 3883 | btrfs_mark_buffer_dirty(parent); |
aa5d6bed | 3884 | return ret; |
be0e5c09 CM |
3885 | } |
3886 | ||
323ac95b CM |
3887 | /* |
3888 | * a helper function to delete the leaf pointed to by path->slots[1] and | |
5d4f98a2 | 3889 | * path->nodes[1]. |
323ac95b CM |
3890 | * |
3891 | * This deletes the pointer in path->nodes[1] and frees the leaf | |
3892 | * block extent. zero is returned if it all worked out, < 0 otherwise. | |
3893 | * | |
3894 | * The path must have already been setup for deleting the leaf, including | |
3895 | * all the proper balancing. path->nodes[1] must be locked. | |
3896 | */ | |
5d4f98a2 YZ |
3897 | static noinline int btrfs_del_leaf(struct btrfs_trans_handle *trans, |
3898 | struct btrfs_root *root, | |
3899 | struct btrfs_path *path, | |
3900 | struct extent_buffer *leaf) | |
323ac95b CM |
3901 | { |
3902 | int ret; | |
323ac95b | 3903 | |
5d4f98a2 | 3904 | WARN_ON(btrfs_header_generation(leaf) != trans->transid); |
323ac95b CM |
3905 | ret = del_ptr(trans, root, path, 1, path->slots[1]); |
3906 | if (ret) | |
3907 | return ret; | |
3908 | ||
4d081c41 CM |
3909 | /* |
3910 | * btrfs_free_extent is expensive, we want to make sure we | |
3911 | * aren't holding any locks when we call it | |
3912 | */ | |
3913 | btrfs_unlock_up_safe(path, 0); | |
3914 | ||
f0486c68 YZ |
3915 | root_sub_used(root, leaf->len); |
3916 | ||
3917 | btrfs_free_tree_block(trans, root, leaf, 0, 1); | |
3918 | return 0; | |
323ac95b | 3919 | } |
74123bd7 CM |
3920 | /* |
3921 | * delete the item at the leaf level in path. If that empties | |
3922 | * the leaf, remove it from the tree | |
3923 | */ | |
85e21bac CM |
3924 | int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root, |
3925 | struct btrfs_path *path, int slot, int nr) | |
be0e5c09 | 3926 | { |
5f39d397 CM |
3927 | struct extent_buffer *leaf; |
3928 | struct btrfs_item *item; | |
85e21bac CM |
3929 | int last_off; |
3930 | int dsize = 0; | |
aa5d6bed CM |
3931 | int ret = 0; |
3932 | int wret; | |
85e21bac | 3933 | int i; |
7518a238 | 3934 | u32 nritems; |
be0e5c09 | 3935 | |
5f39d397 | 3936 | leaf = path->nodes[0]; |
85e21bac CM |
3937 | last_off = btrfs_item_offset_nr(leaf, slot + nr - 1); |
3938 | ||
3939 | for (i = 0; i < nr; i++) | |
3940 | dsize += btrfs_item_size_nr(leaf, slot + i); | |
3941 | ||
5f39d397 | 3942 | nritems = btrfs_header_nritems(leaf); |
be0e5c09 | 3943 | |
85e21bac | 3944 | if (slot + nr != nritems) { |
123abc88 | 3945 | int data_end = leaf_data_end(root, leaf); |
5f39d397 CM |
3946 | |
3947 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
d6025579 CM |
3948 | data_end + dsize, |
3949 | btrfs_leaf_data(leaf) + data_end, | |
85e21bac | 3950 | last_off - data_end); |
5f39d397 | 3951 | |
85e21bac | 3952 | for (i = slot + nr; i < nritems; i++) { |
5f39d397 | 3953 | u32 ioff; |
db94535d | 3954 | |
5f39d397 | 3955 | item = btrfs_item_nr(leaf, i); |
db94535d CM |
3956 | if (!leaf->map_token) { |
3957 | map_extent_buffer(leaf, (unsigned long)item, | |
3958 | sizeof(struct btrfs_item), | |
3959 | &leaf->map_token, &leaf->kaddr, | |
3960 | &leaf->map_start, &leaf->map_len, | |
3961 | KM_USER1); | |
3962 | } | |
5f39d397 CM |
3963 | ioff = btrfs_item_offset(leaf, item); |
3964 | btrfs_set_item_offset(leaf, item, ioff + dsize); | |
0783fcfc | 3965 | } |
db94535d CM |
3966 | |
3967 | if (leaf->map_token) { | |
3968 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3969 | leaf->map_token = NULL; | |
3970 | } | |
3971 | ||
5f39d397 | 3972 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot), |
85e21bac | 3973 | btrfs_item_nr_offset(slot + nr), |
d6025579 | 3974 | sizeof(struct btrfs_item) * |
85e21bac | 3975 | (nritems - slot - nr)); |
be0e5c09 | 3976 | } |
85e21bac CM |
3977 | btrfs_set_header_nritems(leaf, nritems - nr); |
3978 | nritems -= nr; | |
5f39d397 | 3979 | |
74123bd7 | 3980 | /* delete the leaf if we've emptied it */ |
7518a238 | 3981 | if (nritems == 0) { |
5f39d397 CM |
3982 | if (leaf == root->node) { |
3983 | btrfs_set_header_level(leaf, 0); | |
9a8dd150 | 3984 | } else { |
f0486c68 YZ |
3985 | btrfs_set_path_blocking(path); |
3986 | clean_tree_block(trans, root, leaf); | |
5d4f98a2 | 3987 | ret = btrfs_del_leaf(trans, root, path, leaf); |
323ac95b | 3988 | BUG_ON(ret); |
9a8dd150 | 3989 | } |
be0e5c09 | 3990 | } else { |
7518a238 | 3991 | int used = leaf_space_used(leaf, 0, nritems); |
aa5d6bed | 3992 | if (slot == 0) { |
5f39d397 CM |
3993 | struct btrfs_disk_key disk_key; |
3994 | ||
3995 | btrfs_item_key(leaf, &disk_key, 0); | |
e089f05c | 3996 | wret = fixup_low_keys(trans, root, path, |
5f39d397 | 3997 | &disk_key, 1); |
aa5d6bed CM |
3998 | if (wret) |
3999 | ret = wret; | |
4000 | } | |
aa5d6bed | 4001 | |
74123bd7 | 4002 | /* delete the leaf if it is mostly empty */ |
d717aa1d | 4003 | if (used < BTRFS_LEAF_DATA_SIZE(root) / 3) { |
be0e5c09 CM |
4004 | /* push_leaf_left fixes the path. |
4005 | * make sure the path still points to our leaf | |
4006 | * for possible call to del_ptr below | |
4007 | */ | |
4920c9ac | 4008 | slot = path->slots[1]; |
5f39d397 CM |
4009 | extent_buffer_get(leaf); |
4010 | ||
b9473439 | 4011 | btrfs_set_path_blocking(path); |
99d8f83c CM |
4012 | wret = push_leaf_left(trans, root, path, 1, 1, |
4013 | 1, (u32)-1); | |
54aa1f4d | 4014 | if (wret < 0 && wret != -ENOSPC) |
aa5d6bed | 4015 | ret = wret; |
5f39d397 CM |
4016 | |
4017 | if (path->nodes[0] == leaf && | |
4018 | btrfs_header_nritems(leaf)) { | |
99d8f83c CM |
4019 | wret = push_leaf_right(trans, root, path, 1, |
4020 | 1, 1, 0); | |
54aa1f4d | 4021 | if (wret < 0 && wret != -ENOSPC) |
aa5d6bed CM |
4022 | ret = wret; |
4023 | } | |
5f39d397 CM |
4024 | |
4025 | if (btrfs_header_nritems(leaf) == 0) { | |
323ac95b | 4026 | path->slots[1] = slot; |
5d4f98a2 | 4027 | ret = btrfs_del_leaf(trans, root, path, leaf); |
323ac95b | 4028 | BUG_ON(ret); |
5f39d397 | 4029 | free_extent_buffer(leaf); |
5de08d7d | 4030 | } else { |
925baedd CM |
4031 | /* if we're still in the path, make sure |
4032 | * we're dirty. Otherwise, one of the | |
4033 | * push_leaf functions must have already | |
4034 | * dirtied this buffer | |
4035 | */ | |
4036 | if (path->nodes[0] == leaf) | |
4037 | btrfs_mark_buffer_dirty(leaf); | |
5f39d397 | 4038 | free_extent_buffer(leaf); |
be0e5c09 | 4039 | } |
d5719762 | 4040 | } else { |
5f39d397 | 4041 | btrfs_mark_buffer_dirty(leaf); |
be0e5c09 CM |
4042 | } |
4043 | } | |
aa5d6bed | 4044 | return ret; |
be0e5c09 CM |
4045 | } |
4046 | ||
7bb86316 | 4047 | /* |
925baedd | 4048 | * search the tree again to find a leaf with lesser keys |
7bb86316 CM |
4049 | * returns 0 if it found something or 1 if there are no lesser leaves. |
4050 | * returns < 0 on io errors. | |
d352ac68 CM |
4051 | * |
4052 | * This may release the path, and so you may lose any locks held at the | |
4053 | * time you call it. | |
7bb86316 CM |
4054 | */ |
4055 | int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path) | |
4056 | { | |
925baedd CM |
4057 | struct btrfs_key key; |
4058 | struct btrfs_disk_key found_key; | |
4059 | int ret; | |
7bb86316 | 4060 | |
925baedd | 4061 | btrfs_item_key_to_cpu(path->nodes[0], &key, 0); |
7bb86316 | 4062 | |
925baedd CM |
4063 | if (key.offset > 0) |
4064 | key.offset--; | |
4065 | else if (key.type > 0) | |
4066 | key.type--; | |
4067 | else if (key.objectid > 0) | |
4068 | key.objectid--; | |
4069 | else | |
4070 | return 1; | |
7bb86316 | 4071 | |
925baedd CM |
4072 | btrfs_release_path(root, path); |
4073 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
4074 | if (ret < 0) | |
4075 | return ret; | |
4076 | btrfs_item_key(path->nodes[0], &found_key, 0); | |
4077 | ret = comp_keys(&found_key, &key); | |
4078 | if (ret < 0) | |
4079 | return 0; | |
4080 | return 1; | |
7bb86316 CM |
4081 | } |
4082 | ||
3f157a2f CM |
4083 | /* |
4084 | * A helper function to walk down the tree starting at min_key, and looking | |
4085 | * for nodes or leaves that are either in cache or have a minimum | |
d352ac68 | 4086 | * transaction id. This is used by the btree defrag code, and tree logging |
3f157a2f CM |
4087 | * |
4088 | * This does not cow, but it does stuff the starting key it finds back | |
4089 | * into min_key, so you can call btrfs_search_slot with cow=1 on the | |
4090 | * key and get a writable path. | |
4091 | * | |
4092 | * This does lock as it descends, and path->keep_locks should be set | |
4093 | * to 1 by the caller. | |
4094 | * | |
4095 | * This honors path->lowest_level to prevent descent past a given level | |
4096 | * of the tree. | |
4097 | * | |
d352ac68 CM |
4098 | * min_trans indicates the oldest transaction that you are interested |
4099 | * in walking through. Any nodes or leaves older than min_trans are | |
4100 | * skipped over (without reading them). | |
4101 | * | |
3f157a2f CM |
4102 | * returns zero if something useful was found, < 0 on error and 1 if there |
4103 | * was nothing in the tree that matched the search criteria. | |
4104 | */ | |
4105 | int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key, | |
e02119d5 | 4106 | struct btrfs_key *max_key, |
3f157a2f CM |
4107 | struct btrfs_path *path, int cache_only, |
4108 | u64 min_trans) | |
4109 | { | |
4110 | struct extent_buffer *cur; | |
4111 | struct btrfs_key found_key; | |
4112 | int slot; | |
9652480b | 4113 | int sret; |
3f157a2f CM |
4114 | u32 nritems; |
4115 | int level; | |
4116 | int ret = 1; | |
4117 | ||
934d375b | 4118 | WARN_ON(!path->keep_locks); |
3f157a2f CM |
4119 | again: |
4120 | cur = btrfs_lock_root_node(root); | |
4121 | level = btrfs_header_level(cur); | |
e02119d5 | 4122 | WARN_ON(path->nodes[level]); |
3f157a2f CM |
4123 | path->nodes[level] = cur; |
4124 | path->locks[level] = 1; | |
4125 | ||
4126 | if (btrfs_header_generation(cur) < min_trans) { | |
4127 | ret = 1; | |
4128 | goto out; | |
4129 | } | |
d397712b | 4130 | while (1) { |
3f157a2f CM |
4131 | nritems = btrfs_header_nritems(cur); |
4132 | level = btrfs_header_level(cur); | |
9652480b | 4133 | sret = bin_search(cur, min_key, level, &slot); |
3f157a2f | 4134 | |
323ac95b CM |
4135 | /* at the lowest level, we're done, setup the path and exit */ |
4136 | if (level == path->lowest_level) { | |
e02119d5 CM |
4137 | if (slot >= nritems) |
4138 | goto find_next_key; | |
3f157a2f CM |
4139 | ret = 0; |
4140 | path->slots[level] = slot; | |
4141 | btrfs_item_key_to_cpu(cur, &found_key, slot); | |
4142 | goto out; | |
4143 | } | |
9652480b Y |
4144 | if (sret && slot > 0) |
4145 | slot--; | |
3f157a2f CM |
4146 | /* |
4147 | * check this node pointer against the cache_only and | |
4148 | * min_trans parameters. If it isn't in cache or is too | |
4149 | * old, skip to the next one. | |
4150 | */ | |
d397712b | 4151 | while (slot < nritems) { |
3f157a2f CM |
4152 | u64 blockptr; |
4153 | u64 gen; | |
4154 | struct extent_buffer *tmp; | |
e02119d5 CM |
4155 | struct btrfs_disk_key disk_key; |
4156 | ||
3f157a2f CM |
4157 | blockptr = btrfs_node_blockptr(cur, slot); |
4158 | gen = btrfs_node_ptr_generation(cur, slot); | |
4159 | if (gen < min_trans) { | |
4160 | slot++; | |
4161 | continue; | |
4162 | } | |
4163 | if (!cache_only) | |
4164 | break; | |
4165 | ||
e02119d5 CM |
4166 | if (max_key) { |
4167 | btrfs_node_key(cur, &disk_key, slot); | |
4168 | if (comp_keys(&disk_key, max_key) >= 0) { | |
4169 | ret = 1; | |
4170 | goto out; | |
4171 | } | |
4172 | } | |
4173 | ||
3f157a2f CM |
4174 | tmp = btrfs_find_tree_block(root, blockptr, |
4175 | btrfs_level_size(root, level - 1)); | |
4176 | ||
4177 | if (tmp && btrfs_buffer_uptodate(tmp, gen)) { | |
4178 | free_extent_buffer(tmp); | |
4179 | break; | |
4180 | } | |
4181 | if (tmp) | |
4182 | free_extent_buffer(tmp); | |
4183 | slot++; | |
4184 | } | |
e02119d5 | 4185 | find_next_key: |
3f157a2f CM |
4186 | /* |
4187 | * we didn't find a candidate key in this node, walk forward | |
4188 | * and find another one | |
4189 | */ | |
4190 | if (slot >= nritems) { | |
e02119d5 | 4191 | path->slots[level] = slot; |
b4ce94de | 4192 | btrfs_set_path_blocking(path); |
e02119d5 | 4193 | sret = btrfs_find_next_key(root, path, min_key, level, |
3f157a2f | 4194 | cache_only, min_trans); |
e02119d5 | 4195 | if (sret == 0) { |
3f157a2f CM |
4196 | btrfs_release_path(root, path); |
4197 | goto again; | |
4198 | } else { | |
4199 | goto out; | |
4200 | } | |
4201 | } | |
4202 | /* save our key for returning back */ | |
4203 | btrfs_node_key_to_cpu(cur, &found_key, slot); | |
4204 | path->slots[level] = slot; | |
4205 | if (level == path->lowest_level) { | |
4206 | ret = 0; | |
4207 | unlock_up(path, level, 1); | |
4208 | goto out; | |
4209 | } | |
b4ce94de | 4210 | btrfs_set_path_blocking(path); |
3f157a2f CM |
4211 | cur = read_node_slot(root, cur, slot); |
4212 | ||
4213 | btrfs_tree_lock(cur); | |
b4ce94de | 4214 | |
3f157a2f CM |
4215 | path->locks[level - 1] = 1; |
4216 | path->nodes[level - 1] = cur; | |
4217 | unlock_up(path, level, 1); | |
4008c04a | 4218 | btrfs_clear_path_blocking(path, NULL); |
3f157a2f CM |
4219 | } |
4220 | out: | |
4221 | if (ret == 0) | |
4222 | memcpy(min_key, &found_key, sizeof(found_key)); | |
b4ce94de | 4223 | btrfs_set_path_blocking(path); |
3f157a2f CM |
4224 | return ret; |
4225 | } | |
4226 | ||
4227 | /* | |
4228 | * this is similar to btrfs_next_leaf, but does not try to preserve | |
4229 | * and fixup the path. It looks for and returns the next key in the | |
4230 | * tree based on the current path and the cache_only and min_trans | |
4231 | * parameters. | |
4232 | * | |
4233 | * 0 is returned if another key is found, < 0 if there are any errors | |
4234 | * and 1 is returned if there are no higher keys in the tree | |
4235 | * | |
4236 | * path->keep_locks should be set to 1 on the search made before | |
4237 | * calling this function. | |
4238 | */ | |
e7a84565 | 4239 | int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path, |
33c66f43 | 4240 | struct btrfs_key *key, int level, |
3f157a2f | 4241 | int cache_only, u64 min_trans) |
e7a84565 | 4242 | { |
e7a84565 CM |
4243 | int slot; |
4244 | struct extent_buffer *c; | |
4245 | ||
934d375b | 4246 | WARN_ON(!path->keep_locks); |
d397712b | 4247 | while (level < BTRFS_MAX_LEVEL) { |
e7a84565 CM |
4248 | if (!path->nodes[level]) |
4249 | return 1; | |
4250 | ||
4251 | slot = path->slots[level] + 1; | |
4252 | c = path->nodes[level]; | |
3f157a2f | 4253 | next: |
e7a84565 | 4254 | if (slot >= btrfs_header_nritems(c)) { |
33c66f43 YZ |
4255 | int ret; |
4256 | int orig_lowest; | |
4257 | struct btrfs_key cur_key; | |
4258 | if (level + 1 >= BTRFS_MAX_LEVEL || | |
4259 | !path->nodes[level + 1]) | |
e7a84565 | 4260 | return 1; |
33c66f43 YZ |
4261 | |
4262 | if (path->locks[level + 1]) { | |
4263 | level++; | |
4264 | continue; | |
4265 | } | |
4266 | ||
4267 | slot = btrfs_header_nritems(c) - 1; | |
4268 | if (level == 0) | |
4269 | btrfs_item_key_to_cpu(c, &cur_key, slot); | |
4270 | else | |
4271 | btrfs_node_key_to_cpu(c, &cur_key, slot); | |
4272 | ||
4273 | orig_lowest = path->lowest_level; | |
4274 | btrfs_release_path(root, path); | |
4275 | path->lowest_level = level; | |
4276 | ret = btrfs_search_slot(NULL, root, &cur_key, path, | |
4277 | 0, 0); | |
4278 | path->lowest_level = orig_lowest; | |
4279 | if (ret < 0) | |
4280 | return ret; | |
4281 | ||
4282 | c = path->nodes[level]; | |
4283 | slot = path->slots[level]; | |
4284 | if (ret == 0) | |
4285 | slot++; | |
4286 | goto next; | |
e7a84565 | 4287 | } |
33c66f43 | 4288 | |
e7a84565 CM |
4289 | if (level == 0) |
4290 | btrfs_item_key_to_cpu(c, key, slot); | |
3f157a2f CM |
4291 | else { |
4292 | u64 blockptr = btrfs_node_blockptr(c, slot); | |
4293 | u64 gen = btrfs_node_ptr_generation(c, slot); | |
4294 | ||
4295 | if (cache_only) { | |
4296 | struct extent_buffer *cur; | |
4297 | cur = btrfs_find_tree_block(root, blockptr, | |
4298 | btrfs_level_size(root, level - 1)); | |
4299 | if (!cur || !btrfs_buffer_uptodate(cur, gen)) { | |
4300 | slot++; | |
4301 | if (cur) | |
4302 | free_extent_buffer(cur); | |
4303 | goto next; | |
4304 | } | |
4305 | free_extent_buffer(cur); | |
4306 | } | |
4307 | if (gen < min_trans) { | |
4308 | slot++; | |
4309 | goto next; | |
4310 | } | |
e7a84565 | 4311 | btrfs_node_key_to_cpu(c, key, slot); |
3f157a2f | 4312 | } |
e7a84565 CM |
4313 | return 0; |
4314 | } | |
4315 | return 1; | |
4316 | } | |
4317 | ||
97571fd0 | 4318 | /* |
925baedd | 4319 | * search the tree again to find a leaf with greater keys |
0f70abe2 CM |
4320 | * returns 0 if it found something or 1 if there are no greater leaves. |
4321 | * returns < 0 on io errors. | |
97571fd0 | 4322 | */ |
234b63a0 | 4323 | int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path) |
d97e63b6 CM |
4324 | { |
4325 | int slot; | |
8e73f275 | 4326 | int level; |
5f39d397 | 4327 | struct extent_buffer *c; |
8e73f275 | 4328 | struct extent_buffer *next; |
925baedd CM |
4329 | struct btrfs_key key; |
4330 | u32 nritems; | |
4331 | int ret; | |
8e73f275 CM |
4332 | int old_spinning = path->leave_spinning; |
4333 | int force_blocking = 0; | |
925baedd CM |
4334 | |
4335 | nritems = btrfs_header_nritems(path->nodes[0]); | |
d397712b | 4336 | if (nritems == 0) |
925baedd | 4337 | return 1; |
925baedd | 4338 | |
8e73f275 CM |
4339 | /* |
4340 | * we take the blocks in an order that upsets lockdep. Using | |
4341 | * blocking mode is the only way around it. | |
4342 | */ | |
4343 | #ifdef CONFIG_DEBUG_LOCK_ALLOC | |
4344 | force_blocking = 1; | |
4345 | #endif | |
925baedd | 4346 | |
8e73f275 CM |
4347 | btrfs_item_key_to_cpu(path->nodes[0], &key, nritems - 1); |
4348 | again: | |
4349 | level = 1; | |
4350 | next = NULL; | |
925baedd | 4351 | btrfs_release_path(root, path); |
8e73f275 | 4352 | |
a2135011 | 4353 | path->keep_locks = 1; |
8e73f275 CM |
4354 | |
4355 | if (!force_blocking) | |
4356 | path->leave_spinning = 1; | |
4357 | ||
925baedd CM |
4358 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); |
4359 | path->keep_locks = 0; | |
4360 | ||
4361 | if (ret < 0) | |
4362 | return ret; | |
4363 | ||
a2135011 | 4364 | nritems = btrfs_header_nritems(path->nodes[0]); |
168fd7d2 CM |
4365 | /* |
4366 | * by releasing the path above we dropped all our locks. A balance | |
4367 | * could have added more items next to the key that used to be | |
4368 | * at the very end of the block. So, check again here and | |
4369 | * advance the path if there are now more items available. | |
4370 | */ | |
a2135011 | 4371 | if (nritems > 0 && path->slots[0] < nritems - 1) { |
e457afec YZ |
4372 | if (ret == 0) |
4373 | path->slots[0]++; | |
8e73f275 | 4374 | ret = 0; |
925baedd CM |
4375 | goto done; |
4376 | } | |
d97e63b6 | 4377 | |
d397712b | 4378 | while (level < BTRFS_MAX_LEVEL) { |
8e73f275 CM |
4379 | if (!path->nodes[level]) { |
4380 | ret = 1; | |
4381 | goto done; | |
4382 | } | |
5f39d397 | 4383 | |
d97e63b6 CM |
4384 | slot = path->slots[level] + 1; |
4385 | c = path->nodes[level]; | |
5f39d397 | 4386 | if (slot >= btrfs_header_nritems(c)) { |
d97e63b6 | 4387 | level++; |
8e73f275 CM |
4388 | if (level == BTRFS_MAX_LEVEL) { |
4389 | ret = 1; | |
4390 | goto done; | |
4391 | } | |
d97e63b6 CM |
4392 | continue; |
4393 | } | |
5f39d397 | 4394 | |
925baedd CM |
4395 | if (next) { |
4396 | btrfs_tree_unlock(next); | |
5f39d397 | 4397 | free_extent_buffer(next); |
925baedd | 4398 | } |
5f39d397 | 4399 | |
8e73f275 CM |
4400 | next = c; |
4401 | ret = read_block_for_search(NULL, root, path, &next, level, | |
4402 | slot, &key); | |
4403 | if (ret == -EAGAIN) | |
4404 | goto again; | |
5f39d397 | 4405 | |
76a05b35 CM |
4406 | if (ret < 0) { |
4407 | btrfs_release_path(root, path); | |
4408 | goto done; | |
4409 | } | |
4410 | ||
5cd57b2c | 4411 | if (!path->skip_locking) { |
8e73f275 CM |
4412 | ret = btrfs_try_spin_lock(next); |
4413 | if (!ret) { | |
4414 | btrfs_set_path_blocking(path); | |
4415 | btrfs_tree_lock(next); | |
4416 | if (!force_blocking) | |
4417 | btrfs_clear_path_blocking(path, next); | |
4418 | } | |
4419 | if (force_blocking) | |
4420 | btrfs_set_lock_blocking(next); | |
5cd57b2c | 4421 | } |
d97e63b6 CM |
4422 | break; |
4423 | } | |
4424 | path->slots[level] = slot; | |
d397712b | 4425 | while (1) { |
d97e63b6 CM |
4426 | level--; |
4427 | c = path->nodes[level]; | |
925baedd CM |
4428 | if (path->locks[level]) |
4429 | btrfs_tree_unlock(c); | |
8e73f275 | 4430 | |
5f39d397 | 4431 | free_extent_buffer(c); |
d97e63b6 CM |
4432 | path->nodes[level] = next; |
4433 | path->slots[level] = 0; | |
a74a4b97 CM |
4434 | if (!path->skip_locking) |
4435 | path->locks[level] = 1; | |
8e73f275 | 4436 | |
d97e63b6 CM |
4437 | if (!level) |
4438 | break; | |
b4ce94de | 4439 | |
8e73f275 CM |
4440 | ret = read_block_for_search(NULL, root, path, &next, level, |
4441 | 0, &key); | |
4442 | if (ret == -EAGAIN) | |
4443 | goto again; | |
4444 | ||
76a05b35 CM |
4445 | if (ret < 0) { |
4446 | btrfs_release_path(root, path); | |
4447 | goto done; | |
4448 | } | |
4449 | ||
5cd57b2c | 4450 | if (!path->skip_locking) { |
b9447ef8 | 4451 | btrfs_assert_tree_locked(path->nodes[level]); |
8e73f275 CM |
4452 | ret = btrfs_try_spin_lock(next); |
4453 | if (!ret) { | |
4454 | btrfs_set_path_blocking(path); | |
4455 | btrfs_tree_lock(next); | |
4456 | if (!force_blocking) | |
4457 | btrfs_clear_path_blocking(path, next); | |
4458 | } | |
4459 | if (force_blocking) | |
4460 | btrfs_set_lock_blocking(next); | |
5cd57b2c | 4461 | } |
d97e63b6 | 4462 | } |
8e73f275 | 4463 | ret = 0; |
925baedd CM |
4464 | done: |
4465 | unlock_up(path, 0, 1); | |
8e73f275 CM |
4466 | path->leave_spinning = old_spinning; |
4467 | if (!old_spinning) | |
4468 | btrfs_set_path_blocking(path); | |
4469 | ||
4470 | return ret; | |
d97e63b6 | 4471 | } |
0b86a832 | 4472 | |
3f157a2f CM |
4473 | /* |
4474 | * this uses btrfs_prev_leaf to walk backwards in the tree, and keeps | |
4475 | * searching until it gets past min_objectid or finds an item of 'type' | |
4476 | * | |
4477 | * returns 0 if something is found, 1 if nothing was found and < 0 on error | |
4478 | */ | |
0b86a832 CM |
4479 | int btrfs_previous_item(struct btrfs_root *root, |
4480 | struct btrfs_path *path, u64 min_objectid, | |
4481 | int type) | |
4482 | { | |
4483 | struct btrfs_key found_key; | |
4484 | struct extent_buffer *leaf; | |
e02119d5 | 4485 | u32 nritems; |
0b86a832 CM |
4486 | int ret; |
4487 | ||
d397712b | 4488 | while (1) { |
0b86a832 | 4489 | if (path->slots[0] == 0) { |
b4ce94de | 4490 | btrfs_set_path_blocking(path); |
0b86a832 CM |
4491 | ret = btrfs_prev_leaf(root, path); |
4492 | if (ret != 0) | |
4493 | return ret; | |
4494 | } else { | |
4495 | path->slots[0]--; | |
4496 | } | |
4497 | leaf = path->nodes[0]; | |
e02119d5 CM |
4498 | nritems = btrfs_header_nritems(leaf); |
4499 | if (nritems == 0) | |
4500 | return 1; | |
4501 | if (path->slots[0] == nritems) | |
4502 | path->slots[0]--; | |
4503 | ||
0b86a832 | 4504 | btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); |
e02119d5 CM |
4505 | if (found_key.objectid < min_objectid) |
4506 | break; | |
0a4eefbb YZ |
4507 | if (found_key.type == type) |
4508 | return 0; | |
e02119d5 CM |
4509 | if (found_key.objectid == min_objectid && |
4510 | found_key.type < type) | |
4511 | break; | |
0b86a832 CM |
4512 | } |
4513 | return 1; | |
4514 | } |