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> |
eb60ceac CM |
20 | #include "ctree.h" |
21 | #include "disk-io.h" | |
7f5c1516 | 22 | #include "transaction.h" |
5f39d397 | 23 | #include "print-tree.h" |
925baedd | 24 | #include "locking.h" |
9a8dd150 | 25 | |
e089f05c CM |
26 | static int split_node(struct btrfs_trans_handle *trans, struct btrfs_root |
27 | *root, struct btrfs_path *path, int level); | |
28 | static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root | |
d4dbff95 | 29 | *root, struct btrfs_key *ins_key, |
cc0c5538 | 30 | struct btrfs_path *path, int data_size, int extend); |
5f39d397 CM |
31 | static int push_node_left(struct btrfs_trans_handle *trans, |
32 | struct btrfs_root *root, struct extent_buffer *dst, | |
971a1f66 | 33 | struct extent_buffer *src, int empty); |
5f39d397 CM |
34 | static int balance_node_right(struct btrfs_trans_handle *trans, |
35 | struct btrfs_root *root, | |
36 | struct extent_buffer *dst_buf, | |
37 | struct extent_buffer *src_buf); | |
e089f05c CM |
38 | static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root, |
39 | struct btrfs_path *path, int level, int slot); | |
d97e63b6 | 40 | |
df24a2b9 | 41 | struct btrfs_path *btrfs_alloc_path(void) |
2c90e5d6 | 42 | { |
df24a2b9 | 43 | struct btrfs_path *path; |
e00f7308 JM |
44 | path = kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS); |
45 | if (path) | |
2cc58cf2 | 46 | path->reada = 1; |
df24a2b9 | 47 | return path; |
2c90e5d6 CM |
48 | } |
49 | ||
b4ce94de CM |
50 | /* |
51 | * set all locked nodes in the path to blocking locks. This should | |
52 | * be done before scheduling | |
53 | */ | |
54 | noinline void btrfs_set_path_blocking(struct btrfs_path *p) | |
55 | { | |
56 | int i; | |
57 | for (i = 0; i < BTRFS_MAX_LEVEL; i++) { | |
58 | if (p->nodes[i] && p->locks[i]) | |
59 | btrfs_set_lock_blocking(p->nodes[i]); | |
60 | } | |
61 | } | |
62 | ||
63 | /* | |
64 | * reset all the locked nodes in the patch to spinning locks. | |
4008c04a CM |
65 | * |
66 | * held is used to keep lockdep happy, when lockdep is enabled | |
67 | * we set held to a blocking lock before we go around and | |
68 | * retake all the spinlocks in the path. You can safely use NULL | |
69 | * for held | |
b4ce94de | 70 | */ |
4008c04a CM |
71 | noinline void btrfs_clear_path_blocking(struct btrfs_path *p, |
72 | struct extent_buffer *held) | |
b4ce94de CM |
73 | { |
74 | int i; | |
4008c04a CM |
75 | |
76 | #ifdef CONFIG_DEBUG_LOCK_ALLOC | |
77 | /* lockdep really cares that we take all of these spinlocks | |
78 | * in the right order. If any of the locks in the path are not | |
79 | * currently blocking, it is going to complain. So, make really | |
80 | * really sure by forcing the path to blocking before we clear | |
81 | * the path blocking. | |
82 | */ | |
83 | if (held) | |
84 | btrfs_set_lock_blocking(held); | |
85 | btrfs_set_path_blocking(p); | |
86 | #endif | |
87 | ||
88 | for (i = BTRFS_MAX_LEVEL - 1; i >= 0; i--) { | |
b4ce94de CM |
89 | if (p->nodes[i] && p->locks[i]) |
90 | btrfs_clear_lock_blocking(p->nodes[i]); | |
91 | } | |
4008c04a CM |
92 | |
93 | #ifdef CONFIG_DEBUG_LOCK_ALLOC | |
94 | if (held) | |
95 | btrfs_clear_lock_blocking(held); | |
96 | #endif | |
b4ce94de CM |
97 | } |
98 | ||
d352ac68 | 99 | /* this also releases the path */ |
df24a2b9 | 100 | void btrfs_free_path(struct btrfs_path *p) |
be0e5c09 | 101 | { |
df24a2b9 CM |
102 | btrfs_release_path(NULL, p); |
103 | kmem_cache_free(btrfs_path_cachep, p); | |
be0e5c09 CM |
104 | } |
105 | ||
d352ac68 CM |
106 | /* |
107 | * path release drops references on the extent buffers in the path | |
108 | * and it drops any locks held by this path | |
109 | * | |
110 | * It is safe to call this on paths that no locks or extent buffers held. | |
111 | */ | |
d397712b | 112 | noinline void btrfs_release_path(struct btrfs_root *root, struct btrfs_path *p) |
eb60ceac CM |
113 | { |
114 | int i; | |
a2135011 | 115 | |
234b63a0 | 116 | for (i = 0; i < BTRFS_MAX_LEVEL; i++) { |
3f157a2f | 117 | p->slots[i] = 0; |
eb60ceac | 118 | if (!p->nodes[i]) |
925baedd CM |
119 | continue; |
120 | if (p->locks[i]) { | |
121 | btrfs_tree_unlock(p->nodes[i]); | |
122 | p->locks[i] = 0; | |
123 | } | |
5f39d397 | 124 | free_extent_buffer(p->nodes[i]); |
3f157a2f | 125 | p->nodes[i] = NULL; |
eb60ceac CM |
126 | } |
127 | } | |
128 | ||
d352ac68 CM |
129 | /* |
130 | * safely gets a reference on the root node of a tree. A lock | |
131 | * is not taken, so a concurrent writer may put a different node | |
132 | * at the root of the tree. See btrfs_lock_root_node for the | |
133 | * looping required. | |
134 | * | |
135 | * The extent buffer returned by this has a reference taken, so | |
136 | * it won't disappear. It may stop being the root of the tree | |
137 | * at any time because there are no locks held. | |
138 | */ | |
925baedd CM |
139 | struct extent_buffer *btrfs_root_node(struct btrfs_root *root) |
140 | { | |
141 | struct extent_buffer *eb; | |
142 | spin_lock(&root->node_lock); | |
143 | eb = root->node; | |
144 | extent_buffer_get(eb); | |
145 | spin_unlock(&root->node_lock); | |
146 | return eb; | |
147 | } | |
148 | ||
d352ac68 CM |
149 | /* loop around taking references on and locking the root node of the |
150 | * tree until you end up with a lock on the root. A locked buffer | |
151 | * is returned, with a reference held. | |
152 | */ | |
925baedd CM |
153 | struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root) |
154 | { | |
155 | struct extent_buffer *eb; | |
156 | ||
d397712b | 157 | while (1) { |
925baedd CM |
158 | eb = btrfs_root_node(root); |
159 | btrfs_tree_lock(eb); | |
160 | ||
161 | spin_lock(&root->node_lock); | |
162 | if (eb == root->node) { | |
163 | spin_unlock(&root->node_lock); | |
164 | break; | |
165 | } | |
166 | spin_unlock(&root->node_lock); | |
167 | ||
168 | btrfs_tree_unlock(eb); | |
169 | free_extent_buffer(eb); | |
170 | } | |
171 | return eb; | |
172 | } | |
173 | ||
d352ac68 CM |
174 | /* cowonly root (everything not a reference counted cow subvolume), just get |
175 | * put onto a simple dirty list. transaction.c walks this to make sure they | |
176 | * get properly updated on disk. | |
177 | */ | |
0b86a832 CM |
178 | static void add_root_to_dirty_list(struct btrfs_root *root) |
179 | { | |
180 | if (root->track_dirty && list_empty(&root->dirty_list)) { | |
181 | list_add(&root->dirty_list, | |
182 | &root->fs_info->dirty_cowonly_roots); | |
183 | } | |
184 | } | |
185 | ||
d352ac68 CM |
186 | /* |
187 | * used by snapshot creation to make a copy of a root for a tree with | |
188 | * a given objectid. The buffer with the new root node is returned in | |
189 | * cow_ret, and this func returns zero on success or a negative error code. | |
190 | */ | |
be20aa9d CM |
191 | int btrfs_copy_root(struct btrfs_trans_handle *trans, |
192 | struct btrfs_root *root, | |
193 | struct extent_buffer *buf, | |
194 | struct extent_buffer **cow_ret, u64 new_root_objectid) | |
195 | { | |
196 | struct extent_buffer *cow; | |
197 | u32 nritems; | |
198 | int ret = 0; | |
199 | int level; | |
4aec2b52 | 200 | struct btrfs_root *new_root; |
be20aa9d | 201 | |
4aec2b52 CM |
202 | new_root = kmalloc(sizeof(*new_root), GFP_NOFS); |
203 | if (!new_root) | |
204 | return -ENOMEM; | |
205 | ||
206 | memcpy(new_root, root, sizeof(*new_root)); | |
207 | new_root->root_key.objectid = new_root_objectid; | |
be20aa9d CM |
208 | |
209 | WARN_ON(root->ref_cows && trans->transid != | |
210 | root->fs_info->running_transaction->transid); | |
211 | WARN_ON(root->ref_cows && trans->transid != root->last_trans); | |
212 | ||
213 | level = btrfs_header_level(buf); | |
214 | nritems = btrfs_header_nritems(buf); | |
31840ae1 ZY |
215 | |
216 | cow = btrfs_alloc_free_block(trans, new_root, buf->len, 0, | |
217 | new_root_objectid, trans->transid, | |
218 | level, buf->start, 0); | |
4aec2b52 CM |
219 | if (IS_ERR(cow)) { |
220 | kfree(new_root); | |
be20aa9d | 221 | return PTR_ERR(cow); |
4aec2b52 | 222 | } |
be20aa9d CM |
223 | |
224 | copy_extent_buffer(cow, buf, 0, 0, cow->len); | |
225 | btrfs_set_header_bytenr(cow, cow->start); | |
226 | btrfs_set_header_generation(cow, trans->transid); | |
227 | btrfs_set_header_owner(cow, new_root_objectid); | |
63b10fc4 | 228 | btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN); |
be20aa9d | 229 | |
2b82032c YZ |
230 | write_extent_buffer(cow, root->fs_info->fsid, |
231 | (unsigned long)btrfs_header_fsid(cow), | |
232 | BTRFS_FSID_SIZE); | |
233 | ||
be20aa9d | 234 | WARN_ON(btrfs_header_generation(buf) > trans->transid); |
31840ae1 | 235 | ret = btrfs_inc_ref(trans, new_root, buf, cow, NULL); |
4aec2b52 CM |
236 | kfree(new_root); |
237 | ||
be20aa9d CM |
238 | if (ret) |
239 | return ret; | |
240 | ||
241 | btrfs_mark_buffer_dirty(cow); | |
242 | *cow_ret = cow; | |
243 | return 0; | |
244 | } | |
245 | ||
d352ac68 | 246 | /* |
d397712b CM |
247 | * does the dirty work in cow of a single block. The parent block (if |
248 | * supplied) is updated to point to the new cow copy. The new buffer is marked | |
249 | * dirty and returned locked. If you modify the block it needs to be marked | |
250 | * dirty again. | |
d352ac68 CM |
251 | * |
252 | * search_start -- an allocation hint for the new block | |
253 | * | |
d397712b CM |
254 | * empty_size -- a hint that you plan on doing more cow. This is the size in |
255 | * bytes the allocator should try to find free next to the block it returns. | |
256 | * This is just a hint and may be ignored by the allocator. | |
d352ac68 CM |
257 | * |
258 | * prealloc_dest -- if you have already reserved a destination for the cow, | |
d397712b CM |
259 | * this uses that block instead of allocating a new one. |
260 | * btrfs_alloc_reserved_extent is used to finish the allocation. | |
d352ac68 | 261 | */ |
d397712b | 262 | static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans, |
5f39d397 CM |
263 | struct btrfs_root *root, |
264 | struct extent_buffer *buf, | |
265 | struct extent_buffer *parent, int parent_slot, | |
266 | struct extent_buffer **cow_ret, | |
65b51a00 CM |
267 | u64 search_start, u64 empty_size, |
268 | u64 prealloc_dest) | |
02217ed2 | 269 | { |
31840ae1 | 270 | u64 parent_start; |
5f39d397 | 271 | struct extent_buffer *cow; |
7bb86316 | 272 | u32 nritems; |
6702ed49 | 273 | int ret = 0; |
7bb86316 | 274 | int level; |
925baedd | 275 | int unlock_orig = 0; |
7bb86316 | 276 | |
925baedd CM |
277 | if (*cow_ret == buf) |
278 | unlock_orig = 1; | |
279 | ||
280 | WARN_ON(!btrfs_tree_locked(buf)); | |
281 | ||
31840ae1 ZY |
282 | if (parent) |
283 | parent_start = parent->start; | |
284 | else | |
285 | parent_start = 0; | |
286 | ||
7bb86316 CM |
287 | WARN_ON(root->ref_cows && trans->transid != |
288 | root->fs_info->running_transaction->transid); | |
6702ed49 | 289 | WARN_ON(root->ref_cows && trans->transid != root->last_trans); |
5f39d397 | 290 | |
7bb86316 CM |
291 | level = btrfs_header_level(buf); |
292 | nritems = btrfs_header_nritems(buf); | |
31840ae1 | 293 | |
65b51a00 CM |
294 | if (prealloc_dest) { |
295 | struct btrfs_key ins; | |
296 | ||
297 | ins.objectid = prealloc_dest; | |
298 | ins.offset = buf->len; | |
299 | ins.type = BTRFS_EXTENT_ITEM_KEY; | |
300 | ||
31840ae1 | 301 | ret = btrfs_alloc_reserved_extent(trans, root, parent_start, |
65b51a00 | 302 | root->root_key.objectid, |
3bb1a1bc | 303 | trans->transid, level, &ins); |
65b51a00 CM |
304 | BUG_ON(ret); |
305 | cow = btrfs_init_new_buffer(trans, root, prealloc_dest, | |
4008c04a | 306 | buf->len, level); |
65b51a00 CM |
307 | } else { |
308 | cow = btrfs_alloc_free_block(trans, root, buf->len, | |
31840ae1 | 309 | parent_start, |
65b51a00 | 310 | root->root_key.objectid, |
31840ae1 ZY |
311 | trans->transid, level, |
312 | search_start, empty_size); | |
65b51a00 | 313 | } |
54aa1f4d CM |
314 | if (IS_ERR(cow)) |
315 | return PTR_ERR(cow); | |
6702ed49 | 316 | |
b4ce94de CM |
317 | /* cow is set to blocking by btrfs_init_new_buffer */ |
318 | ||
5f39d397 | 319 | copy_extent_buffer(cow, buf, 0, 0, cow->len); |
db94535d | 320 | btrfs_set_header_bytenr(cow, cow->start); |
5f39d397 CM |
321 | btrfs_set_header_generation(cow, trans->transid); |
322 | btrfs_set_header_owner(cow, root->root_key.objectid); | |
63b10fc4 | 323 | btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN); |
6702ed49 | 324 | |
2b82032c YZ |
325 | write_extent_buffer(cow, root->fs_info->fsid, |
326 | (unsigned long)btrfs_header_fsid(cow), | |
327 | BTRFS_FSID_SIZE); | |
328 | ||
5f39d397 CM |
329 | WARN_ON(btrfs_header_generation(buf) > trans->transid); |
330 | if (btrfs_header_generation(buf) != trans->transid) { | |
31840ae1 | 331 | u32 nr_extents; |
31840ae1 | 332 | ret = btrfs_inc_ref(trans, root, buf, cow, &nr_extents); |
6702ed49 CM |
333 | if (ret) |
334 | return ret; | |
31840ae1 ZY |
335 | |
336 | ret = btrfs_cache_ref(trans, root, buf, nr_extents); | |
337 | WARN_ON(ret); | |
1a40e23b ZY |
338 | } else if (btrfs_header_owner(buf) == BTRFS_TREE_RELOC_OBJECTID) { |
339 | /* | |
340 | * There are only two places that can drop reference to | |
341 | * tree blocks owned by living reloc trees, one is here, | |
f82d02d9 | 342 | * the other place is btrfs_drop_subtree. In both places, |
1a40e23b ZY |
343 | * we check reference count while tree block is locked. |
344 | * Furthermore, if reference count is one, it won't get | |
345 | * increased by someone else. | |
346 | */ | |
347 | u32 refs; | |
348 | ret = btrfs_lookup_extent_ref(trans, root, buf->start, | |
349 | buf->len, &refs); | |
350 | BUG_ON(ret); | |
351 | if (refs == 1) { | |
352 | ret = btrfs_update_ref(trans, root, buf, cow, | |
353 | 0, nritems); | |
354 | clean_tree_block(trans, root, buf); | |
355 | } else { | |
356 | ret = btrfs_inc_ref(trans, root, buf, cow, NULL); | |
357 | } | |
358 | BUG_ON(ret); | |
6702ed49 | 359 | } else { |
31840ae1 ZY |
360 | ret = btrfs_update_ref(trans, root, buf, cow, 0, nritems); |
361 | if (ret) | |
362 | return ret; | |
6702ed49 CM |
363 | clean_tree_block(trans, root, buf); |
364 | } | |
365 | ||
1a40e23b | 366 | if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) { |
1a40e23b ZY |
367 | ret = btrfs_reloc_tree_cache_ref(trans, root, cow, buf->start); |
368 | WARN_ON(ret); | |
369 | } | |
370 | ||
02217ed2 | 371 | if (buf == root->node) { |
925baedd | 372 | WARN_ON(parent && parent != buf); |
925baedd CM |
373 | |
374 | spin_lock(&root->node_lock); | |
02217ed2 | 375 | root->node = cow; |
5f39d397 | 376 | extent_buffer_get(cow); |
925baedd CM |
377 | spin_unlock(&root->node_lock); |
378 | ||
2c90e5d6 | 379 | if (buf != root->commit_root) { |
db94535d | 380 | btrfs_free_extent(trans, root, buf->start, |
31840ae1 ZY |
381 | buf->len, buf->start, |
382 | root->root_key.objectid, | |
383 | btrfs_header_generation(buf), | |
3bb1a1bc | 384 | level, 1); |
2c90e5d6 | 385 | } |
5f39d397 | 386 | free_extent_buffer(buf); |
0b86a832 | 387 | add_root_to_dirty_list(root); |
02217ed2 | 388 | } else { |
5f39d397 | 389 | btrfs_set_node_blockptr(parent, parent_slot, |
db94535d | 390 | cow->start); |
74493f7a CM |
391 | WARN_ON(trans->transid == 0); |
392 | btrfs_set_node_ptr_generation(parent, parent_slot, | |
393 | trans->transid); | |
d6025579 | 394 | btrfs_mark_buffer_dirty(parent); |
5f39d397 | 395 | WARN_ON(btrfs_header_generation(parent) != trans->transid); |
7bb86316 | 396 | btrfs_free_extent(trans, root, buf->start, buf->len, |
31840ae1 | 397 | parent_start, btrfs_header_owner(parent), |
3bb1a1bc | 398 | btrfs_header_generation(parent), level, 1); |
02217ed2 | 399 | } |
925baedd CM |
400 | if (unlock_orig) |
401 | btrfs_tree_unlock(buf); | |
5f39d397 | 402 | free_extent_buffer(buf); |
ccd467d6 | 403 | btrfs_mark_buffer_dirty(cow); |
2c90e5d6 | 404 | *cow_ret = cow; |
02217ed2 CM |
405 | return 0; |
406 | } | |
407 | ||
d352ac68 CM |
408 | /* |
409 | * cows a single block, see __btrfs_cow_block for the real work. | |
410 | * This version of it has extra checks so that a block isn't cow'd more than | |
411 | * once per transaction, as long as it hasn't been written yet | |
412 | */ | |
d397712b | 413 | noinline int btrfs_cow_block(struct btrfs_trans_handle *trans, |
5f39d397 CM |
414 | struct btrfs_root *root, struct extent_buffer *buf, |
415 | struct extent_buffer *parent, int parent_slot, | |
65b51a00 | 416 | struct extent_buffer **cow_ret, u64 prealloc_dest) |
6702ed49 CM |
417 | { |
418 | u64 search_start; | |
f510cfec | 419 | int ret; |
dc17ff8f | 420 | |
6702ed49 | 421 | if (trans->transaction != root->fs_info->running_transaction) { |
d397712b CM |
422 | printk(KERN_CRIT "trans %llu running %llu\n", |
423 | (unsigned long long)trans->transid, | |
424 | (unsigned long long) | |
6702ed49 CM |
425 | root->fs_info->running_transaction->transid); |
426 | WARN_ON(1); | |
427 | } | |
428 | if (trans->transid != root->fs_info->generation) { | |
d397712b CM |
429 | printk(KERN_CRIT "trans %llu running %llu\n", |
430 | (unsigned long long)trans->transid, | |
431 | (unsigned long long)root->fs_info->generation); | |
6702ed49 CM |
432 | WARN_ON(1); |
433 | } | |
dc17ff8f | 434 | |
5b21f2ed ZY |
435 | if (btrfs_header_generation(buf) == trans->transid && |
436 | btrfs_header_owner(buf) == root->root_key.objectid && | |
63b10fc4 | 437 | !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) { |
6702ed49 | 438 | *cow_ret = buf; |
65b51a00 | 439 | WARN_ON(prealloc_dest); |
6702ed49 CM |
440 | return 0; |
441 | } | |
c487685d | 442 | |
0b86a832 | 443 | search_start = buf->start & ~((u64)(1024 * 1024 * 1024) - 1); |
b4ce94de CM |
444 | |
445 | if (parent) | |
446 | btrfs_set_lock_blocking(parent); | |
447 | btrfs_set_lock_blocking(buf); | |
448 | ||
f510cfec | 449 | ret = __btrfs_cow_block(trans, root, buf, parent, |
65b51a00 CM |
450 | parent_slot, cow_ret, search_start, 0, |
451 | prealloc_dest); | |
f510cfec | 452 | return ret; |
6702ed49 CM |
453 | } |
454 | ||
d352ac68 CM |
455 | /* |
456 | * helper function for defrag to decide if two blocks pointed to by a | |
457 | * node are actually close by | |
458 | */ | |
6b80053d | 459 | static int close_blocks(u64 blocknr, u64 other, u32 blocksize) |
6702ed49 | 460 | { |
6b80053d | 461 | if (blocknr < other && other - (blocknr + blocksize) < 32768) |
6702ed49 | 462 | return 1; |
6b80053d | 463 | if (blocknr > other && blocknr - (other + blocksize) < 32768) |
6702ed49 CM |
464 | return 1; |
465 | return 0; | |
466 | } | |
467 | ||
081e9573 CM |
468 | /* |
469 | * compare two keys in a memcmp fashion | |
470 | */ | |
471 | static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2) | |
472 | { | |
473 | struct btrfs_key k1; | |
474 | ||
475 | btrfs_disk_key_to_cpu(&k1, disk); | |
476 | ||
477 | if (k1.objectid > k2->objectid) | |
478 | return 1; | |
479 | if (k1.objectid < k2->objectid) | |
480 | return -1; | |
481 | if (k1.type > k2->type) | |
482 | return 1; | |
483 | if (k1.type < k2->type) | |
484 | return -1; | |
485 | if (k1.offset > k2->offset) | |
486 | return 1; | |
487 | if (k1.offset < k2->offset) | |
488 | return -1; | |
489 | return 0; | |
490 | } | |
491 | ||
f3465ca4 JB |
492 | /* |
493 | * same as comp_keys only with two btrfs_key's | |
494 | */ | |
495 | static int comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2) | |
496 | { | |
497 | if (k1->objectid > k2->objectid) | |
498 | return 1; | |
499 | if (k1->objectid < k2->objectid) | |
500 | return -1; | |
501 | if (k1->type > k2->type) | |
502 | return 1; | |
503 | if (k1->type < k2->type) | |
504 | return -1; | |
505 | if (k1->offset > k2->offset) | |
506 | return 1; | |
507 | if (k1->offset < k2->offset) | |
508 | return -1; | |
509 | return 0; | |
510 | } | |
081e9573 | 511 | |
d352ac68 CM |
512 | /* |
513 | * this is used by the defrag code to go through all the | |
514 | * leaves pointed to by a node and reallocate them so that | |
515 | * disk order is close to key order | |
516 | */ | |
6702ed49 | 517 | int btrfs_realloc_node(struct btrfs_trans_handle *trans, |
5f39d397 | 518 | struct btrfs_root *root, struct extent_buffer *parent, |
a6b6e75e CM |
519 | int start_slot, int cache_only, u64 *last_ret, |
520 | struct btrfs_key *progress) | |
6702ed49 | 521 | { |
6b80053d | 522 | struct extent_buffer *cur; |
6702ed49 | 523 | u64 blocknr; |
ca7a79ad | 524 | u64 gen; |
e9d0b13b CM |
525 | u64 search_start = *last_ret; |
526 | u64 last_block = 0; | |
6702ed49 CM |
527 | u64 other; |
528 | u32 parent_nritems; | |
6702ed49 CM |
529 | int end_slot; |
530 | int i; | |
531 | int err = 0; | |
f2183bde | 532 | int parent_level; |
6b80053d CM |
533 | int uptodate; |
534 | u32 blocksize; | |
081e9573 CM |
535 | int progress_passed = 0; |
536 | struct btrfs_disk_key disk_key; | |
6702ed49 | 537 | |
5708b959 CM |
538 | parent_level = btrfs_header_level(parent); |
539 | if (cache_only && parent_level != 1) | |
540 | return 0; | |
541 | ||
d397712b | 542 | if (trans->transaction != root->fs_info->running_transaction) |
6702ed49 | 543 | WARN_ON(1); |
d397712b | 544 | if (trans->transid != root->fs_info->generation) |
6702ed49 | 545 | WARN_ON(1); |
86479a04 | 546 | |
6b80053d | 547 | parent_nritems = btrfs_header_nritems(parent); |
6b80053d | 548 | blocksize = btrfs_level_size(root, parent_level - 1); |
6702ed49 CM |
549 | end_slot = parent_nritems; |
550 | ||
551 | if (parent_nritems == 1) | |
552 | return 0; | |
553 | ||
b4ce94de CM |
554 | btrfs_set_lock_blocking(parent); |
555 | ||
6702ed49 CM |
556 | for (i = start_slot; i < end_slot; i++) { |
557 | int close = 1; | |
a6b6e75e | 558 | |
5708b959 CM |
559 | if (!parent->map_token) { |
560 | map_extent_buffer(parent, | |
561 | btrfs_node_key_ptr_offset(i), | |
562 | sizeof(struct btrfs_key_ptr), | |
563 | &parent->map_token, &parent->kaddr, | |
564 | &parent->map_start, &parent->map_len, | |
565 | KM_USER1); | |
566 | } | |
081e9573 CM |
567 | btrfs_node_key(parent, &disk_key, i); |
568 | if (!progress_passed && comp_keys(&disk_key, progress) < 0) | |
569 | continue; | |
570 | ||
571 | progress_passed = 1; | |
6b80053d | 572 | blocknr = btrfs_node_blockptr(parent, i); |
ca7a79ad | 573 | gen = btrfs_node_ptr_generation(parent, i); |
e9d0b13b CM |
574 | if (last_block == 0) |
575 | last_block = blocknr; | |
5708b959 | 576 | |
6702ed49 | 577 | if (i > 0) { |
6b80053d CM |
578 | other = btrfs_node_blockptr(parent, i - 1); |
579 | close = close_blocks(blocknr, other, blocksize); | |
6702ed49 | 580 | } |
0ef3e66b | 581 | if (!close && i < end_slot - 2) { |
6b80053d CM |
582 | other = btrfs_node_blockptr(parent, i + 1); |
583 | close = close_blocks(blocknr, other, blocksize); | |
6702ed49 | 584 | } |
e9d0b13b CM |
585 | if (close) { |
586 | last_block = blocknr; | |
6702ed49 | 587 | continue; |
e9d0b13b | 588 | } |
5708b959 CM |
589 | if (parent->map_token) { |
590 | unmap_extent_buffer(parent, parent->map_token, | |
591 | KM_USER1); | |
592 | parent->map_token = NULL; | |
593 | } | |
6702ed49 | 594 | |
6b80053d CM |
595 | cur = btrfs_find_tree_block(root, blocknr, blocksize); |
596 | if (cur) | |
1259ab75 | 597 | uptodate = btrfs_buffer_uptodate(cur, gen); |
6b80053d CM |
598 | else |
599 | uptodate = 0; | |
5708b959 | 600 | if (!cur || !uptodate) { |
6702ed49 | 601 | if (cache_only) { |
6b80053d | 602 | free_extent_buffer(cur); |
6702ed49 CM |
603 | continue; |
604 | } | |
6b80053d CM |
605 | if (!cur) { |
606 | cur = read_tree_block(root, blocknr, | |
ca7a79ad | 607 | blocksize, gen); |
6b80053d | 608 | } else if (!uptodate) { |
ca7a79ad | 609 | btrfs_read_buffer(cur, gen); |
f2183bde | 610 | } |
6702ed49 | 611 | } |
e9d0b13b | 612 | if (search_start == 0) |
6b80053d | 613 | search_start = last_block; |
e9d0b13b | 614 | |
e7a84565 | 615 | btrfs_tree_lock(cur); |
b4ce94de | 616 | btrfs_set_lock_blocking(cur); |
6b80053d | 617 | err = __btrfs_cow_block(trans, root, cur, parent, i, |
e7a84565 | 618 | &cur, search_start, |
6b80053d | 619 | min(16 * blocksize, |
65b51a00 | 620 | (end_slot - i) * blocksize), 0); |
252c38f0 | 621 | if (err) { |
e7a84565 | 622 | btrfs_tree_unlock(cur); |
6b80053d | 623 | free_extent_buffer(cur); |
6702ed49 | 624 | break; |
252c38f0 | 625 | } |
e7a84565 CM |
626 | search_start = cur->start; |
627 | last_block = cur->start; | |
f2183bde | 628 | *last_ret = search_start; |
e7a84565 CM |
629 | btrfs_tree_unlock(cur); |
630 | free_extent_buffer(cur); | |
6702ed49 | 631 | } |
5708b959 CM |
632 | if (parent->map_token) { |
633 | unmap_extent_buffer(parent, parent->map_token, | |
634 | KM_USER1); | |
635 | parent->map_token = NULL; | |
636 | } | |
6702ed49 CM |
637 | return err; |
638 | } | |
639 | ||
74123bd7 CM |
640 | /* |
641 | * The leaf data grows from end-to-front in the node. | |
642 | * this returns the address of the start of the last item, | |
643 | * which is the stop of the leaf data stack | |
644 | */ | |
123abc88 | 645 | static inline unsigned int leaf_data_end(struct btrfs_root *root, |
5f39d397 | 646 | struct extent_buffer *leaf) |
be0e5c09 | 647 | { |
5f39d397 | 648 | u32 nr = btrfs_header_nritems(leaf); |
be0e5c09 | 649 | if (nr == 0) |
123abc88 | 650 | return BTRFS_LEAF_DATA_SIZE(root); |
5f39d397 | 651 | return btrfs_item_offset_nr(leaf, nr - 1); |
be0e5c09 CM |
652 | } |
653 | ||
d352ac68 CM |
654 | /* |
655 | * extra debugging checks to make sure all the items in a key are | |
656 | * well formed and in the proper order | |
657 | */ | |
123abc88 CM |
658 | static int check_node(struct btrfs_root *root, struct btrfs_path *path, |
659 | int level) | |
aa5d6bed | 660 | { |
5f39d397 CM |
661 | struct extent_buffer *parent = NULL; |
662 | struct extent_buffer *node = path->nodes[level]; | |
663 | struct btrfs_disk_key parent_key; | |
664 | struct btrfs_disk_key node_key; | |
aa5d6bed | 665 | int parent_slot; |
8d7be552 CM |
666 | int slot; |
667 | struct btrfs_key cpukey; | |
5f39d397 | 668 | u32 nritems = btrfs_header_nritems(node); |
aa5d6bed CM |
669 | |
670 | if (path->nodes[level + 1]) | |
5f39d397 | 671 | parent = path->nodes[level + 1]; |
a1f39630 | 672 | |
8d7be552 | 673 | slot = path->slots[level]; |
7518a238 CM |
674 | BUG_ON(nritems == 0); |
675 | if (parent) { | |
a1f39630 | 676 | parent_slot = path->slots[level + 1]; |
5f39d397 CM |
677 | btrfs_node_key(parent, &parent_key, parent_slot); |
678 | btrfs_node_key(node, &node_key, 0); | |
679 | BUG_ON(memcmp(&parent_key, &node_key, | |
e2fa7227 | 680 | sizeof(struct btrfs_disk_key))); |
1d4f8a0c | 681 | BUG_ON(btrfs_node_blockptr(parent, parent_slot) != |
db94535d | 682 | btrfs_header_bytenr(node)); |
aa5d6bed | 683 | } |
123abc88 | 684 | BUG_ON(nritems > BTRFS_NODEPTRS_PER_BLOCK(root)); |
8d7be552 | 685 | if (slot != 0) { |
5f39d397 CM |
686 | btrfs_node_key_to_cpu(node, &cpukey, slot - 1); |
687 | btrfs_node_key(node, &node_key, slot); | |
688 | BUG_ON(comp_keys(&node_key, &cpukey) <= 0); | |
8d7be552 CM |
689 | } |
690 | if (slot < nritems - 1) { | |
5f39d397 CM |
691 | btrfs_node_key_to_cpu(node, &cpukey, slot + 1); |
692 | btrfs_node_key(node, &node_key, slot); | |
693 | BUG_ON(comp_keys(&node_key, &cpukey) >= 0); | |
aa5d6bed CM |
694 | } |
695 | return 0; | |
696 | } | |
697 | ||
d352ac68 CM |
698 | /* |
699 | * extra checking to make sure all the items in a leaf are | |
700 | * well formed and in the proper order | |
701 | */ | |
123abc88 CM |
702 | static int check_leaf(struct btrfs_root *root, struct btrfs_path *path, |
703 | int level) | |
aa5d6bed | 704 | { |
5f39d397 CM |
705 | struct extent_buffer *leaf = path->nodes[level]; |
706 | struct extent_buffer *parent = NULL; | |
aa5d6bed | 707 | int parent_slot; |
8d7be552 | 708 | struct btrfs_key cpukey; |
5f39d397 CM |
709 | struct btrfs_disk_key parent_key; |
710 | struct btrfs_disk_key leaf_key; | |
711 | int slot = path->slots[0]; | |
8d7be552 | 712 | |
5f39d397 | 713 | u32 nritems = btrfs_header_nritems(leaf); |
aa5d6bed CM |
714 | |
715 | if (path->nodes[level + 1]) | |
5f39d397 | 716 | parent = path->nodes[level + 1]; |
7518a238 CM |
717 | |
718 | if (nritems == 0) | |
719 | return 0; | |
720 | ||
721 | if (parent) { | |
a1f39630 | 722 | parent_slot = path->slots[level + 1]; |
5f39d397 CM |
723 | btrfs_node_key(parent, &parent_key, parent_slot); |
724 | btrfs_item_key(leaf, &leaf_key, 0); | |
6702ed49 | 725 | |
5f39d397 | 726 | BUG_ON(memcmp(&parent_key, &leaf_key, |
e2fa7227 | 727 | sizeof(struct btrfs_disk_key))); |
1d4f8a0c | 728 | BUG_ON(btrfs_node_blockptr(parent, parent_slot) != |
db94535d | 729 | btrfs_header_bytenr(leaf)); |
5f39d397 | 730 | } |
5f39d397 CM |
731 | if (slot != 0 && slot < nritems - 1) { |
732 | btrfs_item_key(leaf, &leaf_key, slot); | |
733 | btrfs_item_key_to_cpu(leaf, &cpukey, slot - 1); | |
734 | if (comp_keys(&leaf_key, &cpukey) <= 0) { | |
735 | btrfs_print_leaf(root, leaf); | |
d397712b | 736 | printk(KERN_CRIT "slot %d offset bad key\n", slot); |
5f39d397 CM |
737 | BUG_ON(1); |
738 | } | |
739 | if (btrfs_item_offset_nr(leaf, slot - 1) != | |
740 | btrfs_item_end_nr(leaf, slot)) { | |
741 | btrfs_print_leaf(root, leaf); | |
d397712b | 742 | printk(KERN_CRIT "slot %d offset bad\n", slot); |
5f39d397 CM |
743 | BUG_ON(1); |
744 | } | |
8d7be552 CM |
745 | } |
746 | if (slot < nritems - 1) { | |
5f39d397 CM |
747 | btrfs_item_key(leaf, &leaf_key, slot); |
748 | btrfs_item_key_to_cpu(leaf, &cpukey, slot + 1); | |
749 | BUG_ON(comp_keys(&leaf_key, &cpukey) >= 0); | |
750 | if (btrfs_item_offset_nr(leaf, slot) != | |
751 | btrfs_item_end_nr(leaf, slot + 1)) { | |
752 | btrfs_print_leaf(root, leaf); | |
d397712b | 753 | printk(KERN_CRIT "slot %d offset bad\n", slot); |
5f39d397 CM |
754 | BUG_ON(1); |
755 | } | |
aa5d6bed | 756 | } |
5f39d397 CM |
757 | BUG_ON(btrfs_item_offset_nr(leaf, 0) + |
758 | btrfs_item_size_nr(leaf, 0) != BTRFS_LEAF_DATA_SIZE(root)); | |
aa5d6bed CM |
759 | return 0; |
760 | } | |
761 | ||
d397712b | 762 | static noinline int check_block(struct btrfs_root *root, |
98ed5174 | 763 | struct btrfs_path *path, int level) |
aa5d6bed | 764 | { |
85d824c4 | 765 | return 0; |
aa5d6bed | 766 | if (level == 0) |
123abc88 CM |
767 | return check_leaf(root, path, level); |
768 | return check_node(root, path, level); | |
aa5d6bed CM |
769 | } |
770 | ||
74123bd7 | 771 | /* |
5f39d397 CM |
772 | * search for key in the extent_buffer. The items start at offset p, |
773 | * and they are item_size apart. There are 'max' items in p. | |
774 | * | |
74123bd7 CM |
775 | * the slot in the array is returned via slot, and it points to |
776 | * the place where you would insert key if it is not found in | |
777 | * the array. | |
778 | * | |
779 | * slot may point to max if the key is bigger than all of the keys | |
780 | */ | |
e02119d5 CM |
781 | static noinline int generic_bin_search(struct extent_buffer *eb, |
782 | unsigned long p, | |
783 | int item_size, struct btrfs_key *key, | |
784 | int max, int *slot) | |
be0e5c09 CM |
785 | { |
786 | int low = 0; | |
787 | int high = max; | |
788 | int mid; | |
789 | int ret; | |
479965d6 | 790 | struct btrfs_disk_key *tmp = NULL; |
5f39d397 CM |
791 | struct btrfs_disk_key unaligned; |
792 | unsigned long offset; | |
793 | char *map_token = NULL; | |
794 | char *kaddr = NULL; | |
795 | unsigned long map_start = 0; | |
796 | unsigned long map_len = 0; | |
479965d6 | 797 | int err; |
be0e5c09 | 798 | |
d397712b | 799 | while (low < high) { |
be0e5c09 | 800 | mid = (low + high) / 2; |
5f39d397 CM |
801 | offset = p + mid * item_size; |
802 | ||
803 | if (!map_token || offset < map_start || | |
804 | (offset + sizeof(struct btrfs_disk_key)) > | |
805 | map_start + map_len) { | |
479965d6 | 806 | if (map_token) { |
5f39d397 | 807 | unmap_extent_buffer(eb, map_token, KM_USER0); |
479965d6 CM |
808 | map_token = NULL; |
809 | } | |
934d375b CM |
810 | |
811 | err = map_private_extent_buffer(eb, offset, | |
479965d6 CM |
812 | sizeof(struct btrfs_disk_key), |
813 | &map_token, &kaddr, | |
814 | &map_start, &map_len, KM_USER0); | |
815 | ||
816 | if (!err) { | |
817 | tmp = (struct btrfs_disk_key *)(kaddr + offset - | |
818 | map_start); | |
819 | } else { | |
820 | read_extent_buffer(eb, &unaligned, | |
821 | offset, sizeof(unaligned)); | |
822 | tmp = &unaligned; | |
823 | } | |
5f39d397 | 824 | |
5f39d397 CM |
825 | } else { |
826 | tmp = (struct btrfs_disk_key *)(kaddr + offset - | |
827 | map_start); | |
828 | } | |
be0e5c09 CM |
829 | ret = comp_keys(tmp, key); |
830 | ||
831 | if (ret < 0) | |
832 | low = mid + 1; | |
833 | else if (ret > 0) | |
834 | high = mid; | |
835 | else { | |
836 | *slot = mid; | |
479965d6 CM |
837 | if (map_token) |
838 | unmap_extent_buffer(eb, map_token, KM_USER0); | |
be0e5c09 CM |
839 | return 0; |
840 | } | |
841 | } | |
842 | *slot = low; | |
5f39d397 CM |
843 | if (map_token) |
844 | unmap_extent_buffer(eb, map_token, KM_USER0); | |
be0e5c09 CM |
845 | return 1; |
846 | } | |
847 | ||
97571fd0 CM |
848 | /* |
849 | * simple bin_search frontend that does the right thing for | |
850 | * leaves vs nodes | |
851 | */ | |
5f39d397 CM |
852 | static int bin_search(struct extent_buffer *eb, struct btrfs_key *key, |
853 | int level, int *slot) | |
be0e5c09 | 854 | { |
5f39d397 CM |
855 | if (level == 0) { |
856 | return generic_bin_search(eb, | |
857 | offsetof(struct btrfs_leaf, items), | |
0783fcfc | 858 | sizeof(struct btrfs_item), |
5f39d397 | 859 | key, btrfs_header_nritems(eb), |
7518a238 | 860 | slot); |
be0e5c09 | 861 | } else { |
5f39d397 CM |
862 | return generic_bin_search(eb, |
863 | offsetof(struct btrfs_node, ptrs), | |
123abc88 | 864 | sizeof(struct btrfs_key_ptr), |
5f39d397 | 865 | key, btrfs_header_nritems(eb), |
7518a238 | 866 | slot); |
be0e5c09 CM |
867 | } |
868 | return -1; | |
869 | } | |
870 | ||
d352ac68 CM |
871 | /* given a node and slot number, this reads the blocks it points to. The |
872 | * extent buffer is returned with a reference taken (but unlocked). | |
873 | * NULL is returned on error. | |
874 | */ | |
e02119d5 | 875 | static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root, |
5f39d397 | 876 | struct extent_buffer *parent, int slot) |
bb803951 | 877 | { |
ca7a79ad | 878 | int level = btrfs_header_level(parent); |
bb803951 CM |
879 | if (slot < 0) |
880 | return NULL; | |
5f39d397 | 881 | if (slot >= btrfs_header_nritems(parent)) |
bb803951 | 882 | return NULL; |
ca7a79ad CM |
883 | |
884 | BUG_ON(level == 0); | |
885 | ||
db94535d | 886 | return read_tree_block(root, btrfs_node_blockptr(parent, slot), |
ca7a79ad CM |
887 | btrfs_level_size(root, level - 1), |
888 | btrfs_node_ptr_generation(parent, slot)); | |
bb803951 CM |
889 | } |
890 | ||
d352ac68 CM |
891 | /* |
892 | * node level balancing, used to make sure nodes are in proper order for | |
893 | * item deletion. We balance from the top down, so we have to make sure | |
894 | * that a deletion won't leave an node completely empty later on. | |
895 | */ | |
e02119d5 | 896 | static noinline int balance_level(struct btrfs_trans_handle *trans, |
98ed5174 CM |
897 | struct btrfs_root *root, |
898 | struct btrfs_path *path, int level) | |
bb803951 | 899 | { |
5f39d397 CM |
900 | struct extent_buffer *right = NULL; |
901 | struct extent_buffer *mid; | |
902 | struct extent_buffer *left = NULL; | |
903 | struct extent_buffer *parent = NULL; | |
bb803951 CM |
904 | int ret = 0; |
905 | int wret; | |
906 | int pslot; | |
bb803951 | 907 | int orig_slot = path->slots[level]; |
54aa1f4d | 908 | int err_on_enospc = 0; |
79f95c82 | 909 | u64 orig_ptr; |
bb803951 CM |
910 | |
911 | if (level == 0) | |
912 | return 0; | |
913 | ||
5f39d397 | 914 | mid = path->nodes[level]; |
b4ce94de | 915 | |
925baedd | 916 | WARN_ON(!path->locks[level]); |
7bb86316 CM |
917 | WARN_ON(btrfs_header_generation(mid) != trans->transid); |
918 | ||
1d4f8a0c | 919 | orig_ptr = btrfs_node_blockptr(mid, orig_slot); |
79f95c82 | 920 | |
234b63a0 | 921 | if (level < BTRFS_MAX_LEVEL - 1) |
5f39d397 | 922 | parent = path->nodes[level + 1]; |
bb803951 CM |
923 | pslot = path->slots[level + 1]; |
924 | ||
40689478 CM |
925 | /* |
926 | * deal with the case where there is only one pointer in the root | |
927 | * by promoting the node below to a root | |
928 | */ | |
5f39d397 CM |
929 | if (!parent) { |
930 | struct extent_buffer *child; | |
bb803951 | 931 | |
5f39d397 | 932 | if (btrfs_header_nritems(mid) != 1) |
bb803951 CM |
933 | return 0; |
934 | ||
935 | /* promote the child to a root */ | |
5f39d397 | 936 | child = read_node_slot(root, mid, 0); |
7951f3ce | 937 | BUG_ON(!child); |
925baedd | 938 | btrfs_tree_lock(child); |
b4ce94de | 939 | btrfs_set_lock_blocking(child); |
65b51a00 | 940 | ret = btrfs_cow_block(trans, root, child, mid, 0, &child, 0); |
2f375ab9 Y |
941 | BUG_ON(ret); |
942 | ||
925baedd | 943 | spin_lock(&root->node_lock); |
bb803951 | 944 | root->node = child; |
925baedd CM |
945 | spin_unlock(&root->node_lock); |
946 | ||
31840ae1 ZY |
947 | ret = btrfs_update_extent_ref(trans, root, child->start, |
948 | mid->start, child->start, | |
949 | root->root_key.objectid, | |
3bb1a1bc | 950 | trans->transid, level - 1); |
31840ae1 ZY |
951 | BUG_ON(ret); |
952 | ||
0b86a832 | 953 | add_root_to_dirty_list(root); |
925baedd | 954 | btrfs_tree_unlock(child); |
b4ce94de | 955 | |
925baedd | 956 | path->locks[level] = 0; |
bb803951 | 957 | path->nodes[level] = NULL; |
5f39d397 | 958 | clean_tree_block(trans, root, mid); |
925baedd | 959 | btrfs_tree_unlock(mid); |
bb803951 | 960 | /* once for the path */ |
5f39d397 | 961 | free_extent_buffer(mid); |
7bb86316 | 962 | ret = btrfs_free_extent(trans, root, mid->start, mid->len, |
31840ae1 | 963 | mid->start, root->root_key.objectid, |
3bb1a1bc YZ |
964 | btrfs_header_generation(mid), |
965 | level, 1); | |
bb803951 | 966 | /* once for the root ptr */ |
5f39d397 | 967 | free_extent_buffer(mid); |
db94535d | 968 | return ret; |
bb803951 | 969 | } |
5f39d397 | 970 | if (btrfs_header_nritems(mid) > |
123abc88 | 971 | BTRFS_NODEPTRS_PER_BLOCK(root) / 4) |
bb803951 CM |
972 | return 0; |
973 | ||
5f39d397 | 974 | if (btrfs_header_nritems(mid) < 2) |
54aa1f4d CM |
975 | err_on_enospc = 1; |
976 | ||
5f39d397 CM |
977 | left = read_node_slot(root, parent, pslot - 1); |
978 | if (left) { | |
925baedd | 979 | btrfs_tree_lock(left); |
b4ce94de | 980 | btrfs_set_lock_blocking(left); |
5f39d397 | 981 | wret = btrfs_cow_block(trans, root, left, |
65b51a00 | 982 | parent, pslot - 1, &left, 0); |
54aa1f4d CM |
983 | if (wret) { |
984 | ret = wret; | |
985 | goto enospc; | |
986 | } | |
2cc58cf2 | 987 | } |
5f39d397 CM |
988 | right = read_node_slot(root, parent, pslot + 1); |
989 | if (right) { | |
925baedd | 990 | btrfs_tree_lock(right); |
b4ce94de | 991 | btrfs_set_lock_blocking(right); |
5f39d397 | 992 | wret = btrfs_cow_block(trans, root, right, |
65b51a00 | 993 | parent, pslot + 1, &right, 0); |
2cc58cf2 CM |
994 | if (wret) { |
995 | ret = wret; | |
996 | goto enospc; | |
997 | } | |
998 | } | |
999 | ||
1000 | /* first, try to make some room in the middle buffer */ | |
5f39d397 CM |
1001 | if (left) { |
1002 | orig_slot += btrfs_header_nritems(left); | |
bce4eae9 | 1003 | wret = push_node_left(trans, root, left, mid, 1); |
79f95c82 CM |
1004 | if (wret < 0) |
1005 | ret = wret; | |
5f39d397 | 1006 | if (btrfs_header_nritems(mid) < 2) |
54aa1f4d | 1007 | err_on_enospc = 1; |
bb803951 | 1008 | } |
79f95c82 CM |
1009 | |
1010 | /* | |
1011 | * then try to empty the right most buffer into the middle | |
1012 | */ | |
5f39d397 | 1013 | if (right) { |
971a1f66 | 1014 | wret = push_node_left(trans, root, mid, right, 1); |
54aa1f4d | 1015 | if (wret < 0 && wret != -ENOSPC) |
79f95c82 | 1016 | ret = wret; |
5f39d397 | 1017 | if (btrfs_header_nritems(right) == 0) { |
db94535d | 1018 | u64 bytenr = right->start; |
7bb86316 | 1019 | u64 generation = btrfs_header_generation(parent); |
db94535d CM |
1020 | u32 blocksize = right->len; |
1021 | ||
5f39d397 | 1022 | clean_tree_block(trans, root, right); |
925baedd | 1023 | btrfs_tree_unlock(right); |
5f39d397 | 1024 | free_extent_buffer(right); |
bb803951 | 1025 | right = NULL; |
e089f05c CM |
1026 | wret = del_ptr(trans, root, path, level + 1, pslot + |
1027 | 1); | |
bb803951 CM |
1028 | if (wret) |
1029 | ret = wret; | |
db94535d | 1030 | wret = btrfs_free_extent(trans, root, bytenr, |
31840ae1 | 1031 | blocksize, parent->start, |
7bb86316 | 1032 | btrfs_header_owner(parent), |
3bb1a1bc | 1033 | generation, level, 1); |
bb803951 CM |
1034 | if (wret) |
1035 | ret = wret; | |
1036 | } else { | |
5f39d397 CM |
1037 | struct btrfs_disk_key right_key; |
1038 | btrfs_node_key(right, &right_key, 0); | |
1039 | btrfs_set_node_key(parent, &right_key, pslot + 1); | |
1040 | btrfs_mark_buffer_dirty(parent); | |
bb803951 CM |
1041 | } |
1042 | } | |
5f39d397 | 1043 | if (btrfs_header_nritems(mid) == 1) { |
79f95c82 CM |
1044 | /* |
1045 | * we're not allowed to leave a node with one item in the | |
1046 | * tree during a delete. A deletion from lower in the tree | |
1047 | * could try to delete the only pointer in this node. | |
1048 | * So, pull some keys from the left. | |
1049 | * There has to be a left pointer at this point because | |
1050 | * otherwise we would have pulled some pointers from the | |
1051 | * right | |
1052 | */ | |
5f39d397 CM |
1053 | BUG_ON(!left); |
1054 | wret = balance_node_right(trans, root, mid, left); | |
54aa1f4d | 1055 | if (wret < 0) { |
79f95c82 | 1056 | ret = wret; |
54aa1f4d CM |
1057 | goto enospc; |
1058 | } | |
bce4eae9 CM |
1059 | if (wret == 1) { |
1060 | wret = push_node_left(trans, root, left, mid, 1); | |
1061 | if (wret < 0) | |
1062 | ret = wret; | |
1063 | } | |
79f95c82 CM |
1064 | BUG_ON(wret == 1); |
1065 | } | |
5f39d397 | 1066 | if (btrfs_header_nritems(mid) == 0) { |
79f95c82 | 1067 | /* we've managed to empty the middle node, drop it */ |
7bb86316 | 1068 | u64 root_gen = btrfs_header_generation(parent); |
db94535d CM |
1069 | u64 bytenr = mid->start; |
1070 | u32 blocksize = mid->len; | |
925baedd | 1071 | |
5f39d397 | 1072 | clean_tree_block(trans, root, mid); |
925baedd | 1073 | btrfs_tree_unlock(mid); |
5f39d397 | 1074 | free_extent_buffer(mid); |
bb803951 | 1075 | mid = NULL; |
e089f05c | 1076 | wret = del_ptr(trans, root, path, level + 1, pslot); |
bb803951 CM |
1077 | if (wret) |
1078 | ret = wret; | |
7bb86316 | 1079 | wret = btrfs_free_extent(trans, root, bytenr, blocksize, |
31840ae1 | 1080 | parent->start, |
7bb86316 | 1081 | btrfs_header_owner(parent), |
3bb1a1bc | 1082 | root_gen, level, 1); |
bb803951 CM |
1083 | if (wret) |
1084 | ret = wret; | |
79f95c82 CM |
1085 | } else { |
1086 | /* update the parent key to reflect our changes */ | |
5f39d397 CM |
1087 | struct btrfs_disk_key mid_key; |
1088 | btrfs_node_key(mid, &mid_key, 0); | |
1089 | btrfs_set_node_key(parent, &mid_key, pslot); | |
1090 | btrfs_mark_buffer_dirty(parent); | |
79f95c82 | 1091 | } |
bb803951 | 1092 | |
79f95c82 | 1093 | /* update the path */ |
5f39d397 CM |
1094 | if (left) { |
1095 | if (btrfs_header_nritems(left) > orig_slot) { | |
1096 | extent_buffer_get(left); | |
925baedd | 1097 | /* left was locked after cow */ |
5f39d397 | 1098 | path->nodes[level] = left; |
bb803951 CM |
1099 | path->slots[level + 1] -= 1; |
1100 | path->slots[level] = orig_slot; | |
925baedd CM |
1101 | if (mid) { |
1102 | btrfs_tree_unlock(mid); | |
5f39d397 | 1103 | free_extent_buffer(mid); |
925baedd | 1104 | } |
bb803951 | 1105 | } else { |
5f39d397 | 1106 | orig_slot -= btrfs_header_nritems(left); |
bb803951 CM |
1107 | path->slots[level] = orig_slot; |
1108 | } | |
1109 | } | |
79f95c82 | 1110 | /* double check we haven't messed things up */ |
123abc88 | 1111 | check_block(root, path, level); |
e20d96d6 | 1112 | if (orig_ptr != |
5f39d397 | 1113 | btrfs_node_blockptr(path->nodes[level], path->slots[level])) |
79f95c82 | 1114 | BUG(); |
54aa1f4d | 1115 | enospc: |
925baedd CM |
1116 | if (right) { |
1117 | btrfs_tree_unlock(right); | |
5f39d397 | 1118 | free_extent_buffer(right); |
925baedd CM |
1119 | } |
1120 | if (left) { | |
1121 | if (path->nodes[level] != left) | |
1122 | btrfs_tree_unlock(left); | |
5f39d397 | 1123 | free_extent_buffer(left); |
925baedd | 1124 | } |
bb803951 CM |
1125 | return ret; |
1126 | } | |
1127 | ||
d352ac68 CM |
1128 | /* Node balancing for insertion. Here we only split or push nodes around |
1129 | * when they are completely full. This is also done top down, so we | |
1130 | * have to be pessimistic. | |
1131 | */ | |
d397712b | 1132 | static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans, |
98ed5174 CM |
1133 | struct btrfs_root *root, |
1134 | struct btrfs_path *path, int level) | |
e66f709b | 1135 | { |
5f39d397 CM |
1136 | struct extent_buffer *right = NULL; |
1137 | struct extent_buffer *mid; | |
1138 | struct extent_buffer *left = NULL; | |
1139 | struct extent_buffer *parent = NULL; | |
e66f709b CM |
1140 | int ret = 0; |
1141 | int wret; | |
1142 | int pslot; | |
1143 | int orig_slot = path->slots[level]; | |
1144 | u64 orig_ptr; | |
1145 | ||
1146 | if (level == 0) | |
1147 | return 1; | |
1148 | ||
5f39d397 | 1149 | mid = path->nodes[level]; |
7bb86316 | 1150 | WARN_ON(btrfs_header_generation(mid) != trans->transid); |
e66f709b CM |
1151 | orig_ptr = btrfs_node_blockptr(mid, orig_slot); |
1152 | ||
1153 | if (level < BTRFS_MAX_LEVEL - 1) | |
5f39d397 | 1154 | parent = path->nodes[level + 1]; |
e66f709b CM |
1155 | pslot = path->slots[level + 1]; |
1156 | ||
5f39d397 | 1157 | if (!parent) |
e66f709b | 1158 | return 1; |
e66f709b | 1159 | |
5f39d397 | 1160 | left = read_node_slot(root, parent, pslot - 1); |
e66f709b CM |
1161 | |
1162 | /* first, try to make some room in the middle buffer */ | |
5f39d397 | 1163 | if (left) { |
e66f709b | 1164 | u32 left_nr; |
925baedd CM |
1165 | |
1166 | btrfs_tree_lock(left); | |
b4ce94de CM |
1167 | btrfs_set_lock_blocking(left); |
1168 | ||
5f39d397 | 1169 | left_nr = btrfs_header_nritems(left); |
33ade1f8 CM |
1170 | if (left_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) { |
1171 | wret = 1; | |
1172 | } else { | |
5f39d397 | 1173 | ret = btrfs_cow_block(trans, root, left, parent, |
65b51a00 | 1174 | pslot - 1, &left, 0); |
54aa1f4d CM |
1175 | if (ret) |
1176 | wret = 1; | |
1177 | else { | |
54aa1f4d | 1178 | wret = push_node_left(trans, root, |
971a1f66 | 1179 | left, mid, 0); |
54aa1f4d | 1180 | } |
33ade1f8 | 1181 | } |
e66f709b CM |
1182 | if (wret < 0) |
1183 | ret = wret; | |
1184 | if (wret == 0) { | |
5f39d397 | 1185 | struct btrfs_disk_key disk_key; |
e66f709b | 1186 | orig_slot += left_nr; |
5f39d397 CM |
1187 | btrfs_node_key(mid, &disk_key, 0); |
1188 | btrfs_set_node_key(parent, &disk_key, pslot); | |
1189 | btrfs_mark_buffer_dirty(parent); | |
1190 | if (btrfs_header_nritems(left) > orig_slot) { | |
1191 | path->nodes[level] = left; | |
e66f709b CM |
1192 | path->slots[level + 1] -= 1; |
1193 | path->slots[level] = orig_slot; | |
925baedd | 1194 | btrfs_tree_unlock(mid); |
5f39d397 | 1195 | free_extent_buffer(mid); |
e66f709b CM |
1196 | } else { |
1197 | orig_slot -= | |
5f39d397 | 1198 | btrfs_header_nritems(left); |
e66f709b | 1199 | path->slots[level] = orig_slot; |
925baedd | 1200 | btrfs_tree_unlock(left); |
5f39d397 | 1201 | free_extent_buffer(left); |
e66f709b | 1202 | } |
e66f709b CM |
1203 | return 0; |
1204 | } | |
925baedd | 1205 | btrfs_tree_unlock(left); |
5f39d397 | 1206 | free_extent_buffer(left); |
e66f709b | 1207 | } |
925baedd | 1208 | right = read_node_slot(root, parent, pslot + 1); |
e66f709b CM |
1209 | |
1210 | /* | |
1211 | * then try to empty the right most buffer into the middle | |
1212 | */ | |
5f39d397 | 1213 | if (right) { |
33ade1f8 | 1214 | u32 right_nr; |
b4ce94de | 1215 | |
925baedd | 1216 | btrfs_tree_lock(right); |
b4ce94de CM |
1217 | btrfs_set_lock_blocking(right); |
1218 | ||
5f39d397 | 1219 | right_nr = btrfs_header_nritems(right); |
33ade1f8 CM |
1220 | if (right_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) { |
1221 | wret = 1; | |
1222 | } else { | |
5f39d397 CM |
1223 | ret = btrfs_cow_block(trans, root, right, |
1224 | parent, pslot + 1, | |
65b51a00 | 1225 | &right, 0); |
54aa1f4d CM |
1226 | if (ret) |
1227 | wret = 1; | |
1228 | else { | |
54aa1f4d | 1229 | wret = balance_node_right(trans, root, |
5f39d397 | 1230 | right, mid); |
54aa1f4d | 1231 | } |
33ade1f8 | 1232 | } |
e66f709b CM |
1233 | if (wret < 0) |
1234 | ret = wret; | |
1235 | if (wret == 0) { | |
5f39d397 CM |
1236 | struct btrfs_disk_key disk_key; |
1237 | ||
1238 | btrfs_node_key(right, &disk_key, 0); | |
1239 | btrfs_set_node_key(parent, &disk_key, pslot + 1); | |
1240 | btrfs_mark_buffer_dirty(parent); | |
1241 | ||
1242 | if (btrfs_header_nritems(mid) <= orig_slot) { | |
1243 | path->nodes[level] = right; | |
e66f709b CM |
1244 | path->slots[level + 1] += 1; |
1245 | path->slots[level] = orig_slot - | |
5f39d397 | 1246 | btrfs_header_nritems(mid); |
925baedd | 1247 | btrfs_tree_unlock(mid); |
5f39d397 | 1248 | free_extent_buffer(mid); |
e66f709b | 1249 | } else { |
925baedd | 1250 | btrfs_tree_unlock(right); |
5f39d397 | 1251 | free_extent_buffer(right); |
e66f709b | 1252 | } |
e66f709b CM |
1253 | return 0; |
1254 | } | |
925baedd | 1255 | btrfs_tree_unlock(right); |
5f39d397 | 1256 | free_extent_buffer(right); |
e66f709b | 1257 | } |
e66f709b CM |
1258 | return 1; |
1259 | } | |
1260 | ||
3c69faec | 1261 | /* |
d352ac68 CM |
1262 | * readahead one full node of leaves, finding things that are close |
1263 | * to the block in 'slot', and triggering ra on them. | |
3c69faec | 1264 | */ |
e02119d5 CM |
1265 | static noinline void reada_for_search(struct btrfs_root *root, |
1266 | struct btrfs_path *path, | |
1267 | int level, int slot, u64 objectid) | |
3c69faec | 1268 | { |
5f39d397 | 1269 | struct extent_buffer *node; |
01f46658 | 1270 | struct btrfs_disk_key disk_key; |
3c69faec | 1271 | u32 nritems; |
3c69faec | 1272 | u64 search; |
a7175319 | 1273 | u64 target; |
6b80053d | 1274 | u64 nread = 0; |
3c69faec | 1275 | int direction = path->reada; |
5f39d397 | 1276 | struct extent_buffer *eb; |
6b80053d CM |
1277 | u32 nr; |
1278 | u32 blocksize; | |
1279 | u32 nscan = 0; | |
db94535d | 1280 | |
a6b6e75e | 1281 | if (level != 1) |
6702ed49 CM |
1282 | return; |
1283 | ||
1284 | if (!path->nodes[level]) | |
3c69faec CM |
1285 | return; |
1286 | ||
5f39d397 | 1287 | node = path->nodes[level]; |
925baedd | 1288 | |
3c69faec | 1289 | search = btrfs_node_blockptr(node, slot); |
6b80053d CM |
1290 | blocksize = btrfs_level_size(root, level - 1); |
1291 | eb = btrfs_find_tree_block(root, search, blocksize); | |
5f39d397 CM |
1292 | if (eb) { |
1293 | free_extent_buffer(eb); | |
3c69faec CM |
1294 | return; |
1295 | } | |
1296 | ||
a7175319 | 1297 | target = search; |
6b80053d | 1298 | |
5f39d397 | 1299 | nritems = btrfs_header_nritems(node); |
6b80053d | 1300 | nr = slot; |
d397712b | 1301 | while (1) { |
6b80053d CM |
1302 | if (direction < 0) { |
1303 | if (nr == 0) | |
1304 | break; | |
1305 | nr--; | |
1306 | } else if (direction > 0) { | |
1307 | nr++; | |
1308 | if (nr >= nritems) | |
1309 | break; | |
3c69faec | 1310 | } |
01f46658 CM |
1311 | if (path->reada < 0 && objectid) { |
1312 | btrfs_node_key(node, &disk_key, nr); | |
1313 | if (btrfs_disk_key_objectid(&disk_key) != objectid) | |
1314 | break; | |
1315 | } | |
6b80053d | 1316 | search = btrfs_node_blockptr(node, nr); |
a7175319 CM |
1317 | if ((search <= target && target - search <= 65536) || |
1318 | (search > target && search - target <= 65536)) { | |
ca7a79ad CM |
1319 | readahead_tree_block(root, search, blocksize, |
1320 | btrfs_node_ptr_generation(node, nr)); | |
6b80053d CM |
1321 | nread += blocksize; |
1322 | } | |
1323 | nscan++; | |
a7175319 | 1324 | if ((nread > 65536 || nscan > 32)) |
6b80053d | 1325 | break; |
3c69faec CM |
1326 | } |
1327 | } | |
925baedd | 1328 | |
b4ce94de CM |
1329 | /* |
1330 | * returns -EAGAIN if it had to drop the path, or zero if everything was in | |
1331 | * cache | |
1332 | */ | |
1333 | static noinline int reada_for_balance(struct btrfs_root *root, | |
1334 | struct btrfs_path *path, int level) | |
1335 | { | |
1336 | int slot; | |
1337 | int nritems; | |
1338 | struct extent_buffer *parent; | |
1339 | struct extent_buffer *eb; | |
1340 | u64 gen; | |
1341 | u64 block1 = 0; | |
1342 | u64 block2 = 0; | |
1343 | int ret = 0; | |
1344 | int blocksize; | |
1345 | ||
1346 | parent = path->nodes[level - 1]; | |
1347 | if (!parent) | |
1348 | return 0; | |
1349 | ||
1350 | nritems = btrfs_header_nritems(parent); | |
1351 | slot = path->slots[level]; | |
1352 | blocksize = btrfs_level_size(root, level); | |
1353 | ||
1354 | if (slot > 0) { | |
1355 | block1 = btrfs_node_blockptr(parent, slot - 1); | |
1356 | gen = btrfs_node_ptr_generation(parent, slot - 1); | |
1357 | eb = btrfs_find_tree_block(root, block1, blocksize); | |
1358 | if (eb && btrfs_buffer_uptodate(eb, gen)) | |
1359 | block1 = 0; | |
1360 | free_extent_buffer(eb); | |
1361 | } | |
1362 | if (slot < nritems) { | |
1363 | block2 = btrfs_node_blockptr(parent, slot + 1); | |
1364 | gen = btrfs_node_ptr_generation(parent, slot + 1); | |
1365 | eb = btrfs_find_tree_block(root, block2, blocksize); | |
1366 | if (eb && btrfs_buffer_uptodate(eb, gen)) | |
1367 | block2 = 0; | |
1368 | free_extent_buffer(eb); | |
1369 | } | |
1370 | if (block1 || block2) { | |
1371 | ret = -EAGAIN; | |
1372 | btrfs_release_path(root, path); | |
1373 | if (block1) | |
1374 | readahead_tree_block(root, block1, blocksize, 0); | |
1375 | if (block2) | |
1376 | readahead_tree_block(root, block2, blocksize, 0); | |
1377 | ||
1378 | if (block1) { | |
1379 | eb = read_tree_block(root, block1, blocksize, 0); | |
1380 | free_extent_buffer(eb); | |
1381 | } | |
1382 | if (block1) { | |
1383 | eb = read_tree_block(root, block2, blocksize, 0); | |
1384 | free_extent_buffer(eb); | |
1385 | } | |
1386 | } | |
1387 | return ret; | |
1388 | } | |
1389 | ||
1390 | ||
d352ac68 | 1391 | /* |
d397712b CM |
1392 | * when we walk down the tree, it is usually safe to unlock the higher layers |
1393 | * in the tree. The exceptions are when our path goes through slot 0, because | |
1394 | * operations on the tree might require changing key pointers higher up in the | |
1395 | * tree. | |
d352ac68 | 1396 | * |
d397712b CM |
1397 | * callers might also have set path->keep_locks, which tells this code to keep |
1398 | * the lock if the path points to the last slot in the block. This is part of | |
1399 | * walking through the tree, and selecting the next slot in the higher block. | |
d352ac68 | 1400 | * |
d397712b CM |
1401 | * lowest_unlock sets the lowest level in the tree we're allowed to unlock. so |
1402 | * if lowest_unlock is 1, level 0 won't be unlocked | |
d352ac68 | 1403 | */ |
e02119d5 CM |
1404 | static noinline void unlock_up(struct btrfs_path *path, int level, |
1405 | int lowest_unlock) | |
925baedd CM |
1406 | { |
1407 | int i; | |
1408 | int skip_level = level; | |
051e1b9f | 1409 | int no_skips = 0; |
925baedd CM |
1410 | struct extent_buffer *t; |
1411 | ||
1412 | for (i = level; i < BTRFS_MAX_LEVEL; i++) { | |
1413 | if (!path->nodes[i]) | |
1414 | break; | |
1415 | if (!path->locks[i]) | |
1416 | break; | |
051e1b9f | 1417 | if (!no_skips && path->slots[i] == 0) { |
925baedd CM |
1418 | skip_level = i + 1; |
1419 | continue; | |
1420 | } | |
051e1b9f | 1421 | if (!no_skips && path->keep_locks) { |
925baedd CM |
1422 | u32 nritems; |
1423 | t = path->nodes[i]; | |
1424 | nritems = btrfs_header_nritems(t); | |
051e1b9f | 1425 | if (nritems < 1 || path->slots[i] >= nritems - 1) { |
925baedd CM |
1426 | skip_level = i + 1; |
1427 | continue; | |
1428 | } | |
1429 | } | |
051e1b9f CM |
1430 | if (skip_level < i && i >= lowest_unlock) |
1431 | no_skips = 1; | |
1432 | ||
925baedd CM |
1433 | t = path->nodes[i]; |
1434 | if (i >= lowest_unlock && i > skip_level && path->locks[i]) { | |
1435 | btrfs_tree_unlock(t); | |
1436 | path->locks[i] = 0; | |
1437 | } | |
1438 | } | |
1439 | } | |
1440 | ||
b4ce94de CM |
1441 | /* |
1442 | * This releases any locks held in the path starting at level and | |
1443 | * going all the way up to the root. | |
1444 | * | |
1445 | * btrfs_search_slot will keep the lock held on higher nodes in a few | |
1446 | * corner cases, such as COW of the block at slot zero in the node. This | |
1447 | * ignores those rules, and it should only be called when there are no | |
1448 | * more updates to be done higher up in the tree. | |
1449 | */ | |
1450 | noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level) | |
1451 | { | |
1452 | int i; | |
1453 | ||
1454 | if (path->keep_locks || path->lowest_level) | |
1455 | return; | |
1456 | ||
1457 | for (i = level; i < BTRFS_MAX_LEVEL; i++) { | |
1458 | if (!path->nodes[i]) | |
12f4dacc | 1459 | continue; |
b4ce94de | 1460 | if (!path->locks[i]) |
12f4dacc | 1461 | continue; |
b4ce94de CM |
1462 | btrfs_tree_unlock(path->nodes[i]); |
1463 | path->locks[i] = 0; | |
1464 | } | |
1465 | } | |
1466 | ||
74123bd7 CM |
1467 | /* |
1468 | * look for key in the tree. path is filled in with nodes along the way | |
1469 | * if key is found, we return zero and you can find the item in the leaf | |
1470 | * level of the path (level 0) | |
1471 | * | |
1472 | * If the key isn't found, the path points to the slot where it should | |
aa5d6bed CM |
1473 | * be inserted, and 1 is returned. If there are other errors during the |
1474 | * search a negative error number is returned. | |
97571fd0 CM |
1475 | * |
1476 | * if ins_len > 0, nodes and leaves will be split as we walk down the | |
1477 | * tree. if ins_len < 0, nodes will be merged as we walk down the tree (if | |
1478 | * possible) | |
74123bd7 | 1479 | */ |
e089f05c CM |
1480 | int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root |
1481 | *root, struct btrfs_key *key, struct btrfs_path *p, int | |
1482 | ins_len, int cow) | |
be0e5c09 | 1483 | { |
5f39d397 | 1484 | struct extent_buffer *b; |
051e1b9f | 1485 | struct extent_buffer *tmp; |
be0e5c09 CM |
1486 | int slot; |
1487 | int ret; | |
1488 | int level; | |
3c69faec | 1489 | int should_reada = p->reada; |
925baedd | 1490 | int lowest_unlock = 1; |
594a24eb | 1491 | int blocksize; |
9f3a7427 | 1492 | u8 lowest_level = 0; |
594a24eb CM |
1493 | u64 blocknr; |
1494 | u64 gen; | |
65b51a00 | 1495 | struct btrfs_key prealloc_block; |
9f3a7427 | 1496 | |
6702ed49 | 1497 | lowest_level = p->lowest_level; |
323ac95b | 1498 | WARN_ON(lowest_level && ins_len > 0); |
22b0ebda | 1499 | WARN_ON(p->nodes[0] != NULL); |
25179201 | 1500 | |
925baedd CM |
1501 | if (ins_len < 0) |
1502 | lowest_unlock = 2; | |
65b51a00 CM |
1503 | |
1504 | prealloc_block.objectid = 0; | |
1505 | ||
bb803951 | 1506 | again: |
5cd57b2c CM |
1507 | if (p->skip_locking) |
1508 | b = btrfs_root_node(root); | |
1509 | else | |
1510 | b = btrfs_lock_root_node(root); | |
925baedd | 1511 | |
eb60ceac | 1512 | while (b) { |
5f39d397 | 1513 | level = btrfs_header_level(b); |
65b51a00 CM |
1514 | |
1515 | /* | |
1516 | * setup the path here so we can release it under lock | |
1517 | * contention with the cow code | |
1518 | */ | |
1519 | p->nodes[level] = b; | |
1520 | if (!p->skip_locking) | |
1521 | p->locks[level] = 1; | |
1522 | ||
02217ed2 CM |
1523 | if (cow) { |
1524 | int wret; | |
65b51a00 CM |
1525 | |
1526 | /* is a cow on this block not required */ | |
65b51a00 | 1527 | if (btrfs_header_generation(b) == trans->transid && |
5b21f2ed | 1528 | btrfs_header_owner(b) == root->root_key.objectid && |
65b51a00 | 1529 | !btrfs_header_flag(b, BTRFS_HEADER_FLAG_WRITTEN)) { |
65b51a00 CM |
1530 | goto cow_done; |
1531 | } | |
65b51a00 CM |
1532 | |
1533 | /* ok, we have to cow, is our old prealloc the right | |
1534 | * size? | |
1535 | */ | |
1536 | if (prealloc_block.objectid && | |
1537 | prealloc_block.offset != b->len) { | |
7b78c170 | 1538 | btrfs_release_path(root, p); |
65b51a00 CM |
1539 | btrfs_free_reserved_extent(root, |
1540 | prealloc_block.objectid, | |
1541 | prealloc_block.offset); | |
1542 | prealloc_block.objectid = 0; | |
7b78c170 | 1543 | goto again; |
65b51a00 CM |
1544 | } |
1545 | ||
1546 | /* | |
1547 | * for higher level blocks, try not to allocate blocks | |
1548 | * with the block and the parent locks held. | |
1549 | */ | |
284b066a | 1550 | if (level > 0 && !prealloc_block.objectid) { |
65b51a00 CM |
1551 | u32 size = b->len; |
1552 | u64 hint = b->start; | |
1553 | ||
1554 | btrfs_release_path(root, p); | |
1555 | ret = btrfs_reserve_extent(trans, root, | |
1556 | size, size, 0, | |
1557 | hint, (u64)-1, | |
1558 | &prealloc_block, 0); | |
1559 | BUG_ON(ret); | |
1560 | goto again; | |
1561 | } | |
1562 | ||
b4ce94de CM |
1563 | btrfs_set_path_blocking(p); |
1564 | ||
e20d96d6 CM |
1565 | wret = btrfs_cow_block(trans, root, b, |
1566 | p->nodes[level + 1], | |
1567 | p->slots[level + 1], | |
65b51a00 CM |
1568 | &b, prealloc_block.objectid); |
1569 | prealloc_block.objectid = 0; | |
54aa1f4d | 1570 | if (wret) { |
5f39d397 | 1571 | free_extent_buffer(b); |
65b51a00 CM |
1572 | ret = wret; |
1573 | goto done; | |
54aa1f4d | 1574 | } |
02217ed2 | 1575 | } |
65b51a00 | 1576 | cow_done: |
02217ed2 | 1577 | BUG_ON(!cow && ins_len); |
5f39d397 | 1578 | if (level != btrfs_header_level(b)) |
2c90e5d6 | 1579 | WARN_ON(1); |
5f39d397 | 1580 | level = btrfs_header_level(b); |
65b51a00 | 1581 | |
eb60ceac | 1582 | p->nodes[level] = b; |
5cd57b2c CM |
1583 | if (!p->skip_locking) |
1584 | p->locks[level] = 1; | |
65b51a00 | 1585 | |
4008c04a | 1586 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de CM |
1587 | |
1588 | /* | |
1589 | * we have a lock on b and as long as we aren't changing | |
1590 | * the tree, there is no way to for the items in b to change. | |
1591 | * It is safe to drop the lock on our parent before we | |
1592 | * go through the expensive btree search on b. | |
1593 | * | |
1594 | * If cow is true, then we might be changing slot zero, | |
1595 | * which may require changing the parent. So, we can't | |
1596 | * drop the lock until after we know which slot we're | |
1597 | * operating on. | |
1598 | */ | |
1599 | if (!cow) | |
1600 | btrfs_unlock_up_safe(p, level + 1); | |
1601 | ||
123abc88 | 1602 | ret = check_block(root, p, level); |
65b51a00 CM |
1603 | if (ret) { |
1604 | ret = -1; | |
1605 | goto done; | |
1606 | } | |
925baedd | 1607 | |
5f39d397 | 1608 | ret = bin_search(b, key, level, &slot); |
b4ce94de | 1609 | |
5f39d397 | 1610 | if (level != 0) { |
be0e5c09 CM |
1611 | if (ret && slot > 0) |
1612 | slot -= 1; | |
1613 | p->slots[level] = slot; | |
459931ec CM |
1614 | if ((p->search_for_split || ins_len > 0) && |
1615 | btrfs_header_nritems(b) >= | |
1514794e | 1616 | BTRFS_NODEPTRS_PER_BLOCK(root) - 3) { |
b4ce94de CM |
1617 | int sret; |
1618 | ||
1619 | sret = reada_for_balance(root, p, level); | |
1620 | if (sret) | |
1621 | goto again; | |
1622 | ||
1623 | btrfs_set_path_blocking(p); | |
1624 | sret = split_node(trans, root, p, level); | |
4008c04a | 1625 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de | 1626 | |
5c680ed6 | 1627 | BUG_ON(sret > 0); |
65b51a00 CM |
1628 | if (sret) { |
1629 | ret = sret; | |
1630 | goto done; | |
1631 | } | |
5c680ed6 | 1632 | b = p->nodes[level]; |
5c680ed6 | 1633 | slot = p->slots[level]; |
7b78c170 CM |
1634 | } else if (ins_len < 0 && |
1635 | btrfs_header_nritems(b) < | |
1636 | BTRFS_NODEPTRS_PER_BLOCK(root) / 4) { | |
b4ce94de CM |
1637 | int sret; |
1638 | ||
1639 | sret = reada_for_balance(root, p, level); | |
1640 | if (sret) | |
1641 | goto again; | |
1642 | ||
1643 | btrfs_set_path_blocking(p); | |
1644 | sret = balance_level(trans, root, p, level); | |
4008c04a | 1645 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de | 1646 | |
65b51a00 CM |
1647 | if (sret) { |
1648 | ret = sret; | |
1649 | goto done; | |
1650 | } | |
bb803951 | 1651 | b = p->nodes[level]; |
f510cfec CM |
1652 | if (!b) { |
1653 | btrfs_release_path(NULL, p); | |
bb803951 | 1654 | goto again; |
f510cfec | 1655 | } |
bb803951 | 1656 | slot = p->slots[level]; |
5f39d397 | 1657 | BUG_ON(btrfs_header_nritems(b) == 1); |
5c680ed6 | 1658 | } |
f9efa9c7 CM |
1659 | unlock_up(p, level, lowest_unlock); |
1660 | ||
9f3a7427 | 1661 | /* this is only true while dropping a snapshot */ |
925baedd | 1662 | if (level == lowest_level) { |
5b21f2ed ZY |
1663 | ret = 0; |
1664 | goto done; | |
925baedd | 1665 | } |
ca7a79ad | 1666 | |
594a24eb CM |
1667 | blocknr = btrfs_node_blockptr(b, slot); |
1668 | gen = btrfs_node_ptr_generation(b, slot); | |
1669 | blocksize = btrfs_level_size(root, level - 1); | |
1670 | ||
1671 | tmp = btrfs_find_tree_block(root, blocknr, blocksize); | |
1672 | if (tmp && btrfs_buffer_uptodate(tmp, gen)) { | |
051e1b9f CM |
1673 | b = tmp; |
1674 | } else { | |
1675 | /* | |
1676 | * reduce lock contention at high levels | |
1677 | * of the btree by dropping locks before | |
1678 | * we read. | |
1679 | */ | |
b4ce94de | 1680 | if (level > 0) { |
051e1b9f | 1681 | btrfs_release_path(NULL, p); |
594a24eb CM |
1682 | if (tmp) |
1683 | free_extent_buffer(tmp); | |
f9efa9c7 CM |
1684 | if (should_reada) |
1685 | reada_for_search(root, p, | |
1686 | level, slot, | |
1687 | key->objectid); | |
1688 | ||
594a24eb CM |
1689 | tmp = read_tree_block(root, blocknr, |
1690 | blocksize, gen); | |
051e1b9f CM |
1691 | if (tmp) |
1692 | free_extent_buffer(tmp); | |
1693 | goto again; | |
1694 | } else { | |
b4ce94de | 1695 | btrfs_set_path_blocking(p); |
a74a4b97 CM |
1696 | if (tmp) |
1697 | free_extent_buffer(tmp); | |
f9efa9c7 CM |
1698 | if (should_reada) |
1699 | reada_for_search(root, p, | |
1700 | level, slot, | |
1701 | key->objectid); | |
051e1b9f CM |
1702 | b = read_node_slot(root, b, slot); |
1703 | } | |
1704 | } | |
b4ce94de CM |
1705 | if (!p->skip_locking) { |
1706 | int lret; | |
1707 | ||
4008c04a | 1708 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de CM |
1709 | lret = btrfs_try_spin_lock(b); |
1710 | ||
1711 | if (!lret) { | |
1712 | btrfs_set_path_blocking(p); | |
1713 | btrfs_tree_lock(b); | |
4008c04a | 1714 | btrfs_clear_path_blocking(p, b); |
b4ce94de CM |
1715 | } |
1716 | } | |
be0e5c09 CM |
1717 | } else { |
1718 | p->slots[level] = slot; | |
87b29b20 YZ |
1719 | if (ins_len > 0 && |
1720 | btrfs_leaf_free_space(root, b) < ins_len) { | |
b4ce94de CM |
1721 | int sret; |
1722 | ||
1723 | btrfs_set_path_blocking(p); | |
1724 | sret = split_leaf(trans, root, key, | |
cc0c5538 | 1725 | p, ins_len, ret == 0); |
4008c04a | 1726 | btrfs_clear_path_blocking(p, NULL); |
b4ce94de | 1727 | |
5c680ed6 | 1728 | BUG_ON(sret > 0); |
65b51a00 CM |
1729 | if (sret) { |
1730 | ret = sret; | |
1731 | goto done; | |
1732 | } | |
5c680ed6 | 1733 | } |
459931ec CM |
1734 | if (!p->search_for_split) |
1735 | unlock_up(p, level, lowest_unlock); | |
65b51a00 | 1736 | goto done; |
be0e5c09 CM |
1737 | } |
1738 | } | |
65b51a00 CM |
1739 | ret = 1; |
1740 | done: | |
b4ce94de CM |
1741 | /* |
1742 | * we don't really know what they plan on doing with the path | |
1743 | * from here on, so for now just mark it as blocking | |
1744 | */ | |
1745 | btrfs_set_path_blocking(p); | |
65b51a00 CM |
1746 | if (prealloc_block.objectid) { |
1747 | btrfs_free_reserved_extent(root, | |
1748 | prealloc_block.objectid, | |
1749 | prealloc_block.offset); | |
1750 | } | |
65b51a00 | 1751 | return ret; |
be0e5c09 CM |
1752 | } |
1753 | ||
1a40e23b ZY |
1754 | int btrfs_merge_path(struct btrfs_trans_handle *trans, |
1755 | struct btrfs_root *root, | |
1756 | struct btrfs_key *node_keys, | |
1757 | u64 *nodes, int lowest_level) | |
1758 | { | |
1759 | struct extent_buffer *eb; | |
1760 | struct extent_buffer *parent; | |
1761 | struct btrfs_key key; | |
1762 | u64 bytenr; | |
1763 | u64 generation; | |
1764 | u32 blocksize; | |
1765 | int level; | |
1766 | int slot; | |
1767 | int key_match; | |
1768 | int ret; | |
1769 | ||
1770 | eb = btrfs_lock_root_node(root); | |
1771 | ret = btrfs_cow_block(trans, root, eb, NULL, 0, &eb, 0); | |
1772 | BUG_ON(ret); | |
1773 | ||
b4ce94de CM |
1774 | btrfs_set_lock_blocking(eb); |
1775 | ||
1a40e23b ZY |
1776 | parent = eb; |
1777 | while (1) { | |
1778 | level = btrfs_header_level(parent); | |
1779 | if (level == 0 || level <= lowest_level) | |
1780 | break; | |
1781 | ||
1782 | ret = bin_search(parent, &node_keys[lowest_level], level, | |
1783 | &slot); | |
1784 | if (ret && slot > 0) | |
1785 | slot--; | |
1786 | ||
1787 | bytenr = btrfs_node_blockptr(parent, slot); | |
1788 | if (nodes[level - 1] == bytenr) | |
1789 | break; | |
1790 | ||
1791 | blocksize = btrfs_level_size(root, level - 1); | |
1792 | generation = btrfs_node_ptr_generation(parent, slot); | |
1793 | btrfs_node_key_to_cpu(eb, &key, slot); | |
1794 | key_match = !memcmp(&key, &node_keys[level - 1], sizeof(key)); | |
1795 | ||
f82d02d9 YZ |
1796 | if (generation == trans->transid) { |
1797 | eb = read_tree_block(root, bytenr, blocksize, | |
1798 | generation); | |
1799 | btrfs_tree_lock(eb); | |
b4ce94de | 1800 | btrfs_set_lock_blocking(eb); |
f82d02d9 YZ |
1801 | } |
1802 | ||
1a40e23b ZY |
1803 | /* |
1804 | * if node keys match and node pointer hasn't been modified | |
1805 | * in the running transaction, we can merge the path. for | |
f82d02d9 YZ |
1806 | * blocks owened by reloc trees, the node pointer check is |
1807 | * skipped, this is because these blocks are fully controlled | |
1808 | * by the space balance code, no one else can modify them. | |
1a40e23b ZY |
1809 | */ |
1810 | if (!nodes[level - 1] || !key_match || | |
1811 | (generation == trans->transid && | |
f82d02d9 YZ |
1812 | btrfs_header_owner(eb) != BTRFS_TREE_RELOC_OBJECTID)) { |
1813 | if (level == 1 || level == lowest_level + 1) { | |
1814 | if (generation == trans->transid) { | |
1815 | btrfs_tree_unlock(eb); | |
1816 | free_extent_buffer(eb); | |
1817 | } | |
1a40e23b | 1818 | break; |
f82d02d9 | 1819 | } |
1a40e23b | 1820 | |
f82d02d9 YZ |
1821 | if (generation != trans->transid) { |
1822 | eb = read_tree_block(root, bytenr, blocksize, | |
1823 | generation); | |
1824 | btrfs_tree_lock(eb); | |
b4ce94de | 1825 | btrfs_set_lock_blocking(eb); |
f82d02d9 | 1826 | } |
1a40e23b ZY |
1827 | |
1828 | ret = btrfs_cow_block(trans, root, eb, parent, slot, | |
1829 | &eb, 0); | |
1830 | BUG_ON(ret); | |
1831 | ||
f82d02d9 YZ |
1832 | if (root->root_key.objectid == |
1833 | BTRFS_TREE_RELOC_OBJECTID) { | |
1834 | if (!nodes[level - 1]) { | |
1835 | nodes[level - 1] = eb->start; | |
1836 | memcpy(&node_keys[level - 1], &key, | |
1837 | sizeof(node_keys[0])); | |
1838 | } else { | |
1839 | WARN_ON(1); | |
1840 | } | |
1841 | } | |
1842 | ||
1a40e23b ZY |
1843 | btrfs_tree_unlock(parent); |
1844 | free_extent_buffer(parent); | |
1845 | parent = eb; | |
1846 | continue; | |
1847 | } | |
1848 | ||
1a40e23b ZY |
1849 | btrfs_set_node_blockptr(parent, slot, nodes[level - 1]); |
1850 | btrfs_set_node_ptr_generation(parent, slot, trans->transid); | |
1851 | btrfs_mark_buffer_dirty(parent); | |
1852 | ||
1853 | ret = btrfs_inc_extent_ref(trans, root, | |
1854 | nodes[level - 1], | |
1855 | blocksize, parent->start, | |
1856 | btrfs_header_owner(parent), | |
1857 | btrfs_header_generation(parent), | |
3bb1a1bc | 1858 | level - 1); |
1a40e23b | 1859 | BUG_ON(ret); |
1a40e23b | 1860 | |
f82d02d9 YZ |
1861 | /* |
1862 | * If the block was created in the running transaction, | |
1863 | * it's possible this is the last reference to it, so we | |
1864 | * should drop the subtree. | |
1865 | */ | |
1a40e23b | 1866 | if (generation == trans->transid) { |
f82d02d9 YZ |
1867 | ret = btrfs_drop_subtree(trans, root, eb, parent); |
1868 | BUG_ON(ret); | |
1a40e23b ZY |
1869 | btrfs_tree_unlock(eb); |
1870 | free_extent_buffer(eb); | |
f82d02d9 YZ |
1871 | } else { |
1872 | ret = btrfs_free_extent(trans, root, bytenr, | |
1873 | blocksize, parent->start, | |
1874 | btrfs_header_owner(parent), | |
1875 | btrfs_header_generation(parent), | |
1876 | level - 1, 1); | |
1877 | BUG_ON(ret); | |
1a40e23b ZY |
1878 | } |
1879 | break; | |
1880 | } | |
1881 | btrfs_tree_unlock(parent); | |
1882 | free_extent_buffer(parent); | |
1883 | return 0; | |
1884 | } | |
1885 | ||
74123bd7 CM |
1886 | /* |
1887 | * adjust the pointers going up the tree, starting at level | |
1888 | * making sure the right key of each node is points to 'key'. | |
1889 | * This is used after shifting pointers to the left, so it stops | |
1890 | * fixing up pointers when a given leaf/node is not in slot 0 of the | |
1891 | * higher levels | |
aa5d6bed CM |
1892 | * |
1893 | * If this fails to write a tree block, it returns -1, but continues | |
1894 | * fixing up the blocks in ram so the tree is consistent. | |
74123bd7 | 1895 | */ |
5f39d397 CM |
1896 | static int fixup_low_keys(struct btrfs_trans_handle *trans, |
1897 | struct btrfs_root *root, struct btrfs_path *path, | |
1898 | struct btrfs_disk_key *key, int level) | |
be0e5c09 CM |
1899 | { |
1900 | int i; | |
aa5d6bed | 1901 | int ret = 0; |
5f39d397 CM |
1902 | struct extent_buffer *t; |
1903 | ||
234b63a0 | 1904 | for (i = level; i < BTRFS_MAX_LEVEL; i++) { |
be0e5c09 | 1905 | int tslot = path->slots[i]; |
eb60ceac | 1906 | if (!path->nodes[i]) |
be0e5c09 | 1907 | break; |
5f39d397 CM |
1908 | t = path->nodes[i]; |
1909 | btrfs_set_node_key(t, key, tslot); | |
d6025579 | 1910 | btrfs_mark_buffer_dirty(path->nodes[i]); |
be0e5c09 CM |
1911 | if (tslot != 0) |
1912 | break; | |
1913 | } | |
aa5d6bed | 1914 | return ret; |
be0e5c09 CM |
1915 | } |
1916 | ||
31840ae1 ZY |
1917 | /* |
1918 | * update item key. | |
1919 | * | |
1920 | * This function isn't completely safe. It's the caller's responsibility | |
1921 | * that the new key won't break the order | |
1922 | */ | |
1923 | int btrfs_set_item_key_safe(struct btrfs_trans_handle *trans, | |
1924 | struct btrfs_root *root, struct btrfs_path *path, | |
1925 | struct btrfs_key *new_key) | |
1926 | { | |
1927 | struct btrfs_disk_key disk_key; | |
1928 | struct extent_buffer *eb; | |
1929 | int slot; | |
1930 | ||
1931 | eb = path->nodes[0]; | |
1932 | slot = path->slots[0]; | |
1933 | if (slot > 0) { | |
1934 | btrfs_item_key(eb, &disk_key, slot - 1); | |
1935 | if (comp_keys(&disk_key, new_key) >= 0) | |
1936 | return -1; | |
1937 | } | |
1938 | if (slot < btrfs_header_nritems(eb) - 1) { | |
1939 | btrfs_item_key(eb, &disk_key, slot + 1); | |
1940 | if (comp_keys(&disk_key, new_key) <= 0) | |
1941 | return -1; | |
1942 | } | |
1943 | ||
1944 | btrfs_cpu_key_to_disk(&disk_key, new_key); | |
1945 | btrfs_set_item_key(eb, &disk_key, slot); | |
1946 | btrfs_mark_buffer_dirty(eb); | |
1947 | if (slot == 0) | |
1948 | fixup_low_keys(trans, root, path, &disk_key, 1); | |
1949 | return 0; | |
1950 | } | |
1951 | ||
74123bd7 CM |
1952 | /* |
1953 | * try to push data from one node into the next node left in the | |
79f95c82 | 1954 | * tree. |
aa5d6bed CM |
1955 | * |
1956 | * returns 0 if some ptrs were pushed left, < 0 if there was some horrible | |
1957 | * error, and > 0 if there was no room in the left hand block. | |
74123bd7 | 1958 | */ |
98ed5174 CM |
1959 | static int push_node_left(struct btrfs_trans_handle *trans, |
1960 | struct btrfs_root *root, struct extent_buffer *dst, | |
971a1f66 | 1961 | struct extent_buffer *src, int empty) |
be0e5c09 | 1962 | { |
be0e5c09 | 1963 | int push_items = 0; |
bb803951 CM |
1964 | int src_nritems; |
1965 | int dst_nritems; | |
aa5d6bed | 1966 | int ret = 0; |
be0e5c09 | 1967 | |
5f39d397 CM |
1968 | src_nritems = btrfs_header_nritems(src); |
1969 | dst_nritems = btrfs_header_nritems(dst); | |
123abc88 | 1970 | push_items = BTRFS_NODEPTRS_PER_BLOCK(root) - dst_nritems; |
7bb86316 CM |
1971 | WARN_ON(btrfs_header_generation(src) != trans->transid); |
1972 | WARN_ON(btrfs_header_generation(dst) != trans->transid); | |
54aa1f4d | 1973 | |
bce4eae9 | 1974 | if (!empty && src_nritems <= 8) |
971a1f66 CM |
1975 | return 1; |
1976 | ||
d397712b | 1977 | if (push_items <= 0) |
be0e5c09 CM |
1978 | return 1; |
1979 | ||
bce4eae9 | 1980 | if (empty) { |
971a1f66 | 1981 | push_items = min(src_nritems, push_items); |
bce4eae9 CM |
1982 | if (push_items < src_nritems) { |
1983 | /* leave at least 8 pointers in the node if | |
1984 | * we aren't going to empty it | |
1985 | */ | |
1986 | if (src_nritems - push_items < 8) { | |
1987 | if (push_items <= 8) | |
1988 | return 1; | |
1989 | push_items -= 8; | |
1990 | } | |
1991 | } | |
1992 | } else | |
1993 | push_items = min(src_nritems - 8, push_items); | |
79f95c82 | 1994 | |
5f39d397 CM |
1995 | copy_extent_buffer(dst, src, |
1996 | btrfs_node_key_ptr_offset(dst_nritems), | |
1997 | btrfs_node_key_ptr_offset(0), | |
d397712b | 1998 | push_items * sizeof(struct btrfs_key_ptr)); |
5f39d397 | 1999 | |
bb803951 | 2000 | if (push_items < src_nritems) { |
5f39d397 CM |
2001 | memmove_extent_buffer(src, btrfs_node_key_ptr_offset(0), |
2002 | btrfs_node_key_ptr_offset(push_items), | |
2003 | (src_nritems - push_items) * | |
2004 | sizeof(struct btrfs_key_ptr)); | |
2005 | } | |
2006 | btrfs_set_header_nritems(src, src_nritems - push_items); | |
2007 | btrfs_set_header_nritems(dst, dst_nritems + push_items); | |
2008 | btrfs_mark_buffer_dirty(src); | |
2009 | btrfs_mark_buffer_dirty(dst); | |
31840ae1 ZY |
2010 | |
2011 | ret = btrfs_update_ref(trans, root, src, dst, dst_nritems, push_items); | |
2012 | BUG_ON(ret); | |
2013 | ||
79f95c82 CM |
2014 | return ret; |
2015 | } | |
2016 | ||
2017 | /* | |
2018 | * try to push data from one node into the next node right in the | |
2019 | * tree. | |
2020 | * | |
2021 | * returns 0 if some ptrs were pushed, < 0 if there was some horrible | |
2022 | * error, and > 0 if there was no room in the right hand block. | |
2023 | * | |
2024 | * this will only push up to 1/2 the contents of the left node over | |
2025 | */ | |
5f39d397 CM |
2026 | static int balance_node_right(struct btrfs_trans_handle *trans, |
2027 | struct btrfs_root *root, | |
2028 | struct extent_buffer *dst, | |
2029 | struct extent_buffer *src) | |
79f95c82 | 2030 | { |
79f95c82 CM |
2031 | int push_items = 0; |
2032 | int max_push; | |
2033 | int src_nritems; | |
2034 | int dst_nritems; | |
2035 | int ret = 0; | |
79f95c82 | 2036 | |
7bb86316 CM |
2037 | WARN_ON(btrfs_header_generation(src) != trans->transid); |
2038 | WARN_ON(btrfs_header_generation(dst) != trans->transid); | |
2039 | ||
5f39d397 CM |
2040 | src_nritems = btrfs_header_nritems(src); |
2041 | dst_nritems = btrfs_header_nritems(dst); | |
123abc88 | 2042 | push_items = BTRFS_NODEPTRS_PER_BLOCK(root) - dst_nritems; |
d397712b | 2043 | if (push_items <= 0) |
79f95c82 | 2044 | return 1; |
bce4eae9 | 2045 | |
d397712b | 2046 | if (src_nritems < 4) |
bce4eae9 | 2047 | return 1; |
79f95c82 CM |
2048 | |
2049 | max_push = src_nritems / 2 + 1; | |
2050 | /* don't try to empty the node */ | |
d397712b | 2051 | if (max_push >= src_nritems) |
79f95c82 | 2052 | return 1; |
252c38f0 | 2053 | |
79f95c82 CM |
2054 | if (max_push < push_items) |
2055 | push_items = max_push; | |
2056 | ||
5f39d397 CM |
2057 | memmove_extent_buffer(dst, btrfs_node_key_ptr_offset(push_items), |
2058 | btrfs_node_key_ptr_offset(0), | |
2059 | (dst_nritems) * | |
2060 | sizeof(struct btrfs_key_ptr)); | |
d6025579 | 2061 | |
5f39d397 CM |
2062 | copy_extent_buffer(dst, src, |
2063 | btrfs_node_key_ptr_offset(0), | |
2064 | btrfs_node_key_ptr_offset(src_nritems - push_items), | |
d397712b | 2065 | push_items * sizeof(struct btrfs_key_ptr)); |
79f95c82 | 2066 | |
5f39d397 CM |
2067 | btrfs_set_header_nritems(src, src_nritems - push_items); |
2068 | btrfs_set_header_nritems(dst, dst_nritems + push_items); | |
79f95c82 | 2069 | |
5f39d397 CM |
2070 | btrfs_mark_buffer_dirty(src); |
2071 | btrfs_mark_buffer_dirty(dst); | |
31840ae1 ZY |
2072 | |
2073 | ret = btrfs_update_ref(trans, root, src, dst, 0, push_items); | |
2074 | BUG_ON(ret); | |
2075 | ||
aa5d6bed | 2076 | return ret; |
be0e5c09 CM |
2077 | } |
2078 | ||
97571fd0 CM |
2079 | /* |
2080 | * helper function to insert a new root level in the tree. | |
2081 | * A new node is allocated, and a single item is inserted to | |
2082 | * point to the existing root | |
aa5d6bed CM |
2083 | * |
2084 | * returns zero on success or < 0 on failure. | |
97571fd0 | 2085 | */ |
d397712b | 2086 | static noinline int insert_new_root(struct btrfs_trans_handle *trans, |
5f39d397 CM |
2087 | struct btrfs_root *root, |
2088 | struct btrfs_path *path, int level) | |
5c680ed6 | 2089 | { |
7bb86316 | 2090 | u64 lower_gen; |
5f39d397 CM |
2091 | struct extent_buffer *lower; |
2092 | struct extent_buffer *c; | |
925baedd | 2093 | struct extent_buffer *old; |
5f39d397 | 2094 | struct btrfs_disk_key lower_key; |
31840ae1 | 2095 | int ret; |
5c680ed6 CM |
2096 | |
2097 | BUG_ON(path->nodes[level]); | |
2098 | BUG_ON(path->nodes[level-1] != root->node); | |
2099 | ||
7bb86316 CM |
2100 | lower = path->nodes[level-1]; |
2101 | if (level == 1) | |
2102 | btrfs_item_key(lower, &lower_key, 0); | |
2103 | else | |
2104 | btrfs_node_key(lower, &lower_key, 0); | |
2105 | ||
31840ae1 ZY |
2106 | c = btrfs_alloc_free_block(trans, root, root->nodesize, 0, |
2107 | root->root_key.objectid, trans->transid, | |
ad3d81ba | 2108 | level, root->node->start, 0); |
5f39d397 CM |
2109 | if (IS_ERR(c)) |
2110 | return PTR_ERR(c); | |
925baedd | 2111 | |
5f39d397 CM |
2112 | memset_extent_buffer(c, 0, 0, root->nodesize); |
2113 | btrfs_set_header_nritems(c, 1); | |
2114 | btrfs_set_header_level(c, level); | |
db94535d | 2115 | btrfs_set_header_bytenr(c, c->start); |
5f39d397 CM |
2116 | btrfs_set_header_generation(c, trans->transid); |
2117 | btrfs_set_header_owner(c, root->root_key.objectid); | |
5f39d397 CM |
2118 | |
2119 | write_extent_buffer(c, root->fs_info->fsid, | |
2120 | (unsigned long)btrfs_header_fsid(c), | |
2121 | BTRFS_FSID_SIZE); | |
e17cade2 CM |
2122 | |
2123 | write_extent_buffer(c, root->fs_info->chunk_tree_uuid, | |
2124 | (unsigned long)btrfs_header_chunk_tree_uuid(c), | |
2125 | BTRFS_UUID_SIZE); | |
2126 | ||
5f39d397 | 2127 | btrfs_set_node_key(c, &lower_key, 0); |
db94535d | 2128 | btrfs_set_node_blockptr(c, 0, lower->start); |
7bb86316 | 2129 | lower_gen = btrfs_header_generation(lower); |
31840ae1 | 2130 | WARN_ON(lower_gen != trans->transid); |
7bb86316 CM |
2131 | |
2132 | btrfs_set_node_ptr_generation(c, 0, lower_gen); | |
d5719762 | 2133 | |
5f39d397 | 2134 | btrfs_mark_buffer_dirty(c); |
d5719762 | 2135 | |
925baedd CM |
2136 | spin_lock(&root->node_lock); |
2137 | old = root->node; | |
5f39d397 | 2138 | root->node = c; |
925baedd CM |
2139 | spin_unlock(&root->node_lock); |
2140 | ||
31840ae1 ZY |
2141 | ret = btrfs_update_extent_ref(trans, root, lower->start, |
2142 | lower->start, c->start, | |
2143 | root->root_key.objectid, | |
3bb1a1bc | 2144 | trans->transid, level - 1); |
31840ae1 ZY |
2145 | BUG_ON(ret); |
2146 | ||
925baedd CM |
2147 | /* the super has an extra ref to root->node */ |
2148 | free_extent_buffer(old); | |
2149 | ||
0b86a832 | 2150 | add_root_to_dirty_list(root); |
5f39d397 CM |
2151 | extent_buffer_get(c); |
2152 | path->nodes[level] = c; | |
925baedd | 2153 | path->locks[level] = 1; |
5c680ed6 CM |
2154 | path->slots[level] = 0; |
2155 | return 0; | |
2156 | } | |
2157 | ||
74123bd7 CM |
2158 | /* |
2159 | * worker function to insert a single pointer in a node. | |
2160 | * the node should have enough room for the pointer already | |
97571fd0 | 2161 | * |
74123bd7 CM |
2162 | * slot and level indicate where you want the key to go, and |
2163 | * blocknr is the block the key points to. | |
aa5d6bed CM |
2164 | * |
2165 | * returns zero on success and < 0 on any error | |
74123bd7 | 2166 | */ |
e089f05c CM |
2167 | static int insert_ptr(struct btrfs_trans_handle *trans, struct btrfs_root |
2168 | *root, struct btrfs_path *path, struct btrfs_disk_key | |
db94535d | 2169 | *key, u64 bytenr, int slot, int level) |
74123bd7 | 2170 | { |
5f39d397 | 2171 | struct extent_buffer *lower; |
74123bd7 | 2172 | int nritems; |
5c680ed6 CM |
2173 | |
2174 | BUG_ON(!path->nodes[level]); | |
5f39d397 CM |
2175 | lower = path->nodes[level]; |
2176 | nritems = btrfs_header_nritems(lower); | |
74123bd7 CM |
2177 | if (slot > nritems) |
2178 | BUG(); | |
123abc88 | 2179 | if (nritems == BTRFS_NODEPTRS_PER_BLOCK(root)) |
74123bd7 CM |
2180 | BUG(); |
2181 | if (slot != nritems) { | |
5f39d397 CM |
2182 | memmove_extent_buffer(lower, |
2183 | btrfs_node_key_ptr_offset(slot + 1), | |
2184 | btrfs_node_key_ptr_offset(slot), | |
d6025579 | 2185 | (nritems - slot) * sizeof(struct btrfs_key_ptr)); |
74123bd7 | 2186 | } |
5f39d397 | 2187 | btrfs_set_node_key(lower, key, slot); |
db94535d | 2188 | btrfs_set_node_blockptr(lower, slot, bytenr); |
74493f7a CM |
2189 | WARN_ON(trans->transid == 0); |
2190 | btrfs_set_node_ptr_generation(lower, slot, trans->transid); | |
5f39d397 CM |
2191 | btrfs_set_header_nritems(lower, nritems + 1); |
2192 | btrfs_mark_buffer_dirty(lower); | |
74123bd7 CM |
2193 | return 0; |
2194 | } | |
2195 | ||
97571fd0 CM |
2196 | /* |
2197 | * split the node at the specified level in path in two. | |
2198 | * The path is corrected to point to the appropriate node after the split | |
2199 | * | |
2200 | * Before splitting this tries to make some room in the node by pushing | |
2201 | * left and right, if either one works, it returns right away. | |
aa5d6bed CM |
2202 | * |
2203 | * returns 0 on success and < 0 on failure | |
97571fd0 | 2204 | */ |
e02119d5 CM |
2205 | static noinline int split_node(struct btrfs_trans_handle *trans, |
2206 | struct btrfs_root *root, | |
2207 | struct btrfs_path *path, int level) | |
be0e5c09 | 2208 | { |
5f39d397 CM |
2209 | struct extent_buffer *c; |
2210 | struct extent_buffer *split; | |
2211 | struct btrfs_disk_key disk_key; | |
be0e5c09 | 2212 | int mid; |
5c680ed6 | 2213 | int ret; |
aa5d6bed | 2214 | int wret; |
7518a238 | 2215 | u32 c_nritems; |
eb60ceac | 2216 | |
5f39d397 | 2217 | c = path->nodes[level]; |
7bb86316 | 2218 | WARN_ON(btrfs_header_generation(c) != trans->transid); |
5f39d397 | 2219 | if (c == root->node) { |
5c680ed6 | 2220 | /* trying to split the root, lets make a new one */ |
e089f05c | 2221 | ret = insert_new_root(trans, root, path, level + 1); |
5c680ed6 CM |
2222 | if (ret) |
2223 | return ret; | |
e66f709b CM |
2224 | } else { |
2225 | ret = push_nodes_for_insert(trans, root, path, level); | |
5f39d397 CM |
2226 | c = path->nodes[level]; |
2227 | if (!ret && btrfs_header_nritems(c) < | |
c448acf0 | 2228 | BTRFS_NODEPTRS_PER_BLOCK(root) - 3) |
e66f709b | 2229 | return 0; |
54aa1f4d CM |
2230 | if (ret < 0) |
2231 | return ret; | |
be0e5c09 | 2232 | } |
e66f709b | 2233 | |
5f39d397 | 2234 | c_nritems = btrfs_header_nritems(c); |
7bb86316 | 2235 | |
925baedd | 2236 | split = btrfs_alloc_free_block(trans, root, root->nodesize, |
31840ae1 ZY |
2237 | path->nodes[level + 1]->start, |
2238 | root->root_key.objectid, | |
2239 | trans->transid, level, c->start, 0); | |
5f39d397 CM |
2240 | if (IS_ERR(split)) |
2241 | return PTR_ERR(split); | |
2242 | ||
2243 | btrfs_set_header_flags(split, btrfs_header_flags(c)); | |
2244 | btrfs_set_header_level(split, btrfs_header_level(c)); | |
db94535d | 2245 | btrfs_set_header_bytenr(split, split->start); |
5f39d397 CM |
2246 | btrfs_set_header_generation(split, trans->transid); |
2247 | btrfs_set_header_owner(split, root->root_key.objectid); | |
63b10fc4 | 2248 | btrfs_set_header_flags(split, 0); |
5f39d397 CM |
2249 | write_extent_buffer(split, root->fs_info->fsid, |
2250 | (unsigned long)btrfs_header_fsid(split), | |
2251 | BTRFS_FSID_SIZE); | |
e17cade2 CM |
2252 | write_extent_buffer(split, root->fs_info->chunk_tree_uuid, |
2253 | (unsigned long)btrfs_header_chunk_tree_uuid(split), | |
2254 | BTRFS_UUID_SIZE); | |
54aa1f4d | 2255 | |
7518a238 | 2256 | mid = (c_nritems + 1) / 2; |
5f39d397 CM |
2257 | |
2258 | copy_extent_buffer(split, c, | |
2259 | btrfs_node_key_ptr_offset(0), | |
2260 | btrfs_node_key_ptr_offset(mid), | |
2261 | (c_nritems - mid) * sizeof(struct btrfs_key_ptr)); | |
2262 | btrfs_set_header_nritems(split, c_nritems - mid); | |
2263 | btrfs_set_header_nritems(c, mid); | |
aa5d6bed CM |
2264 | ret = 0; |
2265 | ||
5f39d397 CM |
2266 | btrfs_mark_buffer_dirty(c); |
2267 | btrfs_mark_buffer_dirty(split); | |
2268 | ||
2269 | btrfs_node_key(split, &disk_key, 0); | |
db94535d | 2270 | wret = insert_ptr(trans, root, path, &disk_key, split->start, |
5f39d397 | 2271 | path->slots[level + 1] + 1, |
123abc88 | 2272 | level + 1); |
aa5d6bed CM |
2273 | if (wret) |
2274 | ret = wret; | |
2275 | ||
31840ae1 ZY |
2276 | ret = btrfs_update_ref(trans, root, c, split, 0, c_nritems - mid); |
2277 | BUG_ON(ret); | |
2278 | ||
5de08d7d | 2279 | if (path->slots[level] >= mid) { |
5c680ed6 | 2280 | path->slots[level] -= mid; |
925baedd | 2281 | btrfs_tree_unlock(c); |
5f39d397 CM |
2282 | free_extent_buffer(c); |
2283 | path->nodes[level] = split; | |
5c680ed6 CM |
2284 | path->slots[level + 1] += 1; |
2285 | } else { | |
925baedd | 2286 | btrfs_tree_unlock(split); |
5f39d397 | 2287 | free_extent_buffer(split); |
be0e5c09 | 2288 | } |
aa5d6bed | 2289 | return ret; |
be0e5c09 CM |
2290 | } |
2291 | ||
74123bd7 CM |
2292 | /* |
2293 | * how many bytes are required to store the items in a leaf. start | |
2294 | * and nr indicate which items in the leaf to check. This totals up the | |
2295 | * space used both by the item structs and the item data | |
2296 | */ | |
5f39d397 | 2297 | static int leaf_space_used(struct extent_buffer *l, int start, int nr) |
be0e5c09 CM |
2298 | { |
2299 | int data_len; | |
5f39d397 | 2300 | int nritems = btrfs_header_nritems(l); |
d4dbff95 | 2301 | int end = min(nritems, start + nr) - 1; |
be0e5c09 CM |
2302 | |
2303 | if (!nr) | |
2304 | return 0; | |
5f39d397 CM |
2305 | data_len = btrfs_item_end_nr(l, start); |
2306 | data_len = data_len - btrfs_item_offset_nr(l, end); | |
0783fcfc | 2307 | data_len += sizeof(struct btrfs_item) * nr; |
d4dbff95 | 2308 | WARN_ON(data_len < 0); |
be0e5c09 CM |
2309 | return data_len; |
2310 | } | |
2311 | ||
d4dbff95 CM |
2312 | /* |
2313 | * The space between the end of the leaf items and | |
2314 | * the start of the leaf data. IOW, how much room | |
2315 | * the leaf has left for both items and data | |
2316 | */ | |
d397712b | 2317 | noinline int btrfs_leaf_free_space(struct btrfs_root *root, |
e02119d5 | 2318 | struct extent_buffer *leaf) |
d4dbff95 | 2319 | { |
5f39d397 CM |
2320 | int nritems = btrfs_header_nritems(leaf); |
2321 | int ret; | |
2322 | ret = BTRFS_LEAF_DATA_SIZE(root) - leaf_space_used(leaf, 0, nritems); | |
2323 | if (ret < 0) { | |
d397712b CM |
2324 | printk(KERN_CRIT "leaf free space ret %d, leaf data size %lu, " |
2325 | "used %d nritems %d\n", | |
ae2f5411 | 2326 | ret, (unsigned long) BTRFS_LEAF_DATA_SIZE(root), |
5f39d397 CM |
2327 | leaf_space_used(leaf, 0, nritems), nritems); |
2328 | } | |
2329 | return ret; | |
d4dbff95 CM |
2330 | } |
2331 | ||
00ec4c51 CM |
2332 | /* |
2333 | * push some data in the path leaf to the right, trying to free up at | |
2334 | * least data_size bytes. returns zero if the push worked, nonzero otherwise | |
aa5d6bed CM |
2335 | * |
2336 | * returns 1 if the push failed because the other node didn't have enough | |
2337 | * room, 0 if everything worked out and < 0 if there were major errors. | |
00ec4c51 | 2338 | */ |
e089f05c | 2339 | static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root |
34a38218 CM |
2340 | *root, struct btrfs_path *path, int data_size, |
2341 | int empty) | |
00ec4c51 | 2342 | { |
5f39d397 CM |
2343 | struct extent_buffer *left = path->nodes[0]; |
2344 | struct extent_buffer *right; | |
2345 | struct extent_buffer *upper; | |
2346 | struct btrfs_disk_key disk_key; | |
00ec4c51 | 2347 | int slot; |
34a38218 | 2348 | u32 i; |
00ec4c51 CM |
2349 | int free_space; |
2350 | int push_space = 0; | |
2351 | int push_items = 0; | |
0783fcfc | 2352 | struct btrfs_item *item; |
7518a238 | 2353 | u32 left_nritems; |
34a38218 | 2354 | u32 nr; |
7518a238 | 2355 | u32 right_nritems; |
5f39d397 | 2356 | u32 data_end; |
db94535d | 2357 | u32 this_item_size; |
54aa1f4d | 2358 | int ret; |
00ec4c51 CM |
2359 | |
2360 | slot = path->slots[1]; | |
d397712b | 2361 | if (!path->nodes[1]) |
00ec4c51 | 2362 | return 1; |
d397712b | 2363 | |
00ec4c51 | 2364 | upper = path->nodes[1]; |
5f39d397 | 2365 | if (slot >= btrfs_header_nritems(upper) - 1) |
00ec4c51 | 2366 | return 1; |
5f39d397 | 2367 | |
a2135011 CM |
2368 | WARN_ON(!btrfs_tree_locked(path->nodes[1])); |
2369 | ||
ca7a79ad | 2370 | right = read_node_slot(root, upper, slot + 1); |
925baedd | 2371 | btrfs_tree_lock(right); |
b4ce94de CM |
2372 | btrfs_set_lock_blocking(right); |
2373 | ||
123abc88 | 2374 | free_space = btrfs_leaf_free_space(root, right); |
87b29b20 | 2375 | if (free_space < data_size) |
925baedd | 2376 | goto out_unlock; |
5f39d397 | 2377 | |
02217ed2 | 2378 | /* cow and double check */ |
5f39d397 | 2379 | ret = btrfs_cow_block(trans, root, right, upper, |
65b51a00 | 2380 | slot + 1, &right, 0); |
925baedd CM |
2381 | if (ret) |
2382 | goto out_unlock; | |
2383 | ||
123abc88 | 2384 | free_space = btrfs_leaf_free_space(root, right); |
87b29b20 | 2385 | if (free_space < data_size) |
925baedd | 2386 | goto out_unlock; |
02217ed2 | 2387 | |
5f39d397 | 2388 | left_nritems = btrfs_header_nritems(left); |
925baedd CM |
2389 | if (left_nritems == 0) |
2390 | goto out_unlock; | |
5f39d397 | 2391 | |
34a38218 CM |
2392 | if (empty) |
2393 | nr = 0; | |
2394 | else | |
2395 | nr = 1; | |
2396 | ||
31840ae1 | 2397 | if (path->slots[0] >= left_nritems) |
87b29b20 | 2398 | push_space += data_size; |
31840ae1 | 2399 | |
34a38218 CM |
2400 | i = left_nritems - 1; |
2401 | while (i >= nr) { | |
5f39d397 | 2402 | item = btrfs_item_nr(left, i); |
db94535d | 2403 | |
31840ae1 ZY |
2404 | if (!empty && push_items > 0) { |
2405 | if (path->slots[0] > i) | |
2406 | break; | |
2407 | if (path->slots[0] == i) { | |
2408 | int space = btrfs_leaf_free_space(root, left); | |
2409 | if (space + push_space * 2 > free_space) | |
2410 | break; | |
2411 | } | |
2412 | } | |
2413 | ||
00ec4c51 | 2414 | if (path->slots[0] == i) |
87b29b20 | 2415 | push_space += data_size; |
db94535d CM |
2416 | |
2417 | if (!left->map_token) { | |
2418 | map_extent_buffer(left, (unsigned long)item, | |
2419 | sizeof(struct btrfs_item), | |
2420 | &left->map_token, &left->kaddr, | |
2421 | &left->map_start, &left->map_len, | |
2422 | KM_USER1); | |
2423 | } | |
2424 | ||
2425 | this_item_size = btrfs_item_size(left, item); | |
2426 | if (this_item_size + sizeof(*item) + push_space > free_space) | |
00ec4c51 | 2427 | break; |
31840ae1 | 2428 | |
00ec4c51 | 2429 | push_items++; |
db94535d | 2430 | push_space += this_item_size + sizeof(*item); |
34a38218 CM |
2431 | if (i == 0) |
2432 | break; | |
2433 | i--; | |
db94535d CM |
2434 | } |
2435 | if (left->map_token) { | |
2436 | unmap_extent_buffer(left, left->map_token, KM_USER1); | |
2437 | left->map_token = NULL; | |
00ec4c51 | 2438 | } |
5f39d397 | 2439 | |
925baedd CM |
2440 | if (push_items == 0) |
2441 | goto out_unlock; | |
5f39d397 | 2442 | |
34a38218 | 2443 | if (!empty && push_items == left_nritems) |
a429e513 | 2444 | WARN_ON(1); |
5f39d397 | 2445 | |
00ec4c51 | 2446 | /* push left to right */ |
5f39d397 | 2447 | right_nritems = btrfs_header_nritems(right); |
34a38218 | 2448 | |
5f39d397 | 2449 | push_space = btrfs_item_end_nr(left, left_nritems - push_items); |
123abc88 | 2450 | push_space -= leaf_data_end(root, left); |
5f39d397 | 2451 | |
00ec4c51 | 2452 | /* make room in the right data area */ |
5f39d397 CM |
2453 | data_end = leaf_data_end(root, right); |
2454 | memmove_extent_buffer(right, | |
2455 | btrfs_leaf_data(right) + data_end - push_space, | |
2456 | btrfs_leaf_data(right) + data_end, | |
2457 | BTRFS_LEAF_DATA_SIZE(root) - data_end); | |
2458 | ||
00ec4c51 | 2459 | /* copy from the left data area */ |
5f39d397 | 2460 | copy_extent_buffer(right, left, btrfs_leaf_data(right) + |
d6025579 CM |
2461 | BTRFS_LEAF_DATA_SIZE(root) - push_space, |
2462 | btrfs_leaf_data(left) + leaf_data_end(root, left), | |
2463 | push_space); | |
5f39d397 CM |
2464 | |
2465 | memmove_extent_buffer(right, btrfs_item_nr_offset(push_items), | |
2466 | btrfs_item_nr_offset(0), | |
2467 | right_nritems * sizeof(struct btrfs_item)); | |
2468 | ||
00ec4c51 | 2469 | /* copy the items from left to right */ |
5f39d397 CM |
2470 | copy_extent_buffer(right, left, btrfs_item_nr_offset(0), |
2471 | btrfs_item_nr_offset(left_nritems - push_items), | |
2472 | push_items * sizeof(struct btrfs_item)); | |
00ec4c51 CM |
2473 | |
2474 | /* update the item pointers */ | |
7518a238 | 2475 | right_nritems += push_items; |
5f39d397 | 2476 | btrfs_set_header_nritems(right, right_nritems); |
123abc88 | 2477 | push_space = BTRFS_LEAF_DATA_SIZE(root); |
7518a238 | 2478 | for (i = 0; i < right_nritems; i++) { |
5f39d397 | 2479 | item = btrfs_item_nr(right, i); |
db94535d CM |
2480 | if (!right->map_token) { |
2481 | map_extent_buffer(right, (unsigned long)item, | |
2482 | sizeof(struct btrfs_item), | |
2483 | &right->map_token, &right->kaddr, | |
2484 | &right->map_start, &right->map_len, | |
2485 | KM_USER1); | |
2486 | } | |
2487 | push_space -= btrfs_item_size(right, item); | |
2488 | btrfs_set_item_offset(right, item, push_space); | |
2489 | } | |
2490 | ||
2491 | if (right->map_token) { | |
2492 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2493 | right->map_token = NULL; | |
00ec4c51 | 2494 | } |
7518a238 | 2495 | left_nritems -= push_items; |
5f39d397 | 2496 | btrfs_set_header_nritems(left, left_nritems); |
00ec4c51 | 2497 | |
34a38218 CM |
2498 | if (left_nritems) |
2499 | btrfs_mark_buffer_dirty(left); | |
5f39d397 | 2500 | btrfs_mark_buffer_dirty(right); |
a429e513 | 2501 | |
31840ae1 ZY |
2502 | ret = btrfs_update_ref(trans, root, left, right, 0, push_items); |
2503 | BUG_ON(ret); | |
2504 | ||
5f39d397 CM |
2505 | btrfs_item_key(right, &disk_key, 0); |
2506 | btrfs_set_node_key(upper, &disk_key, slot + 1); | |
d6025579 | 2507 | btrfs_mark_buffer_dirty(upper); |
02217ed2 | 2508 | |
00ec4c51 | 2509 | /* then fixup the leaf pointer in the path */ |
7518a238 CM |
2510 | if (path->slots[0] >= left_nritems) { |
2511 | path->slots[0] -= left_nritems; | |
925baedd CM |
2512 | if (btrfs_header_nritems(path->nodes[0]) == 0) |
2513 | clean_tree_block(trans, root, path->nodes[0]); | |
2514 | btrfs_tree_unlock(path->nodes[0]); | |
5f39d397 CM |
2515 | free_extent_buffer(path->nodes[0]); |
2516 | path->nodes[0] = right; | |
00ec4c51 CM |
2517 | path->slots[1] += 1; |
2518 | } else { | |
925baedd | 2519 | btrfs_tree_unlock(right); |
5f39d397 | 2520 | free_extent_buffer(right); |
00ec4c51 CM |
2521 | } |
2522 | return 0; | |
925baedd CM |
2523 | |
2524 | out_unlock: | |
2525 | btrfs_tree_unlock(right); | |
2526 | free_extent_buffer(right); | |
2527 | return 1; | |
00ec4c51 | 2528 | } |
925baedd | 2529 | |
74123bd7 CM |
2530 | /* |
2531 | * push some data in the path leaf to the left, trying to free up at | |
2532 | * least data_size bytes. returns zero if the push worked, nonzero otherwise | |
2533 | */ | |
e089f05c | 2534 | static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root |
34a38218 CM |
2535 | *root, struct btrfs_path *path, int data_size, |
2536 | int empty) | |
be0e5c09 | 2537 | { |
5f39d397 CM |
2538 | struct btrfs_disk_key disk_key; |
2539 | struct extent_buffer *right = path->nodes[0]; | |
2540 | struct extent_buffer *left; | |
be0e5c09 CM |
2541 | int slot; |
2542 | int i; | |
2543 | int free_space; | |
2544 | int push_space = 0; | |
2545 | int push_items = 0; | |
0783fcfc | 2546 | struct btrfs_item *item; |
7518a238 | 2547 | u32 old_left_nritems; |
5f39d397 | 2548 | u32 right_nritems; |
34a38218 | 2549 | u32 nr; |
aa5d6bed CM |
2550 | int ret = 0; |
2551 | int wret; | |
db94535d CM |
2552 | u32 this_item_size; |
2553 | u32 old_left_item_size; | |
be0e5c09 CM |
2554 | |
2555 | slot = path->slots[1]; | |
5f39d397 | 2556 | if (slot == 0) |
be0e5c09 | 2557 | return 1; |
5f39d397 | 2558 | if (!path->nodes[1]) |
be0e5c09 | 2559 | return 1; |
5f39d397 | 2560 | |
3685f791 | 2561 | right_nritems = btrfs_header_nritems(right); |
d397712b | 2562 | if (right_nritems == 0) |
3685f791 | 2563 | return 1; |
3685f791 | 2564 | |
a2135011 CM |
2565 | WARN_ON(!btrfs_tree_locked(path->nodes[1])); |
2566 | ||
ca7a79ad | 2567 | left = read_node_slot(root, path->nodes[1], slot - 1); |
925baedd | 2568 | btrfs_tree_lock(left); |
b4ce94de CM |
2569 | btrfs_set_lock_blocking(left); |
2570 | ||
123abc88 | 2571 | free_space = btrfs_leaf_free_space(root, left); |
87b29b20 | 2572 | if (free_space < data_size) { |
925baedd CM |
2573 | ret = 1; |
2574 | goto out; | |
be0e5c09 | 2575 | } |
02217ed2 CM |
2576 | |
2577 | /* cow and double check */ | |
5f39d397 | 2578 | ret = btrfs_cow_block(trans, root, left, |
65b51a00 | 2579 | path->nodes[1], slot - 1, &left, 0); |
54aa1f4d CM |
2580 | if (ret) { |
2581 | /* we hit -ENOSPC, but it isn't fatal here */ | |
925baedd CM |
2582 | ret = 1; |
2583 | goto out; | |
54aa1f4d | 2584 | } |
3685f791 | 2585 | |
123abc88 | 2586 | free_space = btrfs_leaf_free_space(root, left); |
87b29b20 | 2587 | if (free_space < data_size) { |
925baedd CM |
2588 | ret = 1; |
2589 | goto out; | |
02217ed2 CM |
2590 | } |
2591 | ||
34a38218 CM |
2592 | if (empty) |
2593 | nr = right_nritems; | |
2594 | else | |
2595 | nr = right_nritems - 1; | |
2596 | ||
2597 | for (i = 0; i < nr; i++) { | |
5f39d397 | 2598 | item = btrfs_item_nr(right, i); |
db94535d CM |
2599 | if (!right->map_token) { |
2600 | map_extent_buffer(right, (unsigned long)item, | |
2601 | sizeof(struct btrfs_item), | |
2602 | &right->map_token, &right->kaddr, | |
2603 | &right->map_start, &right->map_len, | |
2604 | KM_USER1); | |
2605 | } | |
2606 | ||
31840ae1 ZY |
2607 | if (!empty && push_items > 0) { |
2608 | if (path->slots[0] < i) | |
2609 | break; | |
2610 | if (path->slots[0] == i) { | |
2611 | int space = btrfs_leaf_free_space(root, right); | |
2612 | if (space + push_space * 2 > free_space) | |
2613 | break; | |
2614 | } | |
2615 | } | |
2616 | ||
be0e5c09 | 2617 | if (path->slots[0] == i) |
87b29b20 | 2618 | push_space += data_size; |
db94535d CM |
2619 | |
2620 | this_item_size = btrfs_item_size(right, item); | |
2621 | if (this_item_size + sizeof(*item) + push_space > free_space) | |
be0e5c09 | 2622 | break; |
db94535d | 2623 | |
be0e5c09 | 2624 | push_items++; |
db94535d CM |
2625 | push_space += this_item_size + sizeof(*item); |
2626 | } | |
2627 | ||
2628 | if (right->map_token) { | |
2629 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2630 | right->map_token = NULL; | |
be0e5c09 | 2631 | } |
db94535d | 2632 | |
be0e5c09 | 2633 | if (push_items == 0) { |
925baedd CM |
2634 | ret = 1; |
2635 | goto out; | |
be0e5c09 | 2636 | } |
34a38218 | 2637 | if (!empty && push_items == btrfs_header_nritems(right)) |
a429e513 | 2638 | WARN_ON(1); |
5f39d397 | 2639 | |
be0e5c09 | 2640 | /* push data from right to left */ |
5f39d397 CM |
2641 | copy_extent_buffer(left, right, |
2642 | btrfs_item_nr_offset(btrfs_header_nritems(left)), | |
2643 | btrfs_item_nr_offset(0), | |
2644 | push_items * sizeof(struct btrfs_item)); | |
2645 | ||
123abc88 | 2646 | push_space = BTRFS_LEAF_DATA_SIZE(root) - |
d397712b | 2647 | btrfs_item_offset_nr(right, push_items - 1); |
5f39d397 CM |
2648 | |
2649 | copy_extent_buffer(left, right, btrfs_leaf_data(left) + | |
d6025579 CM |
2650 | leaf_data_end(root, left) - push_space, |
2651 | btrfs_leaf_data(right) + | |
5f39d397 | 2652 | btrfs_item_offset_nr(right, push_items - 1), |
d6025579 | 2653 | push_space); |
5f39d397 | 2654 | old_left_nritems = btrfs_header_nritems(left); |
87b29b20 | 2655 | BUG_ON(old_left_nritems <= 0); |
eb60ceac | 2656 | |
db94535d | 2657 | old_left_item_size = btrfs_item_offset_nr(left, old_left_nritems - 1); |
0783fcfc | 2658 | for (i = old_left_nritems; i < old_left_nritems + push_items; i++) { |
5f39d397 | 2659 | u32 ioff; |
db94535d | 2660 | |
5f39d397 | 2661 | item = btrfs_item_nr(left, i); |
db94535d CM |
2662 | if (!left->map_token) { |
2663 | map_extent_buffer(left, (unsigned long)item, | |
2664 | sizeof(struct btrfs_item), | |
2665 | &left->map_token, &left->kaddr, | |
2666 | &left->map_start, &left->map_len, | |
2667 | KM_USER1); | |
2668 | } | |
2669 | ||
5f39d397 CM |
2670 | ioff = btrfs_item_offset(left, item); |
2671 | btrfs_set_item_offset(left, item, | |
db94535d | 2672 | ioff - (BTRFS_LEAF_DATA_SIZE(root) - old_left_item_size)); |
be0e5c09 | 2673 | } |
5f39d397 | 2674 | btrfs_set_header_nritems(left, old_left_nritems + push_items); |
db94535d CM |
2675 | if (left->map_token) { |
2676 | unmap_extent_buffer(left, left->map_token, KM_USER1); | |
2677 | left->map_token = NULL; | |
2678 | } | |
be0e5c09 CM |
2679 | |
2680 | /* fixup right node */ | |
34a38218 | 2681 | if (push_items > right_nritems) { |
d397712b CM |
2682 | printk(KERN_CRIT "push items %d nr %u\n", push_items, |
2683 | right_nritems); | |
34a38218 CM |
2684 | WARN_ON(1); |
2685 | } | |
2686 | ||
2687 | if (push_items < right_nritems) { | |
2688 | push_space = btrfs_item_offset_nr(right, push_items - 1) - | |
2689 | leaf_data_end(root, right); | |
2690 | memmove_extent_buffer(right, btrfs_leaf_data(right) + | |
2691 | BTRFS_LEAF_DATA_SIZE(root) - push_space, | |
2692 | btrfs_leaf_data(right) + | |
2693 | leaf_data_end(root, right), push_space); | |
2694 | ||
2695 | memmove_extent_buffer(right, btrfs_item_nr_offset(0), | |
5f39d397 CM |
2696 | btrfs_item_nr_offset(push_items), |
2697 | (btrfs_header_nritems(right) - push_items) * | |
2698 | sizeof(struct btrfs_item)); | |
34a38218 | 2699 | } |
eef1c494 Y |
2700 | right_nritems -= push_items; |
2701 | btrfs_set_header_nritems(right, right_nritems); | |
123abc88 | 2702 | push_space = BTRFS_LEAF_DATA_SIZE(root); |
5f39d397 CM |
2703 | for (i = 0; i < right_nritems; i++) { |
2704 | item = btrfs_item_nr(right, i); | |
db94535d CM |
2705 | |
2706 | if (!right->map_token) { | |
2707 | map_extent_buffer(right, (unsigned long)item, | |
2708 | sizeof(struct btrfs_item), | |
2709 | &right->map_token, &right->kaddr, | |
2710 | &right->map_start, &right->map_len, | |
2711 | KM_USER1); | |
2712 | } | |
2713 | ||
2714 | push_space = push_space - btrfs_item_size(right, item); | |
2715 | btrfs_set_item_offset(right, item, push_space); | |
2716 | } | |
2717 | if (right->map_token) { | |
2718 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2719 | right->map_token = NULL; | |
be0e5c09 | 2720 | } |
eb60ceac | 2721 | |
5f39d397 | 2722 | btrfs_mark_buffer_dirty(left); |
34a38218 CM |
2723 | if (right_nritems) |
2724 | btrfs_mark_buffer_dirty(right); | |
098f59c2 | 2725 | |
31840ae1 ZY |
2726 | ret = btrfs_update_ref(trans, root, right, left, |
2727 | old_left_nritems, push_items); | |
2728 | BUG_ON(ret); | |
2729 | ||
5f39d397 CM |
2730 | btrfs_item_key(right, &disk_key, 0); |
2731 | wret = fixup_low_keys(trans, root, path, &disk_key, 1); | |
aa5d6bed CM |
2732 | if (wret) |
2733 | ret = wret; | |
be0e5c09 CM |
2734 | |
2735 | /* then fixup the leaf pointer in the path */ | |
2736 | if (path->slots[0] < push_items) { | |
2737 | path->slots[0] += old_left_nritems; | |
925baedd CM |
2738 | if (btrfs_header_nritems(path->nodes[0]) == 0) |
2739 | clean_tree_block(trans, root, path->nodes[0]); | |
2740 | btrfs_tree_unlock(path->nodes[0]); | |
5f39d397 CM |
2741 | free_extent_buffer(path->nodes[0]); |
2742 | path->nodes[0] = left; | |
be0e5c09 CM |
2743 | path->slots[1] -= 1; |
2744 | } else { | |
925baedd | 2745 | btrfs_tree_unlock(left); |
5f39d397 | 2746 | free_extent_buffer(left); |
be0e5c09 CM |
2747 | path->slots[0] -= push_items; |
2748 | } | |
eb60ceac | 2749 | BUG_ON(path->slots[0] < 0); |
aa5d6bed | 2750 | return ret; |
925baedd CM |
2751 | out: |
2752 | btrfs_tree_unlock(left); | |
2753 | free_extent_buffer(left); | |
2754 | return ret; | |
be0e5c09 CM |
2755 | } |
2756 | ||
74123bd7 CM |
2757 | /* |
2758 | * split the path's leaf in two, making sure there is at least data_size | |
2759 | * available for the resulting leaf level of the path. | |
aa5d6bed CM |
2760 | * |
2761 | * returns 0 if all went well and < 0 on failure. | |
74123bd7 | 2762 | */ |
e02119d5 CM |
2763 | static noinline int split_leaf(struct btrfs_trans_handle *trans, |
2764 | struct btrfs_root *root, | |
2765 | struct btrfs_key *ins_key, | |
2766 | struct btrfs_path *path, int data_size, | |
2767 | int extend) | |
be0e5c09 | 2768 | { |
5f39d397 | 2769 | struct extent_buffer *l; |
7518a238 | 2770 | u32 nritems; |
eb60ceac CM |
2771 | int mid; |
2772 | int slot; | |
5f39d397 | 2773 | struct extent_buffer *right; |
be0e5c09 CM |
2774 | int data_copy_size; |
2775 | int rt_data_off; | |
2776 | int i; | |
d4dbff95 | 2777 | int ret = 0; |
aa5d6bed | 2778 | int wret; |
cc0c5538 CM |
2779 | int double_split; |
2780 | int num_doubles = 0; | |
d4dbff95 | 2781 | struct btrfs_disk_key disk_key; |
aa5d6bed | 2782 | |
40689478 | 2783 | /* first try to make some room by pushing left and right */ |
459931ec | 2784 | if (data_size && ins_key->type != BTRFS_DIR_ITEM_KEY) { |
34a38218 | 2785 | wret = push_leaf_right(trans, root, path, data_size, 0); |
d397712b | 2786 | if (wret < 0) |
eaee50e8 | 2787 | return wret; |
3685f791 | 2788 | if (wret) { |
34a38218 | 2789 | wret = push_leaf_left(trans, root, path, data_size, 0); |
3685f791 CM |
2790 | if (wret < 0) |
2791 | return wret; | |
2792 | } | |
2793 | l = path->nodes[0]; | |
aa5d6bed | 2794 | |
3685f791 | 2795 | /* did the pushes work? */ |
87b29b20 | 2796 | if (btrfs_leaf_free_space(root, l) >= data_size) |
3685f791 | 2797 | return 0; |
3326d1b0 | 2798 | } |
aa5d6bed | 2799 | |
5c680ed6 | 2800 | if (!path->nodes[1]) { |
e089f05c | 2801 | ret = insert_new_root(trans, root, path, 1); |
5c680ed6 CM |
2802 | if (ret) |
2803 | return ret; | |
2804 | } | |
cc0c5538 CM |
2805 | again: |
2806 | double_split = 0; | |
2807 | l = path->nodes[0]; | |
eb60ceac | 2808 | slot = path->slots[0]; |
5f39d397 | 2809 | nritems = btrfs_header_nritems(l); |
d397712b | 2810 | mid = (nritems + 1) / 2; |
54aa1f4d | 2811 | |
925baedd | 2812 | right = btrfs_alloc_free_block(trans, root, root->leafsize, |
31840ae1 ZY |
2813 | path->nodes[1]->start, |
2814 | root->root_key.objectid, | |
2815 | trans->transid, 0, l->start, 0); | |
cea9e445 CM |
2816 | if (IS_ERR(right)) { |
2817 | BUG_ON(1); | |
5f39d397 | 2818 | return PTR_ERR(right); |
cea9e445 | 2819 | } |
5f39d397 CM |
2820 | |
2821 | memset_extent_buffer(right, 0, 0, sizeof(struct btrfs_header)); | |
db94535d | 2822 | btrfs_set_header_bytenr(right, right->start); |
5f39d397 CM |
2823 | btrfs_set_header_generation(right, trans->transid); |
2824 | btrfs_set_header_owner(right, root->root_key.objectid); | |
2825 | btrfs_set_header_level(right, 0); | |
2826 | write_extent_buffer(right, root->fs_info->fsid, | |
2827 | (unsigned long)btrfs_header_fsid(right), | |
2828 | BTRFS_FSID_SIZE); | |
e17cade2 CM |
2829 | |
2830 | write_extent_buffer(right, root->fs_info->chunk_tree_uuid, | |
2831 | (unsigned long)btrfs_header_chunk_tree_uuid(right), | |
2832 | BTRFS_UUID_SIZE); | |
d4dbff95 CM |
2833 | if (mid <= slot) { |
2834 | if (nritems == 1 || | |
87b29b20 | 2835 | leaf_space_used(l, mid, nritems - mid) + data_size > |
d4dbff95 CM |
2836 | BTRFS_LEAF_DATA_SIZE(root)) { |
2837 | if (slot >= nritems) { | |
2838 | btrfs_cpu_key_to_disk(&disk_key, ins_key); | |
5f39d397 | 2839 | btrfs_set_header_nritems(right, 0); |
d4dbff95 | 2840 | wret = insert_ptr(trans, root, path, |
db94535d | 2841 | &disk_key, right->start, |
d4dbff95 CM |
2842 | path->slots[1] + 1, 1); |
2843 | if (wret) | |
2844 | ret = wret; | |
925baedd CM |
2845 | |
2846 | btrfs_tree_unlock(path->nodes[0]); | |
5f39d397 CM |
2847 | free_extent_buffer(path->nodes[0]); |
2848 | path->nodes[0] = right; | |
d4dbff95 CM |
2849 | path->slots[0] = 0; |
2850 | path->slots[1] += 1; | |
0ef8b242 | 2851 | btrfs_mark_buffer_dirty(right); |
d4dbff95 CM |
2852 | return ret; |
2853 | } | |
2854 | mid = slot; | |
3326d1b0 CM |
2855 | if (mid != nritems && |
2856 | leaf_space_used(l, mid, nritems - mid) + | |
87b29b20 | 2857 | data_size > BTRFS_LEAF_DATA_SIZE(root)) { |
3326d1b0 CM |
2858 | double_split = 1; |
2859 | } | |
d4dbff95 CM |
2860 | } |
2861 | } else { | |
87b29b20 | 2862 | if (leaf_space_used(l, 0, mid) + data_size > |
d4dbff95 | 2863 | BTRFS_LEAF_DATA_SIZE(root)) { |
459931ec | 2864 | if (!extend && data_size && slot == 0) { |
d4dbff95 | 2865 | btrfs_cpu_key_to_disk(&disk_key, ins_key); |
5f39d397 | 2866 | btrfs_set_header_nritems(right, 0); |
d4dbff95 CM |
2867 | wret = insert_ptr(trans, root, path, |
2868 | &disk_key, | |
db94535d | 2869 | right->start, |
098f59c2 | 2870 | path->slots[1], 1); |
d4dbff95 CM |
2871 | if (wret) |
2872 | ret = wret; | |
925baedd | 2873 | btrfs_tree_unlock(path->nodes[0]); |
5f39d397 CM |
2874 | free_extent_buffer(path->nodes[0]); |
2875 | path->nodes[0] = right; | |
d4dbff95 | 2876 | path->slots[0] = 0; |
a429e513 CM |
2877 | if (path->slots[1] == 0) { |
2878 | wret = fixup_low_keys(trans, root, | |
d397712b | 2879 | path, &disk_key, 1); |
a429e513 CM |
2880 | if (wret) |
2881 | ret = wret; | |
2882 | } | |
0ef8b242 | 2883 | btrfs_mark_buffer_dirty(right); |
d4dbff95 | 2884 | return ret; |
459931ec | 2885 | } else if ((extend || !data_size) && slot == 0) { |
cc0c5538 CM |
2886 | mid = 1; |
2887 | } else { | |
2888 | mid = slot; | |
2889 | if (mid != nritems && | |
2890 | leaf_space_used(l, mid, nritems - mid) + | |
87b29b20 | 2891 | data_size > BTRFS_LEAF_DATA_SIZE(root)) { |
cc0c5538 CM |
2892 | double_split = 1; |
2893 | } | |
5ee78ac7 | 2894 | } |
d4dbff95 CM |
2895 | } |
2896 | } | |
5f39d397 CM |
2897 | nritems = nritems - mid; |
2898 | btrfs_set_header_nritems(right, nritems); | |
2899 | data_copy_size = btrfs_item_end_nr(l, mid) - leaf_data_end(root, l); | |
2900 | ||
2901 | copy_extent_buffer(right, l, btrfs_item_nr_offset(0), | |
2902 | btrfs_item_nr_offset(mid), | |
2903 | nritems * sizeof(struct btrfs_item)); | |
2904 | ||
2905 | copy_extent_buffer(right, l, | |
d6025579 CM |
2906 | btrfs_leaf_data(right) + BTRFS_LEAF_DATA_SIZE(root) - |
2907 | data_copy_size, btrfs_leaf_data(l) + | |
2908 | leaf_data_end(root, l), data_copy_size); | |
5f39d397 | 2909 | |
123abc88 | 2910 | rt_data_off = BTRFS_LEAF_DATA_SIZE(root) - |
5f39d397 | 2911 | btrfs_item_end_nr(l, mid); |
74123bd7 | 2912 | |
5f39d397 CM |
2913 | for (i = 0; i < nritems; i++) { |
2914 | struct btrfs_item *item = btrfs_item_nr(right, i); | |
db94535d CM |
2915 | u32 ioff; |
2916 | ||
2917 | if (!right->map_token) { | |
2918 | map_extent_buffer(right, (unsigned long)item, | |
2919 | sizeof(struct btrfs_item), | |
2920 | &right->map_token, &right->kaddr, | |
2921 | &right->map_start, &right->map_len, | |
2922 | KM_USER1); | |
2923 | } | |
2924 | ||
2925 | ioff = btrfs_item_offset(right, item); | |
5f39d397 | 2926 | btrfs_set_item_offset(right, item, ioff + rt_data_off); |
0783fcfc | 2927 | } |
74123bd7 | 2928 | |
db94535d CM |
2929 | if (right->map_token) { |
2930 | unmap_extent_buffer(right, right->map_token, KM_USER1); | |
2931 | right->map_token = NULL; | |
2932 | } | |
2933 | ||
5f39d397 | 2934 | btrfs_set_header_nritems(l, mid); |
aa5d6bed | 2935 | ret = 0; |
5f39d397 | 2936 | btrfs_item_key(right, &disk_key, 0); |
db94535d CM |
2937 | wret = insert_ptr(trans, root, path, &disk_key, right->start, |
2938 | path->slots[1] + 1, 1); | |
aa5d6bed CM |
2939 | if (wret) |
2940 | ret = wret; | |
5f39d397 CM |
2941 | |
2942 | btrfs_mark_buffer_dirty(right); | |
2943 | btrfs_mark_buffer_dirty(l); | |
eb60ceac | 2944 | BUG_ON(path->slots[0] != slot); |
5f39d397 | 2945 | |
31840ae1 ZY |
2946 | ret = btrfs_update_ref(trans, root, l, right, 0, nritems); |
2947 | BUG_ON(ret); | |
2948 | ||
be0e5c09 | 2949 | if (mid <= slot) { |
925baedd | 2950 | btrfs_tree_unlock(path->nodes[0]); |
5f39d397 CM |
2951 | free_extent_buffer(path->nodes[0]); |
2952 | path->nodes[0] = right; | |
be0e5c09 CM |
2953 | path->slots[0] -= mid; |
2954 | path->slots[1] += 1; | |
925baedd CM |
2955 | } else { |
2956 | btrfs_tree_unlock(right); | |
5f39d397 | 2957 | free_extent_buffer(right); |
925baedd | 2958 | } |
5f39d397 | 2959 | |
eb60ceac | 2960 | BUG_ON(path->slots[0] < 0); |
d4dbff95 | 2961 | |
cc0c5538 CM |
2962 | if (double_split) { |
2963 | BUG_ON(num_doubles != 0); | |
2964 | num_doubles++; | |
2965 | goto again; | |
a429e513 | 2966 | } |
be0e5c09 CM |
2967 | return ret; |
2968 | } | |
2969 | ||
459931ec CM |
2970 | /* |
2971 | * This function splits a single item into two items, | |
2972 | * giving 'new_key' to the new item and splitting the | |
2973 | * old one at split_offset (from the start of the item). | |
2974 | * | |
2975 | * The path may be released by this operation. After | |
2976 | * the split, the path is pointing to the old item. The | |
2977 | * new item is going to be in the same node as the old one. | |
2978 | * | |
2979 | * Note, the item being split must be smaller enough to live alone on | |
2980 | * a tree block with room for one extra struct btrfs_item | |
2981 | * | |
2982 | * This allows us to split the item in place, keeping a lock on the | |
2983 | * leaf the entire time. | |
2984 | */ | |
2985 | int btrfs_split_item(struct btrfs_trans_handle *trans, | |
2986 | struct btrfs_root *root, | |
2987 | struct btrfs_path *path, | |
2988 | struct btrfs_key *new_key, | |
2989 | unsigned long split_offset) | |
2990 | { | |
2991 | u32 item_size; | |
2992 | struct extent_buffer *leaf; | |
2993 | struct btrfs_key orig_key; | |
2994 | struct btrfs_item *item; | |
2995 | struct btrfs_item *new_item; | |
2996 | int ret = 0; | |
2997 | int slot; | |
2998 | u32 nritems; | |
2999 | u32 orig_offset; | |
3000 | struct btrfs_disk_key disk_key; | |
3001 | char *buf; | |
3002 | ||
3003 | leaf = path->nodes[0]; | |
3004 | btrfs_item_key_to_cpu(leaf, &orig_key, path->slots[0]); | |
3005 | if (btrfs_leaf_free_space(root, leaf) >= sizeof(struct btrfs_item)) | |
3006 | goto split; | |
3007 | ||
3008 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
3009 | btrfs_release_path(root, path); | |
3010 | ||
3011 | path->search_for_split = 1; | |
3012 | path->keep_locks = 1; | |
3013 | ||
3014 | ret = btrfs_search_slot(trans, root, &orig_key, path, 0, 1); | |
3015 | path->search_for_split = 0; | |
3016 | ||
3017 | /* if our item isn't there or got smaller, return now */ | |
3018 | if (ret != 0 || item_size != btrfs_item_size_nr(path->nodes[0], | |
3019 | path->slots[0])) { | |
3020 | path->keep_locks = 0; | |
3021 | return -EAGAIN; | |
3022 | } | |
3023 | ||
87b29b20 YZ |
3024 | ret = split_leaf(trans, root, &orig_key, path, |
3025 | sizeof(struct btrfs_item), 1); | |
459931ec CM |
3026 | path->keep_locks = 0; |
3027 | BUG_ON(ret); | |
3028 | ||
b4ce94de CM |
3029 | /* |
3030 | * make sure any changes to the path from split_leaf leave it | |
3031 | * in a blocking state | |
3032 | */ | |
3033 | btrfs_set_path_blocking(path); | |
3034 | ||
459931ec | 3035 | leaf = path->nodes[0]; |
42dc7bab | 3036 | BUG_ON(btrfs_leaf_free_space(root, leaf) < sizeof(struct btrfs_item)); |
459931ec CM |
3037 | |
3038 | split: | |
3039 | item = btrfs_item_nr(leaf, path->slots[0]); | |
3040 | orig_offset = btrfs_item_offset(leaf, item); | |
3041 | item_size = btrfs_item_size(leaf, item); | |
3042 | ||
3043 | ||
3044 | buf = kmalloc(item_size, GFP_NOFS); | |
3045 | read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, | |
3046 | path->slots[0]), item_size); | |
3047 | slot = path->slots[0] + 1; | |
3048 | leaf = path->nodes[0]; | |
3049 | ||
3050 | nritems = btrfs_header_nritems(leaf); | |
3051 | ||
3052 | if (slot != nritems) { | |
3053 | /* shift the items */ | |
3054 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + 1), | |
3055 | btrfs_item_nr_offset(slot), | |
3056 | (nritems - slot) * sizeof(struct btrfs_item)); | |
3057 | ||
3058 | } | |
3059 | ||
3060 | btrfs_cpu_key_to_disk(&disk_key, new_key); | |
3061 | btrfs_set_item_key(leaf, &disk_key, slot); | |
3062 | ||
3063 | new_item = btrfs_item_nr(leaf, slot); | |
3064 | ||
3065 | btrfs_set_item_offset(leaf, new_item, orig_offset); | |
3066 | btrfs_set_item_size(leaf, new_item, item_size - split_offset); | |
3067 | ||
3068 | btrfs_set_item_offset(leaf, item, | |
3069 | orig_offset + item_size - split_offset); | |
3070 | btrfs_set_item_size(leaf, item, split_offset); | |
3071 | ||
3072 | btrfs_set_header_nritems(leaf, nritems + 1); | |
3073 | ||
3074 | /* write the data for the start of the original item */ | |
3075 | write_extent_buffer(leaf, buf, | |
3076 | btrfs_item_ptr_offset(leaf, path->slots[0]), | |
3077 | split_offset); | |
3078 | ||
3079 | /* write the data for the new item */ | |
3080 | write_extent_buffer(leaf, buf + split_offset, | |
3081 | btrfs_item_ptr_offset(leaf, slot), | |
3082 | item_size - split_offset); | |
3083 | btrfs_mark_buffer_dirty(leaf); | |
3084 | ||
3085 | ret = 0; | |
3086 | if (btrfs_leaf_free_space(root, leaf) < 0) { | |
3087 | btrfs_print_leaf(root, leaf); | |
3088 | BUG(); | |
3089 | } | |
3090 | kfree(buf); | |
3091 | return ret; | |
3092 | } | |
3093 | ||
d352ac68 CM |
3094 | /* |
3095 | * make the item pointed to by the path smaller. new_size indicates | |
3096 | * how small to make it, and from_end tells us if we just chop bytes | |
3097 | * off the end of the item or if we shift the item to chop bytes off | |
3098 | * the front. | |
3099 | */ | |
b18c6685 CM |
3100 | int btrfs_truncate_item(struct btrfs_trans_handle *trans, |
3101 | struct btrfs_root *root, | |
3102 | struct btrfs_path *path, | |
179e29e4 | 3103 | u32 new_size, int from_end) |
b18c6685 CM |
3104 | { |
3105 | int ret = 0; | |
3106 | int slot; | |
3107 | int slot_orig; | |
5f39d397 CM |
3108 | struct extent_buffer *leaf; |
3109 | struct btrfs_item *item; | |
b18c6685 CM |
3110 | u32 nritems; |
3111 | unsigned int data_end; | |
3112 | unsigned int old_data_start; | |
3113 | unsigned int old_size; | |
3114 | unsigned int size_diff; | |
3115 | int i; | |
3116 | ||
3117 | slot_orig = path->slots[0]; | |
5f39d397 | 3118 | leaf = path->nodes[0]; |
179e29e4 CM |
3119 | slot = path->slots[0]; |
3120 | ||
3121 | old_size = btrfs_item_size_nr(leaf, slot); | |
3122 | if (old_size == new_size) | |
3123 | return 0; | |
b18c6685 | 3124 | |
5f39d397 | 3125 | nritems = btrfs_header_nritems(leaf); |
b18c6685 CM |
3126 | data_end = leaf_data_end(root, leaf); |
3127 | ||
5f39d397 | 3128 | old_data_start = btrfs_item_offset_nr(leaf, slot); |
179e29e4 | 3129 | |
b18c6685 CM |
3130 | size_diff = old_size - new_size; |
3131 | ||
3132 | BUG_ON(slot < 0); | |
3133 | BUG_ON(slot >= nritems); | |
3134 | ||
3135 | /* | |
3136 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3137 | */ | |
3138 | /* first correct the data pointers */ | |
3139 | for (i = slot; i < nritems; i++) { | |
5f39d397 CM |
3140 | u32 ioff; |
3141 | item = btrfs_item_nr(leaf, i); | |
db94535d CM |
3142 | |
3143 | if (!leaf->map_token) { | |
3144 | map_extent_buffer(leaf, (unsigned long)item, | |
3145 | sizeof(struct btrfs_item), | |
3146 | &leaf->map_token, &leaf->kaddr, | |
3147 | &leaf->map_start, &leaf->map_len, | |
3148 | KM_USER1); | |
3149 | } | |
3150 | ||
5f39d397 CM |
3151 | ioff = btrfs_item_offset(leaf, item); |
3152 | btrfs_set_item_offset(leaf, item, ioff + size_diff); | |
b18c6685 | 3153 | } |
db94535d CM |
3154 | |
3155 | if (leaf->map_token) { | |
3156 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3157 | leaf->map_token = NULL; | |
3158 | } | |
3159 | ||
b18c6685 | 3160 | /* shift the data */ |
179e29e4 CM |
3161 | if (from_end) { |
3162 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
3163 | data_end + size_diff, btrfs_leaf_data(leaf) + | |
3164 | data_end, old_data_start + new_size - data_end); | |
3165 | } else { | |
3166 | struct btrfs_disk_key disk_key; | |
3167 | u64 offset; | |
3168 | ||
3169 | btrfs_item_key(leaf, &disk_key, slot); | |
3170 | ||
3171 | if (btrfs_disk_key_type(&disk_key) == BTRFS_EXTENT_DATA_KEY) { | |
3172 | unsigned long ptr; | |
3173 | struct btrfs_file_extent_item *fi; | |
3174 | ||
3175 | fi = btrfs_item_ptr(leaf, slot, | |
3176 | struct btrfs_file_extent_item); | |
3177 | fi = (struct btrfs_file_extent_item *)( | |
3178 | (unsigned long)fi - size_diff); | |
3179 | ||
3180 | if (btrfs_file_extent_type(leaf, fi) == | |
3181 | BTRFS_FILE_EXTENT_INLINE) { | |
3182 | ptr = btrfs_item_ptr_offset(leaf, slot); | |
3183 | memmove_extent_buffer(leaf, ptr, | |
d397712b CM |
3184 | (unsigned long)fi, |
3185 | offsetof(struct btrfs_file_extent_item, | |
179e29e4 CM |
3186 | disk_bytenr)); |
3187 | } | |
3188 | } | |
3189 | ||
3190 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
3191 | data_end + size_diff, btrfs_leaf_data(leaf) + | |
3192 | data_end, old_data_start - data_end); | |
3193 | ||
3194 | offset = btrfs_disk_key_offset(&disk_key); | |
3195 | btrfs_set_disk_key_offset(&disk_key, offset + size_diff); | |
3196 | btrfs_set_item_key(leaf, &disk_key, slot); | |
3197 | if (slot == 0) | |
3198 | fixup_low_keys(trans, root, path, &disk_key, 1); | |
3199 | } | |
5f39d397 CM |
3200 | |
3201 | item = btrfs_item_nr(leaf, slot); | |
3202 | btrfs_set_item_size(leaf, item, new_size); | |
3203 | btrfs_mark_buffer_dirty(leaf); | |
b18c6685 CM |
3204 | |
3205 | ret = 0; | |
5f39d397 CM |
3206 | if (btrfs_leaf_free_space(root, leaf) < 0) { |
3207 | btrfs_print_leaf(root, leaf); | |
b18c6685 | 3208 | BUG(); |
5f39d397 | 3209 | } |
b18c6685 CM |
3210 | return ret; |
3211 | } | |
3212 | ||
d352ac68 CM |
3213 | /* |
3214 | * make the item pointed to by the path bigger, data_size is the new size. | |
3215 | */ | |
5f39d397 CM |
3216 | int btrfs_extend_item(struct btrfs_trans_handle *trans, |
3217 | struct btrfs_root *root, struct btrfs_path *path, | |
3218 | u32 data_size) | |
6567e837 CM |
3219 | { |
3220 | int ret = 0; | |
3221 | int slot; | |
3222 | int slot_orig; | |
5f39d397 CM |
3223 | struct extent_buffer *leaf; |
3224 | struct btrfs_item *item; | |
6567e837 CM |
3225 | u32 nritems; |
3226 | unsigned int data_end; | |
3227 | unsigned int old_data; | |
3228 | unsigned int old_size; | |
3229 | int i; | |
3230 | ||
3231 | slot_orig = path->slots[0]; | |
5f39d397 | 3232 | leaf = path->nodes[0]; |
6567e837 | 3233 | |
5f39d397 | 3234 | nritems = btrfs_header_nritems(leaf); |
6567e837 CM |
3235 | data_end = leaf_data_end(root, leaf); |
3236 | ||
5f39d397 CM |
3237 | if (btrfs_leaf_free_space(root, leaf) < data_size) { |
3238 | btrfs_print_leaf(root, leaf); | |
6567e837 | 3239 | BUG(); |
5f39d397 | 3240 | } |
6567e837 | 3241 | slot = path->slots[0]; |
5f39d397 | 3242 | old_data = btrfs_item_end_nr(leaf, slot); |
6567e837 CM |
3243 | |
3244 | BUG_ON(slot < 0); | |
3326d1b0 CM |
3245 | if (slot >= nritems) { |
3246 | btrfs_print_leaf(root, leaf); | |
d397712b CM |
3247 | printk(KERN_CRIT "slot %d too large, nritems %d\n", |
3248 | slot, nritems); | |
3326d1b0 CM |
3249 | BUG_ON(1); |
3250 | } | |
6567e837 CM |
3251 | |
3252 | /* | |
3253 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3254 | */ | |
3255 | /* first correct the data pointers */ | |
3256 | for (i = slot; i < nritems; i++) { | |
5f39d397 CM |
3257 | u32 ioff; |
3258 | item = btrfs_item_nr(leaf, i); | |
db94535d CM |
3259 | |
3260 | if (!leaf->map_token) { | |
3261 | map_extent_buffer(leaf, (unsigned long)item, | |
3262 | sizeof(struct btrfs_item), | |
3263 | &leaf->map_token, &leaf->kaddr, | |
3264 | &leaf->map_start, &leaf->map_len, | |
3265 | KM_USER1); | |
3266 | } | |
5f39d397 CM |
3267 | ioff = btrfs_item_offset(leaf, item); |
3268 | btrfs_set_item_offset(leaf, item, ioff - data_size); | |
6567e837 | 3269 | } |
5f39d397 | 3270 | |
db94535d CM |
3271 | if (leaf->map_token) { |
3272 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3273 | leaf->map_token = NULL; | |
3274 | } | |
3275 | ||
6567e837 | 3276 | /* shift the data */ |
5f39d397 | 3277 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + |
6567e837 CM |
3278 | data_end - data_size, btrfs_leaf_data(leaf) + |
3279 | data_end, old_data - data_end); | |
5f39d397 | 3280 | |
6567e837 | 3281 | data_end = old_data; |
5f39d397 CM |
3282 | old_size = btrfs_item_size_nr(leaf, slot); |
3283 | item = btrfs_item_nr(leaf, slot); | |
3284 | btrfs_set_item_size(leaf, item, old_size + data_size); | |
3285 | btrfs_mark_buffer_dirty(leaf); | |
6567e837 CM |
3286 | |
3287 | ret = 0; | |
5f39d397 CM |
3288 | if (btrfs_leaf_free_space(root, leaf) < 0) { |
3289 | btrfs_print_leaf(root, leaf); | |
6567e837 | 3290 | BUG(); |
5f39d397 | 3291 | } |
6567e837 CM |
3292 | return ret; |
3293 | } | |
3294 | ||
f3465ca4 JB |
3295 | /* |
3296 | * Given a key and some data, insert items into the tree. | |
3297 | * This does all the path init required, making room in the tree if needed. | |
3298 | * Returns the number of keys that were inserted. | |
3299 | */ | |
3300 | int btrfs_insert_some_items(struct btrfs_trans_handle *trans, | |
3301 | struct btrfs_root *root, | |
3302 | struct btrfs_path *path, | |
3303 | struct btrfs_key *cpu_key, u32 *data_size, | |
3304 | int nr) | |
3305 | { | |
3306 | struct extent_buffer *leaf; | |
3307 | struct btrfs_item *item; | |
3308 | int ret = 0; | |
3309 | int slot; | |
f3465ca4 JB |
3310 | int i; |
3311 | u32 nritems; | |
3312 | u32 total_data = 0; | |
3313 | u32 total_size = 0; | |
3314 | unsigned int data_end; | |
3315 | struct btrfs_disk_key disk_key; | |
3316 | struct btrfs_key found_key; | |
3317 | ||
87b29b20 YZ |
3318 | for (i = 0; i < nr; i++) { |
3319 | if (total_size + data_size[i] + sizeof(struct btrfs_item) > | |
3320 | BTRFS_LEAF_DATA_SIZE(root)) { | |
3321 | break; | |
3322 | nr = i; | |
3323 | } | |
f3465ca4 | 3324 | total_data += data_size[i]; |
87b29b20 YZ |
3325 | total_size += data_size[i] + sizeof(struct btrfs_item); |
3326 | } | |
3327 | BUG_ON(nr == 0); | |
f3465ca4 | 3328 | |
f3465ca4 JB |
3329 | ret = btrfs_search_slot(trans, root, cpu_key, path, total_size, 1); |
3330 | if (ret == 0) | |
3331 | return -EEXIST; | |
3332 | if (ret < 0) | |
3333 | goto out; | |
3334 | ||
f3465ca4 JB |
3335 | leaf = path->nodes[0]; |
3336 | ||
3337 | nritems = btrfs_header_nritems(leaf); | |
3338 | data_end = leaf_data_end(root, leaf); | |
3339 | ||
3340 | if (btrfs_leaf_free_space(root, leaf) < total_size) { | |
3341 | for (i = nr; i >= 0; i--) { | |
3342 | total_data -= data_size[i]; | |
3343 | total_size -= data_size[i] + sizeof(struct btrfs_item); | |
3344 | if (total_size < btrfs_leaf_free_space(root, leaf)) | |
3345 | break; | |
3346 | } | |
3347 | nr = i; | |
3348 | } | |
3349 | ||
3350 | slot = path->slots[0]; | |
3351 | BUG_ON(slot < 0); | |
3352 | ||
3353 | if (slot != nritems) { | |
3354 | unsigned int old_data = btrfs_item_end_nr(leaf, slot); | |
3355 | ||
3356 | item = btrfs_item_nr(leaf, slot); | |
3357 | btrfs_item_key_to_cpu(leaf, &found_key, slot); | |
3358 | ||
3359 | /* figure out how many keys we can insert in here */ | |
3360 | total_data = data_size[0]; | |
3361 | for (i = 1; i < nr; i++) { | |
3362 | if (comp_cpu_keys(&found_key, cpu_key + i) <= 0) | |
3363 | break; | |
3364 | total_data += data_size[i]; | |
3365 | } | |
3366 | nr = i; | |
3367 | ||
3368 | if (old_data < data_end) { | |
3369 | btrfs_print_leaf(root, leaf); | |
d397712b | 3370 | printk(KERN_CRIT "slot %d old_data %d data_end %d\n", |
f3465ca4 JB |
3371 | slot, old_data, data_end); |
3372 | BUG_ON(1); | |
3373 | } | |
3374 | /* | |
3375 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3376 | */ | |
3377 | /* first correct the data pointers */ | |
3378 | WARN_ON(leaf->map_token); | |
3379 | for (i = slot; i < nritems; i++) { | |
3380 | u32 ioff; | |
3381 | ||
3382 | item = btrfs_item_nr(leaf, i); | |
3383 | if (!leaf->map_token) { | |
3384 | map_extent_buffer(leaf, (unsigned long)item, | |
3385 | sizeof(struct btrfs_item), | |
3386 | &leaf->map_token, &leaf->kaddr, | |
3387 | &leaf->map_start, &leaf->map_len, | |
3388 | KM_USER1); | |
3389 | } | |
3390 | ||
3391 | ioff = btrfs_item_offset(leaf, item); | |
3392 | btrfs_set_item_offset(leaf, item, ioff - total_data); | |
3393 | } | |
3394 | if (leaf->map_token) { | |
3395 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3396 | leaf->map_token = NULL; | |
3397 | } | |
3398 | ||
3399 | /* shift the items */ | |
3400 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), | |
3401 | btrfs_item_nr_offset(slot), | |
3402 | (nritems - slot) * sizeof(struct btrfs_item)); | |
3403 | ||
3404 | /* shift the data */ | |
3405 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
3406 | data_end - total_data, btrfs_leaf_data(leaf) + | |
3407 | data_end, old_data - data_end); | |
3408 | data_end = old_data; | |
3409 | } else { | |
3410 | /* | |
3411 | * this sucks but it has to be done, if we are inserting at | |
3412 | * the end of the leaf only insert 1 of the items, since we | |
3413 | * have no way of knowing whats on the next leaf and we'd have | |
3414 | * to drop our current locks to figure it out | |
3415 | */ | |
3416 | nr = 1; | |
3417 | } | |
3418 | ||
3419 | /* setup the item for the new data */ | |
3420 | for (i = 0; i < nr; i++) { | |
3421 | btrfs_cpu_key_to_disk(&disk_key, cpu_key + i); | |
3422 | btrfs_set_item_key(leaf, &disk_key, slot + i); | |
3423 | item = btrfs_item_nr(leaf, slot + i); | |
3424 | btrfs_set_item_offset(leaf, item, data_end - data_size[i]); | |
3425 | data_end -= data_size[i]; | |
3426 | btrfs_set_item_size(leaf, item, data_size[i]); | |
3427 | } | |
3428 | btrfs_set_header_nritems(leaf, nritems + nr); | |
3429 | btrfs_mark_buffer_dirty(leaf); | |
3430 | ||
3431 | ret = 0; | |
3432 | if (slot == 0) { | |
3433 | btrfs_cpu_key_to_disk(&disk_key, cpu_key); | |
3434 | ret = fixup_low_keys(trans, root, path, &disk_key, 1); | |
3435 | } | |
3436 | ||
3437 | if (btrfs_leaf_free_space(root, leaf) < 0) { | |
3438 | btrfs_print_leaf(root, leaf); | |
3439 | BUG(); | |
3440 | } | |
3441 | out: | |
3442 | if (!ret) | |
3443 | ret = nr; | |
3444 | return ret; | |
3445 | } | |
3446 | ||
74123bd7 | 3447 | /* |
d352ac68 | 3448 | * Given a key and some data, insert items into the tree. |
74123bd7 CM |
3449 | * This does all the path init required, making room in the tree if needed. |
3450 | */ | |
9c58309d | 3451 | int btrfs_insert_empty_items(struct btrfs_trans_handle *trans, |
5f39d397 CM |
3452 | struct btrfs_root *root, |
3453 | struct btrfs_path *path, | |
9c58309d CM |
3454 | struct btrfs_key *cpu_key, u32 *data_size, |
3455 | int nr) | |
be0e5c09 | 3456 | { |
5f39d397 CM |
3457 | struct extent_buffer *leaf; |
3458 | struct btrfs_item *item; | |
aa5d6bed | 3459 | int ret = 0; |
be0e5c09 | 3460 | int slot; |
eb60ceac | 3461 | int slot_orig; |
9c58309d | 3462 | int i; |
7518a238 | 3463 | u32 nritems; |
9c58309d CM |
3464 | u32 total_size = 0; |
3465 | u32 total_data = 0; | |
be0e5c09 | 3466 | unsigned int data_end; |
e2fa7227 CM |
3467 | struct btrfs_disk_key disk_key; |
3468 | ||
d397712b | 3469 | for (i = 0; i < nr; i++) |
9c58309d | 3470 | total_data += data_size[i]; |
be0e5c09 | 3471 | |
7b128766 | 3472 | total_size = total_data + (nr * sizeof(struct btrfs_item)); |
9c58309d | 3473 | ret = btrfs_search_slot(trans, root, cpu_key, path, total_size, 1); |
0f9dd46c | 3474 | if (ret == 0) |
f0930a37 | 3475 | return -EEXIST; |
ed2ff2cb CM |
3476 | if (ret < 0) |
3477 | goto out; | |
be0e5c09 | 3478 | |
62e2749e | 3479 | slot_orig = path->slots[0]; |
5f39d397 | 3480 | leaf = path->nodes[0]; |
74123bd7 | 3481 | |
5f39d397 | 3482 | nritems = btrfs_header_nritems(leaf); |
123abc88 | 3483 | data_end = leaf_data_end(root, leaf); |
eb60ceac | 3484 | |
f25956cc | 3485 | if (btrfs_leaf_free_space(root, leaf) < total_size) { |
3326d1b0 | 3486 | btrfs_print_leaf(root, leaf); |
d397712b | 3487 | printk(KERN_CRIT "not enough freespace need %u have %d\n", |
9c58309d | 3488 | total_size, btrfs_leaf_free_space(root, leaf)); |
be0e5c09 | 3489 | BUG(); |
d4dbff95 | 3490 | } |
5f39d397 | 3491 | |
62e2749e | 3492 | slot = path->slots[0]; |
eb60ceac | 3493 | BUG_ON(slot < 0); |
5f39d397 | 3494 | |
be0e5c09 | 3495 | if (slot != nritems) { |
5f39d397 | 3496 | unsigned int old_data = btrfs_item_end_nr(leaf, slot); |
be0e5c09 | 3497 | |
5f39d397 CM |
3498 | if (old_data < data_end) { |
3499 | btrfs_print_leaf(root, leaf); | |
d397712b | 3500 | printk(KERN_CRIT "slot %d old_data %d data_end %d\n", |
5f39d397 CM |
3501 | slot, old_data, data_end); |
3502 | BUG_ON(1); | |
3503 | } | |
be0e5c09 CM |
3504 | /* |
3505 | * item0..itemN ... dataN.offset..dataN.size .. data0.size | |
3506 | */ | |
3507 | /* first correct the data pointers */ | |
db94535d | 3508 | WARN_ON(leaf->map_token); |
0783fcfc | 3509 | for (i = slot; i < nritems; i++) { |
5f39d397 | 3510 | u32 ioff; |
db94535d | 3511 | |
5f39d397 | 3512 | item = btrfs_item_nr(leaf, i); |
db94535d CM |
3513 | if (!leaf->map_token) { |
3514 | map_extent_buffer(leaf, (unsigned long)item, | |
3515 | sizeof(struct btrfs_item), | |
3516 | &leaf->map_token, &leaf->kaddr, | |
3517 | &leaf->map_start, &leaf->map_len, | |
3518 | KM_USER1); | |
3519 | } | |
3520 | ||
5f39d397 | 3521 | ioff = btrfs_item_offset(leaf, item); |
9c58309d | 3522 | btrfs_set_item_offset(leaf, item, ioff - total_data); |
0783fcfc | 3523 | } |
db94535d CM |
3524 | if (leaf->map_token) { |
3525 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3526 | leaf->map_token = NULL; | |
3527 | } | |
be0e5c09 CM |
3528 | |
3529 | /* shift the items */ | |
9c58309d | 3530 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), |
5f39d397 | 3531 | btrfs_item_nr_offset(slot), |
d6025579 | 3532 | (nritems - slot) * sizeof(struct btrfs_item)); |
be0e5c09 CM |
3533 | |
3534 | /* shift the data */ | |
5f39d397 | 3535 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + |
9c58309d | 3536 | data_end - total_data, btrfs_leaf_data(leaf) + |
d6025579 | 3537 | data_end, old_data - data_end); |
be0e5c09 CM |
3538 | data_end = old_data; |
3539 | } | |
5f39d397 | 3540 | |
62e2749e | 3541 | /* setup the item for the new data */ |
9c58309d CM |
3542 | for (i = 0; i < nr; i++) { |
3543 | btrfs_cpu_key_to_disk(&disk_key, cpu_key + i); | |
3544 | btrfs_set_item_key(leaf, &disk_key, slot + i); | |
3545 | item = btrfs_item_nr(leaf, slot + i); | |
3546 | btrfs_set_item_offset(leaf, item, data_end - data_size[i]); | |
3547 | data_end -= data_size[i]; | |
3548 | btrfs_set_item_size(leaf, item, data_size[i]); | |
3549 | } | |
3550 | btrfs_set_header_nritems(leaf, nritems + nr); | |
5f39d397 | 3551 | btrfs_mark_buffer_dirty(leaf); |
aa5d6bed CM |
3552 | |
3553 | ret = 0; | |
5a01a2e3 CM |
3554 | if (slot == 0) { |
3555 | btrfs_cpu_key_to_disk(&disk_key, cpu_key); | |
e089f05c | 3556 | ret = fixup_low_keys(trans, root, path, &disk_key, 1); |
5a01a2e3 | 3557 | } |
aa5d6bed | 3558 | |
5f39d397 CM |
3559 | if (btrfs_leaf_free_space(root, leaf) < 0) { |
3560 | btrfs_print_leaf(root, leaf); | |
be0e5c09 | 3561 | BUG(); |
5f39d397 | 3562 | } |
ed2ff2cb | 3563 | out: |
b4ce94de | 3564 | btrfs_unlock_up_safe(path, 1); |
62e2749e CM |
3565 | return ret; |
3566 | } | |
3567 | ||
3568 | /* | |
3569 | * Given a key and some data, insert an item into the tree. | |
3570 | * This does all the path init required, making room in the tree if needed. | |
3571 | */ | |
e089f05c CM |
3572 | int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root |
3573 | *root, struct btrfs_key *cpu_key, void *data, u32 | |
3574 | data_size) | |
62e2749e CM |
3575 | { |
3576 | int ret = 0; | |
2c90e5d6 | 3577 | struct btrfs_path *path; |
5f39d397 CM |
3578 | struct extent_buffer *leaf; |
3579 | unsigned long ptr; | |
62e2749e | 3580 | |
2c90e5d6 CM |
3581 | path = btrfs_alloc_path(); |
3582 | BUG_ON(!path); | |
2c90e5d6 | 3583 | ret = btrfs_insert_empty_item(trans, root, path, cpu_key, data_size); |
62e2749e | 3584 | if (!ret) { |
5f39d397 CM |
3585 | leaf = path->nodes[0]; |
3586 | ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
3587 | write_extent_buffer(leaf, data, ptr, data_size); | |
3588 | btrfs_mark_buffer_dirty(leaf); | |
62e2749e | 3589 | } |
2c90e5d6 | 3590 | btrfs_free_path(path); |
aa5d6bed | 3591 | return ret; |
be0e5c09 CM |
3592 | } |
3593 | ||
74123bd7 | 3594 | /* |
5de08d7d | 3595 | * delete the pointer from a given node. |
74123bd7 | 3596 | * |
d352ac68 CM |
3597 | * the tree should have been previously balanced so the deletion does not |
3598 | * empty a node. | |
74123bd7 | 3599 | */ |
e089f05c CM |
3600 | static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root, |
3601 | struct btrfs_path *path, int level, int slot) | |
be0e5c09 | 3602 | { |
5f39d397 | 3603 | struct extent_buffer *parent = path->nodes[level]; |
7518a238 | 3604 | u32 nritems; |
aa5d6bed | 3605 | int ret = 0; |
bb803951 | 3606 | int wret; |
be0e5c09 | 3607 | |
5f39d397 | 3608 | nritems = btrfs_header_nritems(parent); |
d397712b | 3609 | if (slot != nritems - 1) { |
5f39d397 CM |
3610 | memmove_extent_buffer(parent, |
3611 | btrfs_node_key_ptr_offset(slot), | |
3612 | btrfs_node_key_ptr_offset(slot + 1), | |
d6025579 CM |
3613 | sizeof(struct btrfs_key_ptr) * |
3614 | (nritems - slot - 1)); | |
bb803951 | 3615 | } |
7518a238 | 3616 | nritems--; |
5f39d397 | 3617 | btrfs_set_header_nritems(parent, nritems); |
7518a238 | 3618 | if (nritems == 0 && parent == root->node) { |
5f39d397 | 3619 | BUG_ON(btrfs_header_level(root->node) != 1); |
bb803951 | 3620 | /* just turn the root into a leaf and break */ |
5f39d397 | 3621 | btrfs_set_header_level(root->node, 0); |
bb803951 | 3622 | } else if (slot == 0) { |
5f39d397 CM |
3623 | struct btrfs_disk_key disk_key; |
3624 | ||
3625 | btrfs_node_key(parent, &disk_key, 0); | |
3626 | wret = fixup_low_keys(trans, root, path, &disk_key, level + 1); | |
0f70abe2 CM |
3627 | if (wret) |
3628 | ret = wret; | |
be0e5c09 | 3629 | } |
d6025579 | 3630 | btrfs_mark_buffer_dirty(parent); |
aa5d6bed | 3631 | return ret; |
be0e5c09 CM |
3632 | } |
3633 | ||
323ac95b CM |
3634 | /* |
3635 | * a helper function to delete the leaf pointed to by path->slots[1] and | |
3636 | * path->nodes[1]. bytenr is the node block pointer, but since the callers | |
3637 | * already know it, it is faster to have them pass it down than to | |
3638 | * read it out of the node again. | |
3639 | * | |
3640 | * This deletes the pointer in path->nodes[1] and frees the leaf | |
3641 | * block extent. zero is returned if it all worked out, < 0 otherwise. | |
3642 | * | |
3643 | * The path must have already been setup for deleting the leaf, including | |
3644 | * all the proper balancing. path->nodes[1] must be locked. | |
3645 | */ | |
3646 | noinline int btrfs_del_leaf(struct btrfs_trans_handle *trans, | |
3647 | struct btrfs_root *root, | |
3648 | struct btrfs_path *path, u64 bytenr) | |
3649 | { | |
3650 | int ret; | |
3651 | u64 root_gen = btrfs_header_generation(path->nodes[1]); | |
4d081c41 CM |
3652 | u64 parent_start = path->nodes[1]->start; |
3653 | u64 parent_owner = btrfs_header_owner(path->nodes[1]); | |
323ac95b CM |
3654 | |
3655 | ret = del_ptr(trans, root, path, 1, path->slots[1]); | |
3656 | if (ret) | |
3657 | return ret; | |
3658 | ||
4d081c41 CM |
3659 | /* |
3660 | * btrfs_free_extent is expensive, we want to make sure we | |
3661 | * aren't holding any locks when we call it | |
3662 | */ | |
3663 | btrfs_unlock_up_safe(path, 0); | |
3664 | ||
323ac95b CM |
3665 | ret = btrfs_free_extent(trans, root, bytenr, |
3666 | btrfs_level_size(root, 0), | |
4d081c41 | 3667 | parent_start, parent_owner, |
3bb1a1bc | 3668 | root_gen, 0, 1); |
323ac95b CM |
3669 | return ret; |
3670 | } | |
74123bd7 CM |
3671 | /* |
3672 | * delete the item at the leaf level in path. If that empties | |
3673 | * the leaf, remove it from the tree | |
3674 | */ | |
85e21bac CM |
3675 | int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root, |
3676 | struct btrfs_path *path, int slot, int nr) | |
be0e5c09 | 3677 | { |
5f39d397 CM |
3678 | struct extent_buffer *leaf; |
3679 | struct btrfs_item *item; | |
85e21bac CM |
3680 | int last_off; |
3681 | int dsize = 0; | |
aa5d6bed CM |
3682 | int ret = 0; |
3683 | int wret; | |
85e21bac | 3684 | int i; |
7518a238 | 3685 | u32 nritems; |
be0e5c09 | 3686 | |
5f39d397 | 3687 | leaf = path->nodes[0]; |
85e21bac CM |
3688 | last_off = btrfs_item_offset_nr(leaf, slot + nr - 1); |
3689 | ||
3690 | for (i = 0; i < nr; i++) | |
3691 | dsize += btrfs_item_size_nr(leaf, slot + i); | |
3692 | ||
5f39d397 | 3693 | nritems = btrfs_header_nritems(leaf); |
be0e5c09 | 3694 | |
85e21bac | 3695 | if (slot + nr != nritems) { |
123abc88 | 3696 | int data_end = leaf_data_end(root, leaf); |
5f39d397 CM |
3697 | |
3698 | memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + | |
d6025579 CM |
3699 | data_end + dsize, |
3700 | btrfs_leaf_data(leaf) + data_end, | |
85e21bac | 3701 | last_off - data_end); |
5f39d397 | 3702 | |
85e21bac | 3703 | for (i = slot + nr; i < nritems; i++) { |
5f39d397 | 3704 | u32 ioff; |
db94535d | 3705 | |
5f39d397 | 3706 | item = btrfs_item_nr(leaf, i); |
db94535d CM |
3707 | if (!leaf->map_token) { |
3708 | map_extent_buffer(leaf, (unsigned long)item, | |
3709 | sizeof(struct btrfs_item), | |
3710 | &leaf->map_token, &leaf->kaddr, | |
3711 | &leaf->map_start, &leaf->map_len, | |
3712 | KM_USER1); | |
3713 | } | |
5f39d397 CM |
3714 | ioff = btrfs_item_offset(leaf, item); |
3715 | btrfs_set_item_offset(leaf, item, ioff + dsize); | |
0783fcfc | 3716 | } |
db94535d CM |
3717 | |
3718 | if (leaf->map_token) { | |
3719 | unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); | |
3720 | leaf->map_token = NULL; | |
3721 | } | |
3722 | ||
5f39d397 | 3723 | memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot), |
85e21bac | 3724 | btrfs_item_nr_offset(slot + nr), |
d6025579 | 3725 | sizeof(struct btrfs_item) * |
85e21bac | 3726 | (nritems - slot - nr)); |
be0e5c09 | 3727 | } |
85e21bac CM |
3728 | btrfs_set_header_nritems(leaf, nritems - nr); |
3729 | nritems -= nr; | |
5f39d397 | 3730 | |
74123bd7 | 3731 | /* delete the leaf if we've emptied it */ |
7518a238 | 3732 | if (nritems == 0) { |
5f39d397 CM |
3733 | if (leaf == root->node) { |
3734 | btrfs_set_header_level(leaf, 0); | |
9a8dd150 | 3735 | } else { |
323ac95b CM |
3736 | ret = btrfs_del_leaf(trans, root, path, leaf->start); |
3737 | BUG_ON(ret); | |
9a8dd150 | 3738 | } |
be0e5c09 | 3739 | } else { |
7518a238 | 3740 | int used = leaf_space_used(leaf, 0, nritems); |
aa5d6bed | 3741 | if (slot == 0) { |
5f39d397 CM |
3742 | struct btrfs_disk_key disk_key; |
3743 | ||
3744 | btrfs_item_key(leaf, &disk_key, 0); | |
e089f05c | 3745 | wret = fixup_low_keys(trans, root, path, |
5f39d397 | 3746 | &disk_key, 1); |
aa5d6bed CM |
3747 | if (wret) |
3748 | ret = wret; | |
3749 | } | |
aa5d6bed | 3750 | |
74123bd7 | 3751 | /* delete the leaf if it is mostly empty */ |
85e21bac | 3752 | if (used < BTRFS_LEAF_DATA_SIZE(root) / 4) { |
be0e5c09 CM |
3753 | /* push_leaf_left fixes the path. |
3754 | * make sure the path still points to our leaf | |
3755 | * for possible call to del_ptr below | |
3756 | */ | |
4920c9ac | 3757 | slot = path->slots[1]; |
5f39d397 CM |
3758 | extent_buffer_get(leaf); |
3759 | ||
85e21bac | 3760 | wret = push_leaf_left(trans, root, path, 1, 1); |
54aa1f4d | 3761 | if (wret < 0 && wret != -ENOSPC) |
aa5d6bed | 3762 | ret = wret; |
5f39d397 CM |
3763 | |
3764 | if (path->nodes[0] == leaf && | |
3765 | btrfs_header_nritems(leaf)) { | |
85e21bac | 3766 | wret = push_leaf_right(trans, root, path, 1, 1); |
54aa1f4d | 3767 | if (wret < 0 && wret != -ENOSPC) |
aa5d6bed CM |
3768 | ret = wret; |
3769 | } | |
5f39d397 CM |
3770 | |
3771 | if (btrfs_header_nritems(leaf) == 0) { | |
323ac95b | 3772 | path->slots[1] = slot; |
d397712b CM |
3773 | ret = btrfs_del_leaf(trans, root, path, |
3774 | leaf->start); | |
323ac95b | 3775 | BUG_ON(ret); |
5f39d397 | 3776 | free_extent_buffer(leaf); |
5de08d7d | 3777 | } else { |
925baedd CM |
3778 | /* if we're still in the path, make sure |
3779 | * we're dirty. Otherwise, one of the | |
3780 | * push_leaf functions must have already | |
3781 | * dirtied this buffer | |
3782 | */ | |
3783 | if (path->nodes[0] == leaf) | |
3784 | btrfs_mark_buffer_dirty(leaf); | |
5f39d397 | 3785 | free_extent_buffer(leaf); |
be0e5c09 | 3786 | } |
d5719762 | 3787 | } else { |
5f39d397 | 3788 | btrfs_mark_buffer_dirty(leaf); |
be0e5c09 CM |
3789 | } |
3790 | } | |
aa5d6bed | 3791 | return ret; |
be0e5c09 CM |
3792 | } |
3793 | ||
7bb86316 | 3794 | /* |
925baedd | 3795 | * search the tree again to find a leaf with lesser keys |
7bb86316 CM |
3796 | * returns 0 if it found something or 1 if there are no lesser leaves. |
3797 | * returns < 0 on io errors. | |
d352ac68 CM |
3798 | * |
3799 | * This may release the path, and so you may lose any locks held at the | |
3800 | * time you call it. | |
7bb86316 CM |
3801 | */ |
3802 | int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path) | |
3803 | { | |
925baedd CM |
3804 | struct btrfs_key key; |
3805 | struct btrfs_disk_key found_key; | |
3806 | int ret; | |
7bb86316 | 3807 | |
925baedd | 3808 | btrfs_item_key_to_cpu(path->nodes[0], &key, 0); |
7bb86316 | 3809 | |
925baedd CM |
3810 | if (key.offset > 0) |
3811 | key.offset--; | |
3812 | else if (key.type > 0) | |
3813 | key.type--; | |
3814 | else if (key.objectid > 0) | |
3815 | key.objectid--; | |
3816 | else | |
3817 | return 1; | |
7bb86316 | 3818 | |
925baedd CM |
3819 | btrfs_release_path(root, path); |
3820 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
3821 | if (ret < 0) | |
3822 | return ret; | |
3823 | btrfs_item_key(path->nodes[0], &found_key, 0); | |
3824 | ret = comp_keys(&found_key, &key); | |
3825 | if (ret < 0) | |
3826 | return 0; | |
3827 | return 1; | |
7bb86316 CM |
3828 | } |
3829 | ||
3f157a2f CM |
3830 | /* |
3831 | * A helper function to walk down the tree starting at min_key, and looking | |
3832 | * for nodes or leaves that are either in cache or have a minimum | |
d352ac68 | 3833 | * transaction id. This is used by the btree defrag code, and tree logging |
3f157a2f CM |
3834 | * |
3835 | * This does not cow, but it does stuff the starting key it finds back | |
3836 | * into min_key, so you can call btrfs_search_slot with cow=1 on the | |
3837 | * key and get a writable path. | |
3838 | * | |
3839 | * This does lock as it descends, and path->keep_locks should be set | |
3840 | * to 1 by the caller. | |
3841 | * | |
3842 | * This honors path->lowest_level to prevent descent past a given level | |
3843 | * of the tree. | |
3844 | * | |
d352ac68 CM |
3845 | * min_trans indicates the oldest transaction that you are interested |
3846 | * in walking through. Any nodes or leaves older than min_trans are | |
3847 | * skipped over (without reading them). | |
3848 | * | |
3f157a2f CM |
3849 | * returns zero if something useful was found, < 0 on error and 1 if there |
3850 | * was nothing in the tree that matched the search criteria. | |
3851 | */ | |
3852 | int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key, | |
e02119d5 | 3853 | struct btrfs_key *max_key, |
3f157a2f CM |
3854 | struct btrfs_path *path, int cache_only, |
3855 | u64 min_trans) | |
3856 | { | |
3857 | struct extent_buffer *cur; | |
3858 | struct btrfs_key found_key; | |
3859 | int slot; | |
9652480b | 3860 | int sret; |
3f157a2f CM |
3861 | u32 nritems; |
3862 | int level; | |
3863 | int ret = 1; | |
3864 | ||
934d375b | 3865 | WARN_ON(!path->keep_locks); |
3f157a2f CM |
3866 | again: |
3867 | cur = btrfs_lock_root_node(root); | |
3868 | level = btrfs_header_level(cur); | |
e02119d5 | 3869 | WARN_ON(path->nodes[level]); |
3f157a2f CM |
3870 | path->nodes[level] = cur; |
3871 | path->locks[level] = 1; | |
3872 | ||
3873 | if (btrfs_header_generation(cur) < min_trans) { | |
3874 | ret = 1; | |
3875 | goto out; | |
3876 | } | |
d397712b | 3877 | while (1) { |
3f157a2f CM |
3878 | nritems = btrfs_header_nritems(cur); |
3879 | level = btrfs_header_level(cur); | |
9652480b | 3880 | sret = bin_search(cur, min_key, level, &slot); |
3f157a2f | 3881 | |
323ac95b CM |
3882 | /* at the lowest level, we're done, setup the path and exit */ |
3883 | if (level == path->lowest_level) { | |
e02119d5 CM |
3884 | if (slot >= nritems) |
3885 | goto find_next_key; | |
3f157a2f CM |
3886 | ret = 0; |
3887 | path->slots[level] = slot; | |
3888 | btrfs_item_key_to_cpu(cur, &found_key, slot); | |
3889 | goto out; | |
3890 | } | |
9652480b Y |
3891 | if (sret && slot > 0) |
3892 | slot--; | |
3f157a2f CM |
3893 | /* |
3894 | * check this node pointer against the cache_only and | |
3895 | * min_trans parameters. If it isn't in cache or is too | |
3896 | * old, skip to the next one. | |
3897 | */ | |
d397712b | 3898 | while (slot < nritems) { |
3f157a2f CM |
3899 | u64 blockptr; |
3900 | u64 gen; | |
3901 | struct extent_buffer *tmp; | |
e02119d5 CM |
3902 | struct btrfs_disk_key disk_key; |
3903 | ||
3f157a2f CM |
3904 | blockptr = btrfs_node_blockptr(cur, slot); |
3905 | gen = btrfs_node_ptr_generation(cur, slot); | |
3906 | if (gen < min_trans) { | |
3907 | slot++; | |
3908 | continue; | |
3909 | } | |
3910 | if (!cache_only) | |
3911 | break; | |
3912 | ||
e02119d5 CM |
3913 | if (max_key) { |
3914 | btrfs_node_key(cur, &disk_key, slot); | |
3915 | if (comp_keys(&disk_key, max_key) >= 0) { | |
3916 | ret = 1; | |
3917 | goto out; | |
3918 | } | |
3919 | } | |
3920 | ||
3f157a2f CM |
3921 | tmp = btrfs_find_tree_block(root, blockptr, |
3922 | btrfs_level_size(root, level - 1)); | |
3923 | ||
3924 | if (tmp && btrfs_buffer_uptodate(tmp, gen)) { | |
3925 | free_extent_buffer(tmp); | |
3926 | break; | |
3927 | } | |
3928 | if (tmp) | |
3929 | free_extent_buffer(tmp); | |
3930 | slot++; | |
3931 | } | |
e02119d5 | 3932 | find_next_key: |
3f157a2f CM |
3933 | /* |
3934 | * we didn't find a candidate key in this node, walk forward | |
3935 | * and find another one | |
3936 | */ | |
3937 | if (slot >= nritems) { | |
e02119d5 | 3938 | path->slots[level] = slot; |
b4ce94de | 3939 | btrfs_set_path_blocking(path); |
e02119d5 | 3940 | sret = btrfs_find_next_key(root, path, min_key, level, |
3f157a2f | 3941 | cache_only, min_trans); |
e02119d5 | 3942 | if (sret == 0) { |
3f157a2f CM |
3943 | btrfs_release_path(root, path); |
3944 | goto again; | |
3945 | } else { | |
3946 | goto out; | |
3947 | } | |
3948 | } | |
3949 | /* save our key for returning back */ | |
3950 | btrfs_node_key_to_cpu(cur, &found_key, slot); | |
3951 | path->slots[level] = slot; | |
3952 | if (level == path->lowest_level) { | |
3953 | ret = 0; | |
3954 | unlock_up(path, level, 1); | |
3955 | goto out; | |
3956 | } | |
b4ce94de | 3957 | btrfs_set_path_blocking(path); |
3f157a2f CM |
3958 | cur = read_node_slot(root, cur, slot); |
3959 | ||
3960 | btrfs_tree_lock(cur); | |
b4ce94de | 3961 | |
3f157a2f CM |
3962 | path->locks[level - 1] = 1; |
3963 | path->nodes[level - 1] = cur; | |
3964 | unlock_up(path, level, 1); | |
4008c04a | 3965 | btrfs_clear_path_blocking(path, NULL); |
3f157a2f CM |
3966 | } |
3967 | out: | |
3968 | if (ret == 0) | |
3969 | memcpy(min_key, &found_key, sizeof(found_key)); | |
b4ce94de | 3970 | btrfs_set_path_blocking(path); |
3f157a2f CM |
3971 | return ret; |
3972 | } | |
3973 | ||
3974 | /* | |
3975 | * this is similar to btrfs_next_leaf, but does not try to preserve | |
3976 | * and fixup the path. It looks for and returns the next key in the | |
3977 | * tree based on the current path and the cache_only and min_trans | |
3978 | * parameters. | |
3979 | * | |
3980 | * 0 is returned if another key is found, < 0 if there are any errors | |
3981 | * and 1 is returned if there are no higher keys in the tree | |
3982 | * | |
3983 | * path->keep_locks should be set to 1 on the search made before | |
3984 | * calling this function. | |
3985 | */ | |
e7a84565 | 3986 | int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path, |
3f157a2f CM |
3987 | struct btrfs_key *key, int lowest_level, |
3988 | int cache_only, u64 min_trans) | |
e7a84565 CM |
3989 | { |
3990 | int level = lowest_level; | |
3991 | int slot; | |
3992 | struct extent_buffer *c; | |
3993 | ||
934d375b | 3994 | WARN_ON(!path->keep_locks); |
d397712b | 3995 | while (level < BTRFS_MAX_LEVEL) { |
e7a84565 CM |
3996 | if (!path->nodes[level]) |
3997 | return 1; | |
3998 | ||
3999 | slot = path->slots[level] + 1; | |
4000 | c = path->nodes[level]; | |
3f157a2f | 4001 | next: |
e7a84565 CM |
4002 | if (slot >= btrfs_header_nritems(c)) { |
4003 | level++; | |
d397712b | 4004 | if (level == BTRFS_MAX_LEVEL) |
e7a84565 | 4005 | return 1; |
e7a84565 CM |
4006 | continue; |
4007 | } | |
4008 | if (level == 0) | |
4009 | btrfs_item_key_to_cpu(c, key, slot); | |
3f157a2f CM |
4010 | else { |
4011 | u64 blockptr = btrfs_node_blockptr(c, slot); | |
4012 | u64 gen = btrfs_node_ptr_generation(c, slot); | |
4013 | ||
4014 | if (cache_only) { | |
4015 | struct extent_buffer *cur; | |
4016 | cur = btrfs_find_tree_block(root, blockptr, | |
4017 | btrfs_level_size(root, level - 1)); | |
4018 | if (!cur || !btrfs_buffer_uptodate(cur, gen)) { | |
4019 | slot++; | |
4020 | if (cur) | |
4021 | free_extent_buffer(cur); | |
4022 | goto next; | |
4023 | } | |
4024 | free_extent_buffer(cur); | |
4025 | } | |
4026 | if (gen < min_trans) { | |
4027 | slot++; | |
4028 | goto next; | |
4029 | } | |
e7a84565 | 4030 | btrfs_node_key_to_cpu(c, key, slot); |
3f157a2f | 4031 | } |
e7a84565 CM |
4032 | return 0; |
4033 | } | |
4034 | return 1; | |
4035 | } | |
4036 | ||
97571fd0 | 4037 | /* |
925baedd | 4038 | * search the tree again to find a leaf with greater keys |
0f70abe2 CM |
4039 | * returns 0 if it found something or 1 if there are no greater leaves. |
4040 | * returns < 0 on io errors. | |
97571fd0 | 4041 | */ |
234b63a0 | 4042 | int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path) |
d97e63b6 CM |
4043 | { |
4044 | int slot; | |
4045 | int level = 1; | |
5f39d397 CM |
4046 | struct extent_buffer *c; |
4047 | struct extent_buffer *next = NULL; | |
925baedd CM |
4048 | struct btrfs_key key; |
4049 | u32 nritems; | |
4050 | int ret; | |
4051 | ||
4052 | nritems = btrfs_header_nritems(path->nodes[0]); | |
d397712b | 4053 | if (nritems == 0) |
925baedd | 4054 | return 1; |
925baedd CM |
4055 | |
4056 | btrfs_item_key_to_cpu(path->nodes[0], &key, nritems - 1); | |
4057 | ||
925baedd | 4058 | btrfs_release_path(root, path); |
a2135011 | 4059 | path->keep_locks = 1; |
925baedd CM |
4060 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); |
4061 | path->keep_locks = 0; | |
4062 | ||
4063 | if (ret < 0) | |
4064 | return ret; | |
4065 | ||
b4ce94de | 4066 | btrfs_set_path_blocking(path); |
a2135011 | 4067 | nritems = btrfs_header_nritems(path->nodes[0]); |
168fd7d2 CM |
4068 | /* |
4069 | * by releasing the path above we dropped all our locks. A balance | |
4070 | * could have added more items next to the key that used to be | |
4071 | * at the very end of the block. So, check again here and | |
4072 | * advance the path if there are now more items available. | |
4073 | */ | |
a2135011 | 4074 | if (nritems > 0 && path->slots[0] < nritems - 1) { |
168fd7d2 | 4075 | path->slots[0]++; |
925baedd CM |
4076 | goto done; |
4077 | } | |
d97e63b6 | 4078 | |
d397712b | 4079 | while (level < BTRFS_MAX_LEVEL) { |
d97e63b6 | 4080 | if (!path->nodes[level]) |
0f70abe2 | 4081 | return 1; |
5f39d397 | 4082 | |
d97e63b6 CM |
4083 | slot = path->slots[level] + 1; |
4084 | c = path->nodes[level]; | |
5f39d397 | 4085 | if (slot >= btrfs_header_nritems(c)) { |
d97e63b6 | 4086 | level++; |
d397712b | 4087 | if (level == BTRFS_MAX_LEVEL) |
7bb86316 | 4088 | return 1; |
d97e63b6 CM |
4089 | continue; |
4090 | } | |
5f39d397 | 4091 | |
925baedd CM |
4092 | if (next) { |
4093 | btrfs_tree_unlock(next); | |
5f39d397 | 4094 | free_extent_buffer(next); |
925baedd | 4095 | } |
5f39d397 | 4096 | |
b4ce94de | 4097 | /* the path was set to blocking above */ |
0bd40a71 CM |
4098 | if (level == 1 && (path->locks[1] || path->skip_locking) && |
4099 | path->reada) | |
01f46658 | 4100 | reada_for_search(root, path, level, slot, 0); |
5f39d397 | 4101 | |
ca7a79ad | 4102 | next = read_node_slot(root, c, slot); |
5cd57b2c CM |
4103 | if (!path->skip_locking) { |
4104 | WARN_ON(!btrfs_tree_locked(c)); | |
4105 | btrfs_tree_lock(next); | |
b4ce94de | 4106 | btrfs_set_lock_blocking(next); |
5cd57b2c | 4107 | } |
d97e63b6 CM |
4108 | break; |
4109 | } | |
4110 | path->slots[level] = slot; | |
d397712b | 4111 | while (1) { |
d97e63b6 CM |
4112 | level--; |
4113 | c = path->nodes[level]; | |
925baedd CM |
4114 | if (path->locks[level]) |
4115 | btrfs_tree_unlock(c); | |
5f39d397 | 4116 | free_extent_buffer(c); |
d97e63b6 CM |
4117 | path->nodes[level] = next; |
4118 | path->slots[level] = 0; | |
a74a4b97 CM |
4119 | if (!path->skip_locking) |
4120 | path->locks[level] = 1; | |
d97e63b6 CM |
4121 | if (!level) |
4122 | break; | |
b4ce94de CM |
4123 | |
4124 | btrfs_set_path_blocking(path); | |
925baedd CM |
4125 | if (level == 1 && path->locks[1] && path->reada) |
4126 | reada_for_search(root, path, level, slot, 0); | |
ca7a79ad | 4127 | next = read_node_slot(root, next, 0); |
5cd57b2c CM |
4128 | if (!path->skip_locking) { |
4129 | WARN_ON(!btrfs_tree_locked(path->nodes[level])); | |
4130 | btrfs_tree_lock(next); | |
b4ce94de | 4131 | btrfs_set_lock_blocking(next); |
5cd57b2c | 4132 | } |
d97e63b6 | 4133 | } |
925baedd CM |
4134 | done: |
4135 | unlock_up(path, 0, 1); | |
d97e63b6 CM |
4136 | return 0; |
4137 | } | |
0b86a832 | 4138 | |
3f157a2f CM |
4139 | /* |
4140 | * this uses btrfs_prev_leaf to walk backwards in the tree, and keeps | |
4141 | * searching until it gets past min_objectid or finds an item of 'type' | |
4142 | * | |
4143 | * returns 0 if something is found, 1 if nothing was found and < 0 on error | |
4144 | */ | |
0b86a832 CM |
4145 | int btrfs_previous_item(struct btrfs_root *root, |
4146 | struct btrfs_path *path, u64 min_objectid, | |
4147 | int type) | |
4148 | { | |
4149 | struct btrfs_key found_key; | |
4150 | struct extent_buffer *leaf; | |
e02119d5 | 4151 | u32 nritems; |
0b86a832 CM |
4152 | int ret; |
4153 | ||
d397712b | 4154 | while (1) { |
0b86a832 | 4155 | if (path->slots[0] == 0) { |
b4ce94de | 4156 | btrfs_set_path_blocking(path); |
0b86a832 CM |
4157 | ret = btrfs_prev_leaf(root, path); |
4158 | if (ret != 0) | |
4159 | return ret; | |
4160 | } else { | |
4161 | path->slots[0]--; | |
4162 | } | |
4163 | leaf = path->nodes[0]; | |
e02119d5 CM |
4164 | nritems = btrfs_header_nritems(leaf); |
4165 | if (nritems == 0) | |
4166 | return 1; | |
4167 | if (path->slots[0] == nritems) | |
4168 | path->slots[0]--; | |
4169 | ||
0b86a832 CM |
4170 | btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); |
4171 | if (found_key.type == type) | |
4172 | return 0; | |
e02119d5 CM |
4173 | if (found_key.objectid < min_objectid) |
4174 | break; | |
4175 | if (found_key.objectid == min_objectid && | |
4176 | found_key.type < type) | |
4177 | break; | |
0b86a832 CM |
4178 | } |
4179 | return 1; | |
4180 | } |