Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
[deliverable/linux.git] / fs / ext4 / extents.c
1 /*
2 * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com
3 * Written by Alex Tomas <alex@clusterfs.com>
4 *
5 * Architecture independence:
6 * Copyright (c) 2005, Bull S.A.
7 * Written by Pierre Peiffer <pierre.peiffer@bull.net>
8 *
9 * This program is free software; you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License version 2 as
11 * published by the Free Software Foundation.
12 *
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
17 *
18 * You should have received a copy of the GNU General Public Licens
19 * along with this program; if not, write to the Free Software
20 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-
21 */
22
23 /*
24 * Extents support for EXT4
25 *
26 * TODO:
27 * - ext4*_error() should be used in some situations
28 * - analyze all BUG()/BUG_ON(), use -EIO where appropriate
29 * - smart tree reduction
30 */
31
32 #include <linux/module.h>
33 #include <linux/fs.h>
34 #include <linux/time.h>
35 #include <linux/jbd2.h>
36 #include <linux/highuid.h>
37 #include <linux/pagemap.h>
38 #include <linux/quotaops.h>
39 #include <linux/string.h>
40 #include <linux/slab.h>
41 #include <linux/falloc.h>
42 #include <asm/uaccess.h>
43 #include <linux/fiemap.h>
44 #include "ext4_jbd2.h"
45 #include "ext4_extents.h"
46
47
48 /*
49 * ext_pblock:
50 * combine low and high parts of physical block number into ext4_fsblk_t
51 */
52 static ext4_fsblk_t ext_pblock(struct ext4_extent *ex)
53 {
54 ext4_fsblk_t block;
55
56 block = le32_to_cpu(ex->ee_start_lo);
57 block |= ((ext4_fsblk_t) le16_to_cpu(ex->ee_start_hi) << 31) << 1;
58 return block;
59 }
60
61 /*
62 * idx_pblock:
63 * combine low and high parts of a leaf physical block number into ext4_fsblk_t
64 */
65 ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix)
66 {
67 ext4_fsblk_t block;
68
69 block = le32_to_cpu(ix->ei_leaf_lo);
70 block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1;
71 return block;
72 }
73
74 /*
75 * ext4_ext_store_pblock:
76 * stores a large physical block number into an extent struct,
77 * breaking it into parts
78 */
79 void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb)
80 {
81 ex->ee_start_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff));
82 ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff);
83 }
84
85 /*
86 * ext4_idx_store_pblock:
87 * stores a large physical block number into an index struct,
88 * breaking it into parts
89 */
90 static void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb)
91 {
92 ix->ei_leaf_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff));
93 ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff);
94 }
95
96 static int ext4_ext_journal_restart(handle_t *handle, int needed)
97 {
98 int err;
99
100 if (!ext4_handle_valid(handle))
101 return 0;
102 if (handle->h_buffer_credits > needed)
103 return 0;
104 err = ext4_journal_extend(handle, needed);
105 if (err <= 0)
106 return err;
107 return ext4_journal_restart(handle, needed);
108 }
109
110 /*
111 * could return:
112 * - EROFS
113 * - ENOMEM
114 */
115 static int ext4_ext_get_access(handle_t *handle, struct inode *inode,
116 struct ext4_ext_path *path)
117 {
118 if (path->p_bh) {
119 /* path points to block */
120 return ext4_journal_get_write_access(handle, path->p_bh);
121 }
122 /* path points to leaf/index in inode body */
123 /* we use in-core data, no need to protect them */
124 return 0;
125 }
126
127 /*
128 * could return:
129 * - EROFS
130 * - ENOMEM
131 * - EIO
132 */
133 static int ext4_ext_dirty(handle_t *handle, struct inode *inode,
134 struct ext4_ext_path *path)
135 {
136 int err;
137 if (path->p_bh) {
138 /* path points to block */
139 err = ext4_handle_dirty_metadata(handle, inode, path->p_bh);
140 } else {
141 /* path points to leaf/index in inode body */
142 err = ext4_mark_inode_dirty(handle, inode);
143 }
144 return err;
145 }
146
147 static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode,
148 struct ext4_ext_path *path,
149 ext4_lblk_t block)
150 {
151 struct ext4_inode_info *ei = EXT4_I(inode);
152 ext4_fsblk_t bg_start;
153 ext4_fsblk_t last_block;
154 ext4_grpblk_t colour;
155 ext4_group_t block_group;
156 int flex_size = ext4_flex_bg_size(EXT4_SB(inode->i_sb));
157 int depth;
158
159 if (path) {
160 struct ext4_extent *ex;
161 depth = path->p_depth;
162
163 /* try to predict block placement */
164 ex = path[depth].p_ext;
165 if (ex)
166 return ext_pblock(ex)+(block-le32_to_cpu(ex->ee_block));
167
168 /* it looks like index is empty;
169 * try to find starting block from index itself */
170 if (path[depth].p_bh)
171 return path[depth].p_bh->b_blocknr;
172 }
173
174 /* OK. use inode's group */
175 block_group = ei->i_block_group;
176 if (flex_size >= EXT4_FLEX_SIZE_DIR_ALLOC_SCHEME) {
177 /*
178 * If there are at least EXT4_FLEX_SIZE_DIR_ALLOC_SCHEME
179 * block groups per flexgroup, reserve the first block
180 * group for directories and special files. Regular
181 * files will start at the second block group. This
182 * tends to speed up directory access and improves
183 * fsck times.
184 */
185 block_group &= ~(flex_size-1);
186 if (S_ISREG(inode->i_mode))
187 block_group++;
188 }
189 bg_start = (block_group * EXT4_BLOCKS_PER_GROUP(inode->i_sb)) +
190 le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_first_data_block);
191 last_block = ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es) - 1;
192
193 /*
194 * If we are doing delayed allocation, we don't need take
195 * colour into account.
196 */
197 if (test_opt(inode->i_sb, DELALLOC))
198 return bg_start;
199
200 if (bg_start + EXT4_BLOCKS_PER_GROUP(inode->i_sb) <= last_block)
201 colour = (current->pid % 16) *
202 (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16);
203 else
204 colour = (current->pid % 16) * ((last_block - bg_start) / 16);
205 return bg_start + colour + block;
206 }
207
208 /*
209 * Allocation for a meta data block
210 */
211 static ext4_fsblk_t
212 ext4_ext_new_meta_block(handle_t *handle, struct inode *inode,
213 struct ext4_ext_path *path,
214 struct ext4_extent *ex, int *err)
215 {
216 ext4_fsblk_t goal, newblock;
217
218 goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block));
219 newblock = ext4_new_meta_blocks(handle, inode, goal, NULL, err);
220 return newblock;
221 }
222
223 static int ext4_ext_space_block(struct inode *inode)
224 {
225 int size;
226
227 size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
228 / sizeof(struct ext4_extent);
229 #ifdef AGGRESSIVE_TEST
230 if (size > 6)
231 size = 6;
232 #endif
233 return size;
234 }
235
236 static int ext4_ext_space_block_idx(struct inode *inode)
237 {
238 int size;
239
240 size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
241 / sizeof(struct ext4_extent_idx);
242 #ifdef AGGRESSIVE_TEST
243 if (size > 5)
244 size = 5;
245 #endif
246 return size;
247 }
248
249 static int ext4_ext_space_root(struct inode *inode)
250 {
251 int size;
252
253 size = sizeof(EXT4_I(inode)->i_data);
254 size -= sizeof(struct ext4_extent_header);
255 size /= sizeof(struct ext4_extent);
256 #ifdef AGGRESSIVE_TEST
257 if (size > 3)
258 size = 3;
259 #endif
260 return size;
261 }
262
263 static int ext4_ext_space_root_idx(struct inode *inode)
264 {
265 int size;
266
267 size = sizeof(EXT4_I(inode)->i_data);
268 size -= sizeof(struct ext4_extent_header);
269 size /= sizeof(struct ext4_extent_idx);
270 #ifdef AGGRESSIVE_TEST
271 if (size > 4)
272 size = 4;
273 #endif
274 return size;
275 }
276
277 /*
278 * Calculate the number of metadata blocks needed
279 * to allocate @blocks
280 * Worse case is one block per extent
281 */
282 int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks)
283 {
284 int lcap, icap, rcap, leafs, idxs, num;
285 int newextents = blocks;
286
287 rcap = ext4_ext_space_root_idx(inode);
288 lcap = ext4_ext_space_block(inode);
289 icap = ext4_ext_space_block_idx(inode);
290
291 /* number of new leaf blocks needed */
292 num = leafs = (newextents + lcap - 1) / lcap;
293
294 /*
295 * Worse case, we need separate index block(s)
296 * to link all new leaf blocks
297 */
298 idxs = (leafs + icap - 1) / icap;
299 do {
300 num += idxs;
301 idxs = (idxs + icap - 1) / icap;
302 } while (idxs > rcap);
303
304 return num;
305 }
306
307 static int
308 ext4_ext_max_entries(struct inode *inode, int depth)
309 {
310 int max;
311
312 if (depth == ext_depth(inode)) {
313 if (depth == 0)
314 max = ext4_ext_space_root(inode);
315 else
316 max = ext4_ext_space_root_idx(inode);
317 } else {
318 if (depth == 0)
319 max = ext4_ext_space_block(inode);
320 else
321 max = ext4_ext_space_block_idx(inode);
322 }
323
324 return max;
325 }
326
327 static int ext4_valid_extent(struct inode *inode, struct ext4_extent *ext)
328 {
329 ext4_fsblk_t block = ext_pblock(ext);
330 int len = ext4_ext_get_actual_len(ext);
331 struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
332 if (unlikely(block < le32_to_cpu(es->s_first_data_block) ||
333 ((block + len) > ext4_blocks_count(es))))
334 return 0;
335 else
336 return 1;
337 }
338
339 static int ext4_valid_extent_idx(struct inode *inode,
340 struct ext4_extent_idx *ext_idx)
341 {
342 ext4_fsblk_t block = idx_pblock(ext_idx);
343 struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
344 if (unlikely(block < le32_to_cpu(es->s_first_data_block) ||
345 (block >= ext4_blocks_count(es))))
346 return 0;
347 else
348 return 1;
349 }
350
351 static int ext4_valid_extent_entries(struct inode *inode,
352 struct ext4_extent_header *eh,
353 int depth)
354 {
355 struct ext4_extent *ext;
356 struct ext4_extent_idx *ext_idx;
357 unsigned short entries;
358 if (eh->eh_entries == 0)
359 return 1;
360
361 entries = le16_to_cpu(eh->eh_entries);
362
363 if (depth == 0) {
364 /* leaf entries */
365 ext = EXT_FIRST_EXTENT(eh);
366 while (entries) {
367 if (!ext4_valid_extent(inode, ext))
368 return 0;
369 ext++;
370 entries--;
371 }
372 } else {
373 ext_idx = EXT_FIRST_INDEX(eh);
374 while (entries) {
375 if (!ext4_valid_extent_idx(inode, ext_idx))
376 return 0;
377 ext_idx++;
378 entries--;
379 }
380 }
381 return 1;
382 }
383
384 static int __ext4_ext_check(const char *function, struct inode *inode,
385 struct ext4_extent_header *eh,
386 int depth)
387 {
388 const char *error_msg;
389 int max = 0;
390
391 if (unlikely(eh->eh_magic != EXT4_EXT_MAGIC)) {
392 error_msg = "invalid magic";
393 goto corrupted;
394 }
395 if (unlikely(le16_to_cpu(eh->eh_depth) != depth)) {
396 error_msg = "unexpected eh_depth";
397 goto corrupted;
398 }
399 if (unlikely(eh->eh_max == 0)) {
400 error_msg = "invalid eh_max";
401 goto corrupted;
402 }
403 max = ext4_ext_max_entries(inode, depth);
404 if (unlikely(le16_to_cpu(eh->eh_max) > max)) {
405 error_msg = "too large eh_max";
406 goto corrupted;
407 }
408 if (unlikely(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max))) {
409 error_msg = "invalid eh_entries";
410 goto corrupted;
411 }
412 if (!ext4_valid_extent_entries(inode, eh, depth)) {
413 error_msg = "invalid extent entries";
414 goto corrupted;
415 }
416 return 0;
417
418 corrupted:
419 ext4_error(inode->i_sb, function,
420 "bad header/extent in inode #%lu: %s - magic %x, "
421 "entries %u, max %u(%u), depth %u(%u)",
422 inode->i_ino, error_msg, le16_to_cpu(eh->eh_magic),
423 le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max),
424 max, le16_to_cpu(eh->eh_depth), depth);
425
426 return -EIO;
427 }
428
429 #define ext4_ext_check(inode, eh, depth) \
430 __ext4_ext_check(__func__, inode, eh, depth)
431
432 int ext4_ext_check_inode(struct inode *inode)
433 {
434 return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode));
435 }
436
437 #ifdef EXT_DEBUG
438 static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
439 {
440 int k, l = path->p_depth;
441
442 ext_debug("path:");
443 for (k = 0; k <= l; k++, path++) {
444 if (path->p_idx) {
445 ext_debug(" %d->%llu", le32_to_cpu(path->p_idx->ei_block),
446 idx_pblock(path->p_idx));
447 } else if (path->p_ext) {
448 ext_debug(" %d:%d:%llu ",
449 le32_to_cpu(path->p_ext->ee_block),
450 ext4_ext_get_actual_len(path->p_ext),
451 ext_pblock(path->p_ext));
452 } else
453 ext_debug(" []");
454 }
455 ext_debug("\n");
456 }
457
458 static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path)
459 {
460 int depth = ext_depth(inode);
461 struct ext4_extent_header *eh;
462 struct ext4_extent *ex;
463 int i;
464
465 if (!path)
466 return;
467
468 eh = path[depth].p_hdr;
469 ex = EXT_FIRST_EXTENT(eh);
470
471 for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) {
472 ext_debug("%d:%d:%llu ", le32_to_cpu(ex->ee_block),
473 ext4_ext_get_actual_len(ex), ext_pblock(ex));
474 }
475 ext_debug("\n");
476 }
477 #else
478 #define ext4_ext_show_path(inode, path)
479 #define ext4_ext_show_leaf(inode, path)
480 #endif
481
482 void ext4_ext_drop_refs(struct ext4_ext_path *path)
483 {
484 int depth = path->p_depth;
485 int i;
486
487 for (i = 0; i <= depth; i++, path++)
488 if (path->p_bh) {
489 brelse(path->p_bh);
490 path->p_bh = NULL;
491 }
492 }
493
494 /*
495 * ext4_ext_binsearch_idx:
496 * binary search for the closest index of the given block
497 * the header must be checked before calling this
498 */
499 static void
500 ext4_ext_binsearch_idx(struct inode *inode,
501 struct ext4_ext_path *path, ext4_lblk_t block)
502 {
503 struct ext4_extent_header *eh = path->p_hdr;
504 struct ext4_extent_idx *r, *l, *m;
505
506
507 ext_debug("binsearch for %u(idx): ", block);
508
509 l = EXT_FIRST_INDEX(eh) + 1;
510 r = EXT_LAST_INDEX(eh);
511 while (l <= r) {
512 m = l + (r - l) / 2;
513 if (block < le32_to_cpu(m->ei_block))
514 r = m - 1;
515 else
516 l = m + 1;
517 ext_debug("%p(%u):%p(%u):%p(%u) ", l, le32_to_cpu(l->ei_block),
518 m, le32_to_cpu(m->ei_block),
519 r, le32_to_cpu(r->ei_block));
520 }
521
522 path->p_idx = l - 1;
523 ext_debug(" -> %d->%lld ", le32_to_cpu(path->p_idx->ei_block),
524 idx_pblock(path->p_idx));
525
526 #ifdef CHECK_BINSEARCH
527 {
528 struct ext4_extent_idx *chix, *ix;
529 int k;
530
531 chix = ix = EXT_FIRST_INDEX(eh);
532 for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) {
533 if (k != 0 &&
534 le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) {
535 printk(KERN_DEBUG "k=%d, ix=0x%p, "
536 "first=0x%p\n", k,
537 ix, EXT_FIRST_INDEX(eh));
538 printk(KERN_DEBUG "%u <= %u\n",
539 le32_to_cpu(ix->ei_block),
540 le32_to_cpu(ix[-1].ei_block));
541 }
542 BUG_ON(k && le32_to_cpu(ix->ei_block)
543 <= le32_to_cpu(ix[-1].ei_block));
544 if (block < le32_to_cpu(ix->ei_block))
545 break;
546 chix = ix;
547 }
548 BUG_ON(chix != path->p_idx);
549 }
550 #endif
551
552 }
553
554 /*
555 * ext4_ext_binsearch:
556 * binary search for closest extent of the given block
557 * the header must be checked before calling this
558 */
559 static void
560 ext4_ext_binsearch(struct inode *inode,
561 struct ext4_ext_path *path, ext4_lblk_t block)
562 {
563 struct ext4_extent_header *eh = path->p_hdr;
564 struct ext4_extent *r, *l, *m;
565
566 if (eh->eh_entries == 0) {
567 /*
568 * this leaf is empty:
569 * we get such a leaf in split/add case
570 */
571 return;
572 }
573
574 ext_debug("binsearch for %u: ", block);
575
576 l = EXT_FIRST_EXTENT(eh) + 1;
577 r = EXT_LAST_EXTENT(eh);
578
579 while (l <= r) {
580 m = l + (r - l) / 2;
581 if (block < le32_to_cpu(m->ee_block))
582 r = m - 1;
583 else
584 l = m + 1;
585 ext_debug("%p(%u):%p(%u):%p(%u) ", l, le32_to_cpu(l->ee_block),
586 m, le32_to_cpu(m->ee_block),
587 r, le32_to_cpu(r->ee_block));
588 }
589
590 path->p_ext = l - 1;
591 ext_debug(" -> %d:%llu:%d ",
592 le32_to_cpu(path->p_ext->ee_block),
593 ext_pblock(path->p_ext),
594 ext4_ext_get_actual_len(path->p_ext));
595
596 #ifdef CHECK_BINSEARCH
597 {
598 struct ext4_extent *chex, *ex;
599 int k;
600
601 chex = ex = EXT_FIRST_EXTENT(eh);
602 for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ex++) {
603 BUG_ON(k && le32_to_cpu(ex->ee_block)
604 <= le32_to_cpu(ex[-1].ee_block));
605 if (block < le32_to_cpu(ex->ee_block))
606 break;
607 chex = ex;
608 }
609 BUG_ON(chex != path->p_ext);
610 }
611 #endif
612
613 }
614
615 int ext4_ext_tree_init(handle_t *handle, struct inode *inode)
616 {
617 struct ext4_extent_header *eh;
618
619 eh = ext_inode_hdr(inode);
620 eh->eh_depth = 0;
621 eh->eh_entries = 0;
622 eh->eh_magic = EXT4_EXT_MAGIC;
623 eh->eh_max = cpu_to_le16(ext4_ext_space_root(inode));
624 ext4_mark_inode_dirty(handle, inode);
625 ext4_ext_invalidate_cache(inode);
626 return 0;
627 }
628
629 struct ext4_ext_path *
630 ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
631 struct ext4_ext_path *path)
632 {
633 struct ext4_extent_header *eh;
634 struct buffer_head *bh;
635 short int depth, i, ppos = 0, alloc = 0;
636
637 eh = ext_inode_hdr(inode);
638 depth = ext_depth(inode);
639
640 /* account possible depth increase */
641 if (!path) {
642 path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 2),
643 GFP_NOFS);
644 if (!path)
645 return ERR_PTR(-ENOMEM);
646 alloc = 1;
647 }
648 path[0].p_hdr = eh;
649 path[0].p_bh = NULL;
650
651 i = depth;
652 /* walk through the tree */
653 while (i) {
654 int need_to_validate = 0;
655
656 ext_debug("depth %d: num %d, max %d\n",
657 ppos, le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max));
658
659 ext4_ext_binsearch_idx(inode, path + ppos, block);
660 path[ppos].p_block = idx_pblock(path[ppos].p_idx);
661 path[ppos].p_depth = i;
662 path[ppos].p_ext = NULL;
663
664 bh = sb_getblk(inode->i_sb, path[ppos].p_block);
665 if (unlikely(!bh))
666 goto err;
667 if (!bh_uptodate_or_lock(bh)) {
668 if (bh_submit_read(bh) < 0) {
669 put_bh(bh);
670 goto err;
671 }
672 /* validate the extent entries */
673 need_to_validate = 1;
674 }
675 eh = ext_block_hdr(bh);
676 ppos++;
677 BUG_ON(ppos > depth);
678 path[ppos].p_bh = bh;
679 path[ppos].p_hdr = eh;
680 i--;
681
682 if (need_to_validate && ext4_ext_check(inode, eh, i))
683 goto err;
684 }
685
686 path[ppos].p_depth = i;
687 path[ppos].p_ext = NULL;
688 path[ppos].p_idx = NULL;
689
690 /* find extent */
691 ext4_ext_binsearch(inode, path + ppos, block);
692 /* if not an empty leaf */
693 if (path[ppos].p_ext)
694 path[ppos].p_block = ext_pblock(path[ppos].p_ext);
695
696 ext4_ext_show_path(inode, path);
697
698 return path;
699
700 err:
701 ext4_ext_drop_refs(path);
702 if (alloc)
703 kfree(path);
704 return ERR_PTR(-EIO);
705 }
706
707 /*
708 * ext4_ext_insert_index:
709 * insert new index [@logical;@ptr] into the block at @curp;
710 * check where to insert: before @curp or after @curp
711 */
712 static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
713 struct ext4_ext_path *curp,
714 int logical, ext4_fsblk_t ptr)
715 {
716 struct ext4_extent_idx *ix;
717 int len, err;
718
719 err = ext4_ext_get_access(handle, inode, curp);
720 if (err)
721 return err;
722
723 BUG_ON(logical == le32_to_cpu(curp->p_idx->ei_block));
724 len = EXT_MAX_INDEX(curp->p_hdr) - curp->p_idx;
725 if (logical > le32_to_cpu(curp->p_idx->ei_block)) {
726 /* insert after */
727 if (curp->p_idx != EXT_LAST_INDEX(curp->p_hdr)) {
728 len = (len - 1) * sizeof(struct ext4_extent_idx);
729 len = len < 0 ? 0 : len;
730 ext_debug("insert new index %d after: %llu. "
731 "move %d from 0x%p to 0x%p\n",
732 logical, ptr, len,
733 (curp->p_idx + 1), (curp->p_idx + 2));
734 memmove(curp->p_idx + 2, curp->p_idx + 1, len);
735 }
736 ix = curp->p_idx + 1;
737 } else {
738 /* insert before */
739 len = len * sizeof(struct ext4_extent_idx);
740 len = len < 0 ? 0 : len;
741 ext_debug("insert new index %d before: %llu. "
742 "move %d from 0x%p to 0x%p\n",
743 logical, ptr, len,
744 curp->p_idx, (curp->p_idx + 1));
745 memmove(curp->p_idx + 1, curp->p_idx, len);
746 ix = curp->p_idx;
747 }
748
749 ix->ei_block = cpu_to_le32(logical);
750 ext4_idx_store_pblock(ix, ptr);
751 le16_add_cpu(&curp->p_hdr->eh_entries, 1);
752
753 BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries)
754 > le16_to_cpu(curp->p_hdr->eh_max));
755 BUG_ON(ix > EXT_LAST_INDEX(curp->p_hdr));
756
757 err = ext4_ext_dirty(handle, inode, curp);
758 ext4_std_error(inode->i_sb, err);
759
760 return err;
761 }
762
763 /*
764 * ext4_ext_split:
765 * inserts new subtree into the path, using free index entry
766 * at depth @at:
767 * - allocates all needed blocks (new leaf and all intermediate index blocks)
768 * - makes decision where to split
769 * - moves remaining extents and index entries (right to the split point)
770 * into the newly allocated blocks
771 * - initializes subtree
772 */
773 static int ext4_ext_split(handle_t *handle, struct inode *inode,
774 struct ext4_ext_path *path,
775 struct ext4_extent *newext, int at)
776 {
777 struct buffer_head *bh = NULL;
778 int depth = ext_depth(inode);
779 struct ext4_extent_header *neh;
780 struct ext4_extent_idx *fidx;
781 struct ext4_extent *ex;
782 int i = at, k, m, a;
783 ext4_fsblk_t newblock, oldblock;
784 __le32 border;
785 ext4_fsblk_t *ablocks = NULL; /* array of allocated blocks */
786 int err = 0;
787
788 /* make decision: where to split? */
789 /* FIXME: now decision is simplest: at current extent */
790
791 /* if current leaf will be split, then we should use
792 * border from split point */
793 BUG_ON(path[depth].p_ext > EXT_MAX_EXTENT(path[depth].p_hdr));
794 if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) {
795 border = path[depth].p_ext[1].ee_block;
796 ext_debug("leaf will be split."
797 " next leaf starts at %d\n",
798 le32_to_cpu(border));
799 } else {
800 border = newext->ee_block;
801 ext_debug("leaf will be added."
802 " next leaf starts at %d\n",
803 le32_to_cpu(border));
804 }
805
806 /*
807 * If error occurs, then we break processing
808 * and mark filesystem read-only. index won't
809 * be inserted and tree will be in consistent
810 * state. Next mount will repair buffers too.
811 */
812
813 /*
814 * Get array to track all allocated blocks.
815 * We need this to handle errors and free blocks
816 * upon them.
817 */
818 ablocks = kzalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS);
819 if (!ablocks)
820 return -ENOMEM;
821
822 /* allocate all needed blocks */
823 ext_debug("allocate %d blocks for indexes/leaf\n", depth - at);
824 for (a = 0; a < depth - at; a++) {
825 newblock = ext4_ext_new_meta_block(handle, inode, path,
826 newext, &err);
827 if (newblock == 0)
828 goto cleanup;
829 ablocks[a] = newblock;
830 }
831
832 /* initialize new leaf */
833 newblock = ablocks[--a];
834 BUG_ON(newblock == 0);
835 bh = sb_getblk(inode->i_sb, newblock);
836 if (!bh) {
837 err = -EIO;
838 goto cleanup;
839 }
840 lock_buffer(bh);
841
842 err = ext4_journal_get_create_access(handle, bh);
843 if (err)
844 goto cleanup;
845
846 neh = ext_block_hdr(bh);
847 neh->eh_entries = 0;
848 neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode));
849 neh->eh_magic = EXT4_EXT_MAGIC;
850 neh->eh_depth = 0;
851 ex = EXT_FIRST_EXTENT(neh);
852
853 /* move remainder of path[depth] to the new leaf */
854 BUG_ON(path[depth].p_hdr->eh_entries != path[depth].p_hdr->eh_max);
855 /* start copy from next extent */
856 /* TODO: we could do it by single memmove */
857 m = 0;
858 path[depth].p_ext++;
859 while (path[depth].p_ext <=
860 EXT_MAX_EXTENT(path[depth].p_hdr)) {
861 ext_debug("move %d:%llu:%d in new leaf %llu\n",
862 le32_to_cpu(path[depth].p_ext->ee_block),
863 ext_pblock(path[depth].p_ext),
864 ext4_ext_get_actual_len(path[depth].p_ext),
865 newblock);
866 /*memmove(ex++, path[depth].p_ext++,
867 sizeof(struct ext4_extent));
868 neh->eh_entries++;*/
869 path[depth].p_ext++;
870 m++;
871 }
872 if (m) {
873 memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m);
874 le16_add_cpu(&neh->eh_entries, m);
875 }
876
877 set_buffer_uptodate(bh);
878 unlock_buffer(bh);
879
880 err = ext4_handle_dirty_metadata(handle, inode, bh);
881 if (err)
882 goto cleanup;
883 brelse(bh);
884 bh = NULL;
885
886 /* correct old leaf */
887 if (m) {
888 err = ext4_ext_get_access(handle, inode, path + depth);
889 if (err)
890 goto cleanup;
891 le16_add_cpu(&path[depth].p_hdr->eh_entries, -m);
892 err = ext4_ext_dirty(handle, inode, path + depth);
893 if (err)
894 goto cleanup;
895
896 }
897
898 /* create intermediate indexes */
899 k = depth - at - 1;
900 BUG_ON(k < 0);
901 if (k)
902 ext_debug("create %d intermediate indices\n", k);
903 /* insert new index into current index block */
904 /* current depth stored in i var */
905 i = depth - 1;
906 while (k--) {
907 oldblock = newblock;
908 newblock = ablocks[--a];
909 bh = sb_getblk(inode->i_sb, newblock);
910 if (!bh) {
911 err = -EIO;
912 goto cleanup;
913 }
914 lock_buffer(bh);
915
916 err = ext4_journal_get_create_access(handle, bh);
917 if (err)
918 goto cleanup;
919
920 neh = ext_block_hdr(bh);
921 neh->eh_entries = cpu_to_le16(1);
922 neh->eh_magic = EXT4_EXT_MAGIC;
923 neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode));
924 neh->eh_depth = cpu_to_le16(depth - i);
925 fidx = EXT_FIRST_INDEX(neh);
926 fidx->ei_block = border;
927 ext4_idx_store_pblock(fidx, oldblock);
928
929 ext_debug("int.index at %d (block %llu): %u -> %llu\n",
930 i, newblock, le32_to_cpu(border), oldblock);
931 /* copy indexes */
932 m = 0;
933 path[i].p_idx++;
934
935 ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx,
936 EXT_MAX_INDEX(path[i].p_hdr));
937 BUG_ON(EXT_MAX_INDEX(path[i].p_hdr) !=
938 EXT_LAST_INDEX(path[i].p_hdr));
939 while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) {
940 ext_debug("%d: move %d:%llu in new index %llu\n", i,
941 le32_to_cpu(path[i].p_idx->ei_block),
942 idx_pblock(path[i].p_idx),
943 newblock);
944 /*memmove(++fidx, path[i].p_idx++,
945 sizeof(struct ext4_extent_idx));
946 neh->eh_entries++;
947 BUG_ON(neh->eh_entries > neh->eh_max);*/
948 path[i].p_idx++;
949 m++;
950 }
951 if (m) {
952 memmove(++fidx, path[i].p_idx - m,
953 sizeof(struct ext4_extent_idx) * m);
954 le16_add_cpu(&neh->eh_entries, m);
955 }
956 set_buffer_uptodate(bh);
957 unlock_buffer(bh);
958
959 err = ext4_handle_dirty_metadata(handle, inode, bh);
960 if (err)
961 goto cleanup;
962 brelse(bh);
963 bh = NULL;
964
965 /* correct old index */
966 if (m) {
967 err = ext4_ext_get_access(handle, inode, path + i);
968 if (err)
969 goto cleanup;
970 le16_add_cpu(&path[i].p_hdr->eh_entries, -m);
971 err = ext4_ext_dirty(handle, inode, path + i);
972 if (err)
973 goto cleanup;
974 }
975
976 i--;
977 }
978
979 /* insert new index */
980 err = ext4_ext_insert_index(handle, inode, path + at,
981 le32_to_cpu(border), newblock);
982
983 cleanup:
984 if (bh) {
985 if (buffer_locked(bh))
986 unlock_buffer(bh);
987 brelse(bh);
988 }
989
990 if (err) {
991 /* free all allocated blocks in error case */
992 for (i = 0; i < depth; i++) {
993 if (!ablocks[i])
994 continue;
995 ext4_free_blocks(handle, inode, ablocks[i], 1, 1);
996 }
997 }
998 kfree(ablocks);
999
1000 return err;
1001 }
1002
1003 /*
1004 * ext4_ext_grow_indepth:
1005 * implements tree growing procedure:
1006 * - allocates new block
1007 * - moves top-level data (index block or leaf) into the new block
1008 * - initializes new top-level, creating index that points to the
1009 * just created block
1010 */
1011 static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
1012 struct ext4_ext_path *path,
1013 struct ext4_extent *newext)
1014 {
1015 struct ext4_ext_path *curp = path;
1016 struct ext4_extent_header *neh;
1017 struct ext4_extent_idx *fidx;
1018 struct buffer_head *bh;
1019 ext4_fsblk_t newblock;
1020 int err = 0;
1021
1022 newblock = ext4_ext_new_meta_block(handle, inode, path, newext, &err);
1023 if (newblock == 0)
1024 return err;
1025
1026 bh = sb_getblk(inode->i_sb, newblock);
1027 if (!bh) {
1028 err = -EIO;
1029 ext4_std_error(inode->i_sb, err);
1030 return err;
1031 }
1032 lock_buffer(bh);
1033
1034 err = ext4_journal_get_create_access(handle, bh);
1035 if (err) {
1036 unlock_buffer(bh);
1037 goto out;
1038 }
1039
1040 /* move top-level index/leaf into new block */
1041 memmove(bh->b_data, curp->p_hdr, sizeof(EXT4_I(inode)->i_data));
1042
1043 /* set size of new block */
1044 neh = ext_block_hdr(bh);
1045 /* old root could have indexes or leaves
1046 * so calculate e_max right way */
1047 if (ext_depth(inode))
1048 neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode));
1049 else
1050 neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode));
1051 neh->eh_magic = EXT4_EXT_MAGIC;
1052 set_buffer_uptodate(bh);
1053 unlock_buffer(bh);
1054
1055 err = ext4_handle_dirty_metadata(handle, inode, bh);
1056 if (err)
1057 goto out;
1058
1059 /* create index in new top-level index: num,max,pointer */
1060 err = ext4_ext_get_access(handle, inode, curp);
1061 if (err)
1062 goto out;
1063
1064 curp->p_hdr->eh_magic = EXT4_EXT_MAGIC;
1065 curp->p_hdr->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode));
1066 curp->p_hdr->eh_entries = cpu_to_le16(1);
1067 curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr);
1068
1069 if (path[0].p_hdr->eh_depth)
1070 curp->p_idx->ei_block =
1071 EXT_FIRST_INDEX(path[0].p_hdr)->ei_block;
1072 else
1073 curp->p_idx->ei_block =
1074 EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block;
1075 ext4_idx_store_pblock(curp->p_idx, newblock);
1076
1077 neh = ext_inode_hdr(inode);
1078 fidx = EXT_FIRST_INDEX(neh);
1079 ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n",
1080 le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max),
1081 le32_to_cpu(fidx->ei_block), idx_pblock(fidx));
1082
1083 neh->eh_depth = cpu_to_le16(path->p_depth + 1);
1084 err = ext4_ext_dirty(handle, inode, curp);
1085 out:
1086 brelse(bh);
1087
1088 return err;
1089 }
1090
1091 /*
1092 * ext4_ext_create_new_leaf:
1093 * finds empty index and adds new leaf.
1094 * if no free index is found, then it requests in-depth growing.
1095 */
1096 static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode,
1097 struct ext4_ext_path *path,
1098 struct ext4_extent *newext)
1099 {
1100 struct ext4_ext_path *curp;
1101 int depth, i, err = 0;
1102
1103 repeat:
1104 i = depth = ext_depth(inode);
1105
1106 /* walk up to the tree and look for free index entry */
1107 curp = path + depth;
1108 while (i > 0 && !EXT_HAS_FREE_INDEX(curp)) {
1109 i--;
1110 curp--;
1111 }
1112
1113 /* we use already allocated block for index block,
1114 * so subsequent data blocks should be contiguous */
1115 if (EXT_HAS_FREE_INDEX(curp)) {
1116 /* if we found index with free entry, then use that
1117 * entry: create all needed subtree and add new leaf */
1118 err = ext4_ext_split(handle, inode, path, newext, i);
1119 if (err)
1120 goto out;
1121
1122 /* refill path */
1123 ext4_ext_drop_refs(path);
1124 path = ext4_ext_find_extent(inode,
1125 (ext4_lblk_t)le32_to_cpu(newext->ee_block),
1126 path);
1127 if (IS_ERR(path))
1128 err = PTR_ERR(path);
1129 } else {
1130 /* tree is full, time to grow in depth */
1131 err = ext4_ext_grow_indepth(handle, inode, path, newext);
1132 if (err)
1133 goto out;
1134
1135 /* refill path */
1136 ext4_ext_drop_refs(path);
1137 path = ext4_ext_find_extent(inode,
1138 (ext4_lblk_t)le32_to_cpu(newext->ee_block),
1139 path);
1140 if (IS_ERR(path)) {
1141 err = PTR_ERR(path);
1142 goto out;
1143 }
1144
1145 /*
1146 * only first (depth 0 -> 1) produces free space;
1147 * in all other cases we have to split the grown tree
1148 */
1149 depth = ext_depth(inode);
1150 if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) {
1151 /* now we need to split */
1152 goto repeat;
1153 }
1154 }
1155
1156 out:
1157 return err;
1158 }
1159
1160 /*
1161 * search the closest allocated block to the left for *logical
1162 * and returns it at @logical + it's physical address at @phys
1163 * if *logical is the smallest allocated block, the function
1164 * returns 0 at @phys
1165 * return value contains 0 (success) or error code
1166 */
1167 int
1168 ext4_ext_search_left(struct inode *inode, struct ext4_ext_path *path,
1169 ext4_lblk_t *logical, ext4_fsblk_t *phys)
1170 {
1171 struct ext4_extent_idx *ix;
1172 struct ext4_extent *ex;
1173 int depth, ee_len;
1174
1175 BUG_ON(path == NULL);
1176 depth = path->p_depth;
1177 *phys = 0;
1178
1179 if (depth == 0 && path->p_ext == NULL)
1180 return 0;
1181
1182 /* usually extent in the path covers blocks smaller
1183 * then *logical, but it can be that extent is the
1184 * first one in the file */
1185
1186 ex = path[depth].p_ext;
1187 ee_len = ext4_ext_get_actual_len(ex);
1188 if (*logical < le32_to_cpu(ex->ee_block)) {
1189 BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex);
1190 while (--depth >= 0) {
1191 ix = path[depth].p_idx;
1192 BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr));
1193 }
1194 return 0;
1195 }
1196
1197 BUG_ON(*logical < (le32_to_cpu(ex->ee_block) + ee_len));
1198
1199 *logical = le32_to_cpu(ex->ee_block) + ee_len - 1;
1200 *phys = ext_pblock(ex) + ee_len - 1;
1201 return 0;
1202 }
1203
1204 /*
1205 * search the closest allocated block to the right for *logical
1206 * and returns it at @logical + it's physical address at @phys
1207 * if *logical is the smallest allocated block, the function
1208 * returns 0 at @phys
1209 * return value contains 0 (success) or error code
1210 */
1211 int
1212 ext4_ext_search_right(struct inode *inode, struct ext4_ext_path *path,
1213 ext4_lblk_t *logical, ext4_fsblk_t *phys)
1214 {
1215 struct buffer_head *bh = NULL;
1216 struct ext4_extent_header *eh;
1217 struct ext4_extent_idx *ix;
1218 struct ext4_extent *ex;
1219 ext4_fsblk_t block;
1220 int depth; /* Note, NOT eh_depth; depth from top of tree */
1221 int ee_len;
1222
1223 BUG_ON(path == NULL);
1224 depth = path->p_depth;
1225 *phys = 0;
1226
1227 if (depth == 0 && path->p_ext == NULL)
1228 return 0;
1229
1230 /* usually extent in the path covers blocks smaller
1231 * then *logical, but it can be that extent is the
1232 * first one in the file */
1233
1234 ex = path[depth].p_ext;
1235 ee_len = ext4_ext_get_actual_len(ex);
1236 if (*logical < le32_to_cpu(ex->ee_block)) {
1237 BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex);
1238 while (--depth >= 0) {
1239 ix = path[depth].p_idx;
1240 BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr));
1241 }
1242 *logical = le32_to_cpu(ex->ee_block);
1243 *phys = ext_pblock(ex);
1244 return 0;
1245 }
1246
1247 BUG_ON(*logical < (le32_to_cpu(ex->ee_block) + ee_len));
1248
1249 if (ex != EXT_LAST_EXTENT(path[depth].p_hdr)) {
1250 /* next allocated block in this leaf */
1251 ex++;
1252 *logical = le32_to_cpu(ex->ee_block);
1253 *phys = ext_pblock(ex);
1254 return 0;
1255 }
1256
1257 /* go up and search for index to the right */
1258 while (--depth >= 0) {
1259 ix = path[depth].p_idx;
1260 if (ix != EXT_LAST_INDEX(path[depth].p_hdr))
1261 goto got_index;
1262 }
1263
1264 /* we've gone up to the root and found no index to the right */
1265 return 0;
1266
1267 got_index:
1268 /* we've found index to the right, let's
1269 * follow it and find the closest allocated
1270 * block to the right */
1271 ix++;
1272 block = idx_pblock(ix);
1273 while (++depth < path->p_depth) {
1274 bh = sb_bread(inode->i_sb, block);
1275 if (bh == NULL)
1276 return -EIO;
1277 eh = ext_block_hdr(bh);
1278 /* subtract from p_depth to get proper eh_depth */
1279 if (ext4_ext_check(inode, eh, path->p_depth - depth)) {
1280 put_bh(bh);
1281 return -EIO;
1282 }
1283 ix = EXT_FIRST_INDEX(eh);
1284 block = idx_pblock(ix);
1285 put_bh(bh);
1286 }
1287
1288 bh = sb_bread(inode->i_sb, block);
1289 if (bh == NULL)
1290 return -EIO;
1291 eh = ext_block_hdr(bh);
1292 if (ext4_ext_check(inode, eh, path->p_depth - depth)) {
1293 put_bh(bh);
1294 return -EIO;
1295 }
1296 ex = EXT_FIRST_EXTENT(eh);
1297 *logical = le32_to_cpu(ex->ee_block);
1298 *phys = ext_pblock(ex);
1299 put_bh(bh);
1300 return 0;
1301 }
1302
1303 /*
1304 * ext4_ext_next_allocated_block:
1305 * returns allocated block in subsequent extent or EXT_MAX_BLOCK.
1306 * NOTE: it considers block number from index entry as
1307 * allocated block. Thus, index entries have to be consistent
1308 * with leaves.
1309 */
1310 static ext4_lblk_t
1311 ext4_ext_next_allocated_block(struct ext4_ext_path *path)
1312 {
1313 int depth;
1314
1315 BUG_ON(path == NULL);
1316 depth = path->p_depth;
1317
1318 if (depth == 0 && path->p_ext == NULL)
1319 return EXT_MAX_BLOCK;
1320
1321 while (depth >= 0) {
1322 if (depth == path->p_depth) {
1323 /* leaf */
1324 if (path[depth].p_ext !=
1325 EXT_LAST_EXTENT(path[depth].p_hdr))
1326 return le32_to_cpu(path[depth].p_ext[1].ee_block);
1327 } else {
1328 /* index */
1329 if (path[depth].p_idx !=
1330 EXT_LAST_INDEX(path[depth].p_hdr))
1331 return le32_to_cpu(path[depth].p_idx[1].ei_block);
1332 }
1333 depth--;
1334 }
1335
1336 return EXT_MAX_BLOCK;
1337 }
1338
1339 /*
1340 * ext4_ext_next_leaf_block:
1341 * returns first allocated block from next leaf or EXT_MAX_BLOCK
1342 */
1343 static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode,
1344 struct ext4_ext_path *path)
1345 {
1346 int depth;
1347
1348 BUG_ON(path == NULL);
1349 depth = path->p_depth;
1350
1351 /* zero-tree has no leaf blocks at all */
1352 if (depth == 0)
1353 return EXT_MAX_BLOCK;
1354
1355 /* go to index block */
1356 depth--;
1357
1358 while (depth >= 0) {
1359 if (path[depth].p_idx !=
1360 EXT_LAST_INDEX(path[depth].p_hdr))
1361 return (ext4_lblk_t)
1362 le32_to_cpu(path[depth].p_idx[1].ei_block);
1363 depth--;
1364 }
1365
1366 return EXT_MAX_BLOCK;
1367 }
1368
1369 /*
1370 * ext4_ext_correct_indexes:
1371 * if leaf gets modified and modified extent is first in the leaf,
1372 * then we have to correct all indexes above.
1373 * TODO: do we need to correct tree in all cases?
1374 */
1375 static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode,
1376 struct ext4_ext_path *path)
1377 {
1378 struct ext4_extent_header *eh;
1379 int depth = ext_depth(inode);
1380 struct ext4_extent *ex;
1381 __le32 border;
1382 int k, err = 0;
1383
1384 eh = path[depth].p_hdr;
1385 ex = path[depth].p_ext;
1386 BUG_ON(ex == NULL);
1387 BUG_ON(eh == NULL);
1388
1389 if (depth == 0) {
1390 /* there is no tree at all */
1391 return 0;
1392 }
1393
1394 if (ex != EXT_FIRST_EXTENT(eh)) {
1395 /* we correct tree if first leaf got modified only */
1396 return 0;
1397 }
1398
1399 /*
1400 * TODO: we need correction if border is smaller than current one
1401 */
1402 k = depth - 1;
1403 border = path[depth].p_ext->ee_block;
1404 err = ext4_ext_get_access(handle, inode, path + k);
1405 if (err)
1406 return err;
1407 path[k].p_idx->ei_block = border;
1408 err = ext4_ext_dirty(handle, inode, path + k);
1409 if (err)
1410 return err;
1411
1412 while (k--) {
1413 /* change all left-side indexes */
1414 if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr))
1415 break;
1416 err = ext4_ext_get_access(handle, inode, path + k);
1417 if (err)
1418 break;
1419 path[k].p_idx->ei_block = border;
1420 err = ext4_ext_dirty(handle, inode, path + k);
1421 if (err)
1422 break;
1423 }
1424
1425 return err;
1426 }
1427
1428 static int
1429 ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
1430 struct ext4_extent *ex2)
1431 {
1432 unsigned short ext1_ee_len, ext2_ee_len, max_len;
1433
1434 /*
1435 * Make sure that either both extents are uninitialized, or
1436 * both are _not_.
1437 */
1438 if (ext4_ext_is_uninitialized(ex1) ^ ext4_ext_is_uninitialized(ex2))
1439 return 0;
1440
1441 if (ext4_ext_is_uninitialized(ex1))
1442 max_len = EXT_UNINIT_MAX_LEN;
1443 else
1444 max_len = EXT_INIT_MAX_LEN;
1445
1446 ext1_ee_len = ext4_ext_get_actual_len(ex1);
1447 ext2_ee_len = ext4_ext_get_actual_len(ex2);
1448
1449 if (le32_to_cpu(ex1->ee_block) + ext1_ee_len !=
1450 le32_to_cpu(ex2->ee_block))
1451 return 0;
1452
1453 /*
1454 * To allow future support for preallocated extents to be added
1455 * as an RO_COMPAT feature, refuse to merge to extents if
1456 * this can result in the top bit of ee_len being set.
1457 */
1458 if (ext1_ee_len + ext2_ee_len > max_len)
1459 return 0;
1460 #ifdef AGGRESSIVE_TEST
1461 if (ext1_ee_len >= 4)
1462 return 0;
1463 #endif
1464
1465 if (ext_pblock(ex1) + ext1_ee_len == ext_pblock(ex2))
1466 return 1;
1467 return 0;
1468 }
1469
1470 /*
1471 * This function tries to merge the "ex" extent to the next extent in the tree.
1472 * It always tries to merge towards right. If you want to merge towards
1473 * left, pass "ex - 1" as argument instead of "ex".
1474 * Returns 0 if the extents (ex and ex+1) were _not_ merged and returns
1475 * 1 if they got merged.
1476 */
1477 int ext4_ext_try_to_merge(struct inode *inode,
1478 struct ext4_ext_path *path,
1479 struct ext4_extent *ex)
1480 {
1481 struct ext4_extent_header *eh;
1482 unsigned int depth, len;
1483 int merge_done = 0;
1484 int uninitialized = 0;
1485
1486 depth = ext_depth(inode);
1487 BUG_ON(path[depth].p_hdr == NULL);
1488 eh = path[depth].p_hdr;
1489
1490 while (ex < EXT_LAST_EXTENT(eh)) {
1491 if (!ext4_can_extents_be_merged(inode, ex, ex + 1))
1492 break;
1493 /* merge with next extent! */
1494 if (ext4_ext_is_uninitialized(ex))
1495 uninitialized = 1;
1496 ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
1497 + ext4_ext_get_actual_len(ex + 1));
1498 if (uninitialized)
1499 ext4_ext_mark_uninitialized(ex);
1500
1501 if (ex + 1 < EXT_LAST_EXTENT(eh)) {
1502 len = (EXT_LAST_EXTENT(eh) - ex - 1)
1503 * sizeof(struct ext4_extent);
1504 memmove(ex + 1, ex + 2, len);
1505 }
1506 le16_add_cpu(&eh->eh_entries, -1);
1507 merge_done = 1;
1508 WARN_ON(eh->eh_entries == 0);
1509 if (!eh->eh_entries)
1510 ext4_error(inode->i_sb, "ext4_ext_try_to_merge",
1511 "inode#%lu, eh->eh_entries = 0!", inode->i_ino);
1512 }
1513
1514 return merge_done;
1515 }
1516
1517 /*
1518 * check if a portion of the "newext" extent overlaps with an
1519 * existing extent.
1520 *
1521 * If there is an overlap discovered, it updates the length of the newext
1522 * such that there will be no overlap, and then returns 1.
1523 * If there is no overlap found, it returns 0.
1524 */
1525 unsigned int ext4_ext_check_overlap(struct inode *inode,
1526 struct ext4_extent *newext,
1527 struct ext4_ext_path *path)
1528 {
1529 ext4_lblk_t b1, b2;
1530 unsigned int depth, len1;
1531 unsigned int ret = 0;
1532
1533 b1 = le32_to_cpu(newext->ee_block);
1534 len1 = ext4_ext_get_actual_len(newext);
1535 depth = ext_depth(inode);
1536 if (!path[depth].p_ext)
1537 goto out;
1538 b2 = le32_to_cpu(path[depth].p_ext->ee_block);
1539
1540 /*
1541 * get the next allocated block if the extent in the path
1542 * is before the requested block(s)
1543 */
1544 if (b2 < b1) {
1545 b2 = ext4_ext_next_allocated_block(path);
1546 if (b2 == EXT_MAX_BLOCK)
1547 goto out;
1548 }
1549
1550 /* check for wrap through zero on extent logical start block*/
1551 if (b1 + len1 < b1) {
1552 len1 = EXT_MAX_BLOCK - b1;
1553 newext->ee_len = cpu_to_le16(len1);
1554 ret = 1;
1555 }
1556
1557 /* check for overlap */
1558 if (b1 + len1 > b2) {
1559 newext->ee_len = cpu_to_le16(b2 - b1);
1560 ret = 1;
1561 }
1562 out:
1563 return ret;
1564 }
1565
1566 /*
1567 * ext4_ext_insert_extent:
1568 * tries to merge requsted extent into the existing extent or
1569 * inserts requested extent as new one into the tree,
1570 * creating new leaf in the no-space case.
1571 */
1572 int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
1573 struct ext4_ext_path *path,
1574 struct ext4_extent *newext)
1575 {
1576 struct ext4_extent_header *eh;
1577 struct ext4_extent *ex, *fex;
1578 struct ext4_extent *nearex; /* nearest extent */
1579 struct ext4_ext_path *npath = NULL;
1580 int depth, len, err;
1581 ext4_lblk_t next;
1582 unsigned uninitialized = 0;
1583
1584 BUG_ON(ext4_ext_get_actual_len(newext) == 0);
1585 depth = ext_depth(inode);
1586 ex = path[depth].p_ext;
1587 BUG_ON(path[depth].p_hdr == NULL);
1588
1589 /* try to insert block into found extent and return */
1590 if (ex && ext4_can_extents_be_merged(inode, ex, newext)) {
1591 ext_debug("append %d block to %d:%d (from %llu)\n",
1592 ext4_ext_get_actual_len(newext),
1593 le32_to_cpu(ex->ee_block),
1594 ext4_ext_get_actual_len(ex), ext_pblock(ex));
1595 err = ext4_ext_get_access(handle, inode, path + depth);
1596 if (err)
1597 return err;
1598
1599 /*
1600 * ext4_can_extents_be_merged should have checked that either
1601 * both extents are uninitialized, or both aren't. Thus we
1602 * need to check only one of them here.
1603 */
1604 if (ext4_ext_is_uninitialized(ex))
1605 uninitialized = 1;
1606 ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
1607 + ext4_ext_get_actual_len(newext));
1608 if (uninitialized)
1609 ext4_ext_mark_uninitialized(ex);
1610 eh = path[depth].p_hdr;
1611 nearex = ex;
1612 goto merge;
1613 }
1614
1615 repeat:
1616 depth = ext_depth(inode);
1617 eh = path[depth].p_hdr;
1618 if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max))
1619 goto has_space;
1620
1621 /* probably next leaf has space for us? */
1622 fex = EXT_LAST_EXTENT(eh);
1623 next = ext4_ext_next_leaf_block(inode, path);
1624 if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block)
1625 && next != EXT_MAX_BLOCK) {
1626 ext_debug("next leaf block - %d\n", next);
1627 BUG_ON(npath != NULL);
1628 npath = ext4_ext_find_extent(inode, next, NULL);
1629 if (IS_ERR(npath))
1630 return PTR_ERR(npath);
1631 BUG_ON(npath->p_depth != path->p_depth);
1632 eh = npath[depth].p_hdr;
1633 if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) {
1634 ext_debug("next leaf isnt full(%d)\n",
1635 le16_to_cpu(eh->eh_entries));
1636 path = npath;
1637 goto repeat;
1638 }
1639 ext_debug("next leaf has no free space(%d,%d)\n",
1640 le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max));
1641 }
1642
1643 /*
1644 * There is no free space in the found leaf.
1645 * We're gonna add a new leaf in the tree.
1646 */
1647 err = ext4_ext_create_new_leaf(handle, inode, path, newext);
1648 if (err)
1649 goto cleanup;
1650 depth = ext_depth(inode);
1651 eh = path[depth].p_hdr;
1652
1653 has_space:
1654 nearex = path[depth].p_ext;
1655
1656 err = ext4_ext_get_access(handle, inode, path + depth);
1657 if (err)
1658 goto cleanup;
1659
1660 if (!nearex) {
1661 /* there is no extent in this leaf, create first one */
1662 ext_debug("first extent in the leaf: %d:%llu:%d\n",
1663 le32_to_cpu(newext->ee_block),
1664 ext_pblock(newext),
1665 ext4_ext_get_actual_len(newext));
1666 path[depth].p_ext = EXT_FIRST_EXTENT(eh);
1667 } else if (le32_to_cpu(newext->ee_block)
1668 > le32_to_cpu(nearex->ee_block)) {
1669 /* BUG_ON(newext->ee_block == nearex->ee_block); */
1670 if (nearex != EXT_LAST_EXTENT(eh)) {
1671 len = EXT_MAX_EXTENT(eh) - nearex;
1672 len = (len - 1) * sizeof(struct ext4_extent);
1673 len = len < 0 ? 0 : len;
1674 ext_debug("insert %d:%llu:%d after: nearest 0x%p, "
1675 "move %d from 0x%p to 0x%p\n",
1676 le32_to_cpu(newext->ee_block),
1677 ext_pblock(newext),
1678 ext4_ext_get_actual_len(newext),
1679 nearex, len, nearex + 1, nearex + 2);
1680 memmove(nearex + 2, nearex + 1, len);
1681 }
1682 path[depth].p_ext = nearex + 1;
1683 } else {
1684 BUG_ON(newext->ee_block == nearex->ee_block);
1685 len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext4_extent);
1686 len = len < 0 ? 0 : len;
1687 ext_debug("insert %d:%llu:%d before: nearest 0x%p, "
1688 "move %d from 0x%p to 0x%p\n",
1689 le32_to_cpu(newext->ee_block),
1690 ext_pblock(newext),
1691 ext4_ext_get_actual_len(newext),
1692 nearex, len, nearex + 1, nearex + 2);
1693 memmove(nearex + 1, nearex, len);
1694 path[depth].p_ext = nearex;
1695 }
1696
1697 le16_add_cpu(&eh->eh_entries, 1);
1698 nearex = path[depth].p_ext;
1699 nearex->ee_block = newext->ee_block;
1700 ext4_ext_store_pblock(nearex, ext_pblock(newext));
1701 nearex->ee_len = newext->ee_len;
1702
1703 merge:
1704 /* try to merge extents to the right */
1705 ext4_ext_try_to_merge(inode, path, nearex);
1706
1707 /* try to merge extents to the left */
1708
1709 /* time to correct all indexes above */
1710 err = ext4_ext_correct_indexes(handle, inode, path);
1711 if (err)
1712 goto cleanup;
1713
1714 err = ext4_ext_dirty(handle, inode, path + depth);
1715
1716 cleanup:
1717 if (npath) {
1718 ext4_ext_drop_refs(npath);
1719 kfree(npath);
1720 }
1721 ext4_ext_invalidate_cache(inode);
1722 return err;
1723 }
1724
1725 int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block,
1726 ext4_lblk_t num, ext_prepare_callback func,
1727 void *cbdata)
1728 {
1729 struct ext4_ext_path *path = NULL;
1730 struct ext4_ext_cache cbex;
1731 struct ext4_extent *ex;
1732 ext4_lblk_t next, start = 0, end = 0;
1733 ext4_lblk_t last = block + num;
1734 int depth, exists, err = 0;
1735
1736 BUG_ON(func == NULL);
1737 BUG_ON(inode == NULL);
1738
1739 while (block < last && block != EXT_MAX_BLOCK) {
1740 num = last - block;
1741 /* find extent for this block */
1742 path = ext4_ext_find_extent(inode, block, path);
1743 if (IS_ERR(path)) {
1744 err = PTR_ERR(path);
1745 path = NULL;
1746 break;
1747 }
1748
1749 depth = ext_depth(inode);
1750 BUG_ON(path[depth].p_hdr == NULL);
1751 ex = path[depth].p_ext;
1752 next = ext4_ext_next_allocated_block(path);
1753
1754 exists = 0;
1755 if (!ex) {
1756 /* there is no extent yet, so try to allocate
1757 * all requested space */
1758 start = block;
1759 end = block + num;
1760 } else if (le32_to_cpu(ex->ee_block) > block) {
1761 /* need to allocate space before found extent */
1762 start = block;
1763 end = le32_to_cpu(ex->ee_block);
1764 if (block + num < end)
1765 end = block + num;
1766 } else if (block >= le32_to_cpu(ex->ee_block)
1767 + ext4_ext_get_actual_len(ex)) {
1768 /* need to allocate space after found extent */
1769 start = block;
1770 end = block + num;
1771 if (end >= next)
1772 end = next;
1773 } else if (block >= le32_to_cpu(ex->ee_block)) {
1774 /*
1775 * some part of requested space is covered
1776 * by found extent
1777 */
1778 start = block;
1779 end = le32_to_cpu(ex->ee_block)
1780 + ext4_ext_get_actual_len(ex);
1781 if (block + num < end)
1782 end = block + num;
1783 exists = 1;
1784 } else {
1785 BUG();
1786 }
1787 BUG_ON(end <= start);
1788
1789 if (!exists) {
1790 cbex.ec_block = start;
1791 cbex.ec_len = end - start;
1792 cbex.ec_start = 0;
1793 cbex.ec_type = EXT4_EXT_CACHE_GAP;
1794 } else {
1795 cbex.ec_block = le32_to_cpu(ex->ee_block);
1796 cbex.ec_len = ext4_ext_get_actual_len(ex);
1797 cbex.ec_start = ext_pblock(ex);
1798 cbex.ec_type = EXT4_EXT_CACHE_EXTENT;
1799 }
1800
1801 BUG_ON(cbex.ec_len == 0);
1802 err = func(inode, path, &cbex, ex, cbdata);
1803 ext4_ext_drop_refs(path);
1804
1805 if (err < 0)
1806 break;
1807
1808 if (err == EXT_REPEAT)
1809 continue;
1810 else if (err == EXT_BREAK) {
1811 err = 0;
1812 break;
1813 }
1814
1815 if (ext_depth(inode) != depth) {
1816 /* depth was changed. we have to realloc path */
1817 kfree(path);
1818 path = NULL;
1819 }
1820
1821 block = cbex.ec_block + cbex.ec_len;
1822 }
1823
1824 if (path) {
1825 ext4_ext_drop_refs(path);
1826 kfree(path);
1827 }
1828
1829 return err;
1830 }
1831
1832 static void
1833 ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block,
1834 __u32 len, ext4_fsblk_t start, int type)
1835 {
1836 struct ext4_ext_cache *cex;
1837 BUG_ON(len == 0);
1838 cex = &EXT4_I(inode)->i_cached_extent;
1839 cex->ec_type = type;
1840 cex->ec_block = block;
1841 cex->ec_len = len;
1842 cex->ec_start = start;
1843 }
1844
1845 /*
1846 * ext4_ext_put_gap_in_cache:
1847 * calculate boundaries of the gap that the requested block fits into
1848 * and cache this gap
1849 */
1850 static void
1851 ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
1852 ext4_lblk_t block)
1853 {
1854 int depth = ext_depth(inode);
1855 unsigned long len;
1856 ext4_lblk_t lblock;
1857 struct ext4_extent *ex;
1858
1859 ex = path[depth].p_ext;
1860 if (ex == NULL) {
1861 /* there is no extent yet, so gap is [0;-] */
1862 lblock = 0;
1863 len = EXT_MAX_BLOCK;
1864 ext_debug("cache gap(whole file):");
1865 } else if (block < le32_to_cpu(ex->ee_block)) {
1866 lblock = block;
1867 len = le32_to_cpu(ex->ee_block) - block;
1868 ext_debug("cache gap(before): %u [%u:%u]",
1869 block,
1870 le32_to_cpu(ex->ee_block),
1871 ext4_ext_get_actual_len(ex));
1872 } else if (block >= le32_to_cpu(ex->ee_block)
1873 + ext4_ext_get_actual_len(ex)) {
1874 ext4_lblk_t next;
1875 lblock = le32_to_cpu(ex->ee_block)
1876 + ext4_ext_get_actual_len(ex);
1877
1878 next = ext4_ext_next_allocated_block(path);
1879 ext_debug("cache gap(after): [%u:%u] %u",
1880 le32_to_cpu(ex->ee_block),
1881 ext4_ext_get_actual_len(ex),
1882 block);
1883 BUG_ON(next == lblock);
1884 len = next - lblock;
1885 } else {
1886 lblock = len = 0;
1887 BUG();
1888 }
1889
1890 ext_debug(" -> %u:%lu\n", lblock, len);
1891 ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP);
1892 }
1893
1894 static int
1895 ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
1896 struct ext4_extent *ex)
1897 {
1898 struct ext4_ext_cache *cex;
1899
1900 cex = &EXT4_I(inode)->i_cached_extent;
1901
1902 /* has cache valid data? */
1903 if (cex->ec_type == EXT4_EXT_CACHE_NO)
1904 return EXT4_EXT_CACHE_NO;
1905
1906 BUG_ON(cex->ec_type != EXT4_EXT_CACHE_GAP &&
1907 cex->ec_type != EXT4_EXT_CACHE_EXTENT);
1908 if (block >= cex->ec_block && block < cex->ec_block + cex->ec_len) {
1909 ex->ee_block = cpu_to_le32(cex->ec_block);
1910 ext4_ext_store_pblock(ex, cex->ec_start);
1911 ex->ee_len = cpu_to_le16(cex->ec_len);
1912 ext_debug("%u cached by %u:%u:%llu\n",
1913 block,
1914 cex->ec_block, cex->ec_len, cex->ec_start);
1915 return cex->ec_type;
1916 }
1917
1918 /* not in cache */
1919 return EXT4_EXT_CACHE_NO;
1920 }
1921
1922 /*
1923 * ext4_ext_rm_idx:
1924 * removes index from the index block.
1925 * It's used in truncate case only, thus all requests are for
1926 * last index in the block only.
1927 */
1928 static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
1929 struct ext4_ext_path *path)
1930 {
1931 struct buffer_head *bh;
1932 int err;
1933 ext4_fsblk_t leaf;
1934
1935 /* free index block */
1936 path--;
1937 leaf = idx_pblock(path->p_idx);
1938 BUG_ON(path->p_hdr->eh_entries == 0);
1939 err = ext4_ext_get_access(handle, inode, path);
1940 if (err)
1941 return err;
1942 le16_add_cpu(&path->p_hdr->eh_entries, -1);
1943 err = ext4_ext_dirty(handle, inode, path);
1944 if (err)
1945 return err;
1946 ext_debug("index is empty, remove it, free block %llu\n", leaf);
1947 bh = sb_find_get_block(inode->i_sb, leaf);
1948 ext4_forget(handle, 1, inode, bh, leaf);
1949 ext4_free_blocks(handle, inode, leaf, 1, 1);
1950 return err;
1951 }
1952
1953 /*
1954 * ext4_ext_calc_credits_for_single_extent:
1955 * This routine returns max. credits that needed to insert an extent
1956 * to the extent tree.
1957 * When pass the actual path, the caller should calculate credits
1958 * under i_data_sem.
1959 */
1960 int ext4_ext_calc_credits_for_single_extent(struct inode *inode, int nrblocks,
1961 struct ext4_ext_path *path)
1962 {
1963 if (path) {
1964 int depth = ext_depth(inode);
1965 int ret = 0;
1966
1967 /* probably there is space in leaf? */
1968 if (le16_to_cpu(path[depth].p_hdr->eh_entries)
1969 < le16_to_cpu(path[depth].p_hdr->eh_max)) {
1970
1971 /*
1972 * There are some space in the leaf tree, no
1973 * need to account for leaf block credit
1974 *
1975 * bitmaps and block group descriptor blocks
1976 * and other metadat blocks still need to be
1977 * accounted.
1978 */
1979 /* 1 bitmap, 1 block group descriptor */
1980 ret = 2 + EXT4_META_TRANS_BLOCKS(inode->i_sb);
1981 }
1982 }
1983
1984 return ext4_chunk_trans_blocks(inode, nrblocks);
1985 }
1986
1987 /*
1988 * How many index/leaf blocks need to change/allocate to modify nrblocks?
1989 *
1990 * if nrblocks are fit in a single extent (chunk flag is 1), then
1991 * in the worse case, each tree level index/leaf need to be changed
1992 * if the tree split due to insert a new extent, then the old tree
1993 * index/leaf need to be updated too
1994 *
1995 * If the nrblocks are discontiguous, they could cause
1996 * the whole tree split more than once, but this is really rare.
1997 */
1998 int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, int chunk)
1999 {
2000 int index;
2001 int depth = ext_depth(inode);
2002
2003 if (chunk)
2004 index = depth * 2;
2005 else
2006 index = depth * 3;
2007
2008 return index;
2009 }
2010
2011 static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
2012 struct ext4_extent *ex,
2013 ext4_lblk_t from, ext4_lblk_t to)
2014 {
2015 struct buffer_head *bh;
2016 unsigned short ee_len = ext4_ext_get_actual_len(ex);
2017 int i, metadata = 0;
2018
2019 if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
2020 metadata = 1;
2021 #ifdef EXTENTS_STATS
2022 {
2023 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
2024 spin_lock(&sbi->s_ext_stats_lock);
2025 sbi->s_ext_blocks += ee_len;
2026 sbi->s_ext_extents++;
2027 if (ee_len < sbi->s_ext_min)
2028 sbi->s_ext_min = ee_len;
2029 if (ee_len > sbi->s_ext_max)
2030 sbi->s_ext_max = ee_len;
2031 if (ext_depth(inode) > sbi->s_depth_max)
2032 sbi->s_depth_max = ext_depth(inode);
2033 spin_unlock(&sbi->s_ext_stats_lock);
2034 }
2035 #endif
2036 if (from >= le32_to_cpu(ex->ee_block)
2037 && to == le32_to_cpu(ex->ee_block) + ee_len - 1) {
2038 /* tail removal */
2039 ext4_lblk_t num;
2040 ext4_fsblk_t start;
2041
2042 num = le32_to_cpu(ex->ee_block) + ee_len - from;
2043 start = ext_pblock(ex) + ee_len - num;
2044 ext_debug("free last %u blocks starting %llu\n", num, start);
2045 for (i = 0; i < num; i++) {
2046 bh = sb_find_get_block(inode->i_sb, start + i);
2047 ext4_forget(handle, 0, inode, bh, start + i);
2048 }
2049 ext4_free_blocks(handle, inode, start, num, metadata);
2050 } else if (from == le32_to_cpu(ex->ee_block)
2051 && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) {
2052 printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n",
2053 from, to, le32_to_cpu(ex->ee_block), ee_len);
2054 } else {
2055 printk(KERN_INFO "strange request: removal(2) "
2056 "%u-%u from %u:%u\n",
2057 from, to, le32_to_cpu(ex->ee_block), ee_len);
2058 }
2059 return 0;
2060 }
2061
2062 static int
2063 ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
2064 struct ext4_ext_path *path, ext4_lblk_t start)
2065 {
2066 int err = 0, correct_index = 0;
2067 int depth = ext_depth(inode), credits;
2068 struct ext4_extent_header *eh;
2069 ext4_lblk_t a, b, block;
2070 unsigned num;
2071 ext4_lblk_t ex_ee_block;
2072 unsigned short ex_ee_len;
2073 unsigned uninitialized = 0;
2074 struct ext4_extent *ex;
2075
2076 /* the header must be checked already in ext4_ext_remove_space() */
2077 ext_debug("truncate since %u in leaf\n", start);
2078 if (!path[depth].p_hdr)
2079 path[depth].p_hdr = ext_block_hdr(path[depth].p_bh);
2080 eh = path[depth].p_hdr;
2081 BUG_ON(eh == NULL);
2082
2083 /* find where to start removing */
2084 ex = EXT_LAST_EXTENT(eh);
2085
2086 ex_ee_block = le32_to_cpu(ex->ee_block);
2087 if (ext4_ext_is_uninitialized(ex))
2088 uninitialized = 1;
2089 ex_ee_len = ext4_ext_get_actual_len(ex);
2090
2091 while (ex >= EXT_FIRST_EXTENT(eh) &&
2092 ex_ee_block + ex_ee_len > start) {
2093 ext_debug("remove ext %lu:%u\n", ex_ee_block, ex_ee_len);
2094 path[depth].p_ext = ex;
2095
2096 a = ex_ee_block > start ? ex_ee_block : start;
2097 b = ex_ee_block + ex_ee_len - 1 < EXT_MAX_BLOCK ?
2098 ex_ee_block + ex_ee_len - 1 : EXT_MAX_BLOCK;
2099
2100 ext_debug(" border %u:%u\n", a, b);
2101
2102 if (a != ex_ee_block && b != ex_ee_block + ex_ee_len - 1) {
2103 block = 0;
2104 num = 0;
2105 BUG();
2106 } else if (a != ex_ee_block) {
2107 /* remove tail of the extent */
2108 block = ex_ee_block;
2109 num = a - block;
2110 } else if (b != ex_ee_block + ex_ee_len - 1) {
2111 /* remove head of the extent */
2112 block = a;
2113 num = b - a;
2114 /* there is no "make a hole" API yet */
2115 BUG();
2116 } else {
2117 /* remove whole extent: excellent! */
2118 block = ex_ee_block;
2119 num = 0;
2120 BUG_ON(a != ex_ee_block);
2121 BUG_ON(b != ex_ee_block + ex_ee_len - 1);
2122 }
2123
2124 /*
2125 * 3 for leaf, sb, and inode plus 2 (bmap and group
2126 * descriptor) for each block group; assume two block
2127 * groups plus ex_ee_len/blocks_per_block_group for
2128 * the worst case
2129 */
2130 credits = 7 + 2*(ex_ee_len/EXT4_BLOCKS_PER_GROUP(inode->i_sb));
2131 if (ex == EXT_FIRST_EXTENT(eh)) {
2132 correct_index = 1;
2133 credits += (ext_depth(inode)) + 1;
2134 }
2135 credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb);
2136
2137 err = ext4_ext_journal_restart(handle, credits);
2138 if (err)
2139 goto out;
2140
2141 err = ext4_ext_get_access(handle, inode, path + depth);
2142 if (err)
2143 goto out;
2144
2145 err = ext4_remove_blocks(handle, inode, ex, a, b);
2146 if (err)
2147 goto out;
2148
2149 if (num == 0) {
2150 /* this extent is removed; mark slot entirely unused */
2151 ext4_ext_store_pblock(ex, 0);
2152 le16_add_cpu(&eh->eh_entries, -1);
2153 }
2154
2155 ex->ee_block = cpu_to_le32(block);
2156 ex->ee_len = cpu_to_le16(num);
2157 /*
2158 * Do not mark uninitialized if all the blocks in the
2159 * extent have been removed.
2160 */
2161 if (uninitialized && num)
2162 ext4_ext_mark_uninitialized(ex);
2163
2164 err = ext4_ext_dirty(handle, inode, path + depth);
2165 if (err)
2166 goto out;
2167
2168 ext_debug("new extent: %u:%u:%llu\n", block, num,
2169 ext_pblock(ex));
2170 ex--;
2171 ex_ee_block = le32_to_cpu(ex->ee_block);
2172 ex_ee_len = ext4_ext_get_actual_len(ex);
2173 }
2174
2175 if (correct_index && eh->eh_entries)
2176 err = ext4_ext_correct_indexes(handle, inode, path);
2177
2178 /* if this leaf is free, then we should
2179 * remove it from index block above */
2180 if (err == 0 && eh->eh_entries == 0 && path[depth].p_bh != NULL)
2181 err = ext4_ext_rm_idx(handle, inode, path + depth);
2182
2183 out:
2184 return err;
2185 }
2186
2187 /*
2188 * ext4_ext_more_to_rm:
2189 * returns 1 if current index has to be freed (even partial)
2190 */
2191 static int
2192 ext4_ext_more_to_rm(struct ext4_ext_path *path)
2193 {
2194 BUG_ON(path->p_idx == NULL);
2195
2196 if (path->p_idx < EXT_FIRST_INDEX(path->p_hdr))
2197 return 0;
2198
2199 /*
2200 * if truncate on deeper level happened, it wasn't partial,
2201 * so we have to consider current index for truncation
2202 */
2203 if (le16_to_cpu(path->p_hdr->eh_entries) == path->p_block)
2204 return 0;
2205 return 1;
2206 }
2207
2208 static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start)
2209 {
2210 struct super_block *sb = inode->i_sb;
2211 int depth = ext_depth(inode);
2212 struct ext4_ext_path *path;
2213 handle_t *handle;
2214 int i = 0, err = 0;
2215
2216 ext_debug("truncate since %u\n", start);
2217
2218 /* probably first extent we're gonna free will be last in block */
2219 handle = ext4_journal_start(inode, depth + 1);
2220 if (IS_ERR(handle))
2221 return PTR_ERR(handle);
2222
2223 ext4_ext_invalidate_cache(inode);
2224
2225 /*
2226 * We start scanning from right side, freeing all the blocks
2227 * after i_size and walking into the tree depth-wise.
2228 */
2229 path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_NOFS);
2230 if (path == NULL) {
2231 ext4_journal_stop(handle);
2232 return -ENOMEM;
2233 }
2234 path[0].p_hdr = ext_inode_hdr(inode);
2235 if (ext4_ext_check(inode, path[0].p_hdr, depth)) {
2236 err = -EIO;
2237 goto out;
2238 }
2239 path[0].p_depth = depth;
2240
2241 while (i >= 0 && err == 0) {
2242 if (i == depth) {
2243 /* this is leaf block */
2244 err = ext4_ext_rm_leaf(handle, inode, path, start);
2245 /* root level has p_bh == NULL, brelse() eats this */
2246 brelse(path[i].p_bh);
2247 path[i].p_bh = NULL;
2248 i--;
2249 continue;
2250 }
2251
2252 /* this is index block */
2253 if (!path[i].p_hdr) {
2254 ext_debug("initialize header\n");
2255 path[i].p_hdr = ext_block_hdr(path[i].p_bh);
2256 }
2257
2258 if (!path[i].p_idx) {
2259 /* this level hasn't been touched yet */
2260 path[i].p_idx = EXT_LAST_INDEX(path[i].p_hdr);
2261 path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries)+1;
2262 ext_debug("init index ptr: hdr 0x%p, num %d\n",
2263 path[i].p_hdr,
2264 le16_to_cpu(path[i].p_hdr->eh_entries));
2265 } else {
2266 /* we were already here, see at next index */
2267 path[i].p_idx--;
2268 }
2269
2270 ext_debug("level %d - index, first 0x%p, cur 0x%p\n",
2271 i, EXT_FIRST_INDEX(path[i].p_hdr),
2272 path[i].p_idx);
2273 if (ext4_ext_more_to_rm(path + i)) {
2274 struct buffer_head *bh;
2275 /* go to the next level */
2276 ext_debug("move to level %d (block %llu)\n",
2277 i + 1, idx_pblock(path[i].p_idx));
2278 memset(path + i + 1, 0, sizeof(*path));
2279 bh = sb_bread(sb, idx_pblock(path[i].p_idx));
2280 if (!bh) {
2281 /* should we reset i_size? */
2282 err = -EIO;
2283 break;
2284 }
2285 if (WARN_ON(i + 1 > depth)) {
2286 err = -EIO;
2287 break;
2288 }
2289 if (ext4_ext_check(inode, ext_block_hdr(bh),
2290 depth - i - 1)) {
2291 err = -EIO;
2292 break;
2293 }
2294 path[i + 1].p_bh = bh;
2295
2296 /* save actual number of indexes since this
2297 * number is changed at the next iteration */
2298 path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries);
2299 i++;
2300 } else {
2301 /* we finished processing this index, go up */
2302 if (path[i].p_hdr->eh_entries == 0 && i > 0) {
2303 /* index is empty, remove it;
2304 * handle must be already prepared by the
2305 * truncatei_leaf() */
2306 err = ext4_ext_rm_idx(handle, inode, path + i);
2307 }
2308 /* root level has p_bh == NULL, brelse() eats this */
2309 brelse(path[i].p_bh);
2310 path[i].p_bh = NULL;
2311 i--;
2312 ext_debug("return to level %d\n", i);
2313 }
2314 }
2315
2316 /* TODO: flexible tree reduction should be here */
2317 if (path->p_hdr->eh_entries == 0) {
2318 /*
2319 * truncate to zero freed all the tree,
2320 * so we need to correct eh_depth
2321 */
2322 err = ext4_ext_get_access(handle, inode, path);
2323 if (err == 0) {
2324 ext_inode_hdr(inode)->eh_depth = 0;
2325 ext_inode_hdr(inode)->eh_max =
2326 cpu_to_le16(ext4_ext_space_root(inode));
2327 err = ext4_ext_dirty(handle, inode, path);
2328 }
2329 }
2330 out:
2331 ext4_ext_drop_refs(path);
2332 kfree(path);
2333 ext4_journal_stop(handle);
2334
2335 return err;
2336 }
2337
2338 /*
2339 * called at mount time
2340 */
2341 void ext4_ext_init(struct super_block *sb)
2342 {
2343 /*
2344 * possible initialization would be here
2345 */
2346
2347 if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) {
2348 printk(KERN_INFO "EXT4-fs: file extents enabled");
2349 #ifdef AGGRESSIVE_TEST
2350 printk(", aggressive tests");
2351 #endif
2352 #ifdef CHECK_BINSEARCH
2353 printk(", check binsearch");
2354 #endif
2355 #ifdef EXTENTS_STATS
2356 printk(", stats");
2357 #endif
2358 printk("\n");
2359 #ifdef EXTENTS_STATS
2360 spin_lock_init(&EXT4_SB(sb)->s_ext_stats_lock);
2361 EXT4_SB(sb)->s_ext_min = 1 << 30;
2362 EXT4_SB(sb)->s_ext_max = 0;
2363 #endif
2364 }
2365 }
2366
2367 /*
2368 * called at umount time
2369 */
2370 void ext4_ext_release(struct super_block *sb)
2371 {
2372 if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS))
2373 return;
2374
2375 #ifdef EXTENTS_STATS
2376 if (EXT4_SB(sb)->s_ext_blocks && EXT4_SB(sb)->s_ext_extents) {
2377 struct ext4_sb_info *sbi = EXT4_SB(sb);
2378 printk(KERN_ERR "EXT4-fs: %lu blocks in %lu extents (%lu ave)\n",
2379 sbi->s_ext_blocks, sbi->s_ext_extents,
2380 sbi->s_ext_blocks / sbi->s_ext_extents);
2381 printk(KERN_ERR "EXT4-fs: extents: %lu min, %lu max, max depth %lu\n",
2382 sbi->s_ext_min, sbi->s_ext_max, sbi->s_depth_max);
2383 }
2384 #endif
2385 }
2386
2387 static void bi_complete(struct bio *bio, int error)
2388 {
2389 complete((struct completion *)bio->bi_private);
2390 }
2391
2392 /* FIXME!! we need to try to merge to left or right after zero-out */
2393 static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex)
2394 {
2395 int ret = -EIO;
2396 struct bio *bio;
2397 int blkbits, blocksize;
2398 sector_t ee_pblock;
2399 struct completion event;
2400 unsigned int ee_len, len, done, offset;
2401
2402
2403 blkbits = inode->i_blkbits;
2404 blocksize = inode->i_sb->s_blocksize;
2405 ee_len = ext4_ext_get_actual_len(ex);
2406 ee_pblock = ext_pblock(ex);
2407
2408 /* convert ee_pblock to 512 byte sectors */
2409 ee_pblock = ee_pblock << (blkbits - 9);
2410
2411 while (ee_len > 0) {
2412
2413 if (ee_len > BIO_MAX_PAGES)
2414 len = BIO_MAX_PAGES;
2415 else
2416 len = ee_len;
2417
2418 bio = bio_alloc(GFP_NOIO, len);
2419 bio->bi_sector = ee_pblock;
2420 bio->bi_bdev = inode->i_sb->s_bdev;
2421
2422 done = 0;
2423 offset = 0;
2424 while (done < len) {
2425 ret = bio_add_page(bio, ZERO_PAGE(0),
2426 blocksize, offset);
2427 if (ret != blocksize) {
2428 /*
2429 * We can't add any more pages because of
2430 * hardware limitations. Start a new bio.
2431 */
2432 break;
2433 }
2434 done++;
2435 offset += blocksize;
2436 if (offset >= PAGE_CACHE_SIZE)
2437 offset = 0;
2438 }
2439
2440 init_completion(&event);
2441 bio->bi_private = &event;
2442 bio->bi_end_io = bi_complete;
2443 submit_bio(WRITE, bio);
2444 wait_for_completion(&event);
2445
2446 if (test_bit(BIO_UPTODATE, &bio->bi_flags))
2447 ret = 0;
2448 else {
2449 ret = -EIO;
2450 break;
2451 }
2452 bio_put(bio);
2453 ee_len -= done;
2454 ee_pblock += done << (blkbits - 9);
2455 }
2456 return ret;
2457 }
2458
2459 #define EXT4_EXT_ZERO_LEN 7
2460
2461 /*
2462 * This function is called by ext4_ext_get_blocks() if someone tries to write
2463 * to an uninitialized extent. It may result in splitting the uninitialized
2464 * extent into multiple extents (upto three - one initialized and two
2465 * uninitialized).
2466 * There are three possibilities:
2467 * a> There is no split required: Entire extent should be initialized
2468 * b> Splits in two extents: Write is happening at either end of the extent
2469 * c> Splits in three extents: Somone is writing in middle of the extent
2470 */
2471 static int ext4_ext_convert_to_initialized(handle_t *handle,
2472 struct inode *inode,
2473 struct ext4_ext_path *path,
2474 ext4_lblk_t iblock,
2475 unsigned int max_blocks)
2476 {
2477 struct ext4_extent *ex, newex, orig_ex;
2478 struct ext4_extent *ex1 = NULL;
2479 struct ext4_extent *ex2 = NULL;
2480 struct ext4_extent *ex3 = NULL;
2481 struct ext4_extent_header *eh;
2482 ext4_lblk_t ee_block;
2483 unsigned int allocated, ee_len, depth;
2484 ext4_fsblk_t newblock;
2485 int err = 0;
2486 int ret = 0;
2487
2488 depth = ext_depth(inode);
2489 eh = path[depth].p_hdr;
2490 ex = path[depth].p_ext;
2491 ee_block = le32_to_cpu(ex->ee_block);
2492 ee_len = ext4_ext_get_actual_len(ex);
2493 allocated = ee_len - (iblock - ee_block);
2494 newblock = iblock - ee_block + ext_pblock(ex);
2495 ex2 = ex;
2496 orig_ex.ee_block = ex->ee_block;
2497 orig_ex.ee_len = cpu_to_le16(ee_len);
2498 ext4_ext_store_pblock(&orig_ex, ext_pblock(ex));
2499
2500 err = ext4_ext_get_access(handle, inode, path + depth);
2501 if (err)
2502 goto out;
2503 /* If extent has less than 2*EXT4_EXT_ZERO_LEN zerout directly */
2504 if (ee_len <= 2*EXT4_EXT_ZERO_LEN) {
2505 err = ext4_ext_zeroout(inode, &orig_ex);
2506 if (err)
2507 goto fix_extent_len;
2508 /* update the extent length and mark as initialized */
2509 ex->ee_block = orig_ex.ee_block;
2510 ex->ee_len = orig_ex.ee_len;
2511 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2512 ext4_ext_dirty(handle, inode, path + depth);
2513 /* zeroed the full extent */
2514 return allocated;
2515 }
2516
2517 /* ex1: ee_block to iblock - 1 : uninitialized */
2518 if (iblock > ee_block) {
2519 ex1 = ex;
2520 ex1->ee_len = cpu_to_le16(iblock - ee_block);
2521 ext4_ext_mark_uninitialized(ex1);
2522 ex2 = &newex;
2523 }
2524 /*
2525 * for sanity, update the length of the ex2 extent before
2526 * we insert ex3, if ex1 is NULL. This is to avoid temporary
2527 * overlap of blocks.
2528 */
2529 if (!ex1 && allocated > max_blocks)
2530 ex2->ee_len = cpu_to_le16(max_blocks);
2531 /* ex3: to ee_block + ee_len : uninitialised */
2532 if (allocated > max_blocks) {
2533 unsigned int newdepth;
2534 /* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */
2535 if (allocated <= EXT4_EXT_ZERO_LEN) {
2536 /*
2537 * iblock == ee_block is handled by the zerouout
2538 * at the beginning.
2539 * Mark first half uninitialized.
2540 * Mark second half initialized and zero out the
2541 * initialized extent
2542 */
2543 ex->ee_block = orig_ex.ee_block;
2544 ex->ee_len = cpu_to_le16(ee_len - allocated);
2545 ext4_ext_mark_uninitialized(ex);
2546 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2547 ext4_ext_dirty(handle, inode, path + depth);
2548
2549 ex3 = &newex;
2550 ex3->ee_block = cpu_to_le32(iblock);
2551 ext4_ext_store_pblock(ex3, newblock);
2552 ex3->ee_len = cpu_to_le16(allocated);
2553 err = ext4_ext_insert_extent(handle, inode, path, ex3);
2554 if (err == -ENOSPC) {
2555 err = ext4_ext_zeroout(inode, &orig_ex);
2556 if (err)
2557 goto fix_extent_len;
2558 ex->ee_block = orig_ex.ee_block;
2559 ex->ee_len = orig_ex.ee_len;
2560 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2561 ext4_ext_dirty(handle, inode, path + depth);
2562 /* blocks available from iblock */
2563 return allocated;
2564
2565 } else if (err)
2566 goto fix_extent_len;
2567
2568 /*
2569 * We need to zero out the second half because
2570 * an fallocate request can update file size and
2571 * converting the second half to initialized extent
2572 * implies that we can leak some junk data to user
2573 * space.
2574 */
2575 err = ext4_ext_zeroout(inode, ex3);
2576 if (err) {
2577 /*
2578 * We should actually mark the
2579 * second half as uninit and return error
2580 * Insert would have changed the extent
2581 */
2582 depth = ext_depth(inode);
2583 ext4_ext_drop_refs(path);
2584 path = ext4_ext_find_extent(inode,
2585 iblock, path);
2586 if (IS_ERR(path)) {
2587 err = PTR_ERR(path);
2588 return err;
2589 }
2590 /* get the second half extent details */
2591 ex = path[depth].p_ext;
2592 err = ext4_ext_get_access(handle, inode,
2593 path + depth);
2594 if (err)
2595 return err;
2596 ext4_ext_mark_uninitialized(ex);
2597 ext4_ext_dirty(handle, inode, path + depth);
2598 return err;
2599 }
2600
2601 /* zeroed the second half */
2602 return allocated;
2603 }
2604 ex3 = &newex;
2605 ex3->ee_block = cpu_to_le32(iblock + max_blocks);
2606 ext4_ext_store_pblock(ex3, newblock + max_blocks);
2607 ex3->ee_len = cpu_to_le16(allocated - max_blocks);
2608 ext4_ext_mark_uninitialized(ex3);
2609 err = ext4_ext_insert_extent(handle, inode, path, ex3);
2610 if (err == -ENOSPC) {
2611 err = ext4_ext_zeroout(inode, &orig_ex);
2612 if (err)
2613 goto fix_extent_len;
2614 /* update the extent length and mark as initialized */
2615 ex->ee_block = orig_ex.ee_block;
2616 ex->ee_len = orig_ex.ee_len;
2617 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2618 ext4_ext_dirty(handle, inode, path + depth);
2619 /* zeroed the full extent */
2620 /* blocks available from iblock */
2621 return allocated;
2622
2623 } else if (err)
2624 goto fix_extent_len;
2625 /*
2626 * The depth, and hence eh & ex might change
2627 * as part of the insert above.
2628 */
2629 newdepth = ext_depth(inode);
2630 /*
2631 * update the extent length after successful insert of the
2632 * split extent
2633 */
2634 orig_ex.ee_len = cpu_to_le16(ee_len -
2635 ext4_ext_get_actual_len(ex3));
2636 depth = newdepth;
2637 ext4_ext_drop_refs(path);
2638 path = ext4_ext_find_extent(inode, iblock, path);
2639 if (IS_ERR(path)) {
2640 err = PTR_ERR(path);
2641 goto out;
2642 }
2643 eh = path[depth].p_hdr;
2644 ex = path[depth].p_ext;
2645 if (ex2 != &newex)
2646 ex2 = ex;
2647
2648 err = ext4_ext_get_access(handle, inode, path + depth);
2649 if (err)
2650 goto out;
2651
2652 allocated = max_blocks;
2653
2654 /* If extent has less than EXT4_EXT_ZERO_LEN and we are trying
2655 * to insert a extent in the middle zerout directly
2656 * otherwise give the extent a chance to merge to left
2657 */
2658 if (le16_to_cpu(orig_ex.ee_len) <= EXT4_EXT_ZERO_LEN &&
2659 iblock != ee_block) {
2660 err = ext4_ext_zeroout(inode, &orig_ex);
2661 if (err)
2662 goto fix_extent_len;
2663 /* update the extent length and mark as initialized */
2664 ex->ee_block = orig_ex.ee_block;
2665 ex->ee_len = orig_ex.ee_len;
2666 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2667 ext4_ext_dirty(handle, inode, path + depth);
2668 /* zero out the first half */
2669 /* blocks available from iblock */
2670 return allocated;
2671 }
2672 }
2673 /*
2674 * If there was a change of depth as part of the
2675 * insertion of ex3 above, we need to update the length
2676 * of the ex1 extent again here
2677 */
2678 if (ex1 && ex1 != ex) {
2679 ex1 = ex;
2680 ex1->ee_len = cpu_to_le16(iblock - ee_block);
2681 ext4_ext_mark_uninitialized(ex1);
2682 ex2 = &newex;
2683 }
2684 /* ex2: iblock to iblock + maxblocks-1 : initialised */
2685 ex2->ee_block = cpu_to_le32(iblock);
2686 ext4_ext_store_pblock(ex2, newblock);
2687 ex2->ee_len = cpu_to_le16(allocated);
2688 if (ex2 != ex)
2689 goto insert;
2690 /*
2691 * New (initialized) extent starts from the first block
2692 * in the current extent. i.e., ex2 == ex
2693 * We have to see if it can be merged with the extent
2694 * on the left.
2695 */
2696 if (ex2 > EXT_FIRST_EXTENT(eh)) {
2697 /*
2698 * To merge left, pass "ex2 - 1" to try_to_merge(),
2699 * since it merges towards right _only_.
2700 */
2701 ret = ext4_ext_try_to_merge(inode, path, ex2 - 1);
2702 if (ret) {
2703 err = ext4_ext_correct_indexes(handle, inode, path);
2704 if (err)
2705 goto out;
2706 depth = ext_depth(inode);
2707 ex2--;
2708 }
2709 }
2710 /*
2711 * Try to Merge towards right. This might be required
2712 * only when the whole extent is being written to.
2713 * i.e. ex2 == ex and ex3 == NULL.
2714 */
2715 if (!ex3) {
2716 ret = ext4_ext_try_to_merge(inode, path, ex2);
2717 if (ret) {
2718 err = ext4_ext_correct_indexes(handle, inode, path);
2719 if (err)
2720 goto out;
2721 }
2722 }
2723 /* Mark modified extent as dirty */
2724 err = ext4_ext_dirty(handle, inode, path + depth);
2725 goto out;
2726 insert:
2727 err = ext4_ext_insert_extent(handle, inode, path, &newex);
2728 if (err == -ENOSPC) {
2729 err = ext4_ext_zeroout(inode, &orig_ex);
2730 if (err)
2731 goto fix_extent_len;
2732 /* update the extent length and mark as initialized */
2733 ex->ee_block = orig_ex.ee_block;
2734 ex->ee_len = orig_ex.ee_len;
2735 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2736 ext4_ext_dirty(handle, inode, path + depth);
2737 /* zero out the first half */
2738 return allocated;
2739 } else if (err)
2740 goto fix_extent_len;
2741 out:
2742 return err ? err : allocated;
2743
2744 fix_extent_len:
2745 ex->ee_block = orig_ex.ee_block;
2746 ex->ee_len = orig_ex.ee_len;
2747 ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
2748 ext4_ext_mark_uninitialized(ex);
2749 ext4_ext_dirty(handle, inode, path + depth);
2750 return err;
2751 }
2752
2753 /*
2754 * Block allocation/map/preallocation routine for extents based files
2755 *
2756 *
2757 * Need to be called with
2758 * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block
2759 * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem)
2760 *
2761 * return > 0, number of of blocks already mapped/allocated
2762 * if create == 0 and these are pre-allocated blocks
2763 * buffer head is unmapped
2764 * otherwise blocks are mapped
2765 *
2766 * return = 0, if plain look up failed (blocks have not been allocated)
2767 * buffer head is unmapped
2768 *
2769 * return < 0, error case.
2770 */
2771 int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
2772 ext4_lblk_t iblock,
2773 unsigned int max_blocks, struct buffer_head *bh_result,
2774 int create, int extend_disksize)
2775 {
2776 struct ext4_ext_path *path = NULL;
2777 struct ext4_extent_header *eh;
2778 struct ext4_extent newex, *ex;
2779 ext4_fsblk_t newblock;
2780 int err = 0, depth, ret, cache_type;
2781 unsigned int allocated = 0;
2782 struct ext4_allocation_request ar;
2783 loff_t disksize;
2784
2785 __clear_bit(BH_New, &bh_result->b_state);
2786 ext_debug("blocks %u/%u requested for inode %u\n",
2787 iblock, max_blocks, inode->i_ino);
2788
2789 /* check in cache */
2790 cache_type = ext4_ext_in_cache(inode, iblock, &newex);
2791 if (cache_type) {
2792 if (cache_type == EXT4_EXT_CACHE_GAP) {
2793 if (!create) {
2794 /*
2795 * block isn't allocated yet and
2796 * user doesn't want to allocate it
2797 */
2798 goto out2;
2799 }
2800 /* we should allocate requested block */
2801 } else if (cache_type == EXT4_EXT_CACHE_EXTENT) {
2802 /* block is already allocated */
2803 newblock = iblock
2804 - le32_to_cpu(newex.ee_block)
2805 + ext_pblock(&newex);
2806 /* number of remaining blocks in the extent */
2807 allocated = ext4_ext_get_actual_len(&newex) -
2808 (iblock - le32_to_cpu(newex.ee_block));
2809 goto out;
2810 } else {
2811 BUG();
2812 }
2813 }
2814
2815 /* find extent for this block */
2816 path = ext4_ext_find_extent(inode, iblock, NULL);
2817 if (IS_ERR(path)) {
2818 err = PTR_ERR(path);
2819 path = NULL;
2820 goto out2;
2821 }
2822
2823 depth = ext_depth(inode);
2824
2825 /*
2826 * consistent leaf must not be empty;
2827 * this situation is possible, though, _during_ tree modification;
2828 * this is why assert can't be put in ext4_ext_find_extent()
2829 */
2830 BUG_ON(path[depth].p_ext == NULL && depth != 0);
2831 eh = path[depth].p_hdr;
2832
2833 ex = path[depth].p_ext;
2834 if (ex) {
2835 ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block);
2836 ext4_fsblk_t ee_start = ext_pblock(ex);
2837 unsigned short ee_len;
2838
2839 /*
2840 * Uninitialized extents are treated as holes, except that
2841 * we split out initialized portions during a write.
2842 */
2843 ee_len = ext4_ext_get_actual_len(ex);
2844 /* if found extent covers block, simply return it */
2845 if (iblock >= ee_block && iblock < ee_block + ee_len) {
2846 newblock = iblock - ee_block + ee_start;
2847 /* number of remaining blocks in the extent */
2848 allocated = ee_len - (iblock - ee_block);
2849 ext_debug("%u fit into %lu:%d -> %llu\n", iblock,
2850 ee_block, ee_len, newblock);
2851
2852 /* Do not put uninitialized extent in the cache */
2853 if (!ext4_ext_is_uninitialized(ex)) {
2854 ext4_ext_put_in_cache(inode, ee_block,
2855 ee_len, ee_start,
2856 EXT4_EXT_CACHE_EXTENT);
2857 goto out;
2858 }
2859 if (create == EXT4_CREATE_UNINITIALIZED_EXT)
2860 goto out;
2861 if (!create) {
2862 /*
2863 * We have blocks reserved already. We
2864 * return allocated blocks so that delalloc
2865 * won't do block reservation for us. But
2866 * the buffer head will be unmapped so that
2867 * a read from the block returns 0s.
2868 */
2869 if (allocated > max_blocks)
2870 allocated = max_blocks;
2871 set_buffer_unwritten(bh_result);
2872 goto out2;
2873 }
2874
2875 ret = ext4_ext_convert_to_initialized(handle, inode,
2876 path, iblock,
2877 max_blocks);
2878 if (ret <= 0) {
2879 err = ret;
2880 goto out2;
2881 } else
2882 allocated = ret;
2883 goto outnew;
2884 }
2885 }
2886
2887 /*
2888 * requested block isn't allocated yet;
2889 * we couldn't try to create block if create flag is zero
2890 */
2891 if (!create) {
2892 /*
2893 * put just found gap into cache to speed up
2894 * subsequent requests
2895 */
2896 ext4_ext_put_gap_in_cache(inode, path, iblock);
2897 goto out2;
2898 }
2899 /*
2900 * Okay, we need to do block allocation.
2901 */
2902
2903 /* find neighbour allocated blocks */
2904 ar.lleft = iblock;
2905 err = ext4_ext_search_left(inode, path, &ar.lleft, &ar.pleft);
2906 if (err)
2907 goto out2;
2908 ar.lright = iblock;
2909 err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright);
2910 if (err)
2911 goto out2;
2912
2913 /*
2914 * See if request is beyond maximum number of blocks we can have in
2915 * a single extent. For an initialized extent this limit is
2916 * EXT_INIT_MAX_LEN and for an uninitialized extent this limit is
2917 * EXT_UNINIT_MAX_LEN.
2918 */
2919 if (max_blocks > EXT_INIT_MAX_LEN &&
2920 create != EXT4_CREATE_UNINITIALIZED_EXT)
2921 max_blocks = EXT_INIT_MAX_LEN;
2922 else if (max_blocks > EXT_UNINIT_MAX_LEN &&
2923 create == EXT4_CREATE_UNINITIALIZED_EXT)
2924 max_blocks = EXT_UNINIT_MAX_LEN;
2925
2926 /* Check if we can really insert (iblock)::(iblock+max_blocks) extent */
2927 newex.ee_block = cpu_to_le32(iblock);
2928 newex.ee_len = cpu_to_le16(max_blocks);
2929 err = ext4_ext_check_overlap(inode, &newex, path);
2930 if (err)
2931 allocated = ext4_ext_get_actual_len(&newex);
2932 else
2933 allocated = max_blocks;
2934
2935 /* allocate new block */
2936 ar.inode = inode;
2937 ar.goal = ext4_ext_find_goal(inode, path, iblock);
2938 ar.logical = iblock;
2939 ar.len = allocated;
2940 if (S_ISREG(inode->i_mode))
2941 ar.flags = EXT4_MB_HINT_DATA;
2942 else
2943 /* disable in-core preallocation for non-regular files */
2944 ar.flags = 0;
2945 newblock = ext4_mb_new_blocks(handle, &ar, &err);
2946 if (!newblock)
2947 goto out2;
2948 ext_debug("allocate new block: goal %llu, found %llu/%lu\n",
2949 ar.goal, newblock, allocated);
2950
2951 /* try to insert new extent into found leaf and return */
2952 ext4_ext_store_pblock(&newex, newblock);
2953 newex.ee_len = cpu_to_le16(ar.len);
2954 if (create == EXT4_CREATE_UNINITIALIZED_EXT) /* Mark uninitialized */
2955 ext4_ext_mark_uninitialized(&newex);
2956 err = ext4_ext_insert_extent(handle, inode, path, &newex);
2957 if (err) {
2958 /* free data blocks we just allocated */
2959 /* not a good idea to call discard here directly,
2960 * but otherwise we'd need to call it every free() */
2961 ext4_discard_preallocations(inode);
2962 ext4_free_blocks(handle, inode, ext_pblock(&newex),
2963 ext4_ext_get_actual_len(&newex), 0);
2964 goto out2;
2965 }
2966
2967 /* previous routine could use block we allocated */
2968 newblock = ext_pblock(&newex);
2969 allocated = ext4_ext_get_actual_len(&newex);
2970 outnew:
2971 if (extend_disksize) {
2972 disksize = ((loff_t) iblock + ar.len) << inode->i_blkbits;
2973 if (disksize > i_size_read(inode))
2974 disksize = i_size_read(inode);
2975 if (disksize > EXT4_I(inode)->i_disksize)
2976 EXT4_I(inode)->i_disksize = disksize;
2977 }
2978
2979 set_buffer_new(bh_result);
2980
2981 /* Cache only when it is _not_ an uninitialized extent */
2982 if (create != EXT4_CREATE_UNINITIALIZED_EXT)
2983 ext4_ext_put_in_cache(inode, iblock, allocated, newblock,
2984 EXT4_EXT_CACHE_EXTENT);
2985 out:
2986 if (allocated > max_blocks)
2987 allocated = max_blocks;
2988 ext4_ext_show_leaf(inode, path);
2989 set_buffer_mapped(bh_result);
2990 bh_result->b_bdev = inode->i_sb->s_bdev;
2991 bh_result->b_blocknr = newblock;
2992 out2:
2993 if (path) {
2994 ext4_ext_drop_refs(path);
2995 kfree(path);
2996 }
2997 return err ? err : allocated;
2998 }
2999
3000 void ext4_ext_truncate(struct inode *inode)
3001 {
3002 struct address_space *mapping = inode->i_mapping;
3003 struct super_block *sb = inode->i_sb;
3004 ext4_lblk_t last_block;
3005 handle_t *handle;
3006 int err = 0;
3007
3008 /*
3009 * probably first extent we're gonna free will be last in block
3010 */
3011 err = ext4_writepage_trans_blocks(inode);
3012 handle = ext4_journal_start(inode, err);
3013 if (IS_ERR(handle))
3014 return;
3015
3016 if (inode->i_size & (sb->s_blocksize - 1))
3017 ext4_block_truncate_page(handle, mapping, inode->i_size);
3018
3019 if (ext4_orphan_add(handle, inode))
3020 goto out_stop;
3021
3022 down_write(&EXT4_I(inode)->i_data_sem);
3023 ext4_ext_invalidate_cache(inode);
3024
3025 ext4_discard_preallocations(inode);
3026
3027 /*
3028 * TODO: optimization is possible here.
3029 * Probably we need not scan at all,
3030 * because page truncation is enough.
3031 */
3032
3033 /* we have to know where to truncate from in crash case */
3034 EXT4_I(inode)->i_disksize = inode->i_size;
3035 ext4_mark_inode_dirty(handle, inode);
3036
3037 last_block = (inode->i_size + sb->s_blocksize - 1)
3038 >> EXT4_BLOCK_SIZE_BITS(sb);
3039 err = ext4_ext_remove_space(inode, last_block);
3040
3041 /* In a multi-transaction truncate, we only make the final
3042 * transaction synchronous.
3043 */
3044 if (IS_SYNC(inode))
3045 ext4_handle_sync(handle);
3046
3047 out_stop:
3048 up_write(&EXT4_I(inode)->i_data_sem);
3049 /*
3050 * If this was a simple ftruncate() and the file will remain alive,
3051 * then we need to clear up the orphan record which we created above.
3052 * However, if this was a real unlink then we were called by
3053 * ext4_delete_inode(), and we allow that function to clean up the
3054 * orphan info for us.
3055 */
3056 if (inode->i_nlink)
3057 ext4_orphan_del(handle, inode);
3058
3059 inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
3060 ext4_mark_inode_dirty(handle, inode);
3061 ext4_journal_stop(handle);
3062 }
3063
3064 static void ext4_falloc_update_inode(struct inode *inode,
3065 int mode, loff_t new_size, int update_ctime)
3066 {
3067 struct timespec now;
3068
3069 if (update_ctime) {
3070 now = current_fs_time(inode->i_sb);
3071 if (!timespec_equal(&inode->i_ctime, &now))
3072 inode->i_ctime = now;
3073 }
3074 /*
3075 * Update only when preallocation was requested beyond
3076 * the file size.
3077 */
3078 if (!(mode & FALLOC_FL_KEEP_SIZE)) {
3079 if (new_size > i_size_read(inode))
3080 i_size_write(inode, new_size);
3081 if (new_size > EXT4_I(inode)->i_disksize)
3082 ext4_update_i_disksize(inode, new_size);
3083 }
3084
3085 }
3086
3087 /*
3088 * preallocate space for a file. This implements ext4's fallocate inode
3089 * operation, which gets called from sys_fallocate system call.
3090 * For block-mapped files, posix_fallocate should fall back to the method
3091 * of writing zeroes to the required new blocks (the same behavior which is
3092 * expected for file systems which do not support fallocate() system call).
3093 */
3094 long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
3095 {
3096 handle_t *handle;
3097 ext4_lblk_t block;
3098 loff_t new_size;
3099 unsigned int max_blocks;
3100 int ret = 0;
3101 int ret2 = 0;
3102 int retries = 0;
3103 struct buffer_head map_bh;
3104 unsigned int credits, blkbits = inode->i_blkbits;
3105
3106 /*
3107 * currently supporting (pre)allocate mode for extent-based
3108 * files _only_
3109 */
3110 if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL))
3111 return -EOPNOTSUPP;
3112
3113 /* preallocation to directories is currently not supported */
3114 if (S_ISDIR(inode->i_mode))
3115 return -ENODEV;
3116
3117 block = offset >> blkbits;
3118 /*
3119 * We can't just convert len to max_blocks because
3120 * If blocksize = 4096 offset = 3072 and len = 2048
3121 */
3122 max_blocks = (EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits)
3123 - block;
3124 /*
3125 * credits to insert 1 extent into extent tree
3126 */
3127 credits = ext4_chunk_trans_blocks(inode, max_blocks);
3128 mutex_lock(&inode->i_mutex);
3129 retry:
3130 while (ret >= 0 && ret < max_blocks) {
3131 block = block + ret;
3132 max_blocks = max_blocks - ret;
3133 handle = ext4_journal_start(inode, credits);
3134 if (IS_ERR(handle)) {
3135 ret = PTR_ERR(handle);
3136 break;
3137 }
3138 ret = ext4_get_blocks_wrap(handle, inode, block,
3139 max_blocks, &map_bh,
3140 EXT4_CREATE_UNINITIALIZED_EXT, 0, 0);
3141 if (ret <= 0) {
3142 #ifdef EXT4FS_DEBUG
3143 WARN_ON(ret <= 0);
3144 printk(KERN_ERR "%s: ext4_ext_get_blocks "
3145 "returned error inode#%lu, block=%u, "
3146 "max_blocks=%u", __func__,
3147 inode->i_ino, block, max_blocks);
3148 #endif
3149 ext4_mark_inode_dirty(handle, inode);
3150 ret2 = ext4_journal_stop(handle);
3151 break;
3152 }
3153 if ((block + ret) >= (EXT4_BLOCK_ALIGN(offset + len,
3154 blkbits) >> blkbits))
3155 new_size = offset + len;
3156 else
3157 new_size = (block + ret) << blkbits;
3158
3159 ext4_falloc_update_inode(inode, mode, new_size,
3160 buffer_new(&map_bh));
3161 ext4_mark_inode_dirty(handle, inode);
3162 ret2 = ext4_journal_stop(handle);
3163 if (ret2)
3164 break;
3165 }
3166 if (ret == -ENOSPC &&
3167 ext4_should_retry_alloc(inode->i_sb, &retries)) {
3168 ret = 0;
3169 goto retry;
3170 }
3171 mutex_unlock(&inode->i_mutex);
3172 return ret > 0 ? ret2 : ret;
3173 }
3174
3175 /*
3176 * Callback function called for each extent to gather FIEMAP information.
3177 */
3178 static int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path,
3179 struct ext4_ext_cache *newex, struct ext4_extent *ex,
3180 void *data)
3181 {
3182 struct fiemap_extent_info *fieinfo = data;
3183 unsigned long blksize_bits = inode->i_sb->s_blocksize_bits;
3184 __u64 logical;
3185 __u64 physical;
3186 __u64 length;
3187 __u32 flags = 0;
3188 int error;
3189
3190 logical = (__u64)newex->ec_block << blksize_bits;
3191
3192 if (newex->ec_type == EXT4_EXT_CACHE_GAP) {
3193 pgoff_t offset;
3194 struct page *page;
3195 struct buffer_head *bh = NULL;
3196
3197 offset = logical >> PAGE_SHIFT;
3198 page = find_get_page(inode->i_mapping, offset);
3199 if (!page || !page_has_buffers(page))
3200 return EXT_CONTINUE;
3201
3202 bh = page_buffers(page);
3203
3204 if (!bh)
3205 return EXT_CONTINUE;
3206
3207 if (buffer_delay(bh)) {
3208 flags |= FIEMAP_EXTENT_DELALLOC;
3209 page_cache_release(page);
3210 } else {
3211 page_cache_release(page);
3212 return EXT_CONTINUE;
3213 }
3214 }
3215
3216 physical = (__u64)newex->ec_start << blksize_bits;
3217 length = (__u64)newex->ec_len << blksize_bits;
3218
3219 if (ex && ext4_ext_is_uninitialized(ex))
3220 flags |= FIEMAP_EXTENT_UNWRITTEN;
3221
3222 /*
3223 * If this extent reaches EXT_MAX_BLOCK, it must be last.
3224 *
3225 * Or if ext4_ext_next_allocated_block is EXT_MAX_BLOCK,
3226 * this also indicates no more allocated blocks.
3227 *
3228 * XXX this might miss a single-block extent at EXT_MAX_BLOCK
3229 */
3230 if (logical + length - 1 == EXT_MAX_BLOCK ||
3231 ext4_ext_next_allocated_block(path) == EXT_MAX_BLOCK)
3232 flags |= FIEMAP_EXTENT_LAST;
3233
3234 error = fiemap_fill_next_extent(fieinfo, logical, physical,
3235 length, flags);
3236 if (error < 0)
3237 return error;
3238 if (error == 1)
3239 return EXT_BREAK;
3240
3241 return EXT_CONTINUE;
3242 }
3243
3244 /* fiemap flags we can handle specified here */
3245 #define EXT4_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
3246
3247 static int ext4_xattr_fiemap(struct inode *inode,
3248 struct fiemap_extent_info *fieinfo)
3249 {
3250 __u64 physical = 0;
3251 __u64 length;
3252 __u32 flags = FIEMAP_EXTENT_LAST;
3253 int blockbits = inode->i_sb->s_blocksize_bits;
3254 int error = 0;
3255
3256 /* in-inode? */
3257 if (EXT4_I(inode)->i_state & EXT4_STATE_XATTR) {
3258 struct ext4_iloc iloc;
3259 int offset; /* offset of xattr in inode */
3260
3261 error = ext4_get_inode_loc(inode, &iloc);
3262 if (error)
3263 return error;
3264 physical = iloc.bh->b_blocknr << blockbits;
3265 offset = EXT4_GOOD_OLD_INODE_SIZE +
3266 EXT4_I(inode)->i_extra_isize;
3267 physical += offset;
3268 length = EXT4_SB(inode->i_sb)->s_inode_size - offset;
3269 flags |= FIEMAP_EXTENT_DATA_INLINE;
3270 } else { /* external block */
3271 physical = EXT4_I(inode)->i_file_acl << blockbits;
3272 length = inode->i_sb->s_blocksize;
3273 }
3274
3275 if (physical)
3276 error = fiemap_fill_next_extent(fieinfo, 0, physical,
3277 length, flags);
3278 return (error < 0 ? error : 0);
3279 }
3280
3281 int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
3282 __u64 start, __u64 len)
3283 {
3284 ext4_lblk_t start_blk;
3285 ext4_lblk_t len_blks;
3286 int error = 0;
3287
3288 /* fallback to generic here if not in extents fmt */
3289 if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL))
3290 return generic_block_fiemap(inode, fieinfo, start, len,
3291 ext4_get_block);
3292
3293 if (fiemap_check_flags(fieinfo, EXT4_FIEMAP_FLAGS))
3294 return -EBADR;
3295
3296 if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) {
3297 error = ext4_xattr_fiemap(inode, fieinfo);
3298 } else {
3299 start_blk = start >> inode->i_sb->s_blocksize_bits;
3300 len_blks = len >> inode->i_sb->s_blocksize_bits;
3301
3302 /*
3303 * Walk the extent tree gathering extent information.
3304 * ext4_ext_fiemap_cb will push extents back to user.
3305 */
3306 down_write(&EXT4_I(inode)->i_data_sem);
3307 error = ext4_ext_walk_space(inode, start_blk, len_blks,
3308 ext4_ext_fiemap_cb, fieinfo);
3309 up_write(&EXT4_I(inode)->i_data_sem);
3310 }
3311
3312 return error;
3313 }
3314
This page took 0.129803 seconds and 6 git commands to generate.