Commit | Line | Data |
---|---|---|
a86c6181 AT |
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 | ||
a86c6181 AT |
32 | #include <linux/fs.h> |
33 | #include <linux/time.h> | |
cd02ff0b | 34 | #include <linux/jbd2.h> |
a86c6181 AT |
35 | #include <linux/highuid.h> |
36 | #include <linux/pagemap.h> | |
37 | #include <linux/quotaops.h> | |
38 | #include <linux/string.h> | |
39 | #include <linux/slab.h> | |
a86c6181 | 40 | #include <asm/uaccess.h> |
6873fa0d | 41 | #include <linux/fiemap.h> |
3dcf5451 | 42 | #include "ext4_jbd2.h" |
4a092d73 | 43 | #include "ext4_extents.h" |
f19d5870 | 44 | #include "xattr.h" |
a86c6181 | 45 | |
0562e0ba JZ |
46 | #include <trace/events/ext4.h> |
47 | ||
5f95d21f LC |
48 | /* |
49 | * used by extent splitting. | |
50 | */ | |
51 | #define EXT4_EXT_MAY_ZEROOUT 0x1 /* safe to zeroout if split fails \ | |
52 | due to ENOSPC */ | |
556615dc LC |
53 | #define EXT4_EXT_MARK_UNWRIT1 0x2 /* mark first half unwritten */ |
54 | #define EXT4_EXT_MARK_UNWRIT2 0x4 /* mark second half unwritten */ | |
5f95d21f | 55 | |
dee1f973 DM |
56 | #define EXT4_EXT_DATA_VALID1 0x8 /* first half contains valid data */ |
57 | #define EXT4_EXT_DATA_VALID2 0x10 /* second half contains valid data */ | |
58 | ||
7ac5990d DW |
59 | static __le32 ext4_extent_block_csum(struct inode *inode, |
60 | struct ext4_extent_header *eh) | |
61 | { | |
62 | struct ext4_inode_info *ei = EXT4_I(inode); | |
63 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); | |
64 | __u32 csum; | |
65 | ||
66 | csum = ext4_chksum(sbi, ei->i_csum_seed, (__u8 *)eh, | |
67 | EXT4_EXTENT_TAIL_OFFSET(eh)); | |
68 | return cpu_to_le32(csum); | |
69 | } | |
70 | ||
71 | static int ext4_extent_block_csum_verify(struct inode *inode, | |
72 | struct ext4_extent_header *eh) | |
73 | { | |
74 | struct ext4_extent_tail *et; | |
75 | ||
76 | if (!EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb, | |
77 | EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) | |
78 | return 1; | |
79 | ||
80 | et = find_ext4_extent_tail(eh); | |
81 | if (et->et_checksum != ext4_extent_block_csum(inode, eh)) | |
82 | return 0; | |
83 | return 1; | |
84 | } | |
85 | ||
86 | static void ext4_extent_block_csum_set(struct inode *inode, | |
87 | struct ext4_extent_header *eh) | |
88 | { | |
89 | struct ext4_extent_tail *et; | |
90 | ||
91 | if (!EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb, | |
92 | EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) | |
93 | return; | |
94 | ||
95 | et = find_ext4_extent_tail(eh); | |
96 | et->et_checksum = ext4_extent_block_csum(inode, eh); | |
97 | } | |
98 | ||
d583fb87 AH |
99 | static int ext4_split_extent(handle_t *handle, |
100 | struct inode *inode, | |
dfe50809 | 101 | struct ext4_ext_path **ppath, |
d583fb87 AH |
102 | struct ext4_map_blocks *map, |
103 | int split_flag, | |
104 | int flags); | |
105 | ||
5f95d21f LC |
106 | static int ext4_split_extent_at(handle_t *handle, |
107 | struct inode *inode, | |
dfe50809 | 108 | struct ext4_ext_path **ppath, |
5f95d21f LC |
109 | ext4_lblk_t split, |
110 | int split_flag, | |
111 | int flags); | |
112 | ||
91dd8c11 | 113 | static int ext4_find_delayed_extent(struct inode *inode, |
69eb33dc | 114 | struct extent_status *newes); |
91dd8c11 | 115 | |
487caeef JK |
116 | static int ext4_ext_truncate_extend_restart(handle_t *handle, |
117 | struct inode *inode, | |
118 | int needed) | |
a86c6181 AT |
119 | { |
120 | int err; | |
121 | ||
0390131b FM |
122 | if (!ext4_handle_valid(handle)) |
123 | return 0; | |
a86c6181 | 124 | if (handle->h_buffer_credits > needed) |
9102e4fa SF |
125 | return 0; |
126 | err = ext4_journal_extend(handle, needed); | |
0123c939 | 127 | if (err <= 0) |
9102e4fa | 128 | return err; |
487caeef | 129 | err = ext4_truncate_restart_trans(handle, inode, needed); |
0617b83f DM |
130 | if (err == 0) |
131 | err = -EAGAIN; | |
487caeef JK |
132 | |
133 | return err; | |
a86c6181 AT |
134 | } |
135 | ||
136 | /* | |
137 | * could return: | |
138 | * - EROFS | |
139 | * - ENOMEM | |
140 | */ | |
141 | static int ext4_ext_get_access(handle_t *handle, struct inode *inode, | |
142 | struct ext4_ext_path *path) | |
143 | { | |
144 | if (path->p_bh) { | |
145 | /* path points to block */ | |
5d601255 | 146 | BUFFER_TRACE(path->p_bh, "get_write_access"); |
a86c6181 AT |
147 | return ext4_journal_get_write_access(handle, path->p_bh); |
148 | } | |
149 | /* path points to leaf/index in inode body */ | |
150 | /* we use in-core data, no need to protect them */ | |
151 | return 0; | |
152 | } | |
153 | ||
154 | /* | |
155 | * could return: | |
156 | * - EROFS | |
157 | * - ENOMEM | |
158 | * - EIO | |
159 | */ | |
2656497b DW |
160 | int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle, |
161 | struct inode *inode, struct ext4_ext_path *path) | |
a86c6181 AT |
162 | { |
163 | int err; | |
4b1f1660 DM |
164 | |
165 | WARN_ON(!rwsem_is_locked(&EXT4_I(inode)->i_data_sem)); | |
a86c6181 | 166 | if (path->p_bh) { |
7ac5990d | 167 | ext4_extent_block_csum_set(inode, ext_block_hdr(path->p_bh)); |
a86c6181 | 168 | /* path points to block */ |
9ea7a0df TT |
169 | err = __ext4_handle_dirty_metadata(where, line, handle, |
170 | inode, path->p_bh); | |
a86c6181 AT |
171 | } else { |
172 | /* path points to leaf/index in inode body */ | |
173 | err = ext4_mark_inode_dirty(handle, inode); | |
174 | } | |
175 | return err; | |
176 | } | |
177 | ||
f65e6fba | 178 | static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, |
a86c6181 | 179 | struct ext4_ext_path *path, |
725d26d3 | 180 | ext4_lblk_t block) |
a86c6181 | 181 | { |
a86c6181 | 182 | if (path) { |
81fdbb4a | 183 | int depth = path->p_depth; |
a86c6181 | 184 | struct ext4_extent *ex; |
a86c6181 | 185 | |
ad4fb9ca KM |
186 | /* |
187 | * Try to predict block placement assuming that we are | |
188 | * filling in a file which will eventually be | |
189 | * non-sparse --- i.e., in the case of libbfd writing | |
190 | * an ELF object sections out-of-order but in a way | |
191 | * the eventually results in a contiguous object or | |
192 | * executable file, or some database extending a table | |
193 | * space file. However, this is actually somewhat | |
194 | * non-ideal if we are writing a sparse file such as | |
195 | * qemu or KVM writing a raw image file that is going | |
196 | * to stay fairly sparse, since it will end up | |
197 | * fragmenting the file system's free space. Maybe we | |
198 | * should have some hueristics or some way to allow | |
199 | * userspace to pass a hint to file system, | |
b8d6568a | 200 | * especially if the latter case turns out to be |
ad4fb9ca KM |
201 | * common. |
202 | */ | |
7e028976 | 203 | ex = path[depth].p_ext; |
ad4fb9ca KM |
204 | if (ex) { |
205 | ext4_fsblk_t ext_pblk = ext4_ext_pblock(ex); | |
206 | ext4_lblk_t ext_block = le32_to_cpu(ex->ee_block); | |
207 | ||
208 | if (block > ext_block) | |
209 | return ext_pblk + (block - ext_block); | |
210 | else | |
211 | return ext_pblk - (ext_block - block); | |
212 | } | |
a86c6181 | 213 | |
d0d856e8 RD |
214 | /* it looks like index is empty; |
215 | * try to find starting block from index itself */ | |
a86c6181 AT |
216 | if (path[depth].p_bh) |
217 | return path[depth].p_bh->b_blocknr; | |
218 | } | |
219 | ||
220 | /* OK. use inode's group */ | |
f86186b4 | 221 | return ext4_inode_to_goal_block(inode); |
a86c6181 AT |
222 | } |
223 | ||
654b4908 AK |
224 | /* |
225 | * Allocation for a meta data block | |
226 | */ | |
f65e6fba | 227 | static ext4_fsblk_t |
654b4908 | 228 | ext4_ext_new_meta_block(handle_t *handle, struct inode *inode, |
a86c6181 | 229 | struct ext4_ext_path *path, |
55f020db | 230 | struct ext4_extent *ex, int *err, unsigned int flags) |
a86c6181 | 231 | { |
f65e6fba | 232 | ext4_fsblk_t goal, newblock; |
a86c6181 AT |
233 | |
234 | goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block)); | |
55f020db AH |
235 | newblock = ext4_new_meta_blocks(handle, inode, goal, flags, |
236 | NULL, err); | |
a86c6181 AT |
237 | return newblock; |
238 | } | |
239 | ||
55ad63bf | 240 | static inline int ext4_ext_space_block(struct inode *inode, int check) |
a86c6181 AT |
241 | { |
242 | int size; | |
243 | ||
244 | size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) | |
245 | / sizeof(struct ext4_extent); | |
bbf2f9fb | 246 | #ifdef AGGRESSIVE_TEST |
02dc62fb YY |
247 | if (!check && size > 6) |
248 | size = 6; | |
a86c6181 AT |
249 | #endif |
250 | return size; | |
251 | } | |
252 | ||
55ad63bf | 253 | static inline int ext4_ext_space_block_idx(struct inode *inode, int check) |
a86c6181 AT |
254 | { |
255 | int size; | |
256 | ||
257 | size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) | |
258 | / sizeof(struct ext4_extent_idx); | |
bbf2f9fb | 259 | #ifdef AGGRESSIVE_TEST |
02dc62fb YY |
260 | if (!check && size > 5) |
261 | size = 5; | |
a86c6181 AT |
262 | #endif |
263 | return size; | |
264 | } | |
265 | ||
55ad63bf | 266 | static inline int ext4_ext_space_root(struct inode *inode, int check) |
a86c6181 AT |
267 | { |
268 | int size; | |
269 | ||
270 | size = sizeof(EXT4_I(inode)->i_data); | |
271 | size -= sizeof(struct ext4_extent_header); | |
272 | size /= sizeof(struct ext4_extent); | |
bbf2f9fb | 273 | #ifdef AGGRESSIVE_TEST |
02dc62fb YY |
274 | if (!check && size > 3) |
275 | size = 3; | |
a86c6181 AT |
276 | #endif |
277 | return size; | |
278 | } | |
279 | ||
55ad63bf | 280 | static inline int ext4_ext_space_root_idx(struct inode *inode, int check) |
a86c6181 AT |
281 | { |
282 | int size; | |
283 | ||
284 | size = sizeof(EXT4_I(inode)->i_data); | |
285 | size -= sizeof(struct ext4_extent_header); | |
286 | size /= sizeof(struct ext4_extent_idx); | |
bbf2f9fb | 287 | #ifdef AGGRESSIVE_TEST |
02dc62fb YY |
288 | if (!check && size > 4) |
289 | size = 4; | |
a86c6181 AT |
290 | #endif |
291 | return size; | |
292 | } | |
293 | ||
fcf6b1b7 DM |
294 | static inline int |
295 | ext4_force_split_extent_at(handle_t *handle, struct inode *inode, | |
dfe50809 | 296 | struct ext4_ext_path **ppath, ext4_lblk_t lblk, |
fcf6b1b7 DM |
297 | int nofail) |
298 | { | |
dfe50809 | 299 | struct ext4_ext_path *path = *ppath; |
fcf6b1b7 DM |
300 | int unwritten = ext4_ext_is_unwritten(path[path->p_depth].p_ext); |
301 | ||
dfe50809 | 302 | return ext4_split_extent_at(handle, inode, ppath, lblk, unwritten ? |
fcf6b1b7 DM |
303 | EXT4_EXT_MARK_UNWRIT1|EXT4_EXT_MARK_UNWRIT2 : 0, |
304 | EXT4_EX_NOCACHE | EXT4_GET_BLOCKS_PRE_IO | | |
305 | (nofail ? EXT4_GET_BLOCKS_METADATA_NOFAIL:0)); | |
306 | } | |
307 | ||
d2a17637 MC |
308 | /* |
309 | * Calculate the number of metadata blocks needed | |
310 | * to allocate @blocks | |
311 | * Worse case is one block per extent | |
312 | */ | |
01f49d0b | 313 | int ext4_ext_calc_metadata_amount(struct inode *inode, ext4_lblk_t lblock) |
d2a17637 | 314 | { |
9d0be502 | 315 | struct ext4_inode_info *ei = EXT4_I(inode); |
81fdbb4a | 316 | int idxs; |
d2a17637 | 317 | |
9d0be502 TT |
318 | idxs = ((inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) |
319 | / sizeof(struct ext4_extent_idx)); | |
d2a17637 MC |
320 | |
321 | /* | |
9d0be502 TT |
322 | * If the new delayed allocation block is contiguous with the |
323 | * previous da block, it can share index blocks with the | |
324 | * previous block, so we only need to allocate a new index | |
325 | * block every idxs leaf blocks. At ldxs**2 blocks, we need | |
326 | * an additional index block, and at ldxs**3 blocks, yet | |
327 | * another index blocks. | |
d2a17637 | 328 | */ |
9d0be502 TT |
329 | if (ei->i_da_metadata_calc_len && |
330 | ei->i_da_metadata_calc_last_lblock+1 == lblock) { | |
81fdbb4a YY |
331 | int num = 0; |
332 | ||
9d0be502 TT |
333 | if ((ei->i_da_metadata_calc_len % idxs) == 0) |
334 | num++; | |
335 | if ((ei->i_da_metadata_calc_len % (idxs*idxs)) == 0) | |
336 | num++; | |
337 | if ((ei->i_da_metadata_calc_len % (idxs*idxs*idxs)) == 0) { | |
338 | num++; | |
339 | ei->i_da_metadata_calc_len = 0; | |
340 | } else | |
341 | ei->i_da_metadata_calc_len++; | |
342 | ei->i_da_metadata_calc_last_lblock++; | |
343 | return num; | |
344 | } | |
d2a17637 | 345 | |
9d0be502 TT |
346 | /* |
347 | * In the worst case we need a new set of index blocks at | |
348 | * every level of the inode's extent tree. | |
349 | */ | |
350 | ei->i_da_metadata_calc_len = 1; | |
351 | ei->i_da_metadata_calc_last_lblock = lblock; | |
352 | return ext_depth(inode) + 1; | |
d2a17637 MC |
353 | } |
354 | ||
c29c0ae7 AT |
355 | static int |
356 | ext4_ext_max_entries(struct inode *inode, int depth) | |
357 | { | |
358 | int max; | |
359 | ||
360 | if (depth == ext_depth(inode)) { | |
361 | if (depth == 0) | |
55ad63bf | 362 | max = ext4_ext_space_root(inode, 1); |
c29c0ae7 | 363 | else |
55ad63bf | 364 | max = ext4_ext_space_root_idx(inode, 1); |
c29c0ae7 AT |
365 | } else { |
366 | if (depth == 0) | |
55ad63bf | 367 | max = ext4_ext_space_block(inode, 1); |
c29c0ae7 | 368 | else |
55ad63bf | 369 | max = ext4_ext_space_block_idx(inode, 1); |
c29c0ae7 AT |
370 | } |
371 | ||
372 | return max; | |
373 | } | |
374 | ||
56b19868 AK |
375 | static int ext4_valid_extent(struct inode *inode, struct ext4_extent *ext) |
376 | { | |
bf89d16f | 377 | ext4_fsblk_t block = ext4_ext_pblock(ext); |
56b19868 | 378 | int len = ext4_ext_get_actual_len(ext); |
5946d089 EG |
379 | ext4_lblk_t lblock = le32_to_cpu(ext->ee_block); |
380 | ext4_lblk_t last = lblock + len - 1; | |
e84a26ce | 381 | |
5946d089 | 382 | if (lblock > last) |
31d4f3a2 | 383 | return 0; |
6fd058f7 | 384 | return ext4_data_block_valid(EXT4_SB(inode->i_sb), block, len); |
56b19868 AK |
385 | } |
386 | ||
387 | static int ext4_valid_extent_idx(struct inode *inode, | |
388 | struct ext4_extent_idx *ext_idx) | |
389 | { | |
bf89d16f | 390 | ext4_fsblk_t block = ext4_idx_pblock(ext_idx); |
e84a26ce | 391 | |
6fd058f7 | 392 | return ext4_data_block_valid(EXT4_SB(inode->i_sb), block, 1); |
56b19868 AK |
393 | } |
394 | ||
395 | static int ext4_valid_extent_entries(struct inode *inode, | |
396 | struct ext4_extent_header *eh, | |
397 | int depth) | |
398 | { | |
56b19868 AK |
399 | unsigned short entries; |
400 | if (eh->eh_entries == 0) | |
401 | return 1; | |
402 | ||
403 | entries = le16_to_cpu(eh->eh_entries); | |
404 | ||
405 | if (depth == 0) { | |
406 | /* leaf entries */ | |
81fdbb4a | 407 | struct ext4_extent *ext = EXT_FIRST_EXTENT(eh); |
5946d089 EG |
408 | struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es; |
409 | ext4_fsblk_t pblock = 0; | |
410 | ext4_lblk_t lblock = 0; | |
411 | ext4_lblk_t prev = 0; | |
412 | int len = 0; | |
56b19868 AK |
413 | while (entries) { |
414 | if (!ext4_valid_extent(inode, ext)) | |
415 | return 0; | |
5946d089 EG |
416 | |
417 | /* Check for overlapping extents */ | |
418 | lblock = le32_to_cpu(ext->ee_block); | |
419 | len = ext4_ext_get_actual_len(ext); | |
420 | if ((lblock <= prev) && prev) { | |
421 | pblock = ext4_ext_pblock(ext); | |
422 | es->s_last_error_block = cpu_to_le64(pblock); | |
423 | return 0; | |
424 | } | |
56b19868 AK |
425 | ext++; |
426 | entries--; | |
5946d089 | 427 | prev = lblock + len - 1; |
56b19868 AK |
428 | } |
429 | } else { | |
81fdbb4a | 430 | struct ext4_extent_idx *ext_idx = EXT_FIRST_INDEX(eh); |
56b19868 AK |
431 | while (entries) { |
432 | if (!ext4_valid_extent_idx(inode, ext_idx)) | |
433 | return 0; | |
434 | ext_idx++; | |
435 | entries--; | |
436 | } | |
437 | } | |
438 | return 1; | |
439 | } | |
440 | ||
c398eda0 TT |
441 | static int __ext4_ext_check(const char *function, unsigned int line, |
442 | struct inode *inode, struct ext4_extent_header *eh, | |
c349179b | 443 | int depth, ext4_fsblk_t pblk) |
c29c0ae7 AT |
444 | { |
445 | const char *error_msg; | |
446 | int max = 0; | |
447 | ||
448 | if (unlikely(eh->eh_magic != EXT4_EXT_MAGIC)) { | |
449 | error_msg = "invalid magic"; | |
450 | goto corrupted; | |
451 | } | |
452 | if (unlikely(le16_to_cpu(eh->eh_depth) != depth)) { | |
453 | error_msg = "unexpected eh_depth"; | |
454 | goto corrupted; | |
455 | } | |
456 | if (unlikely(eh->eh_max == 0)) { | |
457 | error_msg = "invalid eh_max"; | |
458 | goto corrupted; | |
459 | } | |
460 | max = ext4_ext_max_entries(inode, depth); | |
461 | if (unlikely(le16_to_cpu(eh->eh_max) > max)) { | |
462 | error_msg = "too large eh_max"; | |
463 | goto corrupted; | |
464 | } | |
465 | if (unlikely(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max))) { | |
466 | error_msg = "invalid eh_entries"; | |
467 | goto corrupted; | |
468 | } | |
56b19868 AK |
469 | if (!ext4_valid_extent_entries(inode, eh, depth)) { |
470 | error_msg = "invalid extent entries"; | |
471 | goto corrupted; | |
472 | } | |
7ac5990d DW |
473 | /* Verify checksum on non-root extent tree nodes */ |
474 | if (ext_depth(inode) != depth && | |
475 | !ext4_extent_block_csum_verify(inode, eh)) { | |
476 | error_msg = "extent tree corrupted"; | |
477 | goto corrupted; | |
478 | } | |
c29c0ae7 AT |
479 | return 0; |
480 | ||
481 | corrupted: | |
c398eda0 | 482 | ext4_error_inode(inode, function, line, 0, |
c349179b TT |
483 | "pblk %llu bad header/extent: %s - magic %x, " |
484 | "entries %u, max %u(%u), depth %u(%u)", | |
485 | (unsigned long long) pblk, error_msg, | |
486 | le16_to_cpu(eh->eh_magic), | |
487 | le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max), | |
488 | max, le16_to_cpu(eh->eh_depth), depth); | |
c29c0ae7 AT |
489 | return -EIO; |
490 | } | |
491 | ||
c349179b TT |
492 | #define ext4_ext_check(inode, eh, depth, pblk) \ |
493 | __ext4_ext_check(__func__, __LINE__, (inode), (eh), (depth), (pblk)) | |
c29c0ae7 | 494 | |
7a262f7c AK |
495 | int ext4_ext_check_inode(struct inode *inode) |
496 | { | |
c349179b | 497 | return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0); |
7a262f7c AK |
498 | } |
499 | ||
7d7ea89e TT |
500 | static struct buffer_head * |
501 | __read_extent_tree_block(const char *function, unsigned int line, | |
107a7bd3 TT |
502 | struct inode *inode, ext4_fsblk_t pblk, int depth, |
503 | int flags) | |
f8489128 | 504 | { |
7d7ea89e TT |
505 | struct buffer_head *bh; |
506 | int err; | |
507 | ||
508 | bh = sb_getblk(inode->i_sb, pblk); | |
509 | if (unlikely(!bh)) | |
510 | return ERR_PTR(-ENOMEM); | |
f8489128 | 511 | |
7d7ea89e TT |
512 | if (!bh_uptodate_or_lock(bh)) { |
513 | trace_ext4_ext_load_extent(inode, pblk, _RET_IP_); | |
514 | err = bh_submit_read(bh); | |
515 | if (err < 0) | |
516 | goto errout; | |
517 | } | |
7869a4a6 | 518 | if (buffer_verified(bh) && !(flags & EXT4_EX_FORCE_CACHE)) |
7d7ea89e TT |
519 | return bh; |
520 | err = __ext4_ext_check(function, line, inode, | |
c349179b | 521 | ext_block_hdr(bh), depth, pblk); |
7d7ea89e TT |
522 | if (err) |
523 | goto errout; | |
f8489128 | 524 | set_buffer_verified(bh); |
107a7bd3 TT |
525 | /* |
526 | * If this is a leaf block, cache all of its entries | |
527 | */ | |
528 | if (!(flags & EXT4_EX_NOCACHE) && depth == 0) { | |
529 | struct ext4_extent_header *eh = ext_block_hdr(bh); | |
530 | struct ext4_extent *ex = EXT_FIRST_EXTENT(eh); | |
531 | ext4_lblk_t prev = 0; | |
532 | int i; | |
533 | ||
534 | for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) { | |
535 | unsigned int status = EXTENT_STATUS_WRITTEN; | |
536 | ext4_lblk_t lblk = le32_to_cpu(ex->ee_block); | |
537 | int len = ext4_ext_get_actual_len(ex); | |
538 | ||
539 | if (prev && (prev != lblk)) | |
540 | ext4_es_cache_extent(inode, prev, | |
541 | lblk - prev, ~0, | |
542 | EXTENT_STATUS_HOLE); | |
543 | ||
556615dc | 544 | if (ext4_ext_is_unwritten(ex)) |
107a7bd3 TT |
545 | status = EXTENT_STATUS_UNWRITTEN; |
546 | ext4_es_cache_extent(inode, lblk, len, | |
547 | ext4_ext_pblock(ex), status); | |
548 | prev = lblk + len; | |
549 | } | |
550 | } | |
7d7ea89e TT |
551 | return bh; |
552 | errout: | |
553 | put_bh(bh); | |
554 | return ERR_PTR(err); | |
555 | ||
f8489128 DW |
556 | } |
557 | ||
107a7bd3 TT |
558 | #define read_extent_tree_block(inode, pblk, depth, flags) \ |
559 | __read_extent_tree_block(__func__, __LINE__, (inode), (pblk), \ | |
560 | (depth), (flags)) | |
f8489128 | 561 | |
7869a4a6 TT |
562 | /* |
563 | * This function is called to cache a file's extent information in the | |
564 | * extent status tree | |
565 | */ | |
566 | int ext4_ext_precache(struct inode *inode) | |
567 | { | |
568 | struct ext4_inode_info *ei = EXT4_I(inode); | |
569 | struct ext4_ext_path *path = NULL; | |
570 | struct buffer_head *bh; | |
571 | int i = 0, depth, ret = 0; | |
572 | ||
573 | if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) | |
574 | return 0; /* not an extent-mapped inode */ | |
575 | ||
576 | down_read(&ei->i_data_sem); | |
577 | depth = ext_depth(inode); | |
578 | ||
579 | path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), | |
580 | GFP_NOFS); | |
581 | if (path == NULL) { | |
582 | up_read(&ei->i_data_sem); | |
583 | return -ENOMEM; | |
584 | } | |
585 | ||
586 | /* Don't cache anything if there are no external extent blocks */ | |
587 | if (depth == 0) | |
588 | goto out; | |
589 | path[0].p_hdr = ext_inode_hdr(inode); | |
590 | ret = ext4_ext_check(inode, path[0].p_hdr, depth, 0); | |
591 | if (ret) | |
592 | goto out; | |
593 | path[0].p_idx = EXT_FIRST_INDEX(path[0].p_hdr); | |
594 | while (i >= 0) { | |
595 | /* | |
596 | * If this is a leaf block or we've reached the end of | |
597 | * the index block, go up | |
598 | */ | |
599 | if ((i == depth) || | |
600 | path[i].p_idx > EXT_LAST_INDEX(path[i].p_hdr)) { | |
601 | brelse(path[i].p_bh); | |
602 | path[i].p_bh = NULL; | |
603 | i--; | |
604 | continue; | |
605 | } | |
606 | bh = read_extent_tree_block(inode, | |
607 | ext4_idx_pblock(path[i].p_idx++), | |
608 | depth - i - 1, | |
609 | EXT4_EX_FORCE_CACHE); | |
610 | if (IS_ERR(bh)) { | |
611 | ret = PTR_ERR(bh); | |
612 | break; | |
613 | } | |
614 | i++; | |
615 | path[i].p_bh = bh; | |
616 | path[i].p_hdr = ext_block_hdr(bh); | |
617 | path[i].p_idx = EXT_FIRST_INDEX(path[i].p_hdr); | |
618 | } | |
619 | ext4_set_inode_state(inode, EXT4_STATE_EXT_PRECACHED); | |
620 | out: | |
621 | up_read(&ei->i_data_sem); | |
622 | ext4_ext_drop_refs(path); | |
623 | kfree(path); | |
624 | return ret; | |
625 | } | |
626 | ||
a86c6181 AT |
627 | #ifdef EXT_DEBUG |
628 | static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) | |
629 | { | |
630 | int k, l = path->p_depth; | |
631 | ||
632 | ext_debug("path:"); | |
633 | for (k = 0; k <= l; k++, path++) { | |
634 | if (path->p_idx) { | |
2ae02107 | 635 | ext_debug(" %d->%llu", le32_to_cpu(path->p_idx->ei_block), |
bf89d16f | 636 | ext4_idx_pblock(path->p_idx)); |
a86c6181 | 637 | } else if (path->p_ext) { |
553f9008 | 638 | ext_debug(" %d:[%d]%d:%llu ", |
a86c6181 | 639 | le32_to_cpu(path->p_ext->ee_block), |
556615dc | 640 | ext4_ext_is_unwritten(path->p_ext), |
a2df2a63 | 641 | ext4_ext_get_actual_len(path->p_ext), |
bf89d16f | 642 | ext4_ext_pblock(path->p_ext)); |
a86c6181 AT |
643 | } else |
644 | ext_debug(" []"); | |
645 | } | |
646 | ext_debug("\n"); | |
647 | } | |
648 | ||
649 | static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) | |
650 | { | |
651 | int depth = ext_depth(inode); | |
652 | struct ext4_extent_header *eh; | |
653 | struct ext4_extent *ex; | |
654 | int i; | |
655 | ||
656 | if (!path) | |
657 | return; | |
658 | ||
659 | eh = path[depth].p_hdr; | |
660 | ex = EXT_FIRST_EXTENT(eh); | |
661 | ||
553f9008 M |
662 | ext_debug("Displaying leaf extents for inode %lu\n", inode->i_ino); |
663 | ||
a86c6181 | 664 | for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) { |
553f9008 | 665 | ext_debug("%d:[%d]%d:%llu ", le32_to_cpu(ex->ee_block), |
556615dc | 666 | ext4_ext_is_unwritten(ex), |
bf89d16f | 667 | ext4_ext_get_actual_len(ex), ext4_ext_pblock(ex)); |
a86c6181 AT |
668 | } |
669 | ext_debug("\n"); | |
670 | } | |
1b16da77 YY |
671 | |
672 | static void ext4_ext_show_move(struct inode *inode, struct ext4_ext_path *path, | |
673 | ext4_fsblk_t newblock, int level) | |
674 | { | |
675 | int depth = ext_depth(inode); | |
676 | struct ext4_extent *ex; | |
677 | ||
678 | if (depth != level) { | |
679 | struct ext4_extent_idx *idx; | |
680 | idx = path[level].p_idx; | |
681 | while (idx <= EXT_MAX_INDEX(path[level].p_hdr)) { | |
682 | ext_debug("%d: move %d:%llu in new index %llu\n", level, | |
683 | le32_to_cpu(idx->ei_block), | |
684 | ext4_idx_pblock(idx), | |
685 | newblock); | |
686 | idx++; | |
687 | } | |
688 | ||
689 | return; | |
690 | } | |
691 | ||
692 | ex = path[depth].p_ext; | |
693 | while (ex <= EXT_MAX_EXTENT(path[depth].p_hdr)) { | |
694 | ext_debug("move %d:%llu:[%d]%d in new leaf %llu\n", | |
695 | le32_to_cpu(ex->ee_block), | |
696 | ext4_ext_pblock(ex), | |
556615dc | 697 | ext4_ext_is_unwritten(ex), |
1b16da77 YY |
698 | ext4_ext_get_actual_len(ex), |
699 | newblock); | |
700 | ex++; | |
701 | } | |
702 | } | |
703 | ||
a86c6181 | 704 | #else |
af5bc92d TT |
705 | #define ext4_ext_show_path(inode, path) |
706 | #define ext4_ext_show_leaf(inode, path) | |
1b16da77 | 707 | #define ext4_ext_show_move(inode, path, newblock, level) |
a86c6181 AT |
708 | #endif |
709 | ||
b35905c1 | 710 | void ext4_ext_drop_refs(struct ext4_ext_path *path) |
a86c6181 | 711 | { |
b7ea89ad | 712 | int depth, i; |
a86c6181 | 713 | |
b7ea89ad TT |
714 | if (!path) |
715 | return; | |
716 | depth = path->p_depth; | |
a86c6181 AT |
717 | for (i = 0; i <= depth; i++, path++) |
718 | if (path->p_bh) { | |
719 | brelse(path->p_bh); | |
720 | path->p_bh = NULL; | |
721 | } | |
722 | } | |
723 | ||
724 | /* | |
d0d856e8 RD |
725 | * ext4_ext_binsearch_idx: |
726 | * binary search for the closest index of the given block | |
c29c0ae7 | 727 | * the header must be checked before calling this |
a86c6181 AT |
728 | */ |
729 | static void | |
725d26d3 AK |
730 | ext4_ext_binsearch_idx(struct inode *inode, |
731 | struct ext4_ext_path *path, ext4_lblk_t block) | |
a86c6181 AT |
732 | { |
733 | struct ext4_extent_header *eh = path->p_hdr; | |
734 | struct ext4_extent_idx *r, *l, *m; | |
735 | ||
a86c6181 | 736 | |
bba90743 | 737 | ext_debug("binsearch for %u(idx): ", block); |
a86c6181 AT |
738 | |
739 | l = EXT_FIRST_INDEX(eh) + 1; | |
e9f410b1 | 740 | r = EXT_LAST_INDEX(eh); |
a86c6181 AT |
741 | while (l <= r) { |
742 | m = l + (r - l) / 2; | |
743 | if (block < le32_to_cpu(m->ei_block)) | |
744 | r = m - 1; | |
745 | else | |
746 | l = m + 1; | |
26d535ed DM |
747 | ext_debug("%p(%u):%p(%u):%p(%u) ", l, le32_to_cpu(l->ei_block), |
748 | m, le32_to_cpu(m->ei_block), | |
749 | r, le32_to_cpu(r->ei_block)); | |
a86c6181 AT |
750 | } |
751 | ||
752 | path->p_idx = l - 1; | |
4a3c3a51 | 753 | ext_debug(" -> %u->%lld ", le32_to_cpu(path->p_idx->ei_block), |
bf89d16f | 754 | ext4_idx_pblock(path->p_idx)); |
a86c6181 AT |
755 | |
756 | #ifdef CHECK_BINSEARCH | |
757 | { | |
758 | struct ext4_extent_idx *chix, *ix; | |
759 | int k; | |
760 | ||
761 | chix = ix = EXT_FIRST_INDEX(eh); | |
762 | for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) { | |
763 | if (k != 0 && | |
764 | le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) { | |
4776004f TT |
765 | printk(KERN_DEBUG "k=%d, ix=0x%p, " |
766 | "first=0x%p\n", k, | |
767 | ix, EXT_FIRST_INDEX(eh)); | |
768 | printk(KERN_DEBUG "%u <= %u\n", | |
a86c6181 AT |
769 | le32_to_cpu(ix->ei_block), |
770 | le32_to_cpu(ix[-1].ei_block)); | |
771 | } | |
772 | BUG_ON(k && le32_to_cpu(ix->ei_block) | |
8c55e204 | 773 | <= le32_to_cpu(ix[-1].ei_block)); |
a86c6181 AT |
774 | if (block < le32_to_cpu(ix->ei_block)) |
775 | break; | |
776 | chix = ix; | |
777 | } | |
778 | BUG_ON(chix != path->p_idx); | |
779 | } | |
780 | #endif | |
781 | ||
782 | } | |
783 | ||
784 | /* | |
d0d856e8 RD |
785 | * ext4_ext_binsearch: |
786 | * binary search for closest extent of the given block | |
c29c0ae7 | 787 | * the header must be checked before calling this |
a86c6181 AT |
788 | */ |
789 | static void | |
725d26d3 AK |
790 | ext4_ext_binsearch(struct inode *inode, |
791 | struct ext4_ext_path *path, ext4_lblk_t block) | |
a86c6181 AT |
792 | { |
793 | struct ext4_extent_header *eh = path->p_hdr; | |
794 | struct ext4_extent *r, *l, *m; | |
795 | ||
a86c6181 AT |
796 | if (eh->eh_entries == 0) { |
797 | /* | |
d0d856e8 RD |
798 | * this leaf is empty: |
799 | * we get such a leaf in split/add case | |
a86c6181 AT |
800 | */ |
801 | return; | |
802 | } | |
803 | ||
bba90743 | 804 | ext_debug("binsearch for %u: ", block); |
a86c6181 AT |
805 | |
806 | l = EXT_FIRST_EXTENT(eh) + 1; | |
e9f410b1 | 807 | r = EXT_LAST_EXTENT(eh); |
a86c6181 AT |
808 | |
809 | while (l <= r) { | |
810 | m = l + (r - l) / 2; | |
811 | if (block < le32_to_cpu(m->ee_block)) | |
812 | r = m - 1; | |
813 | else | |
814 | l = m + 1; | |
26d535ed DM |
815 | ext_debug("%p(%u):%p(%u):%p(%u) ", l, le32_to_cpu(l->ee_block), |
816 | m, le32_to_cpu(m->ee_block), | |
817 | r, le32_to_cpu(r->ee_block)); | |
a86c6181 AT |
818 | } |
819 | ||
820 | path->p_ext = l - 1; | |
553f9008 | 821 | ext_debug(" -> %d:%llu:[%d]%d ", |
8c55e204 | 822 | le32_to_cpu(path->p_ext->ee_block), |
bf89d16f | 823 | ext4_ext_pblock(path->p_ext), |
556615dc | 824 | ext4_ext_is_unwritten(path->p_ext), |
a2df2a63 | 825 | ext4_ext_get_actual_len(path->p_ext)); |
a86c6181 AT |
826 | |
827 | #ifdef CHECK_BINSEARCH | |
828 | { | |
829 | struct ext4_extent *chex, *ex; | |
830 | int k; | |
831 | ||
832 | chex = ex = EXT_FIRST_EXTENT(eh); | |
833 | for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ex++) { | |
834 | BUG_ON(k && le32_to_cpu(ex->ee_block) | |
8c55e204 | 835 | <= le32_to_cpu(ex[-1].ee_block)); |
a86c6181 AT |
836 | if (block < le32_to_cpu(ex->ee_block)) |
837 | break; | |
838 | chex = ex; | |
839 | } | |
840 | BUG_ON(chex != path->p_ext); | |
841 | } | |
842 | #endif | |
843 | ||
844 | } | |
845 | ||
846 | int ext4_ext_tree_init(handle_t *handle, struct inode *inode) | |
847 | { | |
848 | struct ext4_extent_header *eh; | |
849 | ||
850 | eh = ext_inode_hdr(inode); | |
851 | eh->eh_depth = 0; | |
852 | eh->eh_entries = 0; | |
853 | eh->eh_magic = EXT4_EXT_MAGIC; | |
55ad63bf | 854 | eh->eh_max = cpu_to_le16(ext4_ext_space_root(inode, 0)); |
a86c6181 | 855 | ext4_mark_inode_dirty(handle, inode); |
a86c6181 AT |
856 | return 0; |
857 | } | |
858 | ||
859 | struct ext4_ext_path * | |
725d26d3 | 860 | ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, |
705912ca | 861 | struct ext4_ext_path **orig_path, int flags) |
a86c6181 AT |
862 | { |
863 | struct ext4_extent_header *eh; | |
864 | struct buffer_head *bh; | |
705912ca TT |
865 | struct ext4_ext_path *path = orig_path ? *orig_path : NULL; |
866 | short int depth, i, ppos = 0; | |
860d21e2 | 867 | int ret; |
a86c6181 AT |
868 | |
869 | eh = ext_inode_hdr(inode); | |
c29c0ae7 | 870 | depth = ext_depth(inode); |
a86c6181 | 871 | |
10809df8 | 872 | if (path) { |
523f431c | 873 | ext4_ext_drop_refs(path); |
10809df8 TT |
874 | if (depth > path[0].p_maxdepth) { |
875 | kfree(path); | |
876 | *orig_path = path = NULL; | |
877 | } | |
878 | } | |
879 | if (!path) { | |
523f431c | 880 | /* account possible depth increase */ |
5d4958f9 | 881 | path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 2), |
a86c6181 | 882 | GFP_NOFS); |
19008f6d | 883 | if (unlikely(!path)) |
a86c6181 | 884 | return ERR_PTR(-ENOMEM); |
10809df8 | 885 | path[0].p_maxdepth = depth + 1; |
a86c6181 | 886 | } |
a86c6181 | 887 | path[0].p_hdr = eh; |
1973adcb | 888 | path[0].p_bh = NULL; |
a86c6181 | 889 | |
c29c0ae7 | 890 | i = depth; |
a86c6181 AT |
891 | /* walk through the tree */ |
892 | while (i) { | |
893 | ext_debug("depth %d: num %d, max %d\n", | |
894 | ppos, le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); | |
c29c0ae7 | 895 | |
a86c6181 | 896 | ext4_ext_binsearch_idx(inode, path + ppos, block); |
bf89d16f | 897 | path[ppos].p_block = ext4_idx_pblock(path[ppos].p_idx); |
a86c6181 AT |
898 | path[ppos].p_depth = i; |
899 | path[ppos].p_ext = NULL; | |
900 | ||
107a7bd3 TT |
901 | bh = read_extent_tree_block(inode, path[ppos].p_block, --i, |
902 | flags); | |
19008f6d | 903 | if (unlikely(IS_ERR(bh))) { |
7d7ea89e | 904 | ret = PTR_ERR(bh); |
a86c6181 | 905 | goto err; |
860d21e2 | 906 | } |
7d7ea89e | 907 | |
a86c6181 AT |
908 | eh = ext_block_hdr(bh); |
909 | ppos++; | |
273df556 FM |
910 | if (unlikely(ppos > depth)) { |
911 | put_bh(bh); | |
912 | EXT4_ERROR_INODE(inode, | |
913 | "ppos %d > depth %d", ppos, depth); | |
860d21e2 | 914 | ret = -EIO; |
273df556 FM |
915 | goto err; |
916 | } | |
a86c6181 AT |
917 | path[ppos].p_bh = bh; |
918 | path[ppos].p_hdr = eh; | |
a86c6181 AT |
919 | } |
920 | ||
921 | path[ppos].p_depth = i; | |
a86c6181 AT |
922 | path[ppos].p_ext = NULL; |
923 | path[ppos].p_idx = NULL; | |
924 | ||
a86c6181 AT |
925 | /* find extent */ |
926 | ext4_ext_binsearch(inode, path + ppos, block); | |
1973adcb SF |
927 | /* if not an empty leaf */ |
928 | if (path[ppos].p_ext) | |
bf89d16f | 929 | path[ppos].p_block = ext4_ext_pblock(path[ppos].p_ext); |
a86c6181 AT |
930 | |
931 | ext4_ext_show_path(inode, path); | |
932 | ||
933 | return path; | |
934 | ||
935 | err: | |
936 | ext4_ext_drop_refs(path); | |
dfe50809 TT |
937 | kfree(path); |
938 | if (orig_path) | |
939 | *orig_path = NULL; | |
860d21e2 | 940 | return ERR_PTR(ret); |
a86c6181 AT |
941 | } |
942 | ||
943 | /* | |
d0d856e8 RD |
944 | * ext4_ext_insert_index: |
945 | * insert new index [@logical;@ptr] into the block at @curp; | |
946 | * check where to insert: before @curp or after @curp | |
a86c6181 | 947 | */ |
1f109d5a TT |
948 | static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, |
949 | struct ext4_ext_path *curp, | |
950 | int logical, ext4_fsblk_t ptr) | |
a86c6181 AT |
951 | { |
952 | struct ext4_extent_idx *ix; | |
953 | int len, err; | |
954 | ||
7e028976 AM |
955 | err = ext4_ext_get_access(handle, inode, curp); |
956 | if (err) | |
a86c6181 AT |
957 | return err; |
958 | ||
273df556 FM |
959 | if (unlikely(logical == le32_to_cpu(curp->p_idx->ei_block))) { |
960 | EXT4_ERROR_INODE(inode, | |
961 | "logical %d == ei_block %d!", | |
962 | logical, le32_to_cpu(curp->p_idx->ei_block)); | |
963 | return -EIO; | |
964 | } | |
d4620315 RD |
965 | |
966 | if (unlikely(le16_to_cpu(curp->p_hdr->eh_entries) | |
967 | >= le16_to_cpu(curp->p_hdr->eh_max))) { | |
968 | EXT4_ERROR_INODE(inode, | |
969 | "eh_entries %d >= eh_max %d!", | |
970 | le16_to_cpu(curp->p_hdr->eh_entries), | |
971 | le16_to_cpu(curp->p_hdr->eh_max)); | |
972 | return -EIO; | |
973 | } | |
974 | ||
a86c6181 AT |
975 | if (logical > le32_to_cpu(curp->p_idx->ei_block)) { |
976 | /* insert after */ | |
80e675f9 | 977 | ext_debug("insert new index %d after: %llu\n", logical, ptr); |
a86c6181 AT |
978 | ix = curp->p_idx + 1; |
979 | } else { | |
980 | /* insert before */ | |
80e675f9 | 981 | ext_debug("insert new index %d before: %llu\n", logical, ptr); |
a86c6181 AT |
982 | ix = curp->p_idx; |
983 | } | |
984 | ||
80e675f9 EG |
985 | len = EXT_LAST_INDEX(curp->p_hdr) - ix + 1; |
986 | BUG_ON(len < 0); | |
987 | if (len > 0) { | |
988 | ext_debug("insert new index %d: " | |
989 | "move %d indices from 0x%p to 0x%p\n", | |
990 | logical, len, ix, ix + 1); | |
991 | memmove(ix + 1, ix, len * sizeof(struct ext4_extent_idx)); | |
992 | } | |
993 | ||
f472e026 TM |
994 | if (unlikely(ix > EXT_MAX_INDEX(curp->p_hdr))) { |
995 | EXT4_ERROR_INODE(inode, "ix > EXT_MAX_INDEX!"); | |
996 | return -EIO; | |
997 | } | |
998 | ||
a86c6181 | 999 | ix->ei_block = cpu_to_le32(logical); |
f65e6fba | 1000 | ext4_idx_store_pblock(ix, ptr); |
e8546d06 | 1001 | le16_add_cpu(&curp->p_hdr->eh_entries, 1); |
a86c6181 | 1002 | |
273df556 FM |
1003 | if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr))) { |
1004 | EXT4_ERROR_INODE(inode, "ix > EXT_LAST_INDEX!"); | |
1005 | return -EIO; | |
1006 | } | |
a86c6181 AT |
1007 | |
1008 | err = ext4_ext_dirty(handle, inode, curp); | |
1009 | ext4_std_error(inode->i_sb, err); | |
1010 | ||
1011 | return err; | |
1012 | } | |
1013 | ||
1014 | /* | |
d0d856e8 RD |
1015 | * ext4_ext_split: |
1016 | * inserts new subtree into the path, using free index entry | |
1017 | * at depth @at: | |
1018 | * - allocates all needed blocks (new leaf and all intermediate index blocks) | |
1019 | * - makes decision where to split | |
1020 | * - moves remaining extents and index entries (right to the split point) | |
1021 | * into the newly allocated blocks | |
1022 | * - initializes subtree | |
a86c6181 AT |
1023 | */ |
1024 | static int ext4_ext_split(handle_t *handle, struct inode *inode, | |
55f020db AH |
1025 | unsigned int flags, |
1026 | struct ext4_ext_path *path, | |
1027 | struct ext4_extent *newext, int at) | |
a86c6181 AT |
1028 | { |
1029 | struct buffer_head *bh = NULL; | |
1030 | int depth = ext_depth(inode); | |
1031 | struct ext4_extent_header *neh; | |
1032 | struct ext4_extent_idx *fidx; | |
a86c6181 | 1033 | int i = at, k, m, a; |
f65e6fba | 1034 | ext4_fsblk_t newblock, oldblock; |
a86c6181 | 1035 | __le32 border; |
f65e6fba | 1036 | ext4_fsblk_t *ablocks = NULL; /* array of allocated blocks */ |
a86c6181 AT |
1037 | int err = 0; |
1038 | ||
1039 | /* make decision: where to split? */ | |
d0d856e8 | 1040 | /* FIXME: now decision is simplest: at current extent */ |
a86c6181 | 1041 | |
d0d856e8 | 1042 | /* if current leaf will be split, then we should use |
a86c6181 | 1043 | * border from split point */ |
273df556 FM |
1044 | if (unlikely(path[depth].p_ext > EXT_MAX_EXTENT(path[depth].p_hdr))) { |
1045 | EXT4_ERROR_INODE(inode, "p_ext > EXT_MAX_EXTENT!"); | |
1046 | return -EIO; | |
1047 | } | |
a86c6181 AT |
1048 | if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) { |
1049 | border = path[depth].p_ext[1].ee_block; | |
d0d856e8 | 1050 | ext_debug("leaf will be split." |
a86c6181 | 1051 | " next leaf starts at %d\n", |
8c55e204 | 1052 | le32_to_cpu(border)); |
a86c6181 AT |
1053 | } else { |
1054 | border = newext->ee_block; | |
1055 | ext_debug("leaf will be added." | |
1056 | " next leaf starts at %d\n", | |
8c55e204 | 1057 | le32_to_cpu(border)); |
a86c6181 AT |
1058 | } |
1059 | ||
1060 | /* | |
d0d856e8 RD |
1061 | * If error occurs, then we break processing |
1062 | * and mark filesystem read-only. index won't | |
a86c6181 | 1063 | * be inserted and tree will be in consistent |
d0d856e8 | 1064 | * state. Next mount will repair buffers too. |
a86c6181 AT |
1065 | */ |
1066 | ||
1067 | /* | |
d0d856e8 RD |
1068 | * Get array to track all allocated blocks. |
1069 | * We need this to handle errors and free blocks | |
1070 | * upon them. | |
a86c6181 | 1071 | */ |
5d4958f9 | 1072 | ablocks = kzalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS); |
a86c6181 AT |
1073 | if (!ablocks) |
1074 | return -ENOMEM; | |
a86c6181 AT |
1075 | |
1076 | /* allocate all needed blocks */ | |
1077 | ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); | |
1078 | for (a = 0; a < depth - at; a++) { | |
654b4908 | 1079 | newblock = ext4_ext_new_meta_block(handle, inode, path, |
55f020db | 1080 | newext, &err, flags); |
a86c6181 AT |
1081 | if (newblock == 0) |
1082 | goto cleanup; | |
1083 | ablocks[a] = newblock; | |
1084 | } | |
1085 | ||
1086 | /* initialize new leaf */ | |
1087 | newblock = ablocks[--a]; | |
273df556 FM |
1088 | if (unlikely(newblock == 0)) { |
1089 | EXT4_ERROR_INODE(inode, "newblock == 0!"); | |
1090 | err = -EIO; | |
1091 | goto cleanup; | |
1092 | } | |
a86c6181 | 1093 | bh = sb_getblk(inode->i_sb, newblock); |
aebf0243 | 1094 | if (unlikely(!bh)) { |
860d21e2 | 1095 | err = -ENOMEM; |
a86c6181 AT |
1096 | goto cleanup; |
1097 | } | |
1098 | lock_buffer(bh); | |
1099 | ||
7e028976 AM |
1100 | err = ext4_journal_get_create_access(handle, bh); |
1101 | if (err) | |
a86c6181 AT |
1102 | goto cleanup; |
1103 | ||
1104 | neh = ext_block_hdr(bh); | |
1105 | neh->eh_entries = 0; | |
55ad63bf | 1106 | neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode, 0)); |
a86c6181 AT |
1107 | neh->eh_magic = EXT4_EXT_MAGIC; |
1108 | neh->eh_depth = 0; | |
a86c6181 | 1109 | |
d0d856e8 | 1110 | /* move remainder of path[depth] to the new leaf */ |
273df556 FM |
1111 | if (unlikely(path[depth].p_hdr->eh_entries != |
1112 | path[depth].p_hdr->eh_max)) { | |
1113 | EXT4_ERROR_INODE(inode, "eh_entries %d != eh_max %d!", | |
1114 | path[depth].p_hdr->eh_entries, | |
1115 | path[depth].p_hdr->eh_max); | |
1116 | err = -EIO; | |
1117 | goto cleanup; | |
1118 | } | |
a86c6181 | 1119 | /* start copy from next extent */ |
1b16da77 YY |
1120 | m = EXT_MAX_EXTENT(path[depth].p_hdr) - path[depth].p_ext++; |
1121 | ext4_ext_show_move(inode, path, newblock, depth); | |
a86c6181 | 1122 | if (m) { |
1b16da77 YY |
1123 | struct ext4_extent *ex; |
1124 | ex = EXT_FIRST_EXTENT(neh); | |
1125 | memmove(ex, path[depth].p_ext, sizeof(struct ext4_extent) * m); | |
e8546d06 | 1126 | le16_add_cpu(&neh->eh_entries, m); |
a86c6181 AT |
1127 | } |
1128 | ||
7ac5990d | 1129 | ext4_extent_block_csum_set(inode, neh); |
a86c6181 AT |
1130 | set_buffer_uptodate(bh); |
1131 | unlock_buffer(bh); | |
1132 | ||
0390131b | 1133 | err = ext4_handle_dirty_metadata(handle, inode, bh); |
7e028976 | 1134 | if (err) |
a86c6181 AT |
1135 | goto cleanup; |
1136 | brelse(bh); | |
1137 | bh = NULL; | |
1138 | ||
1139 | /* correct old leaf */ | |
1140 | if (m) { | |
7e028976 AM |
1141 | err = ext4_ext_get_access(handle, inode, path + depth); |
1142 | if (err) | |
a86c6181 | 1143 | goto cleanup; |
e8546d06 | 1144 | le16_add_cpu(&path[depth].p_hdr->eh_entries, -m); |
7e028976 AM |
1145 | err = ext4_ext_dirty(handle, inode, path + depth); |
1146 | if (err) | |
a86c6181 AT |
1147 | goto cleanup; |
1148 | ||
1149 | } | |
1150 | ||
1151 | /* create intermediate indexes */ | |
1152 | k = depth - at - 1; | |
273df556 FM |
1153 | if (unlikely(k < 0)) { |
1154 | EXT4_ERROR_INODE(inode, "k %d < 0!", k); | |
1155 | err = -EIO; | |
1156 | goto cleanup; | |
1157 | } | |
a86c6181 AT |
1158 | if (k) |
1159 | ext_debug("create %d intermediate indices\n", k); | |
1160 | /* insert new index into current index block */ | |
1161 | /* current depth stored in i var */ | |
1162 | i = depth - 1; | |
1163 | while (k--) { | |
1164 | oldblock = newblock; | |
1165 | newblock = ablocks[--a]; | |
bba90743 | 1166 | bh = sb_getblk(inode->i_sb, newblock); |
aebf0243 | 1167 | if (unlikely(!bh)) { |
860d21e2 | 1168 | err = -ENOMEM; |
a86c6181 AT |
1169 | goto cleanup; |
1170 | } | |
1171 | lock_buffer(bh); | |
1172 | ||
7e028976 AM |
1173 | err = ext4_journal_get_create_access(handle, bh); |
1174 | if (err) | |
a86c6181 AT |
1175 | goto cleanup; |
1176 | ||
1177 | neh = ext_block_hdr(bh); | |
1178 | neh->eh_entries = cpu_to_le16(1); | |
1179 | neh->eh_magic = EXT4_EXT_MAGIC; | |
55ad63bf | 1180 | neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode, 0)); |
a86c6181 AT |
1181 | neh->eh_depth = cpu_to_le16(depth - i); |
1182 | fidx = EXT_FIRST_INDEX(neh); | |
1183 | fidx->ei_block = border; | |
f65e6fba | 1184 | ext4_idx_store_pblock(fidx, oldblock); |
a86c6181 | 1185 | |
bba90743 ES |
1186 | ext_debug("int.index at %d (block %llu): %u -> %llu\n", |
1187 | i, newblock, le32_to_cpu(border), oldblock); | |
a86c6181 | 1188 | |
1b16da77 | 1189 | /* move remainder of path[i] to the new index block */ |
273df556 FM |
1190 | if (unlikely(EXT_MAX_INDEX(path[i].p_hdr) != |
1191 | EXT_LAST_INDEX(path[i].p_hdr))) { | |
1192 | EXT4_ERROR_INODE(inode, | |
1193 | "EXT_MAX_INDEX != EXT_LAST_INDEX ee_block %d!", | |
1194 | le32_to_cpu(path[i].p_ext->ee_block)); | |
1195 | err = -EIO; | |
1196 | goto cleanup; | |
1197 | } | |
1b16da77 YY |
1198 | /* start copy indexes */ |
1199 | m = EXT_MAX_INDEX(path[i].p_hdr) - path[i].p_idx++; | |
1200 | ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx, | |
1201 | EXT_MAX_INDEX(path[i].p_hdr)); | |
1202 | ext4_ext_show_move(inode, path, newblock, i); | |
a86c6181 | 1203 | if (m) { |
1b16da77 | 1204 | memmove(++fidx, path[i].p_idx, |
a86c6181 | 1205 | sizeof(struct ext4_extent_idx) * m); |
e8546d06 | 1206 | le16_add_cpu(&neh->eh_entries, m); |
a86c6181 | 1207 | } |
7ac5990d | 1208 | ext4_extent_block_csum_set(inode, neh); |
a86c6181 AT |
1209 | set_buffer_uptodate(bh); |
1210 | unlock_buffer(bh); | |
1211 | ||
0390131b | 1212 | err = ext4_handle_dirty_metadata(handle, inode, bh); |
7e028976 | 1213 | if (err) |
a86c6181 AT |
1214 | goto cleanup; |
1215 | brelse(bh); | |
1216 | bh = NULL; | |
1217 | ||
1218 | /* correct old index */ | |
1219 | if (m) { | |
1220 | err = ext4_ext_get_access(handle, inode, path + i); | |
1221 | if (err) | |
1222 | goto cleanup; | |
e8546d06 | 1223 | le16_add_cpu(&path[i].p_hdr->eh_entries, -m); |
a86c6181 AT |
1224 | err = ext4_ext_dirty(handle, inode, path + i); |
1225 | if (err) | |
1226 | goto cleanup; | |
1227 | } | |
1228 | ||
1229 | i--; | |
1230 | } | |
1231 | ||
1232 | /* insert new index */ | |
a86c6181 AT |
1233 | err = ext4_ext_insert_index(handle, inode, path + at, |
1234 | le32_to_cpu(border), newblock); | |
1235 | ||
1236 | cleanup: | |
1237 | if (bh) { | |
1238 | if (buffer_locked(bh)) | |
1239 | unlock_buffer(bh); | |
1240 | brelse(bh); | |
1241 | } | |
1242 | ||
1243 | if (err) { | |
1244 | /* free all allocated blocks in error case */ | |
1245 | for (i = 0; i < depth; i++) { | |
1246 | if (!ablocks[i]) | |
1247 | continue; | |
7dc57615 | 1248 | ext4_free_blocks(handle, inode, NULL, ablocks[i], 1, |
e6362609 | 1249 | EXT4_FREE_BLOCKS_METADATA); |
a86c6181 AT |
1250 | } |
1251 | } | |
1252 | kfree(ablocks); | |
1253 | ||
1254 | return err; | |
1255 | } | |
1256 | ||
1257 | /* | |
d0d856e8 RD |
1258 | * ext4_ext_grow_indepth: |
1259 | * implements tree growing procedure: | |
1260 | * - allocates new block | |
1261 | * - moves top-level data (index block or leaf) into the new block | |
1262 | * - initializes new top-level, creating index that points to the | |
1263 | * just created block | |
a86c6181 AT |
1264 | */ |
1265 | static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, | |
55f020db | 1266 | unsigned int flags, |
55f020db | 1267 | struct ext4_extent *newext) |
a86c6181 | 1268 | { |
a86c6181 | 1269 | struct ext4_extent_header *neh; |
a86c6181 | 1270 | struct buffer_head *bh; |
f65e6fba | 1271 | ext4_fsblk_t newblock; |
a86c6181 AT |
1272 | int err = 0; |
1273 | ||
1939dd84 | 1274 | newblock = ext4_ext_new_meta_block(handle, inode, NULL, |
55f020db | 1275 | newext, &err, flags); |
a86c6181 AT |
1276 | if (newblock == 0) |
1277 | return err; | |
1278 | ||
1279 | bh = sb_getblk(inode->i_sb, newblock); | |
aebf0243 | 1280 | if (unlikely(!bh)) |
860d21e2 | 1281 | return -ENOMEM; |
a86c6181 AT |
1282 | lock_buffer(bh); |
1283 | ||
7e028976 AM |
1284 | err = ext4_journal_get_create_access(handle, bh); |
1285 | if (err) { | |
a86c6181 AT |
1286 | unlock_buffer(bh); |
1287 | goto out; | |
1288 | } | |
1289 | ||
1290 | /* move top-level index/leaf into new block */ | |
1939dd84 DM |
1291 | memmove(bh->b_data, EXT4_I(inode)->i_data, |
1292 | sizeof(EXT4_I(inode)->i_data)); | |
a86c6181 AT |
1293 | |
1294 | /* set size of new block */ | |
1295 | neh = ext_block_hdr(bh); | |
1296 | /* old root could have indexes or leaves | |
1297 | * so calculate e_max right way */ | |
1298 | if (ext_depth(inode)) | |
55ad63bf | 1299 | neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode, 0)); |
a86c6181 | 1300 | else |
55ad63bf | 1301 | neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode, 0)); |
a86c6181 | 1302 | neh->eh_magic = EXT4_EXT_MAGIC; |
7ac5990d | 1303 | ext4_extent_block_csum_set(inode, neh); |
a86c6181 AT |
1304 | set_buffer_uptodate(bh); |
1305 | unlock_buffer(bh); | |
1306 | ||
0390131b | 1307 | err = ext4_handle_dirty_metadata(handle, inode, bh); |
7e028976 | 1308 | if (err) |
a86c6181 AT |
1309 | goto out; |
1310 | ||
1939dd84 | 1311 | /* Update top-level index: num,max,pointer */ |
a86c6181 | 1312 | neh = ext_inode_hdr(inode); |
1939dd84 DM |
1313 | neh->eh_entries = cpu_to_le16(1); |
1314 | ext4_idx_store_pblock(EXT_FIRST_INDEX(neh), newblock); | |
1315 | if (neh->eh_depth == 0) { | |
1316 | /* Root extent block becomes index block */ | |
1317 | neh->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode, 0)); | |
1318 | EXT_FIRST_INDEX(neh)->ei_block = | |
1319 | EXT_FIRST_EXTENT(neh)->ee_block; | |
1320 | } | |
2ae02107 | 1321 | ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n", |
a86c6181 | 1322 | le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max), |
5a0790c2 | 1323 | le32_to_cpu(EXT_FIRST_INDEX(neh)->ei_block), |
bf89d16f | 1324 | ext4_idx_pblock(EXT_FIRST_INDEX(neh))); |
a86c6181 | 1325 | |
ba39ebb6 | 1326 | le16_add_cpu(&neh->eh_depth, 1); |
1939dd84 | 1327 | ext4_mark_inode_dirty(handle, inode); |
a86c6181 AT |
1328 | out: |
1329 | brelse(bh); | |
1330 | ||
1331 | return err; | |
1332 | } | |
1333 | ||
1334 | /* | |
d0d856e8 RD |
1335 | * ext4_ext_create_new_leaf: |
1336 | * finds empty index and adds new leaf. | |
1337 | * if no free index is found, then it requests in-depth growing. | |
a86c6181 AT |
1338 | */ |
1339 | static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode, | |
107a7bd3 TT |
1340 | unsigned int mb_flags, |
1341 | unsigned int gb_flags, | |
dfe50809 | 1342 | struct ext4_ext_path **ppath, |
55f020db | 1343 | struct ext4_extent *newext) |
a86c6181 | 1344 | { |
dfe50809 | 1345 | struct ext4_ext_path *path = *ppath; |
a86c6181 AT |
1346 | struct ext4_ext_path *curp; |
1347 | int depth, i, err = 0; | |
1348 | ||
1349 | repeat: | |
1350 | i = depth = ext_depth(inode); | |
1351 | ||
1352 | /* walk up to the tree and look for free index entry */ | |
1353 | curp = path + depth; | |
1354 | while (i > 0 && !EXT_HAS_FREE_INDEX(curp)) { | |
1355 | i--; | |
1356 | curp--; | |
1357 | } | |
1358 | ||
d0d856e8 RD |
1359 | /* we use already allocated block for index block, |
1360 | * so subsequent data blocks should be contiguous */ | |
a86c6181 AT |
1361 | if (EXT_HAS_FREE_INDEX(curp)) { |
1362 | /* if we found index with free entry, then use that | |
1363 | * entry: create all needed subtree and add new leaf */ | |
107a7bd3 | 1364 | err = ext4_ext_split(handle, inode, mb_flags, path, newext, i); |
787e0981 SF |
1365 | if (err) |
1366 | goto out; | |
a86c6181 AT |
1367 | |
1368 | /* refill path */ | |
a86c6181 | 1369 | path = ext4_ext_find_extent(inode, |
725d26d3 | 1370 | (ext4_lblk_t)le32_to_cpu(newext->ee_block), |
dfe50809 | 1371 | ppath, gb_flags); |
a86c6181 AT |
1372 | if (IS_ERR(path)) |
1373 | err = PTR_ERR(path); | |
1374 | } else { | |
1375 | /* tree is full, time to grow in depth */ | |
107a7bd3 | 1376 | err = ext4_ext_grow_indepth(handle, inode, mb_flags, newext); |
a86c6181 AT |
1377 | if (err) |
1378 | goto out; | |
1379 | ||
1380 | /* refill path */ | |
a86c6181 | 1381 | path = ext4_ext_find_extent(inode, |
725d26d3 | 1382 | (ext4_lblk_t)le32_to_cpu(newext->ee_block), |
dfe50809 | 1383 | ppath, gb_flags); |
a86c6181 AT |
1384 | if (IS_ERR(path)) { |
1385 | err = PTR_ERR(path); | |
1386 | goto out; | |
1387 | } | |
1388 | ||
1389 | /* | |
d0d856e8 RD |
1390 | * only first (depth 0 -> 1) produces free space; |
1391 | * in all other cases we have to split the grown tree | |
a86c6181 AT |
1392 | */ |
1393 | depth = ext_depth(inode); | |
1394 | if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) { | |
d0d856e8 | 1395 | /* now we need to split */ |
a86c6181 AT |
1396 | goto repeat; |
1397 | } | |
1398 | } | |
1399 | ||
1400 | out: | |
1401 | return err; | |
1402 | } | |
1403 | ||
1988b51e AT |
1404 | /* |
1405 | * search the closest allocated block to the left for *logical | |
1406 | * and returns it at @logical + it's physical address at @phys | |
1407 | * if *logical is the smallest allocated block, the function | |
1408 | * returns 0 at @phys | |
1409 | * return value contains 0 (success) or error code | |
1410 | */ | |
1f109d5a TT |
1411 | static int ext4_ext_search_left(struct inode *inode, |
1412 | struct ext4_ext_path *path, | |
1413 | ext4_lblk_t *logical, ext4_fsblk_t *phys) | |
1988b51e AT |
1414 | { |
1415 | struct ext4_extent_idx *ix; | |
1416 | struct ext4_extent *ex; | |
b939e376 | 1417 | int depth, ee_len; |
1988b51e | 1418 | |
273df556 FM |
1419 | if (unlikely(path == NULL)) { |
1420 | EXT4_ERROR_INODE(inode, "path == NULL *logical %d!", *logical); | |
1421 | return -EIO; | |
1422 | } | |
1988b51e AT |
1423 | depth = path->p_depth; |
1424 | *phys = 0; | |
1425 | ||
1426 | if (depth == 0 && path->p_ext == NULL) | |
1427 | return 0; | |
1428 | ||
1429 | /* usually extent in the path covers blocks smaller | |
1430 | * then *logical, but it can be that extent is the | |
1431 | * first one in the file */ | |
1432 | ||
1433 | ex = path[depth].p_ext; | |
b939e376 | 1434 | ee_len = ext4_ext_get_actual_len(ex); |
1988b51e | 1435 | if (*logical < le32_to_cpu(ex->ee_block)) { |
273df556 FM |
1436 | if (unlikely(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex)) { |
1437 | EXT4_ERROR_INODE(inode, | |
1438 | "EXT_FIRST_EXTENT != ex *logical %d ee_block %d!", | |
1439 | *logical, le32_to_cpu(ex->ee_block)); | |
1440 | return -EIO; | |
1441 | } | |
1988b51e AT |
1442 | while (--depth >= 0) { |
1443 | ix = path[depth].p_idx; | |
273df556 FM |
1444 | if (unlikely(ix != EXT_FIRST_INDEX(path[depth].p_hdr))) { |
1445 | EXT4_ERROR_INODE(inode, | |
1446 | "ix (%d) != EXT_FIRST_INDEX (%d) (depth %d)!", | |
6ee3b212 | 1447 | ix != NULL ? le32_to_cpu(ix->ei_block) : 0, |
273df556 | 1448 | EXT_FIRST_INDEX(path[depth].p_hdr) != NULL ? |
6ee3b212 | 1449 | le32_to_cpu(EXT_FIRST_INDEX(path[depth].p_hdr)->ei_block) : 0, |
273df556 FM |
1450 | depth); |
1451 | return -EIO; | |
1452 | } | |
1988b51e AT |
1453 | } |
1454 | return 0; | |
1455 | } | |
1456 | ||
273df556 FM |
1457 | if (unlikely(*logical < (le32_to_cpu(ex->ee_block) + ee_len))) { |
1458 | EXT4_ERROR_INODE(inode, | |
1459 | "logical %d < ee_block %d + ee_len %d!", | |
1460 | *logical, le32_to_cpu(ex->ee_block), ee_len); | |
1461 | return -EIO; | |
1462 | } | |
1988b51e | 1463 | |
b939e376 | 1464 | *logical = le32_to_cpu(ex->ee_block) + ee_len - 1; |
bf89d16f | 1465 | *phys = ext4_ext_pblock(ex) + ee_len - 1; |
1988b51e AT |
1466 | return 0; |
1467 | } | |
1468 | ||
1469 | /* | |
1470 | * search the closest allocated block to the right for *logical | |
1471 | * and returns it at @logical + it's physical address at @phys | |
df3ab170 | 1472 | * if *logical is the largest allocated block, the function |
1988b51e AT |
1473 | * returns 0 at @phys |
1474 | * return value contains 0 (success) or error code | |
1475 | */ | |
1f109d5a TT |
1476 | static int ext4_ext_search_right(struct inode *inode, |
1477 | struct ext4_ext_path *path, | |
4d33b1ef TT |
1478 | ext4_lblk_t *logical, ext4_fsblk_t *phys, |
1479 | struct ext4_extent **ret_ex) | |
1988b51e AT |
1480 | { |
1481 | struct buffer_head *bh = NULL; | |
1482 | struct ext4_extent_header *eh; | |
1483 | struct ext4_extent_idx *ix; | |
1484 | struct ext4_extent *ex; | |
1485 | ext4_fsblk_t block; | |
395a87bf ES |
1486 | int depth; /* Note, NOT eh_depth; depth from top of tree */ |
1487 | int ee_len; | |
1988b51e | 1488 | |
273df556 FM |
1489 | if (unlikely(path == NULL)) { |
1490 | EXT4_ERROR_INODE(inode, "path == NULL *logical %d!", *logical); | |
1491 | return -EIO; | |
1492 | } | |
1988b51e AT |
1493 | depth = path->p_depth; |
1494 | *phys = 0; | |
1495 | ||
1496 | if (depth == 0 && path->p_ext == NULL) | |
1497 | return 0; | |
1498 | ||
1499 | /* usually extent in the path covers blocks smaller | |
1500 | * then *logical, but it can be that extent is the | |
1501 | * first one in the file */ | |
1502 | ||
1503 | ex = path[depth].p_ext; | |
b939e376 | 1504 | ee_len = ext4_ext_get_actual_len(ex); |
1988b51e | 1505 | if (*logical < le32_to_cpu(ex->ee_block)) { |
273df556 FM |
1506 | if (unlikely(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex)) { |
1507 | EXT4_ERROR_INODE(inode, | |
1508 | "first_extent(path[%d].p_hdr) != ex", | |
1509 | depth); | |
1510 | return -EIO; | |
1511 | } | |
1988b51e AT |
1512 | while (--depth >= 0) { |
1513 | ix = path[depth].p_idx; | |
273df556 FM |
1514 | if (unlikely(ix != EXT_FIRST_INDEX(path[depth].p_hdr))) { |
1515 | EXT4_ERROR_INODE(inode, | |
1516 | "ix != EXT_FIRST_INDEX *logical %d!", | |
1517 | *logical); | |
1518 | return -EIO; | |
1519 | } | |
1988b51e | 1520 | } |
4d33b1ef | 1521 | goto found_extent; |
1988b51e AT |
1522 | } |
1523 | ||
273df556 FM |
1524 | if (unlikely(*logical < (le32_to_cpu(ex->ee_block) + ee_len))) { |
1525 | EXT4_ERROR_INODE(inode, | |
1526 | "logical %d < ee_block %d + ee_len %d!", | |
1527 | *logical, le32_to_cpu(ex->ee_block), ee_len); | |
1528 | return -EIO; | |
1529 | } | |
1988b51e AT |
1530 | |
1531 | if (ex != EXT_LAST_EXTENT(path[depth].p_hdr)) { | |
1532 | /* next allocated block in this leaf */ | |
1533 | ex++; | |
4d33b1ef | 1534 | goto found_extent; |
1988b51e AT |
1535 | } |
1536 | ||
1537 | /* go up and search for index to the right */ | |
1538 | while (--depth >= 0) { | |
1539 | ix = path[depth].p_idx; | |
1540 | if (ix != EXT_LAST_INDEX(path[depth].p_hdr)) | |
25f1ee3a | 1541 | goto got_index; |
1988b51e AT |
1542 | } |
1543 | ||
25f1ee3a WF |
1544 | /* we've gone up to the root and found no index to the right */ |
1545 | return 0; | |
1988b51e | 1546 | |
25f1ee3a | 1547 | got_index: |
1988b51e AT |
1548 | /* we've found index to the right, let's |
1549 | * follow it and find the closest allocated | |
1550 | * block to the right */ | |
1551 | ix++; | |
bf89d16f | 1552 | block = ext4_idx_pblock(ix); |
1988b51e | 1553 | while (++depth < path->p_depth) { |
395a87bf | 1554 | /* subtract from p_depth to get proper eh_depth */ |
7d7ea89e | 1555 | bh = read_extent_tree_block(inode, block, |
107a7bd3 | 1556 | path->p_depth - depth, 0); |
7d7ea89e TT |
1557 | if (IS_ERR(bh)) |
1558 | return PTR_ERR(bh); | |
1559 | eh = ext_block_hdr(bh); | |
1988b51e | 1560 | ix = EXT_FIRST_INDEX(eh); |
bf89d16f | 1561 | block = ext4_idx_pblock(ix); |
1988b51e AT |
1562 | put_bh(bh); |
1563 | } | |
1564 | ||
107a7bd3 | 1565 | bh = read_extent_tree_block(inode, block, path->p_depth - depth, 0); |
7d7ea89e TT |
1566 | if (IS_ERR(bh)) |
1567 | return PTR_ERR(bh); | |
1988b51e | 1568 | eh = ext_block_hdr(bh); |
1988b51e | 1569 | ex = EXT_FIRST_EXTENT(eh); |
4d33b1ef | 1570 | found_extent: |
1988b51e | 1571 | *logical = le32_to_cpu(ex->ee_block); |
bf89d16f | 1572 | *phys = ext4_ext_pblock(ex); |
4d33b1ef TT |
1573 | *ret_ex = ex; |
1574 | if (bh) | |
1575 | put_bh(bh); | |
1988b51e | 1576 | return 0; |
1988b51e AT |
1577 | } |
1578 | ||
a86c6181 | 1579 | /* |
d0d856e8 | 1580 | * ext4_ext_next_allocated_block: |
f17722f9 | 1581 | * returns allocated block in subsequent extent or EXT_MAX_BLOCKS. |
d0d856e8 RD |
1582 | * NOTE: it considers block number from index entry as |
1583 | * allocated block. Thus, index entries have to be consistent | |
1584 | * with leaves. | |
a86c6181 | 1585 | */ |
fcf6b1b7 | 1586 | ext4_lblk_t |
a86c6181 AT |
1587 | ext4_ext_next_allocated_block(struct ext4_ext_path *path) |
1588 | { | |
1589 | int depth; | |
1590 | ||
1591 | BUG_ON(path == NULL); | |
1592 | depth = path->p_depth; | |
1593 | ||
1594 | if (depth == 0 && path->p_ext == NULL) | |
f17722f9 | 1595 | return EXT_MAX_BLOCKS; |
a86c6181 AT |
1596 | |
1597 | while (depth >= 0) { | |
1598 | if (depth == path->p_depth) { | |
1599 | /* leaf */ | |
6f8ff537 CW |
1600 | if (path[depth].p_ext && |
1601 | path[depth].p_ext != | |
a86c6181 AT |
1602 | EXT_LAST_EXTENT(path[depth].p_hdr)) |
1603 | return le32_to_cpu(path[depth].p_ext[1].ee_block); | |
1604 | } else { | |
1605 | /* index */ | |
1606 | if (path[depth].p_idx != | |
1607 | EXT_LAST_INDEX(path[depth].p_hdr)) | |
1608 | return le32_to_cpu(path[depth].p_idx[1].ei_block); | |
1609 | } | |
1610 | depth--; | |
1611 | } | |
1612 | ||
f17722f9 | 1613 | return EXT_MAX_BLOCKS; |
a86c6181 AT |
1614 | } |
1615 | ||
1616 | /* | |
d0d856e8 | 1617 | * ext4_ext_next_leaf_block: |
f17722f9 | 1618 | * returns first allocated block from next leaf or EXT_MAX_BLOCKS |
a86c6181 | 1619 | */ |
5718789d | 1620 | static ext4_lblk_t ext4_ext_next_leaf_block(struct ext4_ext_path *path) |
a86c6181 AT |
1621 | { |
1622 | int depth; | |
1623 | ||
1624 | BUG_ON(path == NULL); | |
1625 | depth = path->p_depth; | |
1626 | ||
1627 | /* zero-tree has no leaf blocks at all */ | |
1628 | if (depth == 0) | |
f17722f9 | 1629 | return EXT_MAX_BLOCKS; |
a86c6181 AT |
1630 | |
1631 | /* go to index block */ | |
1632 | depth--; | |
1633 | ||
1634 | while (depth >= 0) { | |
1635 | if (path[depth].p_idx != | |
1636 | EXT_LAST_INDEX(path[depth].p_hdr)) | |
725d26d3 AK |
1637 | return (ext4_lblk_t) |
1638 | le32_to_cpu(path[depth].p_idx[1].ei_block); | |
a86c6181 AT |
1639 | depth--; |
1640 | } | |
1641 | ||
f17722f9 | 1642 | return EXT_MAX_BLOCKS; |
a86c6181 AT |
1643 | } |
1644 | ||
1645 | /* | |
d0d856e8 RD |
1646 | * ext4_ext_correct_indexes: |
1647 | * if leaf gets modified and modified extent is first in the leaf, | |
1648 | * then we have to correct all indexes above. | |
a86c6181 AT |
1649 | * TODO: do we need to correct tree in all cases? |
1650 | */ | |
1d03ec98 | 1651 | static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, |
a86c6181 AT |
1652 | struct ext4_ext_path *path) |
1653 | { | |
1654 | struct ext4_extent_header *eh; | |
1655 | int depth = ext_depth(inode); | |
1656 | struct ext4_extent *ex; | |
1657 | __le32 border; | |
1658 | int k, err = 0; | |
1659 | ||
1660 | eh = path[depth].p_hdr; | |
1661 | ex = path[depth].p_ext; | |
273df556 FM |
1662 | |
1663 | if (unlikely(ex == NULL || eh == NULL)) { | |
1664 | EXT4_ERROR_INODE(inode, | |
1665 | "ex %p == NULL or eh %p == NULL", ex, eh); | |
1666 | return -EIO; | |
1667 | } | |
a86c6181 AT |
1668 | |
1669 | if (depth == 0) { | |
1670 | /* there is no tree at all */ | |
1671 | return 0; | |
1672 | } | |
1673 | ||
1674 | if (ex != EXT_FIRST_EXTENT(eh)) { | |
1675 | /* we correct tree if first leaf got modified only */ | |
1676 | return 0; | |
1677 | } | |
1678 | ||
1679 | /* | |
d0d856e8 | 1680 | * TODO: we need correction if border is smaller than current one |
a86c6181 AT |
1681 | */ |
1682 | k = depth - 1; | |
1683 | border = path[depth].p_ext->ee_block; | |
7e028976 AM |
1684 | err = ext4_ext_get_access(handle, inode, path + k); |
1685 | if (err) | |
a86c6181 AT |
1686 | return err; |
1687 | path[k].p_idx->ei_block = border; | |
7e028976 AM |
1688 | err = ext4_ext_dirty(handle, inode, path + k); |
1689 | if (err) | |
a86c6181 AT |
1690 | return err; |
1691 | ||
1692 | while (k--) { | |
1693 | /* change all left-side indexes */ | |
1694 | if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr)) | |
1695 | break; | |
7e028976 AM |
1696 | err = ext4_ext_get_access(handle, inode, path + k); |
1697 | if (err) | |
a86c6181 AT |
1698 | break; |
1699 | path[k].p_idx->ei_block = border; | |
7e028976 AM |
1700 | err = ext4_ext_dirty(handle, inode, path + k); |
1701 | if (err) | |
a86c6181 AT |
1702 | break; |
1703 | } | |
1704 | ||
1705 | return err; | |
1706 | } | |
1707 | ||
748de673 | 1708 | int |
a86c6181 AT |
1709 | ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, |
1710 | struct ext4_extent *ex2) | |
1711 | { | |
da0169b3 | 1712 | unsigned short ext1_ee_len, ext2_ee_len; |
a2df2a63 AA |
1713 | |
1714 | /* | |
ec22ba8e | 1715 | * Make sure that both extents are initialized. We don't merge |
556615dc | 1716 | * unwritten extents so that we can be sure that end_io code has |
ec22ba8e DM |
1717 | * the extent that was written properly split out and conversion to |
1718 | * initialized is trivial. | |
a2df2a63 | 1719 | */ |
556615dc | 1720 | if (ext4_ext_is_unwritten(ex1) != ext4_ext_is_unwritten(ex2)) |
a2df2a63 AA |
1721 | return 0; |
1722 | ||
1723 | ext1_ee_len = ext4_ext_get_actual_len(ex1); | |
1724 | ext2_ee_len = ext4_ext_get_actual_len(ex2); | |
1725 | ||
1726 | if (le32_to_cpu(ex1->ee_block) + ext1_ee_len != | |
63f57933 | 1727 | le32_to_cpu(ex2->ee_block)) |
a86c6181 AT |
1728 | return 0; |
1729 | ||
471d4011 SB |
1730 | /* |
1731 | * To allow future support for preallocated extents to be added | |
1732 | * as an RO_COMPAT feature, refuse to merge to extents if | |
d0d856e8 | 1733 | * this can result in the top bit of ee_len being set. |
471d4011 | 1734 | */ |
da0169b3 | 1735 | if (ext1_ee_len + ext2_ee_len > EXT_INIT_MAX_LEN) |
471d4011 | 1736 | return 0; |
556615dc | 1737 | if (ext4_ext_is_unwritten(ex1) && |
a9b82415 DW |
1738 | (ext4_test_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN) || |
1739 | atomic_read(&EXT4_I(inode)->i_unwritten) || | |
556615dc | 1740 | (ext1_ee_len + ext2_ee_len > EXT_UNWRITTEN_MAX_LEN))) |
a9b82415 | 1741 | return 0; |
bbf2f9fb | 1742 | #ifdef AGGRESSIVE_TEST |
b939e376 | 1743 | if (ext1_ee_len >= 4) |
a86c6181 AT |
1744 | return 0; |
1745 | #endif | |
1746 | ||
bf89d16f | 1747 | if (ext4_ext_pblock(ex1) + ext1_ee_len == ext4_ext_pblock(ex2)) |
a86c6181 AT |
1748 | return 1; |
1749 | return 0; | |
1750 | } | |
1751 | ||
56055d3a AA |
1752 | /* |
1753 | * This function tries to merge the "ex" extent to the next extent in the tree. | |
1754 | * It always tries to merge towards right. If you want to merge towards | |
1755 | * left, pass "ex - 1" as argument instead of "ex". | |
1756 | * Returns 0 if the extents (ex and ex+1) were _not_ merged and returns | |
1757 | * 1 if they got merged. | |
1758 | */ | |
197217a5 | 1759 | static int ext4_ext_try_to_merge_right(struct inode *inode, |
1f109d5a TT |
1760 | struct ext4_ext_path *path, |
1761 | struct ext4_extent *ex) | |
56055d3a AA |
1762 | { |
1763 | struct ext4_extent_header *eh; | |
1764 | unsigned int depth, len; | |
556615dc | 1765 | int merge_done = 0, unwritten; |
56055d3a AA |
1766 | |
1767 | depth = ext_depth(inode); | |
1768 | BUG_ON(path[depth].p_hdr == NULL); | |
1769 | eh = path[depth].p_hdr; | |
1770 | ||
1771 | while (ex < EXT_LAST_EXTENT(eh)) { | |
1772 | if (!ext4_can_extents_be_merged(inode, ex, ex + 1)) | |
1773 | break; | |
1774 | /* merge with next extent! */ | |
556615dc | 1775 | unwritten = ext4_ext_is_unwritten(ex); |
56055d3a AA |
1776 | ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex) |
1777 | + ext4_ext_get_actual_len(ex + 1)); | |
556615dc LC |
1778 | if (unwritten) |
1779 | ext4_ext_mark_unwritten(ex); | |
56055d3a AA |
1780 | |
1781 | if (ex + 1 < EXT_LAST_EXTENT(eh)) { | |
1782 | len = (EXT_LAST_EXTENT(eh) - ex - 1) | |
1783 | * sizeof(struct ext4_extent); | |
1784 | memmove(ex + 1, ex + 2, len); | |
1785 | } | |
e8546d06 | 1786 | le16_add_cpu(&eh->eh_entries, -1); |
56055d3a AA |
1787 | merge_done = 1; |
1788 | WARN_ON(eh->eh_entries == 0); | |
1789 | if (!eh->eh_entries) | |
24676da4 | 1790 | EXT4_ERROR_INODE(inode, "eh->eh_entries = 0!"); |
56055d3a AA |
1791 | } |
1792 | ||
1793 | return merge_done; | |
1794 | } | |
1795 | ||
ecb94f5f TT |
1796 | /* |
1797 | * This function does a very simple check to see if we can collapse | |
1798 | * an extent tree with a single extent tree leaf block into the inode. | |
1799 | */ | |
1800 | static void ext4_ext_try_to_merge_up(handle_t *handle, | |
1801 | struct inode *inode, | |
1802 | struct ext4_ext_path *path) | |
1803 | { | |
1804 | size_t s; | |
1805 | unsigned max_root = ext4_ext_space_root(inode, 0); | |
1806 | ext4_fsblk_t blk; | |
1807 | ||
1808 | if ((path[0].p_depth != 1) || | |
1809 | (le16_to_cpu(path[0].p_hdr->eh_entries) != 1) || | |
1810 | (le16_to_cpu(path[1].p_hdr->eh_entries) > max_root)) | |
1811 | return; | |
1812 | ||
1813 | /* | |
1814 | * We need to modify the block allocation bitmap and the block | |
1815 | * group descriptor to release the extent tree block. If we | |
1816 | * can't get the journal credits, give up. | |
1817 | */ | |
1818 | if (ext4_journal_extend(handle, 2)) | |
1819 | return; | |
1820 | ||
1821 | /* | |
1822 | * Copy the extent data up to the inode | |
1823 | */ | |
1824 | blk = ext4_idx_pblock(path[0].p_idx); | |
1825 | s = le16_to_cpu(path[1].p_hdr->eh_entries) * | |
1826 | sizeof(struct ext4_extent_idx); | |
1827 | s += sizeof(struct ext4_extent_header); | |
1828 | ||
10809df8 | 1829 | path[1].p_maxdepth = path[0].p_maxdepth; |
ecb94f5f TT |
1830 | memcpy(path[0].p_hdr, path[1].p_hdr, s); |
1831 | path[0].p_depth = 0; | |
1832 | path[0].p_ext = EXT_FIRST_EXTENT(path[0].p_hdr) + | |
1833 | (path[1].p_ext - EXT_FIRST_EXTENT(path[1].p_hdr)); | |
1834 | path[0].p_hdr->eh_max = cpu_to_le16(max_root); | |
1835 | ||
1836 | brelse(path[1].p_bh); | |
1837 | ext4_free_blocks(handle, inode, NULL, blk, 1, | |
71d4f7d0 | 1838 | EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET); |
ecb94f5f TT |
1839 | } |
1840 | ||
197217a5 YY |
1841 | /* |
1842 | * This function tries to merge the @ex extent to neighbours in the tree. | |
1843 | * return 1 if merge left else 0. | |
1844 | */ | |
ecb94f5f TT |
1845 | static void ext4_ext_try_to_merge(handle_t *handle, |
1846 | struct inode *inode, | |
197217a5 YY |
1847 | struct ext4_ext_path *path, |
1848 | struct ext4_extent *ex) { | |
1849 | struct ext4_extent_header *eh; | |
1850 | unsigned int depth; | |
1851 | int merge_done = 0; | |
197217a5 YY |
1852 | |
1853 | depth = ext_depth(inode); | |
1854 | BUG_ON(path[depth].p_hdr == NULL); | |
1855 | eh = path[depth].p_hdr; | |
1856 | ||
1857 | if (ex > EXT_FIRST_EXTENT(eh)) | |
1858 | merge_done = ext4_ext_try_to_merge_right(inode, path, ex - 1); | |
1859 | ||
1860 | if (!merge_done) | |
ecb94f5f | 1861 | (void) ext4_ext_try_to_merge_right(inode, path, ex); |
197217a5 | 1862 | |
ecb94f5f | 1863 | ext4_ext_try_to_merge_up(handle, inode, path); |
197217a5 YY |
1864 | } |
1865 | ||
25d14f98 AA |
1866 | /* |
1867 | * check if a portion of the "newext" extent overlaps with an | |
1868 | * existing extent. | |
1869 | * | |
1870 | * If there is an overlap discovered, it updates the length of the newext | |
1871 | * such that there will be no overlap, and then returns 1. | |
1872 | * If there is no overlap found, it returns 0. | |
1873 | */ | |
4d33b1ef TT |
1874 | static unsigned int ext4_ext_check_overlap(struct ext4_sb_info *sbi, |
1875 | struct inode *inode, | |
1f109d5a TT |
1876 | struct ext4_extent *newext, |
1877 | struct ext4_ext_path *path) | |
25d14f98 | 1878 | { |
725d26d3 | 1879 | ext4_lblk_t b1, b2; |
25d14f98 AA |
1880 | unsigned int depth, len1; |
1881 | unsigned int ret = 0; | |
1882 | ||
1883 | b1 = le32_to_cpu(newext->ee_block); | |
a2df2a63 | 1884 | len1 = ext4_ext_get_actual_len(newext); |
25d14f98 AA |
1885 | depth = ext_depth(inode); |
1886 | if (!path[depth].p_ext) | |
1887 | goto out; | |
f5a44db5 | 1888 | b2 = EXT4_LBLK_CMASK(sbi, le32_to_cpu(path[depth].p_ext->ee_block)); |
25d14f98 AA |
1889 | |
1890 | /* | |
1891 | * get the next allocated block if the extent in the path | |
2b2d6d01 | 1892 | * is before the requested block(s) |
25d14f98 AA |
1893 | */ |
1894 | if (b2 < b1) { | |
1895 | b2 = ext4_ext_next_allocated_block(path); | |
f17722f9 | 1896 | if (b2 == EXT_MAX_BLOCKS) |
25d14f98 | 1897 | goto out; |
f5a44db5 | 1898 | b2 = EXT4_LBLK_CMASK(sbi, b2); |
25d14f98 AA |
1899 | } |
1900 | ||
725d26d3 | 1901 | /* check for wrap through zero on extent logical start block*/ |
25d14f98 | 1902 | if (b1 + len1 < b1) { |
f17722f9 | 1903 | len1 = EXT_MAX_BLOCKS - b1; |
25d14f98 AA |
1904 | newext->ee_len = cpu_to_le16(len1); |
1905 | ret = 1; | |
1906 | } | |
1907 | ||
1908 | /* check for overlap */ | |
1909 | if (b1 + len1 > b2) { | |
1910 | newext->ee_len = cpu_to_le16(b2 - b1); | |
1911 | ret = 1; | |
1912 | } | |
1913 | out: | |
1914 | return ret; | |
1915 | } | |
1916 | ||
a86c6181 | 1917 | /* |
d0d856e8 RD |
1918 | * ext4_ext_insert_extent: |
1919 | * tries to merge requsted extent into the existing extent or | |
1920 | * inserts requested extent as new one into the tree, | |
1921 | * creating new leaf in the no-space case. | |
a86c6181 AT |
1922 | */ |
1923 | int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, | |
dfe50809 | 1924 | struct ext4_ext_path **ppath, |
107a7bd3 | 1925 | struct ext4_extent *newext, int gb_flags) |
a86c6181 | 1926 | { |
dfe50809 | 1927 | struct ext4_ext_path *path = *ppath; |
af5bc92d | 1928 | struct ext4_extent_header *eh; |
a86c6181 AT |
1929 | struct ext4_extent *ex, *fex; |
1930 | struct ext4_extent *nearex; /* nearest extent */ | |
1931 | struct ext4_ext_path *npath = NULL; | |
725d26d3 AK |
1932 | int depth, len, err; |
1933 | ext4_lblk_t next; | |
556615dc | 1934 | int mb_flags = 0, unwritten; |
a86c6181 | 1935 | |
273df556 FM |
1936 | if (unlikely(ext4_ext_get_actual_len(newext) == 0)) { |
1937 | EXT4_ERROR_INODE(inode, "ext4_ext_get_actual_len(newext) == 0"); | |
1938 | return -EIO; | |
1939 | } | |
a86c6181 AT |
1940 | depth = ext_depth(inode); |
1941 | ex = path[depth].p_ext; | |
be8981be | 1942 | eh = path[depth].p_hdr; |
273df556 FM |
1943 | if (unlikely(path[depth].p_hdr == NULL)) { |
1944 | EXT4_ERROR_INODE(inode, "path[%d].p_hdr == NULL", depth); | |
1945 | return -EIO; | |
1946 | } | |
a86c6181 AT |
1947 | |
1948 | /* try to insert block into found extent and return */ | |
107a7bd3 | 1949 | if (ex && !(gb_flags & EXT4_GET_BLOCKS_PRE_IO)) { |
a2df2a63 AA |
1950 | |
1951 | /* | |
be8981be LC |
1952 | * Try to see whether we should rather test the extent on |
1953 | * right from ex, or from the left of ex. This is because | |
1954 | * ext4_ext_find_extent() can return either extent on the | |
1955 | * left, or on the right from the searched position. This | |
1956 | * will make merging more effective. | |
a2df2a63 | 1957 | */ |
be8981be LC |
1958 | if (ex < EXT_LAST_EXTENT(eh) && |
1959 | (le32_to_cpu(ex->ee_block) + | |
1960 | ext4_ext_get_actual_len(ex) < | |
1961 | le32_to_cpu(newext->ee_block))) { | |
1962 | ex += 1; | |
1963 | goto prepend; | |
1964 | } else if ((ex > EXT_FIRST_EXTENT(eh)) && | |
1965 | (le32_to_cpu(newext->ee_block) + | |
1966 | ext4_ext_get_actual_len(newext) < | |
1967 | le32_to_cpu(ex->ee_block))) | |
1968 | ex -= 1; | |
1969 | ||
1970 | /* Try to append newex to the ex */ | |
1971 | if (ext4_can_extents_be_merged(inode, ex, newext)) { | |
1972 | ext_debug("append [%d]%d block to %u:[%d]%d" | |
1973 | "(from %llu)\n", | |
556615dc | 1974 | ext4_ext_is_unwritten(newext), |
be8981be LC |
1975 | ext4_ext_get_actual_len(newext), |
1976 | le32_to_cpu(ex->ee_block), | |
556615dc | 1977 | ext4_ext_is_unwritten(ex), |
be8981be LC |
1978 | ext4_ext_get_actual_len(ex), |
1979 | ext4_ext_pblock(ex)); | |
1980 | err = ext4_ext_get_access(handle, inode, | |
1981 | path + depth); | |
1982 | if (err) | |
1983 | return err; | |
556615dc | 1984 | unwritten = ext4_ext_is_unwritten(ex); |
be8981be | 1985 | ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex) |
a2df2a63 | 1986 | + ext4_ext_get_actual_len(newext)); |
556615dc LC |
1987 | if (unwritten) |
1988 | ext4_ext_mark_unwritten(ex); | |
be8981be LC |
1989 | eh = path[depth].p_hdr; |
1990 | nearex = ex; | |
1991 | goto merge; | |
1992 | } | |
1993 | ||
1994 | prepend: | |
1995 | /* Try to prepend newex to the ex */ | |
1996 | if (ext4_can_extents_be_merged(inode, newext, ex)) { | |
1997 | ext_debug("prepend %u[%d]%d block to %u:[%d]%d" | |
1998 | "(from %llu)\n", | |
1999 | le32_to_cpu(newext->ee_block), | |
556615dc | 2000 | ext4_ext_is_unwritten(newext), |
be8981be LC |
2001 | ext4_ext_get_actual_len(newext), |
2002 | le32_to_cpu(ex->ee_block), | |
556615dc | 2003 | ext4_ext_is_unwritten(ex), |
be8981be LC |
2004 | ext4_ext_get_actual_len(ex), |
2005 | ext4_ext_pblock(ex)); | |
2006 | err = ext4_ext_get_access(handle, inode, | |
2007 | path + depth); | |
2008 | if (err) | |
2009 | return err; | |
2010 | ||
556615dc | 2011 | unwritten = ext4_ext_is_unwritten(ex); |
be8981be LC |
2012 | ex->ee_block = newext->ee_block; |
2013 | ext4_ext_store_pblock(ex, ext4_ext_pblock(newext)); | |
2014 | ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex) | |
2015 | + ext4_ext_get_actual_len(newext)); | |
556615dc LC |
2016 | if (unwritten) |
2017 | ext4_ext_mark_unwritten(ex); | |
be8981be LC |
2018 | eh = path[depth].p_hdr; |
2019 | nearex = ex; | |
2020 | goto merge; | |
2021 | } | |
a86c6181 AT |
2022 | } |
2023 | ||
a86c6181 AT |
2024 | depth = ext_depth(inode); |
2025 | eh = path[depth].p_hdr; | |
2026 | if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) | |
2027 | goto has_space; | |
2028 | ||
2029 | /* probably next leaf has space for us? */ | |
2030 | fex = EXT_LAST_EXTENT(eh); | |
598dbdf2 RD |
2031 | next = EXT_MAX_BLOCKS; |
2032 | if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block)) | |
5718789d | 2033 | next = ext4_ext_next_leaf_block(path); |
598dbdf2 | 2034 | if (next != EXT_MAX_BLOCKS) { |
32de6756 | 2035 | ext_debug("next leaf block - %u\n", next); |
a86c6181 | 2036 | BUG_ON(npath != NULL); |
107a7bd3 | 2037 | npath = ext4_ext_find_extent(inode, next, NULL, 0); |
a86c6181 AT |
2038 | if (IS_ERR(npath)) |
2039 | return PTR_ERR(npath); | |
2040 | BUG_ON(npath->p_depth != path->p_depth); | |
2041 | eh = npath[depth].p_hdr; | |
2042 | if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) { | |
25985edc | 2043 | ext_debug("next leaf isn't full(%d)\n", |
a86c6181 AT |
2044 | le16_to_cpu(eh->eh_entries)); |
2045 | path = npath; | |
ffb505ff | 2046 | goto has_space; |
a86c6181 AT |
2047 | } |
2048 | ext_debug("next leaf has no free space(%d,%d)\n", | |
2049 | le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); | |
2050 | } | |
2051 | ||
2052 | /* | |
d0d856e8 RD |
2053 | * There is no free space in the found leaf. |
2054 | * We're gonna add a new leaf in the tree. | |
a86c6181 | 2055 | */ |
107a7bd3 TT |
2056 | if (gb_flags & EXT4_GET_BLOCKS_METADATA_NOFAIL) |
2057 | mb_flags = EXT4_MB_USE_RESERVED; | |
2058 | err = ext4_ext_create_new_leaf(handle, inode, mb_flags, gb_flags, | |
dfe50809 | 2059 | ppath, newext); |
a86c6181 AT |
2060 | if (err) |
2061 | goto cleanup; | |
2062 | depth = ext_depth(inode); | |
2063 | eh = path[depth].p_hdr; | |
2064 | ||
2065 | has_space: | |
2066 | nearex = path[depth].p_ext; | |
2067 | ||
7e028976 AM |
2068 | err = ext4_ext_get_access(handle, inode, path + depth); |
2069 | if (err) | |
a86c6181 AT |
2070 | goto cleanup; |
2071 | ||
2072 | if (!nearex) { | |
2073 | /* there is no extent in this leaf, create first one */ | |
32de6756 | 2074 | ext_debug("first extent in the leaf: %u:%llu:[%d]%d\n", |
8c55e204 | 2075 | le32_to_cpu(newext->ee_block), |
bf89d16f | 2076 | ext4_ext_pblock(newext), |
556615dc | 2077 | ext4_ext_is_unwritten(newext), |
a2df2a63 | 2078 | ext4_ext_get_actual_len(newext)); |
80e675f9 EG |
2079 | nearex = EXT_FIRST_EXTENT(eh); |
2080 | } else { | |
2081 | if (le32_to_cpu(newext->ee_block) | |
8c55e204 | 2082 | > le32_to_cpu(nearex->ee_block)) { |
80e675f9 | 2083 | /* Insert after */ |
32de6756 YY |
2084 | ext_debug("insert %u:%llu:[%d]%d before: " |
2085 | "nearest %p\n", | |
80e675f9 EG |
2086 | le32_to_cpu(newext->ee_block), |
2087 | ext4_ext_pblock(newext), | |
556615dc | 2088 | ext4_ext_is_unwritten(newext), |
80e675f9 EG |
2089 | ext4_ext_get_actual_len(newext), |
2090 | nearex); | |
2091 | nearex++; | |
2092 | } else { | |
2093 | /* Insert before */ | |
2094 | BUG_ON(newext->ee_block == nearex->ee_block); | |
32de6756 YY |
2095 | ext_debug("insert %u:%llu:[%d]%d after: " |
2096 | "nearest %p\n", | |
8c55e204 | 2097 | le32_to_cpu(newext->ee_block), |
bf89d16f | 2098 | ext4_ext_pblock(newext), |
556615dc | 2099 | ext4_ext_is_unwritten(newext), |
a2df2a63 | 2100 | ext4_ext_get_actual_len(newext), |
80e675f9 EG |
2101 | nearex); |
2102 | } | |
2103 | len = EXT_LAST_EXTENT(eh) - nearex + 1; | |
2104 | if (len > 0) { | |
32de6756 | 2105 | ext_debug("insert %u:%llu:[%d]%d: " |
80e675f9 EG |
2106 | "move %d extents from 0x%p to 0x%p\n", |
2107 | le32_to_cpu(newext->ee_block), | |
2108 | ext4_ext_pblock(newext), | |
556615dc | 2109 | ext4_ext_is_unwritten(newext), |
80e675f9 EG |
2110 | ext4_ext_get_actual_len(newext), |
2111 | len, nearex, nearex + 1); | |
2112 | memmove(nearex + 1, nearex, | |
2113 | len * sizeof(struct ext4_extent)); | |
a86c6181 | 2114 | } |
a86c6181 AT |
2115 | } |
2116 | ||
e8546d06 | 2117 | le16_add_cpu(&eh->eh_entries, 1); |
80e675f9 | 2118 | path[depth].p_ext = nearex; |
a86c6181 | 2119 | nearex->ee_block = newext->ee_block; |
bf89d16f | 2120 | ext4_ext_store_pblock(nearex, ext4_ext_pblock(newext)); |
a86c6181 | 2121 | nearex->ee_len = newext->ee_len; |
a86c6181 AT |
2122 | |
2123 | merge: | |
e7bcf823 | 2124 | /* try to merge extents */ |
107a7bd3 | 2125 | if (!(gb_flags & EXT4_GET_BLOCKS_PRE_IO)) |
ecb94f5f | 2126 | ext4_ext_try_to_merge(handle, inode, path, nearex); |
a86c6181 | 2127 | |
a86c6181 AT |
2128 | |
2129 | /* time to correct all indexes above */ | |
2130 | err = ext4_ext_correct_indexes(handle, inode, path); | |
2131 | if (err) | |
2132 | goto cleanup; | |
2133 | ||
ecb94f5f | 2134 | err = ext4_ext_dirty(handle, inode, path + path->p_depth); |
a86c6181 AT |
2135 | |
2136 | cleanup: | |
b7ea89ad TT |
2137 | ext4_ext_drop_refs(npath); |
2138 | kfree(npath); | |
a86c6181 AT |
2139 | return err; |
2140 | } | |
2141 | ||
91dd8c11 LC |
2142 | static int ext4_fill_fiemap_extents(struct inode *inode, |
2143 | ext4_lblk_t block, ext4_lblk_t num, | |
2144 | struct fiemap_extent_info *fieinfo) | |
6873fa0d ES |
2145 | { |
2146 | struct ext4_ext_path *path = NULL; | |
6873fa0d | 2147 | struct ext4_extent *ex; |
69eb33dc | 2148 | struct extent_status es; |
91dd8c11 | 2149 | ext4_lblk_t next, next_del, start = 0, end = 0; |
6873fa0d | 2150 | ext4_lblk_t last = block + num; |
91dd8c11 LC |
2151 | int exists, depth = 0, err = 0; |
2152 | unsigned int flags = 0; | |
2153 | unsigned char blksize_bits = inode->i_sb->s_blocksize_bits; | |
6873fa0d | 2154 | |
f17722f9 | 2155 | while (block < last && block != EXT_MAX_BLOCKS) { |
6873fa0d ES |
2156 | num = last - block; |
2157 | /* find extent for this block */ | |
fab3a549 | 2158 | down_read(&EXT4_I(inode)->i_data_sem); |
91dd8c11 | 2159 | |
705912ca | 2160 | path = ext4_ext_find_extent(inode, block, &path, 0); |
6873fa0d | 2161 | if (IS_ERR(path)) { |
91dd8c11 | 2162 | up_read(&EXT4_I(inode)->i_data_sem); |
6873fa0d ES |
2163 | err = PTR_ERR(path); |
2164 | path = NULL; | |
2165 | break; | |
2166 | } | |
2167 | ||
2168 | depth = ext_depth(inode); | |
273df556 | 2169 | if (unlikely(path[depth].p_hdr == NULL)) { |
91dd8c11 | 2170 | up_read(&EXT4_I(inode)->i_data_sem); |
273df556 FM |
2171 | EXT4_ERROR_INODE(inode, "path[%d].p_hdr == NULL", depth); |
2172 | err = -EIO; | |
2173 | break; | |
2174 | } | |
6873fa0d ES |
2175 | ex = path[depth].p_ext; |
2176 | next = ext4_ext_next_allocated_block(path); | |
2177 | ||
91dd8c11 | 2178 | flags = 0; |
6873fa0d ES |
2179 | exists = 0; |
2180 | if (!ex) { | |
2181 | /* there is no extent yet, so try to allocate | |
2182 | * all requested space */ | |
2183 | start = block; | |
2184 | end = block + num; | |
2185 | } else if (le32_to_cpu(ex->ee_block) > block) { | |
2186 | /* need to allocate space before found extent */ | |
2187 | start = block; | |
2188 | end = le32_to_cpu(ex->ee_block); | |
2189 | if (block + num < end) | |
2190 | end = block + num; | |
2191 | } else if (block >= le32_to_cpu(ex->ee_block) | |
2192 | + ext4_ext_get_actual_len(ex)) { | |
2193 | /* need to allocate space after found extent */ | |
2194 | start = block; | |
2195 | end = block + num; | |
2196 | if (end >= next) | |
2197 | end = next; | |
2198 | } else if (block >= le32_to_cpu(ex->ee_block)) { | |
2199 | /* | |
2200 | * some part of requested space is covered | |
2201 | * by found extent | |
2202 | */ | |
2203 | start = block; | |
2204 | end = le32_to_cpu(ex->ee_block) | |
2205 | + ext4_ext_get_actual_len(ex); | |
2206 | if (block + num < end) | |
2207 | end = block + num; | |
2208 | exists = 1; | |
2209 | } else { | |
2210 | BUG(); | |
2211 | } | |
2212 | BUG_ON(end <= start); | |
2213 | ||
2214 | if (!exists) { | |
69eb33dc ZL |
2215 | es.es_lblk = start; |
2216 | es.es_len = end - start; | |
2217 | es.es_pblk = 0; | |
6873fa0d | 2218 | } else { |
69eb33dc ZL |
2219 | es.es_lblk = le32_to_cpu(ex->ee_block); |
2220 | es.es_len = ext4_ext_get_actual_len(ex); | |
2221 | es.es_pblk = ext4_ext_pblock(ex); | |
556615dc | 2222 | if (ext4_ext_is_unwritten(ex)) |
91dd8c11 | 2223 | flags |= FIEMAP_EXTENT_UNWRITTEN; |
6873fa0d ES |
2224 | } |
2225 | ||
91dd8c11 | 2226 | /* |
69eb33dc ZL |
2227 | * Find delayed extent and update es accordingly. We call |
2228 | * it even in !exists case to find out whether es is the | |
91dd8c11 LC |
2229 | * last existing extent or not. |
2230 | */ | |
69eb33dc | 2231 | next_del = ext4_find_delayed_extent(inode, &es); |
91dd8c11 LC |
2232 | if (!exists && next_del) { |
2233 | exists = 1; | |
72dac95d JL |
2234 | flags |= (FIEMAP_EXTENT_DELALLOC | |
2235 | FIEMAP_EXTENT_UNKNOWN); | |
91dd8c11 LC |
2236 | } |
2237 | up_read(&EXT4_I(inode)->i_data_sem); | |
2238 | ||
69eb33dc ZL |
2239 | if (unlikely(es.es_len == 0)) { |
2240 | EXT4_ERROR_INODE(inode, "es.es_len == 0"); | |
273df556 FM |
2241 | err = -EIO; |
2242 | break; | |
2243 | } | |
6873fa0d | 2244 | |
f7fec032 ZL |
2245 | /* |
2246 | * This is possible iff next == next_del == EXT_MAX_BLOCKS. | |
2247 | * we need to check next == EXT_MAX_BLOCKS because it is | |
2248 | * possible that an extent is with unwritten and delayed | |
2249 | * status due to when an extent is delayed allocated and | |
2250 | * is allocated by fallocate status tree will track both of | |
2251 | * them in a extent. | |
2252 | * | |
2253 | * So we could return a unwritten and delayed extent, and | |
2254 | * its block is equal to 'next'. | |
2255 | */ | |
2256 | if (next == next_del && next == EXT_MAX_BLOCKS) { | |
91dd8c11 LC |
2257 | flags |= FIEMAP_EXTENT_LAST; |
2258 | if (unlikely(next_del != EXT_MAX_BLOCKS || | |
2259 | next != EXT_MAX_BLOCKS)) { | |
2260 | EXT4_ERROR_INODE(inode, | |
2261 | "next extent == %u, next " | |
2262 | "delalloc extent = %u", | |
2263 | next, next_del); | |
2264 | err = -EIO; | |
2265 | break; | |
2266 | } | |
6873fa0d ES |
2267 | } |
2268 | ||
91dd8c11 LC |
2269 | if (exists) { |
2270 | err = fiemap_fill_next_extent(fieinfo, | |
69eb33dc ZL |
2271 | (__u64)es.es_lblk << blksize_bits, |
2272 | (__u64)es.es_pblk << blksize_bits, | |
2273 | (__u64)es.es_len << blksize_bits, | |
91dd8c11 LC |
2274 | flags); |
2275 | if (err < 0) | |
2276 | break; | |
2277 | if (err == 1) { | |
2278 | err = 0; | |
2279 | break; | |
2280 | } | |
6873fa0d ES |
2281 | } |
2282 | ||
69eb33dc | 2283 | block = es.es_lblk + es.es_len; |
6873fa0d ES |
2284 | } |
2285 | ||
b7ea89ad TT |
2286 | ext4_ext_drop_refs(path); |
2287 | kfree(path); | |
6873fa0d ES |
2288 | return err; |
2289 | } | |
2290 | ||
a86c6181 | 2291 | /* |
d0d856e8 RD |
2292 | * ext4_ext_put_gap_in_cache: |
2293 | * calculate boundaries of the gap that the requested block fits into | |
a86c6181 AT |
2294 | * and cache this gap |
2295 | */ | |
09b88252 | 2296 | static void |
a86c6181 | 2297 | ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, |
725d26d3 | 2298 | ext4_lblk_t block) |
a86c6181 AT |
2299 | { |
2300 | int depth = ext_depth(inode); | |
27b1b228 AS |
2301 | unsigned long len = 0; |
2302 | ext4_lblk_t lblock = 0; | |
a86c6181 AT |
2303 | struct ext4_extent *ex; |
2304 | ||
2305 | ex = path[depth].p_ext; | |
2306 | if (ex == NULL) { | |
69eb33dc ZL |
2307 | /* |
2308 | * there is no extent yet, so gap is [0;-] and we | |
2309 | * don't cache it | |
2310 | */ | |
a86c6181 AT |
2311 | ext_debug("cache gap(whole file):"); |
2312 | } else if (block < le32_to_cpu(ex->ee_block)) { | |
2313 | lblock = block; | |
2314 | len = le32_to_cpu(ex->ee_block) - block; | |
bba90743 ES |
2315 | ext_debug("cache gap(before): %u [%u:%u]", |
2316 | block, | |
2317 | le32_to_cpu(ex->ee_block), | |
2318 | ext4_ext_get_actual_len(ex)); | |
d100eef2 ZL |
2319 | if (!ext4_find_delalloc_range(inode, lblock, lblock + len - 1)) |
2320 | ext4_es_insert_extent(inode, lblock, len, ~0, | |
2321 | EXTENT_STATUS_HOLE); | |
a86c6181 | 2322 | } else if (block >= le32_to_cpu(ex->ee_block) |
a2df2a63 | 2323 | + ext4_ext_get_actual_len(ex)) { |
725d26d3 | 2324 | ext4_lblk_t next; |
8c55e204 | 2325 | lblock = le32_to_cpu(ex->ee_block) |
a2df2a63 | 2326 | + ext4_ext_get_actual_len(ex); |
725d26d3 AK |
2327 | |
2328 | next = ext4_ext_next_allocated_block(path); | |
bba90743 ES |
2329 | ext_debug("cache gap(after): [%u:%u] %u", |
2330 | le32_to_cpu(ex->ee_block), | |
2331 | ext4_ext_get_actual_len(ex), | |
2332 | block); | |
725d26d3 AK |
2333 | BUG_ON(next == lblock); |
2334 | len = next - lblock; | |
d100eef2 ZL |
2335 | if (!ext4_find_delalloc_range(inode, lblock, lblock + len - 1)) |
2336 | ext4_es_insert_extent(inode, lblock, len, ~0, | |
2337 | EXTENT_STATUS_HOLE); | |
a86c6181 | 2338 | } else { |
a86c6181 AT |
2339 | BUG(); |
2340 | } | |
2341 | ||
bba90743 | 2342 | ext_debug(" -> %u:%lu\n", lblock, len); |
a86c6181 AT |
2343 | } |
2344 | ||
2345 | /* | |
d0d856e8 RD |
2346 | * ext4_ext_rm_idx: |
2347 | * removes index from the index block. | |
a86c6181 | 2348 | */ |
1d03ec98 | 2349 | static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, |
c36575e6 | 2350 | struct ext4_ext_path *path, int depth) |
a86c6181 | 2351 | { |
a86c6181 | 2352 | int err; |
f65e6fba | 2353 | ext4_fsblk_t leaf; |
a86c6181 AT |
2354 | |
2355 | /* free index block */ | |
c36575e6 FL |
2356 | depth--; |
2357 | path = path + depth; | |
bf89d16f | 2358 | leaf = ext4_idx_pblock(path->p_idx); |
273df556 FM |
2359 | if (unlikely(path->p_hdr->eh_entries == 0)) { |
2360 | EXT4_ERROR_INODE(inode, "path->p_hdr->eh_entries == 0"); | |
2361 | return -EIO; | |
2362 | } | |
7e028976 AM |
2363 | err = ext4_ext_get_access(handle, inode, path); |
2364 | if (err) | |
a86c6181 | 2365 | return err; |
0e1147b0 RD |
2366 | |
2367 | if (path->p_idx != EXT_LAST_INDEX(path->p_hdr)) { | |
2368 | int len = EXT_LAST_INDEX(path->p_hdr) - path->p_idx; | |
2369 | len *= sizeof(struct ext4_extent_idx); | |
2370 | memmove(path->p_idx, path->p_idx + 1, len); | |
2371 | } | |
2372 | ||
e8546d06 | 2373 | le16_add_cpu(&path->p_hdr->eh_entries, -1); |
7e028976 AM |
2374 | err = ext4_ext_dirty(handle, inode, path); |
2375 | if (err) | |
a86c6181 | 2376 | return err; |
2ae02107 | 2377 | ext_debug("index is empty, remove it, free block %llu\n", leaf); |
d8990240 AK |
2378 | trace_ext4_ext_rm_idx(inode, leaf); |
2379 | ||
7dc57615 | 2380 | ext4_free_blocks(handle, inode, NULL, leaf, 1, |
e6362609 | 2381 | EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET); |
c36575e6 FL |
2382 | |
2383 | while (--depth >= 0) { | |
2384 | if (path->p_idx != EXT_FIRST_INDEX(path->p_hdr)) | |
2385 | break; | |
2386 | path--; | |
2387 | err = ext4_ext_get_access(handle, inode, path); | |
2388 | if (err) | |
2389 | break; | |
2390 | path->p_idx->ei_block = (path+1)->p_idx->ei_block; | |
2391 | err = ext4_ext_dirty(handle, inode, path); | |
2392 | if (err) | |
2393 | break; | |
2394 | } | |
a86c6181 AT |
2395 | return err; |
2396 | } | |
2397 | ||
2398 | /* | |
ee12b630 MC |
2399 | * ext4_ext_calc_credits_for_single_extent: |
2400 | * This routine returns max. credits that needed to insert an extent | |
2401 | * to the extent tree. | |
2402 | * When pass the actual path, the caller should calculate credits | |
2403 | * under i_data_sem. | |
a86c6181 | 2404 | */ |
525f4ed8 | 2405 | int ext4_ext_calc_credits_for_single_extent(struct inode *inode, int nrblocks, |
a86c6181 AT |
2406 | struct ext4_ext_path *path) |
2407 | { | |
a86c6181 | 2408 | if (path) { |
ee12b630 | 2409 | int depth = ext_depth(inode); |
f3bd1f3f | 2410 | int ret = 0; |
ee12b630 | 2411 | |
a86c6181 | 2412 | /* probably there is space in leaf? */ |
a86c6181 | 2413 | if (le16_to_cpu(path[depth].p_hdr->eh_entries) |
ee12b630 | 2414 | < le16_to_cpu(path[depth].p_hdr->eh_max)) { |
a86c6181 | 2415 | |
ee12b630 MC |
2416 | /* |
2417 | * There are some space in the leaf tree, no | |
2418 | * need to account for leaf block credit | |
2419 | * | |
2420 | * bitmaps and block group descriptor blocks | |
df3ab170 | 2421 | * and other metadata blocks still need to be |
ee12b630 MC |
2422 | * accounted. |
2423 | */ | |
525f4ed8 | 2424 | /* 1 bitmap, 1 block group descriptor */ |
ee12b630 | 2425 | ret = 2 + EXT4_META_TRANS_BLOCKS(inode->i_sb); |
5887e98b | 2426 | return ret; |
ee12b630 MC |
2427 | } |
2428 | } | |
a86c6181 | 2429 | |
525f4ed8 | 2430 | return ext4_chunk_trans_blocks(inode, nrblocks); |
ee12b630 | 2431 | } |
a86c6181 | 2432 | |
ee12b630 | 2433 | /* |
fffb2739 | 2434 | * How many index/leaf blocks need to change/allocate to add @extents extents? |
ee12b630 | 2435 | * |
fffb2739 JK |
2436 | * If we add a single extent, then in the worse case, each tree level |
2437 | * index/leaf need to be changed in case of the tree split. | |
ee12b630 | 2438 | * |
fffb2739 JK |
2439 | * If more extents are inserted, they could cause the whole tree split more |
2440 | * than once, but this is really rare. | |
ee12b630 | 2441 | */ |
fffb2739 | 2442 | int ext4_ext_index_trans_blocks(struct inode *inode, int extents) |
ee12b630 MC |
2443 | { |
2444 | int index; | |
f19d5870 TM |
2445 | int depth; |
2446 | ||
2447 | /* If we are converting the inline data, only one is needed here. */ | |
2448 | if (ext4_has_inline_data(inode)) | |
2449 | return 1; | |
2450 | ||
2451 | depth = ext_depth(inode); | |
a86c6181 | 2452 | |
fffb2739 | 2453 | if (extents <= 1) |
ee12b630 MC |
2454 | index = depth * 2; |
2455 | else | |
2456 | index = depth * 3; | |
a86c6181 | 2457 | |
ee12b630 | 2458 | return index; |
a86c6181 AT |
2459 | } |
2460 | ||
981250ca TT |
2461 | static inline int get_default_free_blocks_flags(struct inode *inode) |
2462 | { | |
2463 | if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) | |
2464 | return EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET; | |
2465 | else if (ext4_should_journal_data(inode)) | |
2466 | return EXT4_FREE_BLOCKS_FORGET; | |
2467 | return 0; | |
2468 | } | |
2469 | ||
a86c6181 | 2470 | static int ext4_remove_blocks(handle_t *handle, struct inode *inode, |
0aa06000 | 2471 | struct ext4_extent *ex, |
d23142c6 | 2472 | long long *partial_cluster, |
0aa06000 | 2473 | ext4_lblk_t from, ext4_lblk_t to) |
a86c6181 | 2474 | { |
0aa06000 | 2475 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); |
a2df2a63 | 2476 | unsigned short ee_len = ext4_ext_get_actual_len(ex); |
0aa06000 | 2477 | ext4_fsblk_t pblk; |
981250ca | 2478 | int flags = get_default_free_blocks_flags(inode); |
18888cf0 | 2479 | |
0aa06000 TT |
2480 | /* |
2481 | * For bigalloc file systems, we never free a partial cluster | |
2482 | * at the beginning of the extent. Instead, we make a note | |
2483 | * that we tried freeing the cluster, and check to see if we | |
2484 | * need to free it on a subsequent call to ext4_remove_blocks, | |
2485 | * or at the end of the ext4_truncate() operation. | |
2486 | */ | |
2487 | flags |= EXT4_FREE_BLOCKS_NOFREE_FIRST_CLUSTER; | |
2488 | ||
d8990240 | 2489 | trace_ext4_remove_blocks(inode, ex, from, to, *partial_cluster); |
0aa06000 TT |
2490 | /* |
2491 | * If we have a partial cluster, and it's different from the | |
2492 | * cluster of the last block, we need to explicitly free the | |
2493 | * partial cluster here. | |
2494 | */ | |
2495 | pblk = ext4_ext_pblock(ex) + ee_len - 1; | |
d23142c6 LC |
2496 | if ((*partial_cluster > 0) && |
2497 | (EXT4_B2C(sbi, pblk) != *partial_cluster)) { | |
0aa06000 TT |
2498 | ext4_free_blocks(handle, inode, NULL, |
2499 | EXT4_C2B(sbi, *partial_cluster), | |
2500 | sbi->s_cluster_ratio, flags); | |
2501 | *partial_cluster = 0; | |
2502 | } | |
2503 | ||
a86c6181 AT |
2504 | #ifdef EXTENTS_STATS |
2505 | { | |
2506 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); | |
a86c6181 AT |
2507 | spin_lock(&sbi->s_ext_stats_lock); |
2508 | sbi->s_ext_blocks += ee_len; | |
2509 | sbi->s_ext_extents++; | |
2510 | if (ee_len < sbi->s_ext_min) | |
2511 | sbi->s_ext_min = ee_len; | |
2512 | if (ee_len > sbi->s_ext_max) | |
2513 | sbi->s_ext_max = ee_len; | |
2514 | if (ext_depth(inode) > sbi->s_depth_max) | |
2515 | sbi->s_depth_max = ext_depth(inode); | |
2516 | spin_unlock(&sbi->s_ext_stats_lock); | |
2517 | } | |
2518 | #endif | |
2519 | if (from >= le32_to_cpu(ex->ee_block) | |
a2df2a63 | 2520 | && to == le32_to_cpu(ex->ee_block) + ee_len - 1) { |
a86c6181 | 2521 | /* tail removal */ |
725d26d3 | 2522 | ext4_lblk_t num; |
d23142c6 | 2523 | unsigned int unaligned; |
725d26d3 | 2524 | |
a2df2a63 | 2525 | num = le32_to_cpu(ex->ee_block) + ee_len - from; |
0aa06000 | 2526 | pblk = ext4_ext_pblock(ex) + ee_len - num; |
d23142c6 LC |
2527 | /* |
2528 | * Usually we want to free partial cluster at the end of the | |
2529 | * extent, except for the situation when the cluster is still | |
2530 | * used by any other extent (partial_cluster is negative). | |
2531 | */ | |
2532 | if (*partial_cluster < 0 && | |
2533 | -(*partial_cluster) == EXT4_B2C(sbi, pblk + num - 1)) | |
2534 | flags |= EXT4_FREE_BLOCKS_NOFREE_LAST_CLUSTER; | |
2535 | ||
2536 | ext_debug("free last %u blocks starting %llu partial %lld\n", | |
2537 | num, pblk, *partial_cluster); | |
0aa06000 TT |
2538 | ext4_free_blocks(handle, inode, NULL, pblk, num, flags); |
2539 | /* | |
2540 | * If the block range to be freed didn't start at the | |
2541 | * beginning of a cluster, and we removed the entire | |
d23142c6 LC |
2542 | * extent and the cluster is not used by any other extent, |
2543 | * save the partial cluster here, since we might need to | |
2544 | * delete if we determine that the truncate operation has | |
2545 | * removed all of the blocks in the cluster. | |
2546 | * | |
2547 | * On the other hand, if we did not manage to free the whole | |
2548 | * extent, we have to mark the cluster as used (store negative | |
2549 | * cluster number in partial_cluster). | |
0aa06000 | 2550 | */ |
f5a44db5 | 2551 | unaligned = EXT4_PBLK_COFF(sbi, pblk); |
d23142c6 LC |
2552 | if (unaligned && (ee_len == num) && |
2553 | (*partial_cluster != -((long long)EXT4_B2C(sbi, pblk)))) | |
0aa06000 | 2554 | *partial_cluster = EXT4_B2C(sbi, pblk); |
d23142c6 LC |
2555 | else if (unaligned) |
2556 | *partial_cluster = -((long long)EXT4_B2C(sbi, pblk)); | |
2557 | else if (*partial_cluster > 0) | |
0aa06000 | 2558 | *partial_cluster = 0; |
78fb9cdf LC |
2559 | } else |
2560 | ext4_error(sbi->s_sb, "strange request: removal(2) " | |
2561 | "%u-%u from %u:%u\n", | |
2562 | from, to, le32_to_cpu(ex->ee_block), ee_len); | |
a86c6181 AT |
2563 | return 0; |
2564 | } | |
2565 | ||
d583fb87 AH |
2566 | |
2567 | /* | |
2568 | * ext4_ext_rm_leaf() Removes the extents associated with the | |
2569 | * blocks appearing between "start" and "end", and splits the extents | |
2570 | * if "start" and "end" appear in the same extent | |
2571 | * | |
2572 | * @handle: The journal handle | |
2573 | * @inode: The files inode | |
2574 | * @path: The path to the leaf | |
d23142c6 LC |
2575 | * @partial_cluster: The cluster which we'll have to free if all extents |
2576 | * has been released from it. It gets negative in case | |
2577 | * that the cluster is still used. | |
d583fb87 AH |
2578 | * @start: The first block to remove |
2579 | * @end: The last block to remove | |
2580 | */ | |
a86c6181 AT |
2581 | static int |
2582 | ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, | |
d23142c6 LC |
2583 | struct ext4_ext_path *path, |
2584 | long long *partial_cluster, | |
0aa06000 | 2585 | ext4_lblk_t start, ext4_lblk_t end) |
a86c6181 | 2586 | { |
0aa06000 | 2587 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); |
a86c6181 AT |
2588 | int err = 0, correct_index = 0; |
2589 | int depth = ext_depth(inode), credits; | |
2590 | struct ext4_extent_header *eh; | |
750c9c47 | 2591 | ext4_lblk_t a, b; |
725d26d3 AK |
2592 | unsigned num; |
2593 | ext4_lblk_t ex_ee_block; | |
a86c6181 | 2594 | unsigned short ex_ee_len; |
556615dc | 2595 | unsigned unwritten = 0; |
a86c6181 | 2596 | struct ext4_extent *ex; |
d23142c6 | 2597 | ext4_fsblk_t pblk; |
a86c6181 | 2598 | |
c29c0ae7 | 2599 | /* the header must be checked already in ext4_ext_remove_space() */ |
5f95d21f | 2600 | ext_debug("truncate since %u in leaf to %u\n", start, end); |
a86c6181 AT |
2601 | if (!path[depth].p_hdr) |
2602 | path[depth].p_hdr = ext_block_hdr(path[depth].p_bh); | |
2603 | eh = path[depth].p_hdr; | |
273df556 FM |
2604 | if (unlikely(path[depth].p_hdr == NULL)) { |
2605 | EXT4_ERROR_INODE(inode, "path[%d].p_hdr == NULL", depth); | |
2606 | return -EIO; | |
2607 | } | |
a86c6181 | 2608 | /* find where to start removing */ |
6ae06ff5 AS |
2609 | ex = path[depth].p_ext; |
2610 | if (!ex) | |
2611 | ex = EXT_LAST_EXTENT(eh); | |
a86c6181 AT |
2612 | |
2613 | ex_ee_block = le32_to_cpu(ex->ee_block); | |
a2df2a63 | 2614 | ex_ee_len = ext4_ext_get_actual_len(ex); |
a86c6181 | 2615 | |
c0634493 EW |
2616 | /* |
2617 | * If we're starting with an extent other than the last one in the | |
2618 | * node, we need to see if it shares a cluster with the extent to | |
2619 | * the right (towards the end of the file). If its leftmost cluster | |
2620 | * is this extent's rightmost cluster and it is not cluster aligned, | |
2621 | * we'll mark it as a partial that is not to be deallocated. | |
2622 | */ | |
2623 | ||
2624 | if (ex != EXT_LAST_EXTENT(eh)) { | |
2625 | ext4_fsblk_t current_pblk, right_pblk; | |
2626 | long long current_cluster, right_cluster; | |
2627 | ||
2628 | current_pblk = ext4_ext_pblock(ex) + ex_ee_len - 1; | |
2629 | current_cluster = (long long)EXT4_B2C(sbi, current_pblk); | |
2630 | right_pblk = ext4_ext_pblock(ex + 1); | |
2631 | right_cluster = (long long)EXT4_B2C(sbi, right_pblk); | |
2632 | if (current_cluster == right_cluster && | |
2633 | EXT4_PBLK_COFF(sbi, right_pblk)) | |
2634 | *partial_cluster = -right_cluster; | |
2635 | } | |
2636 | ||
d8990240 AK |
2637 | trace_ext4_ext_rm_leaf(inode, start, ex, *partial_cluster); |
2638 | ||
a86c6181 AT |
2639 | while (ex >= EXT_FIRST_EXTENT(eh) && |
2640 | ex_ee_block + ex_ee_len > start) { | |
a41f2071 | 2641 | |
556615dc LC |
2642 | if (ext4_ext_is_unwritten(ex)) |
2643 | unwritten = 1; | |
a41f2071 | 2644 | else |
556615dc | 2645 | unwritten = 0; |
a41f2071 | 2646 | |
553f9008 | 2647 | ext_debug("remove ext %u:[%d]%d\n", ex_ee_block, |
556615dc | 2648 | unwritten, ex_ee_len); |
a86c6181 AT |
2649 | path[depth].p_ext = ex; |
2650 | ||
2651 | a = ex_ee_block > start ? ex_ee_block : start; | |
d583fb87 AH |
2652 | b = ex_ee_block+ex_ee_len - 1 < end ? |
2653 | ex_ee_block+ex_ee_len - 1 : end; | |
a86c6181 AT |
2654 | |
2655 | ext_debug(" border %u:%u\n", a, b); | |
2656 | ||
d583fb87 | 2657 | /* If this extent is beyond the end of the hole, skip it */ |
5f95d21f | 2658 | if (end < ex_ee_block) { |
d23142c6 LC |
2659 | /* |
2660 | * We're going to skip this extent and move to another, | |
2661 | * so if this extent is not cluster aligned we have | |
2662 | * to mark the current cluster as used to avoid | |
2663 | * accidentally freeing it later on | |
2664 | */ | |
2665 | pblk = ext4_ext_pblock(ex); | |
f5a44db5 | 2666 | if (EXT4_PBLK_COFF(sbi, pblk)) |
d23142c6 LC |
2667 | *partial_cluster = |
2668 | -((long long)EXT4_B2C(sbi, pblk)); | |
d583fb87 AH |
2669 | ex--; |
2670 | ex_ee_block = le32_to_cpu(ex->ee_block); | |
2671 | ex_ee_len = ext4_ext_get_actual_len(ex); | |
2672 | continue; | |
750c9c47 | 2673 | } else if (b != ex_ee_block + ex_ee_len - 1) { |
dc1841d6 LC |
2674 | EXT4_ERROR_INODE(inode, |
2675 | "can not handle truncate %u:%u " | |
2676 | "on extent %u:%u", | |
2677 | start, end, ex_ee_block, | |
2678 | ex_ee_block + ex_ee_len - 1); | |
750c9c47 DM |
2679 | err = -EIO; |
2680 | goto out; | |
a86c6181 AT |
2681 | } else if (a != ex_ee_block) { |
2682 | /* remove tail of the extent */ | |
750c9c47 | 2683 | num = a - ex_ee_block; |
a86c6181 AT |
2684 | } else { |
2685 | /* remove whole extent: excellent! */ | |
a86c6181 | 2686 | num = 0; |
a86c6181 | 2687 | } |
34071da7 TT |
2688 | /* |
2689 | * 3 for leaf, sb, and inode plus 2 (bmap and group | |
2690 | * descriptor) for each block group; assume two block | |
2691 | * groups plus ex_ee_len/blocks_per_block_group for | |
2692 | * the worst case | |
2693 | */ | |
2694 | credits = 7 + 2*(ex_ee_len/EXT4_BLOCKS_PER_GROUP(inode->i_sb)); | |
a86c6181 AT |
2695 | if (ex == EXT_FIRST_EXTENT(eh)) { |
2696 | correct_index = 1; | |
2697 | credits += (ext_depth(inode)) + 1; | |
2698 | } | |
5aca07eb | 2699 | credits += EXT4_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb); |
a86c6181 | 2700 | |
487caeef | 2701 | err = ext4_ext_truncate_extend_restart(handle, inode, credits); |
9102e4fa | 2702 | if (err) |
a86c6181 | 2703 | goto out; |
a86c6181 AT |
2704 | |
2705 | err = ext4_ext_get_access(handle, inode, path + depth); | |
2706 | if (err) | |
2707 | goto out; | |
2708 | ||
0aa06000 TT |
2709 | err = ext4_remove_blocks(handle, inode, ex, partial_cluster, |
2710 | a, b); | |
a86c6181 AT |
2711 | if (err) |
2712 | goto out; | |
2713 | ||
750c9c47 | 2714 | if (num == 0) |
d0d856e8 | 2715 | /* this extent is removed; mark slot entirely unused */ |
f65e6fba | 2716 | ext4_ext_store_pblock(ex, 0); |
a86c6181 | 2717 | |
a86c6181 | 2718 | ex->ee_len = cpu_to_le16(num); |
749269fa | 2719 | /* |
556615dc | 2720 | * Do not mark unwritten if all the blocks in the |
749269fa AA |
2721 | * extent have been removed. |
2722 | */ | |
556615dc LC |
2723 | if (unwritten && num) |
2724 | ext4_ext_mark_unwritten(ex); | |
d583fb87 AH |
2725 | /* |
2726 | * If the extent was completely released, | |
2727 | * we need to remove it from the leaf | |
2728 | */ | |
2729 | if (num == 0) { | |
f17722f9 | 2730 | if (end != EXT_MAX_BLOCKS - 1) { |
d583fb87 AH |
2731 | /* |
2732 | * For hole punching, we need to scoot all the | |
2733 | * extents up when an extent is removed so that | |
2734 | * we dont have blank extents in the middle | |
2735 | */ | |
2736 | memmove(ex, ex+1, (EXT_LAST_EXTENT(eh) - ex) * | |
2737 | sizeof(struct ext4_extent)); | |
2738 | ||
2739 | /* Now get rid of the one at the end */ | |
2740 | memset(EXT_LAST_EXTENT(eh), 0, | |
2741 | sizeof(struct ext4_extent)); | |
2742 | } | |
2743 | le16_add_cpu(&eh->eh_entries, -1); | |
d23142c6 | 2744 | } else if (*partial_cluster > 0) |
0aa06000 | 2745 | *partial_cluster = 0; |
d583fb87 | 2746 | |
750c9c47 DM |
2747 | err = ext4_ext_dirty(handle, inode, path + depth); |
2748 | if (err) | |
2749 | goto out; | |
2750 | ||
bf52c6f7 | 2751 | ext_debug("new extent: %u:%u:%llu\n", ex_ee_block, num, |
bf89d16f | 2752 | ext4_ext_pblock(ex)); |
a86c6181 AT |
2753 | ex--; |
2754 | ex_ee_block = le32_to_cpu(ex->ee_block); | |
a2df2a63 | 2755 | ex_ee_len = ext4_ext_get_actual_len(ex); |
a86c6181 AT |
2756 | } |
2757 | ||
2758 | if (correct_index && eh->eh_entries) | |
2759 | err = ext4_ext_correct_indexes(handle, inode, path); | |
2760 | ||
0aa06000 | 2761 | /* |
ad6599ab EW |
2762 | * If there's a partial cluster and at least one extent remains in |
2763 | * the leaf, free the partial cluster if it isn't shared with the | |
2764 | * current extent. If there's a partial cluster and no extents | |
2765 | * remain in the leaf, it can't be freed here. It can only be | |
2766 | * freed when it's possible to determine if it's not shared with | |
2767 | * any other extent - when the next leaf is processed or when space | |
2768 | * removal is complete. | |
0aa06000 | 2769 | */ |
ad6599ab | 2770 | if (*partial_cluster > 0 && eh->eh_entries && |
0aa06000 TT |
2771 | (EXT4_B2C(sbi, ext4_ext_pblock(ex) + ex_ee_len - 1) != |
2772 | *partial_cluster)) { | |
981250ca | 2773 | int flags = get_default_free_blocks_flags(inode); |
0aa06000 TT |
2774 | |
2775 | ext4_free_blocks(handle, inode, NULL, | |
2776 | EXT4_C2B(sbi, *partial_cluster), | |
2777 | sbi->s_cluster_ratio, flags); | |
2778 | *partial_cluster = 0; | |
2779 | } | |
2780 | ||
a86c6181 AT |
2781 | /* if this leaf is free, then we should |
2782 | * remove it from index block above */ | |
2783 | if (err == 0 && eh->eh_entries == 0 && path[depth].p_bh != NULL) | |
c36575e6 | 2784 | err = ext4_ext_rm_idx(handle, inode, path, depth); |
a86c6181 AT |
2785 | |
2786 | out: | |
2787 | return err; | |
2788 | } | |
2789 | ||
2790 | /* | |
d0d856e8 RD |
2791 | * ext4_ext_more_to_rm: |
2792 | * returns 1 if current index has to be freed (even partial) | |
a86c6181 | 2793 | */ |
09b88252 | 2794 | static int |
a86c6181 AT |
2795 | ext4_ext_more_to_rm(struct ext4_ext_path *path) |
2796 | { | |
2797 | BUG_ON(path->p_idx == NULL); | |
2798 | ||
2799 | if (path->p_idx < EXT_FIRST_INDEX(path->p_hdr)) | |
2800 | return 0; | |
2801 | ||
2802 | /* | |
d0d856e8 | 2803 | * if truncate on deeper level happened, it wasn't partial, |
a86c6181 AT |
2804 | * so we have to consider current index for truncation |
2805 | */ | |
2806 | if (le16_to_cpu(path->p_hdr->eh_entries) == path->p_block) | |
2807 | return 0; | |
2808 | return 1; | |
2809 | } | |
2810 | ||
26a4c0c6 TT |
2811 | int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start, |
2812 | ext4_lblk_t end) | |
a86c6181 AT |
2813 | { |
2814 | struct super_block *sb = inode->i_sb; | |
2815 | int depth = ext_depth(inode); | |
968dee77 | 2816 | struct ext4_ext_path *path = NULL; |
d23142c6 | 2817 | long long partial_cluster = 0; |
a86c6181 | 2818 | handle_t *handle; |
6f2080e6 | 2819 | int i = 0, err = 0; |
a86c6181 | 2820 | |
5f95d21f | 2821 | ext_debug("truncate since %u to %u\n", start, end); |
a86c6181 AT |
2822 | |
2823 | /* probably first extent we're gonna free will be last in block */ | |
9924a92a | 2824 | handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE, depth + 1); |
a86c6181 AT |
2825 | if (IS_ERR(handle)) |
2826 | return PTR_ERR(handle); | |
2827 | ||
0617b83f | 2828 | again: |
61801325 | 2829 | trace_ext4_ext_remove_space(inode, start, end, depth); |
d8990240 | 2830 | |
5f95d21f LC |
2831 | /* |
2832 | * Check if we are removing extents inside the extent tree. If that | |
2833 | * is the case, we are going to punch a hole inside the extent tree | |
2834 | * so we have to check whether we need to split the extent covering | |
2835 | * the last block to remove so we can easily remove the part of it | |
2836 | * in ext4_ext_rm_leaf(). | |
2837 | */ | |
2838 | if (end < EXT_MAX_BLOCKS - 1) { | |
2839 | struct ext4_extent *ex; | |
2840 | ext4_lblk_t ee_block; | |
2841 | ||
2842 | /* find extent for this block */ | |
107a7bd3 | 2843 | path = ext4_ext_find_extent(inode, end, NULL, EXT4_EX_NOCACHE); |
5f95d21f LC |
2844 | if (IS_ERR(path)) { |
2845 | ext4_journal_stop(handle); | |
2846 | return PTR_ERR(path); | |
2847 | } | |
2848 | depth = ext_depth(inode); | |
6f2080e6 | 2849 | /* Leaf not may not exist only if inode has no blocks at all */ |
5f95d21f | 2850 | ex = path[depth].p_ext; |
968dee77 | 2851 | if (!ex) { |
6f2080e6 DM |
2852 | if (depth) { |
2853 | EXT4_ERROR_INODE(inode, | |
2854 | "path[%d].p_hdr == NULL", | |
2855 | depth); | |
2856 | err = -EIO; | |
2857 | } | |
2858 | goto out; | |
968dee77 | 2859 | } |
5f95d21f LC |
2860 | |
2861 | ee_block = le32_to_cpu(ex->ee_block); | |
2862 | ||
2863 | /* | |
2864 | * See if the last block is inside the extent, if so split | |
2865 | * the extent at 'end' block so we can easily remove the | |
2866 | * tail of the first part of the split extent in | |
2867 | * ext4_ext_rm_leaf(). | |
2868 | */ | |
2869 | if (end >= ee_block && | |
2870 | end < ee_block + ext4_ext_get_actual_len(ex) - 1) { | |
5f95d21f LC |
2871 | /* |
2872 | * Split the extent in two so that 'end' is the last | |
27dd4385 LC |
2873 | * block in the first new extent. Also we should not |
2874 | * fail removing space due to ENOSPC so try to use | |
2875 | * reserved block if that happens. | |
5f95d21f | 2876 | */ |
dfe50809 | 2877 | err = ext4_force_split_extent_at(handle, inode, &path, |
fcf6b1b7 | 2878 | end + 1, 1); |
5f95d21f LC |
2879 | if (err < 0) |
2880 | goto out; | |
2881 | } | |
5f95d21f | 2882 | } |
a86c6181 | 2883 | /* |
d0d856e8 RD |
2884 | * We start scanning from right side, freeing all the blocks |
2885 | * after i_size and walking into the tree depth-wise. | |
a86c6181 | 2886 | */ |
0617b83f | 2887 | depth = ext_depth(inode); |
968dee77 AS |
2888 | if (path) { |
2889 | int k = i = depth; | |
2890 | while (--k > 0) | |
2891 | path[k].p_block = | |
2892 | le16_to_cpu(path[k].p_hdr->eh_entries)+1; | |
2893 | } else { | |
2894 | path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), | |
2895 | GFP_NOFS); | |
2896 | if (path == NULL) { | |
2897 | ext4_journal_stop(handle); | |
2898 | return -ENOMEM; | |
2899 | } | |
10809df8 | 2900 | path[0].p_maxdepth = path[0].p_depth = depth; |
968dee77 | 2901 | path[0].p_hdr = ext_inode_hdr(inode); |
89a4e48f | 2902 | i = 0; |
5f95d21f | 2903 | |
c349179b | 2904 | if (ext4_ext_check(inode, path[0].p_hdr, depth, 0)) { |
968dee77 AS |
2905 | err = -EIO; |
2906 | goto out; | |
2907 | } | |
a86c6181 | 2908 | } |
968dee77 | 2909 | err = 0; |
a86c6181 AT |
2910 | |
2911 | while (i >= 0 && err == 0) { | |
2912 | if (i == depth) { | |
2913 | /* this is leaf block */ | |
d583fb87 | 2914 | err = ext4_ext_rm_leaf(handle, inode, path, |
0aa06000 | 2915 | &partial_cluster, start, |
5f95d21f | 2916 | end); |
d0d856e8 | 2917 | /* root level has p_bh == NULL, brelse() eats this */ |
a86c6181 AT |
2918 | brelse(path[i].p_bh); |
2919 | path[i].p_bh = NULL; | |
2920 | i--; | |
2921 | continue; | |
2922 | } | |
2923 | ||
2924 | /* this is index block */ | |
2925 | if (!path[i].p_hdr) { | |
2926 | ext_debug("initialize header\n"); | |
2927 | path[i].p_hdr = ext_block_hdr(path[i].p_bh); | |
a86c6181 AT |
2928 | } |
2929 | ||
a86c6181 | 2930 | if (!path[i].p_idx) { |
d0d856e8 | 2931 | /* this level hasn't been touched yet */ |
a86c6181 AT |
2932 | path[i].p_idx = EXT_LAST_INDEX(path[i].p_hdr); |
2933 | path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries)+1; | |
2934 | ext_debug("init index ptr: hdr 0x%p, num %d\n", | |
2935 | path[i].p_hdr, | |
2936 | le16_to_cpu(path[i].p_hdr->eh_entries)); | |
2937 | } else { | |
d0d856e8 | 2938 | /* we were already here, see at next index */ |
a86c6181 AT |
2939 | path[i].p_idx--; |
2940 | } | |
2941 | ||
2942 | ext_debug("level %d - index, first 0x%p, cur 0x%p\n", | |
2943 | i, EXT_FIRST_INDEX(path[i].p_hdr), | |
2944 | path[i].p_idx); | |
2945 | if (ext4_ext_more_to_rm(path + i)) { | |
c29c0ae7 | 2946 | struct buffer_head *bh; |
a86c6181 | 2947 | /* go to the next level */ |
2ae02107 | 2948 | ext_debug("move to level %d (block %llu)\n", |
bf89d16f | 2949 | i + 1, ext4_idx_pblock(path[i].p_idx)); |
a86c6181 | 2950 | memset(path + i + 1, 0, sizeof(*path)); |
7d7ea89e | 2951 | bh = read_extent_tree_block(inode, |
107a7bd3 TT |
2952 | ext4_idx_pblock(path[i].p_idx), depth - i - 1, |
2953 | EXT4_EX_NOCACHE); | |
7d7ea89e | 2954 | if (IS_ERR(bh)) { |
a86c6181 | 2955 | /* should we reset i_size? */ |
7d7ea89e | 2956 | err = PTR_ERR(bh); |
a86c6181 AT |
2957 | break; |
2958 | } | |
76828c88 TT |
2959 | /* Yield here to deal with large extent trees. |
2960 | * Should be a no-op if we did IO above. */ | |
2961 | cond_resched(); | |
c29c0ae7 AT |
2962 | if (WARN_ON(i + 1 > depth)) { |
2963 | err = -EIO; | |
c29c0ae7 AT |
2964 | break; |
2965 | } | |
2966 | path[i + 1].p_bh = bh; | |
a86c6181 | 2967 | |
d0d856e8 RD |
2968 | /* save actual number of indexes since this |
2969 | * number is changed at the next iteration */ | |
a86c6181 AT |
2970 | path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries); |
2971 | i++; | |
2972 | } else { | |
d0d856e8 | 2973 | /* we finished processing this index, go up */ |
a86c6181 | 2974 | if (path[i].p_hdr->eh_entries == 0 && i > 0) { |
d0d856e8 | 2975 | /* index is empty, remove it; |
a86c6181 AT |
2976 | * handle must be already prepared by the |
2977 | * truncatei_leaf() */ | |
c36575e6 | 2978 | err = ext4_ext_rm_idx(handle, inode, path, i); |
a86c6181 | 2979 | } |
d0d856e8 | 2980 | /* root level has p_bh == NULL, brelse() eats this */ |
a86c6181 AT |
2981 | brelse(path[i].p_bh); |
2982 | path[i].p_bh = NULL; | |
2983 | i--; | |
2984 | ext_debug("return to level %d\n", i); | |
2985 | } | |
2986 | } | |
2987 | ||
61801325 LC |
2988 | trace_ext4_ext_remove_space_done(inode, start, end, depth, |
2989 | partial_cluster, path->p_hdr->eh_entries); | |
d8990240 | 2990 | |
7b415bf6 AK |
2991 | /* If we still have something in the partial cluster and we have removed |
2992 | * even the first extent, then we should free the blocks in the partial | |
2993 | * cluster as well. */ | |
d23142c6 | 2994 | if (partial_cluster > 0 && path->p_hdr->eh_entries == 0) { |
981250ca | 2995 | int flags = get_default_free_blocks_flags(inode); |
7b415bf6 AK |
2996 | |
2997 | ext4_free_blocks(handle, inode, NULL, | |
2998 | EXT4_C2B(EXT4_SB(sb), partial_cluster), | |
2999 | EXT4_SB(sb)->s_cluster_ratio, flags); | |
3000 | partial_cluster = 0; | |
3001 | } | |
3002 | ||
a86c6181 AT |
3003 | /* TODO: flexible tree reduction should be here */ |
3004 | if (path->p_hdr->eh_entries == 0) { | |
3005 | /* | |
d0d856e8 RD |
3006 | * truncate to zero freed all the tree, |
3007 | * so we need to correct eh_depth | |
a86c6181 AT |
3008 | */ |
3009 | err = ext4_ext_get_access(handle, inode, path); | |
3010 | if (err == 0) { | |
3011 | ext_inode_hdr(inode)->eh_depth = 0; | |
3012 | ext_inode_hdr(inode)->eh_max = | |
55ad63bf | 3013 | cpu_to_le16(ext4_ext_space_root(inode, 0)); |
a86c6181 AT |
3014 | err = ext4_ext_dirty(handle, inode, path); |
3015 | } | |
3016 | } | |
3017 | out: | |
b7ea89ad TT |
3018 | ext4_ext_drop_refs(path); |
3019 | kfree(path); | |
3020 | path = NULL; | |
dfe50809 TT |
3021 | if (err == -EAGAIN) |
3022 | goto again; | |
a86c6181 AT |
3023 | ext4_journal_stop(handle); |
3024 | ||
3025 | return err; | |
3026 | } | |
3027 | ||
3028 | /* | |
3029 | * called at mount time | |
3030 | */ | |
3031 | void ext4_ext_init(struct super_block *sb) | |
3032 | { | |
3033 | /* | |
3034 | * possible initialization would be here | |
3035 | */ | |
3036 | ||
83982b6f | 3037 | if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) { |
90576c0b | 3038 | #if defined(AGGRESSIVE_TEST) || defined(CHECK_BINSEARCH) || defined(EXTENTS_STATS) |
92b97816 | 3039 | printk(KERN_INFO "EXT4-fs: file extents enabled" |
bbf2f9fb | 3040 | #ifdef AGGRESSIVE_TEST |
92b97816 | 3041 | ", aggressive tests" |
a86c6181 AT |
3042 | #endif |
3043 | #ifdef CHECK_BINSEARCH | |
92b97816 | 3044 | ", check binsearch" |
a86c6181 AT |
3045 | #endif |
3046 | #ifdef EXTENTS_STATS | |
92b97816 | 3047 | ", stats" |
a86c6181 | 3048 | #endif |
92b97816 | 3049 | "\n"); |
90576c0b | 3050 | #endif |
a86c6181 AT |
3051 | #ifdef EXTENTS_STATS |
3052 | spin_lock_init(&EXT4_SB(sb)->s_ext_stats_lock); | |
3053 | EXT4_SB(sb)->s_ext_min = 1 << 30; | |
3054 | EXT4_SB(sb)->s_ext_max = 0; | |
3055 | #endif | |
3056 | } | |
3057 | } | |
3058 | ||
3059 | /* | |
3060 | * called at umount time | |
3061 | */ | |
3062 | void ext4_ext_release(struct super_block *sb) | |
3063 | { | |
83982b6f | 3064 | if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) |
a86c6181 AT |
3065 | return; |
3066 | ||
3067 | #ifdef EXTENTS_STATS | |
3068 | if (EXT4_SB(sb)->s_ext_blocks && EXT4_SB(sb)->s_ext_extents) { | |
3069 | struct ext4_sb_info *sbi = EXT4_SB(sb); | |
3070 | printk(KERN_ERR "EXT4-fs: %lu blocks in %lu extents (%lu ave)\n", | |
3071 | sbi->s_ext_blocks, sbi->s_ext_extents, | |
3072 | sbi->s_ext_blocks / sbi->s_ext_extents); | |
3073 | printk(KERN_ERR "EXT4-fs: extents: %lu min, %lu max, max depth %lu\n", | |
3074 | sbi->s_ext_min, sbi->s_ext_max, sbi->s_depth_max); | |
3075 | } | |
3076 | #endif | |
3077 | } | |
3078 | ||
d7b2a00c ZL |
3079 | static int ext4_zeroout_es(struct inode *inode, struct ext4_extent *ex) |
3080 | { | |
3081 | ext4_lblk_t ee_block; | |
3082 | ext4_fsblk_t ee_pblock; | |
3083 | unsigned int ee_len; | |
3084 | ||
3085 | ee_block = le32_to_cpu(ex->ee_block); | |
3086 | ee_len = ext4_ext_get_actual_len(ex); | |
3087 | ee_pblock = ext4_ext_pblock(ex); | |
3088 | ||
3089 | if (ee_len == 0) | |
3090 | return 0; | |
3091 | ||
3092 | return ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock, | |
3093 | EXTENT_STATUS_WRITTEN); | |
3094 | } | |
3095 | ||
093a088b AK |
3096 | /* FIXME!! we need to try to merge to left or right after zero-out */ |
3097 | static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) | |
3098 | { | |
2407518d LC |
3099 | ext4_fsblk_t ee_pblock; |
3100 | unsigned int ee_len; | |
b720303d | 3101 | int ret; |
093a088b | 3102 | |
093a088b | 3103 | ee_len = ext4_ext_get_actual_len(ex); |
bf89d16f | 3104 | ee_pblock = ext4_ext_pblock(ex); |
b720303d | 3105 | |
a107e5a3 | 3106 | ret = sb_issue_zeroout(inode->i_sb, ee_pblock, ee_len, GFP_NOFS); |
2407518d LC |
3107 | if (ret > 0) |
3108 | ret = 0; | |
093a088b | 3109 | |
2407518d | 3110 | return ret; |
093a088b AK |
3111 | } |
3112 | ||
47ea3bb5 YY |
3113 | /* |
3114 | * ext4_split_extent_at() splits an extent at given block. | |
3115 | * | |
3116 | * @handle: the journal handle | |
3117 | * @inode: the file inode | |
3118 | * @path: the path to the extent | |
3119 | * @split: the logical block where the extent is splitted. | |
3120 | * @split_flags: indicates if the extent could be zeroout if split fails, and | |
556615dc | 3121 | * the states(init or unwritten) of new extents. |
47ea3bb5 YY |
3122 | * @flags: flags used to insert new extent to extent tree. |
3123 | * | |
3124 | * | |
3125 | * Splits extent [a, b] into two extents [a, @split) and [@split, b], states | |
3126 | * of which are deterimined by split_flag. | |
3127 | * | |
3128 | * There are two cases: | |
3129 | * a> the extent are splitted into two extent. | |
3130 | * b> split is not needed, and just mark the extent. | |
3131 | * | |
3132 | * return 0 on success. | |
3133 | */ | |
3134 | static int ext4_split_extent_at(handle_t *handle, | |
3135 | struct inode *inode, | |
dfe50809 | 3136 | struct ext4_ext_path **ppath, |
47ea3bb5 YY |
3137 | ext4_lblk_t split, |
3138 | int split_flag, | |
3139 | int flags) | |
3140 | { | |
dfe50809 | 3141 | struct ext4_ext_path *path = *ppath; |
47ea3bb5 YY |
3142 | ext4_fsblk_t newblock; |
3143 | ext4_lblk_t ee_block; | |
adb23551 | 3144 | struct ext4_extent *ex, newex, orig_ex, zero_ex; |
47ea3bb5 YY |
3145 | struct ext4_extent *ex2 = NULL; |
3146 | unsigned int ee_len, depth; | |
3147 | int err = 0; | |
3148 | ||
dee1f973 DM |
3149 | BUG_ON((split_flag & (EXT4_EXT_DATA_VALID1 | EXT4_EXT_DATA_VALID2)) == |
3150 | (EXT4_EXT_DATA_VALID1 | EXT4_EXT_DATA_VALID2)); | |
3151 | ||
47ea3bb5 YY |
3152 | ext_debug("ext4_split_extents_at: inode %lu, logical" |
3153 | "block %llu\n", inode->i_ino, (unsigned long long)split); | |
3154 | ||
3155 | ext4_ext_show_leaf(inode, path); | |
3156 | ||
3157 | depth = ext_depth(inode); | |
3158 | ex = path[depth].p_ext; | |
3159 | ee_block = le32_to_cpu(ex->ee_block); | |
3160 | ee_len = ext4_ext_get_actual_len(ex); | |
3161 | newblock = split - ee_block + ext4_ext_pblock(ex); | |
3162 | ||
3163 | BUG_ON(split < ee_block || split >= (ee_block + ee_len)); | |
556615dc | 3164 | BUG_ON(!ext4_ext_is_unwritten(ex) && |
357b66fd | 3165 | split_flag & (EXT4_EXT_MAY_ZEROOUT | |
556615dc LC |
3166 | EXT4_EXT_MARK_UNWRIT1 | |
3167 | EXT4_EXT_MARK_UNWRIT2)); | |
47ea3bb5 YY |
3168 | |
3169 | err = ext4_ext_get_access(handle, inode, path + depth); | |
3170 | if (err) | |
3171 | goto out; | |
3172 | ||
3173 | if (split == ee_block) { | |
3174 | /* | |
3175 | * case b: block @split is the block that the extent begins with | |
3176 | * then we just change the state of the extent, and splitting | |
3177 | * is not needed. | |
3178 | */ | |
556615dc LC |
3179 | if (split_flag & EXT4_EXT_MARK_UNWRIT2) |
3180 | ext4_ext_mark_unwritten(ex); | |
47ea3bb5 YY |
3181 | else |
3182 | ext4_ext_mark_initialized(ex); | |
3183 | ||
3184 | if (!(flags & EXT4_GET_BLOCKS_PRE_IO)) | |
ecb94f5f | 3185 | ext4_ext_try_to_merge(handle, inode, path, ex); |
47ea3bb5 | 3186 | |
ecb94f5f | 3187 | err = ext4_ext_dirty(handle, inode, path + path->p_depth); |
47ea3bb5 YY |
3188 | goto out; |
3189 | } | |
3190 | ||
3191 | /* case a */ | |
3192 | memcpy(&orig_ex, ex, sizeof(orig_ex)); | |
3193 | ex->ee_len = cpu_to_le16(split - ee_block); | |
556615dc LC |
3194 | if (split_flag & EXT4_EXT_MARK_UNWRIT1) |
3195 | ext4_ext_mark_unwritten(ex); | |
47ea3bb5 YY |
3196 | |
3197 | /* | |
3198 | * path may lead to new leaf, not to original leaf any more | |
3199 | * after ext4_ext_insert_extent() returns, | |
3200 | */ | |
3201 | err = ext4_ext_dirty(handle, inode, path + depth); | |
3202 | if (err) | |
3203 | goto fix_extent_len; | |
3204 | ||
3205 | ex2 = &newex; | |
3206 | ex2->ee_block = cpu_to_le32(split); | |
3207 | ex2->ee_len = cpu_to_le16(ee_len - (split - ee_block)); | |
3208 | ext4_ext_store_pblock(ex2, newblock); | |
556615dc LC |
3209 | if (split_flag & EXT4_EXT_MARK_UNWRIT2) |
3210 | ext4_ext_mark_unwritten(ex2); | |
47ea3bb5 | 3211 | |
dfe50809 | 3212 | err = ext4_ext_insert_extent(handle, inode, ppath, &newex, flags); |
47ea3bb5 | 3213 | if (err == -ENOSPC && (EXT4_EXT_MAY_ZEROOUT & split_flag)) { |
dee1f973 | 3214 | if (split_flag & (EXT4_EXT_DATA_VALID1|EXT4_EXT_DATA_VALID2)) { |
adb23551 | 3215 | if (split_flag & EXT4_EXT_DATA_VALID1) { |
dee1f973 | 3216 | err = ext4_ext_zeroout(inode, ex2); |
adb23551 | 3217 | zero_ex.ee_block = ex2->ee_block; |
8cde7ad1 ZL |
3218 | zero_ex.ee_len = cpu_to_le16( |
3219 | ext4_ext_get_actual_len(ex2)); | |
adb23551 ZL |
3220 | ext4_ext_store_pblock(&zero_ex, |
3221 | ext4_ext_pblock(ex2)); | |
3222 | } else { | |
dee1f973 | 3223 | err = ext4_ext_zeroout(inode, ex); |
adb23551 | 3224 | zero_ex.ee_block = ex->ee_block; |
8cde7ad1 ZL |
3225 | zero_ex.ee_len = cpu_to_le16( |
3226 | ext4_ext_get_actual_len(ex)); | |
adb23551 ZL |
3227 | ext4_ext_store_pblock(&zero_ex, |
3228 | ext4_ext_pblock(ex)); | |
3229 | } | |
3230 | } else { | |
dee1f973 | 3231 | err = ext4_ext_zeroout(inode, &orig_ex); |
adb23551 | 3232 | zero_ex.ee_block = orig_ex.ee_block; |
8cde7ad1 ZL |
3233 | zero_ex.ee_len = cpu_to_le16( |
3234 | ext4_ext_get_actual_len(&orig_ex)); | |
adb23551 ZL |
3235 | ext4_ext_store_pblock(&zero_ex, |
3236 | ext4_ext_pblock(&orig_ex)); | |
3237 | } | |
dee1f973 | 3238 | |
47ea3bb5 YY |
3239 | if (err) |
3240 | goto fix_extent_len; | |
3241 | /* update the extent length and mark as initialized */ | |
af1584f5 | 3242 | ex->ee_len = cpu_to_le16(ee_len); |
ecb94f5f TT |
3243 | ext4_ext_try_to_merge(handle, inode, path, ex); |
3244 | err = ext4_ext_dirty(handle, inode, path + path->p_depth); | |
adb23551 ZL |
3245 | if (err) |
3246 | goto fix_extent_len; | |
3247 | ||
3248 | /* update extent status tree */ | |
d7b2a00c | 3249 | err = ext4_zeroout_es(inode, &zero_ex); |
adb23551 | 3250 | |
47ea3bb5 YY |
3251 | goto out; |
3252 | } else if (err) | |
3253 | goto fix_extent_len; | |
3254 | ||
3255 | out: | |
3256 | ext4_ext_show_leaf(inode, path); | |
3257 | return err; | |
3258 | ||
3259 | fix_extent_len: | |
3260 | ex->ee_len = orig_ex.ee_len; | |
29faed16 | 3261 | ext4_ext_dirty(handle, inode, path + path->p_depth); |
47ea3bb5 YY |
3262 | return err; |
3263 | } | |
3264 | ||
3265 | /* | |
3266 | * ext4_split_extents() splits an extent and mark extent which is covered | |
3267 | * by @map as split_flags indicates | |
3268 | * | |
70261f56 | 3269 | * It may result in splitting the extent into multiple extents (up to three) |
47ea3bb5 YY |
3270 | * There are three possibilities: |
3271 | * a> There is no split required | |
3272 | * b> Splits in two extents: Split is happening at either end of the extent | |
3273 | * c> Splits in three extents: Somone is splitting in middle of the extent | |
3274 | * | |
3275 | */ | |
3276 | static int ext4_split_extent(handle_t *handle, | |
3277 | struct inode *inode, | |
dfe50809 | 3278 | struct ext4_ext_path **ppath, |
47ea3bb5 YY |
3279 | struct ext4_map_blocks *map, |
3280 | int split_flag, | |
3281 | int flags) | |
3282 | { | |
dfe50809 | 3283 | struct ext4_ext_path *path = *ppath; |
47ea3bb5 YY |
3284 | ext4_lblk_t ee_block; |
3285 | struct ext4_extent *ex; | |
3286 | unsigned int ee_len, depth; | |
3287 | int err = 0; | |
556615dc | 3288 | int unwritten; |
47ea3bb5 | 3289 | int split_flag1, flags1; |
3a225670 | 3290 | int allocated = map->m_len; |
47ea3bb5 YY |
3291 | |
3292 | depth = ext_depth(inode); | |
3293 | ex = path[depth].p_ext; | |
3294 | ee_block = le32_to_cpu(ex->ee_block); | |
3295 | ee_len = ext4_ext_get_actual_len(ex); | |
556615dc | 3296 | unwritten = ext4_ext_is_unwritten(ex); |
47ea3bb5 YY |
3297 | |
3298 | if (map->m_lblk + map->m_len < ee_block + ee_len) { | |
dee1f973 | 3299 | split_flag1 = split_flag & EXT4_EXT_MAY_ZEROOUT; |
47ea3bb5 | 3300 | flags1 = flags | EXT4_GET_BLOCKS_PRE_IO; |
556615dc LC |
3301 | if (unwritten) |
3302 | split_flag1 |= EXT4_EXT_MARK_UNWRIT1 | | |
3303 | EXT4_EXT_MARK_UNWRIT2; | |
dee1f973 DM |
3304 | if (split_flag & EXT4_EXT_DATA_VALID2) |
3305 | split_flag1 |= EXT4_EXT_DATA_VALID1; | |
dfe50809 | 3306 | err = ext4_split_extent_at(handle, inode, ppath, |
47ea3bb5 | 3307 | map->m_lblk + map->m_len, split_flag1, flags1); |
93917411 YY |
3308 | if (err) |
3309 | goto out; | |
3a225670 ZL |
3310 | } else { |
3311 | allocated = ee_len - (map->m_lblk - ee_block); | |
47ea3bb5 | 3312 | } |
357b66fd DM |
3313 | /* |
3314 | * Update path is required because previous ext4_split_extent_at() may | |
3315 | * result in split of original leaf or extent zeroout. | |
3316 | */ | |
dfe50809 | 3317 | path = ext4_ext_find_extent(inode, map->m_lblk, ppath, 0); |
47ea3bb5 YY |
3318 | if (IS_ERR(path)) |
3319 | return PTR_ERR(path); | |
357b66fd DM |
3320 | depth = ext_depth(inode); |
3321 | ex = path[depth].p_ext; | |
a18ed359 DM |
3322 | if (!ex) { |
3323 | EXT4_ERROR_INODE(inode, "unexpected hole at %lu", | |
3324 | (unsigned long) map->m_lblk); | |
3325 | return -EIO; | |
3326 | } | |
556615dc | 3327 | unwritten = ext4_ext_is_unwritten(ex); |
357b66fd | 3328 | split_flag1 = 0; |
47ea3bb5 YY |
3329 | |
3330 | if (map->m_lblk >= ee_block) { | |
357b66fd | 3331 | split_flag1 = split_flag & EXT4_EXT_DATA_VALID2; |
556615dc LC |
3332 | if (unwritten) { |
3333 | split_flag1 |= EXT4_EXT_MARK_UNWRIT1; | |
357b66fd | 3334 | split_flag1 |= split_flag & (EXT4_EXT_MAY_ZEROOUT | |
556615dc | 3335 | EXT4_EXT_MARK_UNWRIT2); |
357b66fd | 3336 | } |
dfe50809 | 3337 | err = ext4_split_extent_at(handle, inode, ppath, |
47ea3bb5 YY |
3338 | map->m_lblk, split_flag1, flags); |
3339 | if (err) | |
3340 | goto out; | |
3341 | } | |
3342 | ||
3343 | ext4_ext_show_leaf(inode, path); | |
3344 | out: | |
3a225670 | 3345 | return err ? err : allocated; |
47ea3bb5 YY |
3346 | } |
3347 | ||
56055d3a | 3348 | /* |
e35fd660 | 3349 | * This function is called by ext4_ext_map_blocks() if someone tries to write |
556615dc | 3350 | * to an unwritten extent. It may result in splitting the unwritten |
25985edc | 3351 | * extent into multiple extents (up to three - one initialized and two |
556615dc | 3352 | * unwritten). |
56055d3a AA |
3353 | * There are three possibilities: |
3354 | * a> There is no split required: Entire extent should be initialized | |
3355 | * b> Splits in two extents: Write is happening at either end of the extent | |
3356 | * c> Splits in three extents: Somone is writing in middle of the extent | |
6f91bc5f EG |
3357 | * |
3358 | * Pre-conditions: | |
556615dc | 3359 | * - The extent pointed to by 'path' is unwritten. |
6f91bc5f EG |
3360 | * - The extent pointed to by 'path' contains a superset |
3361 | * of the logical span [map->m_lblk, map->m_lblk + map->m_len). | |
3362 | * | |
3363 | * Post-conditions on success: | |
3364 | * - the returned value is the number of blocks beyond map->l_lblk | |
3365 | * that are allocated and initialized. | |
3366 | * It is guaranteed to be >= map->m_len. | |
56055d3a | 3367 | */ |
725d26d3 | 3368 | static int ext4_ext_convert_to_initialized(handle_t *handle, |
e35fd660 TT |
3369 | struct inode *inode, |
3370 | struct ext4_map_blocks *map, | |
dfe50809 | 3371 | struct ext4_ext_path **ppath, |
27dd4385 | 3372 | int flags) |
56055d3a | 3373 | { |
dfe50809 | 3374 | struct ext4_ext_path *path = *ppath; |
67a5da56 | 3375 | struct ext4_sb_info *sbi; |
6f91bc5f | 3376 | struct ext4_extent_header *eh; |
667eff35 YY |
3377 | struct ext4_map_blocks split_map; |
3378 | struct ext4_extent zero_ex; | |
bc2d9db4 | 3379 | struct ext4_extent *ex, *abut_ex; |
21ca087a | 3380 | ext4_lblk_t ee_block, eof_block; |
bc2d9db4 LC |
3381 | unsigned int ee_len, depth, map_len = map->m_len; |
3382 | int allocated = 0, max_zeroout = 0; | |
56055d3a | 3383 | int err = 0; |
667eff35 | 3384 | int split_flag = 0; |
21ca087a DM |
3385 | |
3386 | ext_debug("ext4_ext_convert_to_initialized: inode %lu, logical" | |
3387 | "block %llu, max_blocks %u\n", inode->i_ino, | |
bc2d9db4 | 3388 | (unsigned long long)map->m_lblk, map_len); |
21ca087a | 3389 | |
67a5da56 | 3390 | sbi = EXT4_SB(inode->i_sb); |
21ca087a DM |
3391 | eof_block = (inode->i_size + inode->i_sb->s_blocksize - 1) >> |
3392 | inode->i_sb->s_blocksize_bits; | |
bc2d9db4 LC |
3393 | if (eof_block < map->m_lblk + map_len) |
3394 | eof_block = map->m_lblk + map_len; | |
56055d3a AA |
3395 | |
3396 | depth = ext_depth(inode); | |
6f91bc5f | 3397 | eh = path[depth].p_hdr; |
56055d3a AA |
3398 | ex = path[depth].p_ext; |
3399 | ee_block = le32_to_cpu(ex->ee_block); | |
3400 | ee_len = ext4_ext_get_actual_len(ex); | |
adb23551 | 3401 | zero_ex.ee_len = 0; |
56055d3a | 3402 | |
6f91bc5f EG |
3403 | trace_ext4_ext_convert_to_initialized_enter(inode, map, ex); |
3404 | ||
3405 | /* Pre-conditions */ | |
556615dc | 3406 | BUG_ON(!ext4_ext_is_unwritten(ex)); |
6f91bc5f | 3407 | BUG_ON(!in_range(map->m_lblk, ee_block, ee_len)); |
6f91bc5f EG |
3408 | |
3409 | /* | |
3410 | * Attempt to transfer newly initialized blocks from the currently | |
556615dc | 3411 | * unwritten extent to its neighbor. This is much cheaper |
6f91bc5f | 3412 | * than an insertion followed by a merge as those involve costly |
bc2d9db4 LC |
3413 | * memmove() calls. Transferring to the left is the common case in |
3414 | * steady state for workloads doing fallocate(FALLOC_FL_KEEP_SIZE) | |
3415 | * followed by append writes. | |
6f91bc5f EG |
3416 | * |
3417 | * Limitations of the current logic: | |
bc2d9db4 | 3418 | * - L1: we do not deal with writes covering the whole extent. |
6f91bc5f EG |
3419 | * This would require removing the extent if the transfer |
3420 | * is possible. | |
bc2d9db4 | 3421 | * - L2: we only attempt to merge with an extent stored in the |
6f91bc5f EG |
3422 | * same extent tree node. |
3423 | */ | |
bc2d9db4 LC |
3424 | if ((map->m_lblk == ee_block) && |
3425 | /* See if we can merge left */ | |
3426 | (map_len < ee_len) && /*L1*/ | |
3427 | (ex > EXT_FIRST_EXTENT(eh))) { /*L2*/ | |
6f91bc5f EG |
3428 | ext4_lblk_t prev_lblk; |
3429 | ext4_fsblk_t prev_pblk, ee_pblk; | |
bc2d9db4 | 3430 | unsigned int prev_len; |
6f91bc5f | 3431 | |
bc2d9db4 LC |
3432 | abut_ex = ex - 1; |
3433 | prev_lblk = le32_to_cpu(abut_ex->ee_block); | |
3434 | prev_len = ext4_ext_get_actual_len(abut_ex); | |
3435 | prev_pblk = ext4_ext_pblock(abut_ex); | |
6f91bc5f | 3436 | ee_pblk = ext4_ext_pblock(ex); |
6f91bc5f EG |
3437 | |
3438 | /* | |
bc2d9db4 | 3439 | * A transfer of blocks from 'ex' to 'abut_ex' is allowed |
6f91bc5f | 3440 | * upon those conditions: |
bc2d9db4 LC |
3441 | * - C1: abut_ex is initialized, |
3442 | * - C2: abut_ex is logically abutting ex, | |
3443 | * - C3: abut_ex is physically abutting ex, | |
3444 | * - C4: abut_ex can receive the additional blocks without | |
6f91bc5f EG |
3445 | * overflowing the (initialized) length limit. |
3446 | */ | |
556615dc | 3447 | if ((!ext4_ext_is_unwritten(abut_ex)) && /*C1*/ |
6f91bc5f EG |
3448 | ((prev_lblk + prev_len) == ee_block) && /*C2*/ |
3449 | ((prev_pblk + prev_len) == ee_pblk) && /*C3*/ | |
bc2d9db4 | 3450 | (prev_len < (EXT_INIT_MAX_LEN - map_len))) { /*C4*/ |
6f91bc5f EG |
3451 | err = ext4_ext_get_access(handle, inode, path + depth); |
3452 | if (err) | |
3453 | goto out; | |
3454 | ||
3455 | trace_ext4_ext_convert_to_initialized_fastpath(inode, | |
bc2d9db4 | 3456 | map, ex, abut_ex); |
6f91bc5f | 3457 | |
bc2d9db4 LC |
3458 | /* Shift the start of ex by 'map_len' blocks */ |
3459 | ex->ee_block = cpu_to_le32(ee_block + map_len); | |
3460 | ext4_ext_store_pblock(ex, ee_pblk + map_len); | |
3461 | ex->ee_len = cpu_to_le16(ee_len - map_len); | |
556615dc | 3462 | ext4_ext_mark_unwritten(ex); /* Restore the flag */ |
6f91bc5f | 3463 | |
bc2d9db4 LC |
3464 | /* Extend abut_ex by 'map_len' blocks */ |
3465 | abut_ex->ee_len = cpu_to_le16(prev_len + map_len); | |
6f91bc5f | 3466 | |
bc2d9db4 LC |
3467 | /* Result: number of initialized blocks past m_lblk */ |
3468 | allocated = map_len; | |
3469 | } | |
3470 | } else if (((map->m_lblk + map_len) == (ee_block + ee_len)) && | |
3471 | (map_len < ee_len) && /*L1*/ | |
3472 | ex < EXT_LAST_EXTENT(eh)) { /*L2*/ | |
3473 | /* See if we can merge right */ | |
3474 | ext4_lblk_t next_lblk; | |
3475 | ext4_fsblk_t next_pblk, ee_pblk; | |
3476 | unsigned int next_len; | |
3477 | ||
3478 | abut_ex = ex + 1; | |
3479 | next_lblk = le32_to_cpu(abut_ex->ee_block); | |
3480 | next_len = ext4_ext_get_actual_len(abut_ex); | |
3481 | next_pblk = ext4_ext_pblock(abut_ex); | |
3482 | ee_pblk = ext4_ext_pblock(ex); | |
6f91bc5f | 3483 | |
bc2d9db4 LC |
3484 | /* |
3485 | * A transfer of blocks from 'ex' to 'abut_ex' is allowed | |
3486 | * upon those conditions: | |
3487 | * - C1: abut_ex is initialized, | |
3488 | * - C2: abut_ex is logically abutting ex, | |
3489 | * - C3: abut_ex is physically abutting ex, | |
3490 | * - C4: abut_ex can receive the additional blocks without | |
3491 | * overflowing the (initialized) length limit. | |
3492 | */ | |
556615dc | 3493 | if ((!ext4_ext_is_unwritten(abut_ex)) && /*C1*/ |
bc2d9db4 LC |
3494 | ((map->m_lblk + map_len) == next_lblk) && /*C2*/ |
3495 | ((ee_pblk + ee_len) == next_pblk) && /*C3*/ | |
3496 | (next_len < (EXT_INIT_MAX_LEN - map_len))) { /*C4*/ | |
3497 | err = ext4_ext_get_access(handle, inode, path + depth); | |
3498 | if (err) | |
3499 | goto out; | |
3500 | ||
3501 | trace_ext4_ext_convert_to_initialized_fastpath(inode, | |
3502 | map, ex, abut_ex); | |
3503 | ||
3504 | /* Shift the start of abut_ex by 'map_len' blocks */ | |
3505 | abut_ex->ee_block = cpu_to_le32(next_lblk - map_len); | |
3506 | ext4_ext_store_pblock(abut_ex, next_pblk - map_len); | |
3507 | ex->ee_len = cpu_to_le16(ee_len - map_len); | |
556615dc | 3508 | ext4_ext_mark_unwritten(ex); /* Restore the flag */ |
bc2d9db4 LC |
3509 | |
3510 | /* Extend abut_ex by 'map_len' blocks */ | |
3511 | abut_ex->ee_len = cpu_to_le16(next_len + map_len); | |
6f91bc5f EG |
3512 | |
3513 | /* Result: number of initialized blocks past m_lblk */ | |
bc2d9db4 | 3514 | allocated = map_len; |
6f91bc5f EG |
3515 | } |
3516 | } | |
bc2d9db4 LC |
3517 | if (allocated) { |
3518 | /* Mark the block containing both extents as dirty */ | |
3519 | ext4_ext_dirty(handle, inode, path + depth); | |
3520 | ||
3521 | /* Update path to point to the right extent */ | |
3522 | path[depth].p_ext = abut_ex; | |
3523 | goto out; | |
3524 | } else | |
3525 | allocated = ee_len - (map->m_lblk - ee_block); | |
6f91bc5f | 3526 | |
667eff35 | 3527 | WARN_ON(map->m_lblk < ee_block); |
21ca087a DM |
3528 | /* |
3529 | * It is safe to convert extent to initialized via explicit | |
9e740568 | 3530 | * zeroout only if extent is fully inside i_size or new_size. |
21ca087a | 3531 | */ |
667eff35 | 3532 | split_flag |= ee_block + ee_len <= eof_block ? EXT4_EXT_MAY_ZEROOUT : 0; |
21ca087a | 3533 | |
67a5da56 ZL |
3534 | if (EXT4_EXT_MAY_ZEROOUT & split_flag) |
3535 | max_zeroout = sbi->s_extent_max_zeroout_kb >> | |
4f42f80a | 3536 | (inode->i_sb->s_blocksize_bits - 10); |
67a5da56 ZL |
3537 | |
3538 | /* If extent is less than s_max_zeroout_kb, zeroout directly */ | |
3539 | if (max_zeroout && (ee_len <= max_zeroout)) { | |
667eff35 | 3540 | err = ext4_ext_zeroout(inode, ex); |
3977c965 | 3541 | if (err) |
d03856bd | 3542 | goto out; |
adb23551 | 3543 | zero_ex.ee_block = ex->ee_block; |
8cde7ad1 | 3544 | zero_ex.ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)); |
adb23551 | 3545 | ext4_ext_store_pblock(&zero_ex, ext4_ext_pblock(ex)); |
d03856bd AK |
3546 | |
3547 | err = ext4_ext_get_access(handle, inode, path + depth); | |
3548 | if (err) | |
3549 | goto out; | |
667eff35 | 3550 | ext4_ext_mark_initialized(ex); |
ecb94f5f TT |
3551 | ext4_ext_try_to_merge(handle, inode, path, ex); |
3552 | err = ext4_ext_dirty(handle, inode, path + path->p_depth); | |
667eff35 | 3553 | goto out; |
56055d3a | 3554 | } |
667eff35 | 3555 | |
56055d3a | 3556 | /* |
667eff35 YY |
3557 | * four cases: |
3558 | * 1. split the extent into three extents. | |
3559 | * 2. split the extent into two extents, zeroout the first half. | |
3560 | * 3. split the extent into two extents, zeroout the second half. | |
3561 | * 4. split the extent into two extents with out zeroout. | |
56055d3a | 3562 | */ |
667eff35 YY |
3563 | split_map.m_lblk = map->m_lblk; |
3564 | split_map.m_len = map->m_len; | |
3565 | ||
67a5da56 ZL |
3566 | if (max_zeroout && (allocated > map->m_len)) { |
3567 | if (allocated <= max_zeroout) { | |
667eff35 YY |
3568 | /* case 3 */ |
3569 | zero_ex.ee_block = | |
9b940f8e AH |
3570 | cpu_to_le32(map->m_lblk); |
3571 | zero_ex.ee_len = cpu_to_le16(allocated); | |
667eff35 YY |
3572 | ext4_ext_store_pblock(&zero_ex, |
3573 | ext4_ext_pblock(ex) + map->m_lblk - ee_block); | |
3574 | err = ext4_ext_zeroout(inode, &zero_ex); | |
56055d3a AA |
3575 | if (err) |
3576 | goto out; | |
667eff35 YY |
3577 | split_map.m_lblk = map->m_lblk; |
3578 | split_map.m_len = allocated; | |
67a5da56 | 3579 | } else if (map->m_lblk - ee_block + map->m_len < max_zeroout) { |
667eff35 YY |
3580 | /* case 2 */ |
3581 | if (map->m_lblk != ee_block) { | |
3582 | zero_ex.ee_block = ex->ee_block; | |
3583 | zero_ex.ee_len = cpu_to_le16(map->m_lblk - | |
3584 | ee_block); | |
3585 | ext4_ext_store_pblock(&zero_ex, | |
3586 | ext4_ext_pblock(ex)); | |
3587 | err = ext4_ext_zeroout(inode, &zero_ex); | |
3588 | if (err) | |
3589 | goto out; | |
3590 | } | |
3591 | ||
667eff35 | 3592 | split_map.m_lblk = ee_block; |
9b940f8e AH |
3593 | split_map.m_len = map->m_lblk - ee_block + map->m_len; |
3594 | allocated = map->m_len; | |
56055d3a AA |
3595 | } |
3596 | } | |
667eff35 | 3597 | |
dfe50809 | 3598 | allocated = ext4_split_extent(handle, inode, ppath, |
27dd4385 | 3599 | &split_map, split_flag, flags); |
667eff35 YY |
3600 | if (allocated < 0) |
3601 | err = allocated; | |
3602 | ||
56055d3a | 3603 | out: |
adb23551 ZL |
3604 | /* If we have gotten a failure, don't zero out status tree */ |
3605 | if (!err) | |
d7b2a00c | 3606 | err = ext4_zeroout_es(inode, &zero_ex); |
56055d3a AA |
3607 | return err ? err : allocated; |
3608 | } | |
3609 | ||
0031462b | 3610 | /* |
e35fd660 | 3611 | * This function is called by ext4_ext_map_blocks() from |
0031462b | 3612 | * ext4_get_blocks_dio_write() when DIO to write |
556615dc | 3613 | * to an unwritten extent. |
0031462b | 3614 | * |
556615dc LC |
3615 | * Writing to an unwritten extent may result in splitting the unwritten |
3616 | * extent into multiple initialized/unwritten extents (up to three) | |
0031462b | 3617 | * There are three possibilities: |
556615dc | 3618 | * a> There is no split required: Entire extent should be unwritten |
0031462b MC |
3619 | * b> Splits in two extents: Write is happening at either end of the extent |
3620 | * c> Splits in three extents: Somone is writing in middle of the extent | |
3621 | * | |
b8a86845 LC |
3622 | * This works the same way in the case of initialized -> unwritten conversion. |
3623 | * | |
0031462b | 3624 | * One of more index blocks maybe needed if the extent tree grow after |
556615dc LC |
3625 | * the unwritten extent split. To prevent ENOSPC occur at the IO |
3626 | * complete, we need to split the unwritten extent before DIO submit | |
3627 | * the IO. The unwritten extent called at this time will be split | |
3628 | * into three unwritten extent(at most). After IO complete, the part | |
0031462b MC |
3629 | * being filled will be convert to initialized by the end_io callback function |
3630 | * via ext4_convert_unwritten_extents(). | |
ba230c3f | 3631 | * |
556615dc | 3632 | * Returns the size of unwritten extent to be written on success. |
0031462b | 3633 | */ |
b8a86845 | 3634 | static int ext4_split_convert_extents(handle_t *handle, |
0031462b | 3635 | struct inode *inode, |
e35fd660 | 3636 | struct ext4_map_blocks *map, |
dfe50809 | 3637 | struct ext4_ext_path **ppath, |
0031462b MC |
3638 | int flags) |
3639 | { | |
dfe50809 | 3640 | struct ext4_ext_path *path = *ppath; |
667eff35 YY |
3641 | ext4_lblk_t eof_block; |
3642 | ext4_lblk_t ee_block; | |
3643 | struct ext4_extent *ex; | |
3644 | unsigned int ee_len; | |
3645 | int split_flag = 0, depth; | |
21ca087a | 3646 | |
b8a86845 LC |
3647 | ext_debug("%s: inode %lu, logical block %llu, max_blocks %u\n", |
3648 | __func__, inode->i_ino, | |
3649 | (unsigned long long)map->m_lblk, map->m_len); | |
21ca087a DM |
3650 | |
3651 | eof_block = (inode->i_size + inode->i_sb->s_blocksize - 1) >> | |
3652 | inode->i_sb->s_blocksize_bits; | |
e35fd660 TT |
3653 | if (eof_block < map->m_lblk + map->m_len) |
3654 | eof_block = map->m_lblk + map->m_len; | |
21ca087a DM |
3655 | /* |
3656 | * It is safe to convert extent to initialized via explicit | |
3657 | * zeroout only if extent is fully insde i_size or new_size. | |
3658 | */ | |
667eff35 YY |
3659 | depth = ext_depth(inode); |
3660 | ex = path[depth].p_ext; | |
3661 | ee_block = le32_to_cpu(ex->ee_block); | |
3662 | ee_len = ext4_ext_get_actual_len(ex); | |
0031462b | 3663 | |
b8a86845 LC |
3664 | /* Convert to unwritten */ |
3665 | if (flags & EXT4_GET_BLOCKS_CONVERT_UNWRITTEN) { | |
3666 | split_flag |= EXT4_EXT_DATA_VALID1; | |
3667 | /* Convert to initialized */ | |
3668 | } else if (flags & EXT4_GET_BLOCKS_CONVERT) { | |
3669 | split_flag |= ee_block + ee_len <= eof_block ? | |
3670 | EXT4_EXT_MAY_ZEROOUT : 0; | |
556615dc | 3671 | split_flag |= (EXT4_EXT_MARK_UNWRIT2 | EXT4_EXT_DATA_VALID2); |
b8a86845 | 3672 | } |
667eff35 | 3673 | flags |= EXT4_GET_BLOCKS_PRE_IO; |
dfe50809 | 3674 | return ext4_split_extent(handle, inode, ppath, map, split_flag, flags); |
0031462b | 3675 | } |
197217a5 | 3676 | |
c7064ef1 | 3677 | static int ext4_convert_unwritten_extents_endio(handle_t *handle, |
dee1f973 DM |
3678 | struct inode *inode, |
3679 | struct ext4_map_blocks *map, | |
dfe50809 | 3680 | struct ext4_ext_path **ppath) |
0031462b | 3681 | { |
dfe50809 | 3682 | struct ext4_ext_path *path = *ppath; |
0031462b | 3683 | struct ext4_extent *ex; |
dee1f973 DM |
3684 | ext4_lblk_t ee_block; |
3685 | unsigned int ee_len; | |
0031462b MC |
3686 | int depth; |
3687 | int err = 0; | |
0031462b MC |
3688 | |
3689 | depth = ext_depth(inode); | |
0031462b | 3690 | ex = path[depth].p_ext; |
dee1f973 DM |
3691 | ee_block = le32_to_cpu(ex->ee_block); |
3692 | ee_len = ext4_ext_get_actual_len(ex); | |
0031462b | 3693 | |
197217a5 YY |
3694 | ext_debug("ext4_convert_unwritten_extents_endio: inode %lu, logical" |
3695 | "block %llu, max_blocks %u\n", inode->i_ino, | |
dee1f973 DM |
3696 | (unsigned long long)ee_block, ee_len); |
3697 | ||
ff95ec22 DM |
3698 | /* If extent is larger than requested it is a clear sign that we still |
3699 | * have some extent state machine issues left. So extent_split is still | |
3700 | * required. | |
3701 | * TODO: Once all related issues will be fixed this situation should be | |
3702 | * illegal. | |
3703 | */ | |
dee1f973 | 3704 | if (ee_block != map->m_lblk || ee_len > map->m_len) { |
ff95ec22 DM |
3705 | #ifdef EXT4_DEBUG |
3706 | ext4_warning("Inode (%ld) finished: extent logical block %llu," | |
3707 | " len %u; IO logical block %llu, len %u\n", | |
3708 | inode->i_ino, (unsigned long long)ee_block, ee_len, | |
3709 | (unsigned long long)map->m_lblk, map->m_len); | |
3710 | #endif | |
dfe50809 | 3711 | err = ext4_split_convert_extents(handle, inode, map, ppath, |
b8a86845 | 3712 | EXT4_GET_BLOCKS_CONVERT); |
dee1f973 | 3713 | if (err < 0) |
dfe50809 | 3714 | return err; |
dfe50809 TT |
3715 | path = ext4_ext_find_extent(inode, map->m_lblk, ppath, 0); |
3716 | if (IS_ERR(path)) | |
3717 | return PTR_ERR(path); | |
dee1f973 DM |
3718 | depth = ext_depth(inode); |
3719 | ex = path[depth].p_ext; | |
3720 | } | |
197217a5 | 3721 | |
0031462b MC |
3722 | err = ext4_ext_get_access(handle, inode, path + depth); |
3723 | if (err) | |
3724 | goto out; | |
3725 | /* first mark the extent as initialized */ | |
3726 | ext4_ext_mark_initialized(ex); | |
3727 | ||
197217a5 YY |
3728 | /* note: ext4_ext_correct_indexes() isn't needed here because |
3729 | * borders are not changed | |
0031462b | 3730 | */ |
ecb94f5f | 3731 | ext4_ext_try_to_merge(handle, inode, path, ex); |
197217a5 | 3732 | |
0031462b | 3733 | /* Mark modified extent as dirty */ |
ecb94f5f | 3734 | err = ext4_ext_dirty(handle, inode, path + path->p_depth); |
0031462b MC |
3735 | out: |
3736 | ext4_ext_show_leaf(inode, path); | |
3737 | return err; | |
3738 | } | |
3739 | ||
515f41c3 AK |
3740 | static void unmap_underlying_metadata_blocks(struct block_device *bdev, |
3741 | sector_t block, int count) | |
3742 | { | |
3743 | int i; | |
3744 | for (i = 0; i < count; i++) | |
3745 | unmap_underlying_metadata(bdev, block + i); | |
3746 | } | |
3747 | ||
58590b06 TT |
3748 | /* |
3749 | * Handle EOFBLOCKS_FL flag, clearing it if necessary | |
3750 | */ | |
3751 | static int check_eofblocks_fl(handle_t *handle, struct inode *inode, | |
d002ebf1 | 3752 | ext4_lblk_t lblk, |
58590b06 TT |
3753 | struct ext4_ext_path *path, |
3754 | unsigned int len) | |
3755 | { | |
3756 | int i, depth; | |
3757 | struct ext4_extent_header *eh; | |
65922cb5 | 3758 | struct ext4_extent *last_ex; |
58590b06 TT |
3759 | |
3760 | if (!ext4_test_inode_flag(inode, EXT4_INODE_EOFBLOCKS)) | |
3761 | return 0; | |
3762 | ||
3763 | depth = ext_depth(inode); | |
3764 | eh = path[depth].p_hdr; | |
58590b06 | 3765 | |
afcff5d8 LC |
3766 | /* |
3767 | * We're going to remove EOFBLOCKS_FL entirely in future so we | |
3768 | * do not care for this case anymore. Simply remove the flag | |
3769 | * if there are no extents. | |
3770 | */ | |
3771 | if (unlikely(!eh->eh_entries)) | |
3772 | goto out; | |
58590b06 TT |
3773 | last_ex = EXT_LAST_EXTENT(eh); |
3774 | /* | |
3775 | * We should clear the EOFBLOCKS_FL flag if we are writing the | |
3776 | * last block in the last extent in the file. We test this by | |
3777 | * first checking to see if the caller to | |
3778 | * ext4_ext_get_blocks() was interested in the last block (or | |
3779 | * a block beyond the last block) in the current extent. If | |
3780 | * this turns out to be false, we can bail out from this | |
3781 | * function immediately. | |
3782 | */ | |
d002ebf1 | 3783 | if (lblk + len < le32_to_cpu(last_ex->ee_block) + |
58590b06 TT |
3784 | ext4_ext_get_actual_len(last_ex)) |
3785 | return 0; | |
3786 | /* | |
3787 | * If the caller does appear to be planning to write at or | |
3788 | * beyond the end of the current extent, we then test to see | |
3789 | * if the current extent is the last extent in the file, by | |
3790 | * checking to make sure it was reached via the rightmost node | |
3791 | * at each level of the tree. | |
3792 | */ | |
3793 | for (i = depth-1; i >= 0; i--) | |
3794 | if (path[i].p_idx != EXT_LAST_INDEX(path[i].p_hdr)) | |
3795 | return 0; | |
afcff5d8 | 3796 | out: |
58590b06 TT |
3797 | ext4_clear_inode_flag(inode, EXT4_INODE_EOFBLOCKS); |
3798 | return ext4_mark_inode_dirty(handle, inode); | |
3799 | } | |
3800 | ||
7b415bf6 AK |
3801 | /** |
3802 | * ext4_find_delalloc_range: find delayed allocated block in the given range. | |
3803 | * | |
7d1b1fbc | 3804 | * Return 1 if there is a delalloc block in the range, otherwise 0. |
7b415bf6 | 3805 | */ |
f7fec032 ZL |
3806 | int ext4_find_delalloc_range(struct inode *inode, |
3807 | ext4_lblk_t lblk_start, | |
3808 | ext4_lblk_t lblk_end) | |
7b415bf6 | 3809 | { |
7d1b1fbc | 3810 | struct extent_status es; |
7b415bf6 | 3811 | |
e30b5dca | 3812 | ext4_es_find_delayed_extent_range(inode, lblk_start, lblk_end, &es); |
06b0c886 | 3813 | if (es.es_len == 0) |
7d1b1fbc | 3814 | return 0; /* there is no delay extent in this tree */ |
06b0c886 ZL |
3815 | else if (es.es_lblk <= lblk_start && |
3816 | lblk_start < es.es_lblk + es.es_len) | |
7d1b1fbc | 3817 | return 1; |
06b0c886 | 3818 | else if (lblk_start <= es.es_lblk && es.es_lblk <= lblk_end) |
7d1b1fbc | 3819 | return 1; |
7b415bf6 | 3820 | else |
7d1b1fbc | 3821 | return 0; |
7b415bf6 AK |
3822 | } |
3823 | ||
7d1b1fbc | 3824 | int ext4_find_delalloc_cluster(struct inode *inode, ext4_lblk_t lblk) |
7b415bf6 AK |
3825 | { |
3826 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); | |
3827 | ext4_lblk_t lblk_start, lblk_end; | |
f5a44db5 | 3828 | lblk_start = EXT4_LBLK_CMASK(sbi, lblk); |
7b415bf6 AK |
3829 | lblk_end = lblk_start + sbi->s_cluster_ratio - 1; |
3830 | ||
7d1b1fbc | 3831 | return ext4_find_delalloc_range(inode, lblk_start, lblk_end); |
7b415bf6 AK |
3832 | } |
3833 | ||
3834 | /** | |
3835 | * Determines how many complete clusters (out of those specified by the 'map') | |
3836 | * are under delalloc and were reserved quota for. | |
3837 | * This function is called when we are writing out the blocks that were | |
3838 | * originally written with their allocation delayed, but then the space was | |
3839 | * allocated using fallocate() before the delayed allocation could be resolved. | |
3840 | * The cases to look for are: | |
3841 | * ('=' indicated delayed allocated blocks | |
3842 | * '-' indicates non-delayed allocated blocks) | |
3843 | * (a) partial clusters towards beginning and/or end outside of allocated range | |
3844 | * are not delalloc'ed. | |
3845 | * Ex: | |
3846 | * |----c---=|====c====|====c====|===-c----| | |
3847 | * |++++++ allocated ++++++| | |
3848 | * ==> 4 complete clusters in above example | |
3849 | * | |
3850 | * (b) partial cluster (outside of allocated range) towards either end is | |
3851 | * marked for delayed allocation. In this case, we will exclude that | |
3852 | * cluster. | |
3853 | * Ex: | |
3854 | * |----====c========|========c========| | |
3855 | * |++++++ allocated ++++++| | |
3856 | * ==> 1 complete clusters in above example | |
3857 | * | |
3858 | * Ex: | |
3859 | * |================c================| | |
3860 | * |++++++ allocated ++++++| | |
3861 | * ==> 0 complete clusters in above example | |
3862 | * | |
3863 | * The ext4_da_update_reserve_space will be called only if we | |
3864 | * determine here that there were some "entire" clusters that span | |
3865 | * this 'allocated' range. | |
3866 | * In the non-bigalloc case, this function will just end up returning num_blks | |
3867 | * without ever calling ext4_find_delalloc_range. | |
3868 | */ | |
3869 | static unsigned int | |
3870 | get_reserved_cluster_alloc(struct inode *inode, ext4_lblk_t lblk_start, | |
3871 | unsigned int num_blks) | |
3872 | { | |
3873 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); | |
3874 | ext4_lblk_t alloc_cluster_start, alloc_cluster_end; | |
3875 | ext4_lblk_t lblk_from, lblk_to, c_offset; | |
3876 | unsigned int allocated_clusters = 0; | |
3877 | ||
3878 | alloc_cluster_start = EXT4_B2C(sbi, lblk_start); | |
3879 | alloc_cluster_end = EXT4_B2C(sbi, lblk_start + num_blks - 1); | |
3880 | ||
3881 | /* max possible clusters for this allocation */ | |
3882 | allocated_clusters = alloc_cluster_end - alloc_cluster_start + 1; | |
3883 | ||
d8990240 AK |
3884 | trace_ext4_get_reserved_cluster_alloc(inode, lblk_start, num_blks); |
3885 | ||
7b415bf6 | 3886 | /* Check towards left side */ |
f5a44db5 | 3887 | c_offset = EXT4_LBLK_COFF(sbi, lblk_start); |
7b415bf6 | 3888 | if (c_offset) { |
f5a44db5 | 3889 | lblk_from = EXT4_LBLK_CMASK(sbi, lblk_start); |
7b415bf6 AK |
3890 | lblk_to = lblk_from + c_offset - 1; |
3891 | ||
7d1b1fbc | 3892 | if (ext4_find_delalloc_range(inode, lblk_from, lblk_to)) |
7b415bf6 AK |
3893 | allocated_clusters--; |
3894 | } | |
3895 | ||
3896 | /* Now check towards right. */ | |
f5a44db5 | 3897 | c_offset = EXT4_LBLK_COFF(sbi, lblk_start + num_blks); |
7b415bf6 AK |
3898 | if (allocated_clusters && c_offset) { |
3899 | lblk_from = lblk_start + num_blks; | |
3900 | lblk_to = lblk_from + (sbi->s_cluster_ratio - c_offset) - 1; | |
3901 | ||
7d1b1fbc | 3902 | if (ext4_find_delalloc_range(inode, lblk_from, lblk_to)) |
7b415bf6 AK |
3903 | allocated_clusters--; |
3904 | } | |
3905 | ||
3906 | return allocated_clusters; | |
3907 | } | |
3908 | ||
b8a86845 | 3909 | static int |
e8b83d93 TT |
3910 | convert_initialized_extent(handle_t *handle, struct inode *inode, |
3911 | struct ext4_map_blocks *map, | |
4f224b8b | 3912 | struct ext4_ext_path **ppath, int flags, |
e8b83d93 | 3913 | unsigned int allocated, ext4_fsblk_t newblock) |
b8a86845 | 3914 | { |
4f224b8b | 3915 | struct ext4_ext_path *path = *ppath; |
e8b83d93 TT |
3916 | struct ext4_extent *ex; |
3917 | ext4_lblk_t ee_block; | |
3918 | unsigned int ee_len; | |
3919 | int depth; | |
b8a86845 LC |
3920 | int err = 0; |
3921 | ||
3922 | /* | |
3923 | * Make sure that the extent is no bigger than we support with | |
556615dc | 3924 | * unwritten extent |
b8a86845 | 3925 | */ |
556615dc LC |
3926 | if (map->m_len > EXT_UNWRITTEN_MAX_LEN) |
3927 | map->m_len = EXT_UNWRITTEN_MAX_LEN / 2; | |
b8a86845 | 3928 | |
e8b83d93 TT |
3929 | depth = ext_depth(inode); |
3930 | ex = path[depth].p_ext; | |
3931 | ee_block = le32_to_cpu(ex->ee_block); | |
3932 | ee_len = ext4_ext_get_actual_len(ex); | |
3933 | ||
3934 | ext_debug("%s: inode %lu, logical" | |
3935 | "block %llu, max_blocks %u\n", __func__, inode->i_ino, | |
3936 | (unsigned long long)ee_block, ee_len); | |
3937 | ||
3938 | if (ee_block != map->m_lblk || ee_len > map->m_len) { | |
dfe50809 | 3939 | err = ext4_split_convert_extents(handle, inode, map, ppath, |
e8b83d93 TT |
3940 | EXT4_GET_BLOCKS_CONVERT_UNWRITTEN); |
3941 | if (err < 0) | |
3942 | return err; | |
4f224b8b | 3943 | path = ext4_ext_find_extent(inode, map->m_lblk, ppath, 0); |
e8b83d93 TT |
3944 | if (IS_ERR(path)) |
3945 | return PTR_ERR(path); | |
3946 | depth = ext_depth(inode); | |
3947 | ex = path[depth].p_ext; | |
3948 | if (!ex) { | |
3949 | EXT4_ERROR_INODE(inode, "unexpected hole at %lu", | |
3950 | (unsigned long) map->m_lblk); | |
3951 | return -EIO; | |
3952 | } | |
3953 | } | |
3954 | ||
3955 | err = ext4_ext_get_access(handle, inode, path + depth); | |
3956 | if (err) | |
3957 | return err; | |
3958 | /* first mark the extent as unwritten */ | |
3959 | ext4_ext_mark_unwritten(ex); | |
3960 | ||
3961 | /* note: ext4_ext_correct_indexes() isn't needed here because | |
3962 | * borders are not changed | |
3963 | */ | |
3964 | ext4_ext_try_to_merge(handle, inode, path, ex); | |
3965 | ||
3966 | /* Mark modified extent as dirty */ | |
3967 | err = ext4_ext_dirty(handle, inode, path + path->p_depth); | |
3968 | if (err) | |
3969 | return err; | |
3970 | ext4_ext_show_leaf(inode, path); | |
3971 | ||
3972 | ext4_update_inode_fsync_trans(handle, inode, 1); | |
3973 | err = check_eofblocks_fl(handle, inode, map->m_lblk, path, map->m_len); | |
3974 | if (err) | |
3975 | return err; | |
b8a86845 LC |
3976 | map->m_flags |= EXT4_MAP_UNWRITTEN; |
3977 | if (allocated > map->m_len) | |
3978 | allocated = map->m_len; | |
3979 | map->m_len = allocated; | |
e8b83d93 | 3980 | return allocated; |
b8a86845 LC |
3981 | } |
3982 | ||
0031462b | 3983 | static int |
556615dc | 3984 | ext4_ext_handle_unwritten_extents(handle_t *handle, struct inode *inode, |
e35fd660 | 3985 | struct ext4_map_blocks *map, |
dfe50809 | 3986 | struct ext4_ext_path **ppath, int flags, |
e35fd660 | 3987 | unsigned int allocated, ext4_fsblk_t newblock) |
0031462b | 3988 | { |
dfe50809 | 3989 | struct ext4_ext_path *path = *ppath; |
0031462b MC |
3990 | int ret = 0; |
3991 | int err = 0; | |
f45ee3a1 | 3992 | ext4_io_end_t *io = ext4_inode_aio(inode); |
0031462b | 3993 | |
556615dc | 3994 | ext_debug("ext4_ext_handle_unwritten_extents: inode %lu, logical " |
88635ca2 | 3995 | "block %llu, max_blocks %u, flags %x, allocated %u\n", |
e35fd660 | 3996 | inode->i_ino, (unsigned long long)map->m_lblk, map->m_len, |
0031462b MC |
3997 | flags, allocated); |
3998 | ext4_ext_show_leaf(inode, path); | |
3999 | ||
27dd4385 | 4000 | /* |
556615dc | 4001 | * When writing into unwritten space, we should not fail to |
27dd4385 LC |
4002 | * allocate metadata blocks for the new extent block if needed. |
4003 | */ | |
4004 | flags |= EXT4_GET_BLOCKS_METADATA_NOFAIL; | |
4005 | ||
556615dc | 4006 | trace_ext4_ext_handle_unwritten_extents(inode, map, flags, |
b5645534 | 4007 | allocated, newblock); |
d8990240 | 4008 | |
c7064ef1 | 4009 | /* get_block() before submit the IO, split the extent */ |
c8b459f4 | 4010 | if (flags & EXT4_GET_BLOCKS_PRE_IO) { |
dfe50809 TT |
4011 | ret = ext4_split_convert_extents(handle, inode, map, ppath, |
4012 | flags | EXT4_GET_BLOCKS_CONVERT); | |
82e54229 DM |
4013 | if (ret <= 0) |
4014 | goto out; | |
5f524950 M |
4015 | /* |
4016 | * Flag the inode(non aio case) or end_io struct (aio case) | |
25985edc | 4017 | * that this IO needs to conversion to written when IO is |
5f524950 M |
4018 | * completed |
4019 | */ | |
0edeb71d TM |
4020 | if (io) |
4021 | ext4_set_io_unwritten_flag(inode, io); | |
4022 | else | |
19f5fb7a | 4023 | ext4_set_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN); |
a25a4e1a | 4024 | map->m_flags |= EXT4_MAP_UNWRITTEN; |
0031462b MC |
4025 | goto out; |
4026 | } | |
c7064ef1 | 4027 | /* IO end_io complete, convert the filled extent to written */ |
c8b459f4 | 4028 | if (flags & EXT4_GET_BLOCKS_CONVERT) { |
dee1f973 | 4029 | ret = ext4_convert_unwritten_extents_endio(handle, inode, map, |
dfe50809 | 4030 | ppath); |
58590b06 | 4031 | if (ret >= 0) { |
b436b9be | 4032 | ext4_update_inode_fsync_trans(handle, inode, 1); |
d002ebf1 ES |
4033 | err = check_eofblocks_fl(handle, inode, map->m_lblk, |
4034 | path, map->m_len); | |
58590b06 TT |
4035 | } else |
4036 | err = ret; | |
cdee7843 | 4037 | map->m_flags |= EXT4_MAP_MAPPED; |
15cc1767 | 4038 | map->m_pblk = newblock; |
cdee7843 ZL |
4039 | if (allocated > map->m_len) |
4040 | allocated = map->m_len; | |
4041 | map->m_len = allocated; | |
0031462b MC |
4042 | goto out2; |
4043 | } | |
4044 | /* buffered IO case */ | |
4045 | /* | |
4046 | * repeat fallocate creation request | |
4047 | * we already have an unwritten extent | |
4048 | */ | |
556615dc | 4049 | if (flags & EXT4_GET_BLOCKS_UNWRIT_EXT) { |
a25a4e1a | 4050 | map->m_flags |= EXT4_MAP_UNWRITTEN; |
0031462b | 4051 | goto map_out; |
a25a4e1a | 4052 | } |
0031462b MC |
4053 | |
4054 | /* buffered READ or buffered write_begin() lookup */ | |
4055 | if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) { | |
4056 | /* | |
4057 | * We have blocks reserved already. We | |
4058 | * return allocated blocks so that delalloc | |
4059 | * won't do block reservation for us. But | |
4060 | * the buffer head will be unmapped so that | |
4061 | * a read from the block returns 0s. | |
4062 | */ | |
e35fd660 | 4063 | map->m_flags |= EXT4_MAP_UNWRITTEN; |
0031462b MC |
4064 | goto out1; |
4065 | } | |
4066 | ||
4067 | /* buffered write, writepage time, convert*/ | |
dfe50809 | 4068 | ret = ext4_ext_convert_to_initialized(handle, inode, map, ppath, flags); |
a4e5d88b | 4069 | if (ret >= 0) |
b436b9be | 4070 | ext4_update_inode_fsync_trans(handle, inode, 1); |
0031462b MC |
4071 | out: |
4072 | if (ret <= 0) { | |
4073 | err = ret; | |
4074 | goto out2; | |
4075 | } else | |
4076 | allocated = ret; | |
e35fd660 | 4077 | map->m_flags |= EXT4_MAP_NEW; |
515f41c3 AK |
4078 | /* |
4079 | * if we allocated more blocks than requested | |
4080 | * we need to make sure we unmap the extra block | |
4081 | * allocated. The actual needed block will get | |
4082 | * unmapped later when we find the buffer_head marked | |
4083 | * new. | |
4084 | */ | |
e35fd660 | 4085 | if (allocated > map->m_len) { |
515f41c3 | 4086 | unmap_underlying_metadata_blocks(inode->i_sb->s_bdev, |
e35fd660 TT |
4087 | newblock + map->m_len, |
4088 | allocated - map->m_len); | |
4089 | allocated = map->m_len; | |
515f41c3 | 4090 | } |
3a225670 | 4091 | map->m_len = allocated; |
5f634d06 AK |
4092 | |
4093 | /* | |
4094 | * If we have done fallocate with the offset that is already | |
4095 | * delayed allocated, we would have block reservation | |
4096 | * and quota reservation done in the delayed write path. | |
4097 | * But fallocate would have already updated quota and block | |
4098 | * count for this offset. So cancel these reservation | |
4099 | */ | |
7b415bf6 AK |
4100 | if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) { |
4101 | unsigned int reserved_clusters; | |
4102 | reserved_clusters = get_reserved_cluster_alloc(inode, | |
4103 | map->m_lblk, map->m_len); | |
4104 | if (reserved_clusters) | |
4105 | ext4_da_update_reserve_space(inode, | |
4106 | reserved_clusters, | |
4107 | 0); | |
4108 | } | |
5f634d06 | 4109 | |
0031462b | 4110 | map_out: |
e35fd660 | 4111 | map->m_flags |= EXT4_MAP_MAPPED; |
a4e5d88b DM |
4112 | if ((flags & EXT4_GET_BLOCKS_KEEP_SIZE) == 0) { |
4113 | err = check_eofblocks_fl(handle, inode, map->m_lblk, path, | |
4114 | map->m_len); | |
4115 | if (err < 0) | |
4116 | goto out2; | |
4117 | } | |
0031462b | 4118 | out1: |
e35fd660 TT |
4119 | if (allocated > map->m_len) |
4120 | allocated = map->m_len; | |
0031462b | 4121 | ext4_ext_show_leaf(inode, path); |
e35fd660 TT |
4122 | map->m_pblk = newblock; |
4123 | map->m_len = allocated; | |
0031462b | 4124 | out2: |
0031462b MC |
4125 | return err ? err : allocated; |
4126 | } | |
58590b06 | 4127 | |
4d33b1ef TT |
4128 | /* |
4129 | * get_implied_cluster_alloc - check to see if the requested | |
4130 | * allocation (in the map structure) overlaps with a cluster already | |
4131 | * allocated in an extent. | |
d8990240 | 4132 | * @sb The filesystem superblock structure |
4d33b1ef TT |
4133 | * @map The requested lblk->pblk mapping |
4134 | * @ex The extent structure which might contain an implied | |
4135 | * cluster allocation | |
4136 | * | |
4137 | * This function is called by ext4_ext_map_blocks() after we failed to | |
4138 | * find blocks that were already in the inode's extent tree. Hence, | |
4139 | * we know that the beginning of the requested region cannot overlap | |
4140 | * the extent from the inode's extent tree. There are three cases we | |
4141 | * want to catch. The first is this case: | |
4142 | * | |
4143 | * |--- cluster # N--| | |
4144 | * |--- extent ---| |---- requested region ---| | |
4145 | * |==========| | |
4146 | * | |
4147 | * The second case that we need to test for is this one: | |
4148 | * | |
4149 | * |--------- cluster # N ----------------| | |
4150 | * |--- requested region --| |------- extent ----| | |
4151 | * |=======================| | |
4152 | * | |
4153 | * The third case is when the requested region lies between two extents | |
4154 | * within the same cluster: | |
4155 | * |------------- cluster # N-------------| | |
4156 | * |----- ex -----| |---- ex_right ----| | |
4157 | * |------ requested region ------| | |
4158 | * |================| | |
4159 | * | |
4160 | * In each of the above cases, we need to set the map->m_pblk and | |
4161 | * map->m_len so it corresponds to the return the extent labelled as | |
4162 | * "|====|" from cluster #N, since it is already in use for data in | |
4163 | * cluster EXT4_B2C(sbi, map->m_lblk). We will then return 1 to | |
4164 | * signal to ext4_ext_map_blocks() that map->m_pblk should be treated | |
4165 | * as a new "allocated" block region. Otherwise, we will return 0 and | |
4166 | * ext4_ext_map_blocks() will then allocate one or more new clusters | |
4167 | * by calling ext4_mb_new_blocks(). | |
4168 | */ | |
d8990240 | 4169 | static int get_implied_cluster_alloc(struct super_block *sb, |
4d33b1ef TT |
4170 | struct ext4_map_blocks *map, |
4171 | struct ext4_extent *ex, | |
4172 | struct ext4_ext_path *path) | |
4173 | { | |
d8990240 | 4174 | struct ext4_sb_info *sbi = EXT4_SB(sb); |
f5a44db5 | 4175 | ext4_lblk_t c_offset = EXT4_LBLK_COFF(sbi, map->m_lblk); |
4d33b1ef | 4176 | ext4_lblk_t ex_cluster_start, ex_cluster_end; |
14d7f3ef | 4177 | ext4_lblk_t rr_cluster_start; |
4d33b1ef TT |
4178 | ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block); |
4179 | ext4_fsblk_t ee_start = ext4_ext_pblock(ex); | |
4180 | unsigned short ee_len = ext4_ext_get_actual_len(ex); | |
4181 | ||
4182 | /* The extent passed in that we are trying to match */ | |
4183 | ex_cluster_start = EXT4_B2C(sbi, ee_block); | |
4184 | ex_cluster_end = EXT4_B2C(sbi, ee_block + ee_len - 1); | |
4185 | ||
4186 | /* The requested region passed into ext4_map_blocks() */ | |
4187 | rr_cluster_start = EXT4_B2C(sbi, map->m_lblk); | |
4d33b1ef TT |
4188 | |
4189 | if ((rr_cluster_start == ex_cluster_end) || | |
4190 | (rr_cluster_start == ex_cluster_start)) { | |
4191 | if (rr_cluster_start == ex_cluster_end) | |
4192 | ee_start += ee_len - 1; | |
f5a44db5 | 4193 | map->m_pblk = EXT4_PBLK_CMASK(sbi, ee_start) + c_offset; |
4d33b1ef TT |
4194 | map->m_len = min(map->m_len, |
4195 | (unsigned) sbi->s_cluster_ratio - c_offset); | |
4196 | /* | |
4197 | * Check for and handle this case: | |
4198 | * | |
4199 | * |--------- cluster # N-------------| | |
4200 | * |------- extent ----| | |
4201 | * |--- requested region ---| | |
4202 | * |===========| | |
4203 | */ | |
4204 | ||
4205 | if (map->m_lblk < ee_block) | |
4206 | map->m_len = min(map->m_len, ee_block - map->m_lblk); | |
4207 | ||
4208 | /* | |
4209 | * Check for the case where there is already another allocated | |
4210 | * block to the right of 'ex' but before the end of the cluster. | |
4211 | * | |
4212 | * |------------- cluster # N-------------| | |
4213 | * |----- ex -----| |---- ex_right ----| | |
4214 | * |------ requested region ------| | |
4215 | * |================| | |
4216 | */ | |
4217 | if (map->m_lblk > ee_block) { | |
4218 | ext4_lblk_t next = ext4_ext_next_allocated_block(path); | |
4219 | map->m_len = min(map->m_len, next - map->m_lblk); | |
4220 | } | |
d8990240 AK |
4221 | |
4222 | trace_ext4_get_implied_cluster_alloc_exit(sb, map, 1); | |
4d33b1ef TT |
4223 | return 1; |
4224 | } | |
d8990240 AK |
4225 | |
4226 | trace_ext4_get_implied_cluster_alloc_exit(sb, map, 0); | |
4d33b1ef TT |
4227 | return 0; |
4228 | } | |
4229 | ||
4230 | ||
c278bfec | 4231 | /* |
f5ab0d1f MC |
4232 | * Block allocation/map/preallocation routine for extents based files |
4233 | * | |
4234 | * | |
c278bfec | 4235 | * Need to be called with |
0e855ac8 AK |
4236 | * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block |
4237 | * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem) | |
f5ab0d1f MC |
4238 | * |
4239 | * return > 0, number of of blocks already mapped/allocated | |
4240 | * if create == 0 and these are pre-allocated blocks | |
4241 | * buffer head is unmapped | |
4242 | * otherwise blocks are mapped | |
4243 | * | |
4244 | * return = 0, if plain look up failed (blocks have not been allocated) | |
4245 | * buffer head is unmapped | |
4246 | * | |
4247 | * return < 0, error case. | |
c278bfec | 4248 | */ |
e35fd660 TT |
4249 | int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, |
4250 | struct ext4_map_blocks *map, int flags) | |
a86c6181 AT |
4251 | { |
4252 | struct ext4_ext_path *path = NULL; | |
4d33b1ef TT |
4253 | struct ext4_extent newex, *ex, *ex2; |
4254 | struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); | |
0562e0ba | 4255 | ext4_fsblk_t newblock = 0; |
ce37c429 | 4256 | int free_on_err = 0, err = 0, depth, ret; |
4d33b1ef | 4257 | unsigned int allocated = 0, offset = 0; |
81fdbb4a | 4258 | unsigned int allocated_clusters = 0; |
c9de560d | 4259 | struct ext4_allocation_request ar; |
f45ee3a1 | 4260 | ext4_io_end_t *io = ext4_inode_aio(inode); |
4d33b1ef | 4261 | ext4_lblk_t cluster_offset; |
82e54229 | 4262 | int set_unwritten = 0; |
a86c6181 | 4263 | |
84fe3bef | 4264 | ext_debug("blocks %u/%u requested for inode %lu\n", |
e35fd660 | 4265 | map->m_lblk, map->m_len, inode->i_ino); |
0562e0ba | 4266 | trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags); |
a86c6181 | 4267 | |
a86c6181 | 4268 | /* find extent for this block */ |
107a7bd3 | 4269 | path = ext4_ext_find_extent(inode, map->m_lblk, NULL, 0); |
a86c6181 AT |
4270 | if (IS_ERR(path)) { |
4271 | err = PTR_ERR(path); | |
4272 | path = NULL; | |
4273 | goto out2; | |
4274 | } | |
4275 | ||
4276 | depth = ext_depth(inode); | |
4277 | ||
4278 | /* | |
d0d856e8 RD |
4279 | * consistent leaf must not be empty; |
4280 | * this situation is possible, though, _during_ tree modification; | |
a86c6181 AT |
4281 | * this is why assert can't be put in ext4_ext_find_extent() |
4282 | */ | |
273df556 FM |
4283 | if (unlikely(path[depth].p_ext == NULL && depth != 0)) { |
4284 | EXT4_ERROR_INODE(inode, "bad extent address " | |
f70f362b TT |
4285 | "lblock: %lu, depth: %d pblock %lld", |
4286 | (unsigned long) map->m_lblk, depth, | |
4287 | path[depth].p_block); | |
034fb4c9 SP |
4288 | err = -EIO; |
4289 | goto out2; | |
4290 | } | |
a86c6181 | 4291 | |
7e028976 AM |
4292 | ex = path[depth].p_ext; |
4293 | if (ex) { | |
725d26d3 | 4294 | ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block); |
bf89d16f | 4295 | ext4_fsblk_t ee_start = ext4_ext_pblock(ex); |
a2df2a63 | 4296 | unsigned short ee_len; |
471d4011 | 4297 | |
b8a86845 | 4298 | |
471d4011 | 4299 | /* |
556615dc | 4300 | * unwritten extents are treated as holes, except that |
56055d3a | 4301 | * we split out initialized portions during a write. |
471d4011 | 4302 | */ |
a2df2a63 | 4303 | ee_len = ext4_ext_get_actual_len(ex); |
d8990240 AK |
4304 | |
4305 | trace_ext4_ext_show_extent(inode, ee_block, ee_start, ee_len); | |
4306 | ||
d0d856e8 | 4307 | /* if found extent covers block, simply return it */ |
e35fd660 TT |
4308 | if (in_range(map->m_lblk, ee_block, ee_len)) { |
4309 | newblock = map->m_lblk - ee_block + ee_start; | |
d0d856e8 | 4310 | /* number of remaining blocks in the extent */ |
e35fd660 TT |
4311 | allocated = ee_len - (map->m_lblk - ee_block); |
4312 | ext_debug("%u fit into %u:%d -> %llu\n", map->m_lblk, | |
4313 | ee_block, ee_len, newblock); | |
56055d3a | 4314 | |
b8a86845 LC |
4315 | /* |
4316 | * If the extent is initialized check whether the | |
4317 | * caller wants to convert it to unwritten. | |
4318 | */ | |
556615dc | 4319 | if ((!ext4_ext_is_unwritten(ex)) && |
b8a86845 | 4320 | (flags & EXT4_GET_BLOCKS_CONVERT_UNWRITTEN)) { |
e8b83d93 | 4321 | allocated = convert_initialized_extent( |
4f224b8b TT |
4322 | handle, inode, map, &path, |
4323 | flags, allocated, newblock); | |
b8a86845 | 4324 | goto out2; |
556615dc | 4325 | } else if (!ext4_ext_is_unwritten(ex)) |
7877191c | 4326 | goto out; |
69eb33dc | 4327 | |
556615dc | 4328 | ret = ext4_ext_handle_unwritten_extents( |
dfe50809 | 4329 | handle, inode, map, &path, flags, |
7877191c | 4330 | allocated, newblock); |
ce37c429 EW |
4331 | if (ret < 0) |
4332 | err = ret; | |
4333 | else | |
4334 | allocated = ret; | |
31cf0f2c | 4335 | goto out2; |
a86c6181 AT |
4336 | } |
4337 | } | |
4338 | ||
7b415bf6 | 4339 | if ((sbi->s_cluster_ratio > 1) && |
7d1b1fbc | 4340 | ext4_find_delalloc_cluster(inode, map->m_lblk)) |
7b415bf6 AK |
4341 | map->m_flags |= EXT4_MAP_FROM_CLUSTER; |
4342 | ||
a86c6181 | 4343 | /* |
d0d856e8 | 4344 | * requested block isn't allocated yet; |
a86c6181 AT |
4345 | * we couldn't try to create block if create flag is zero |
4346 | */ | |
c2177057 | 4347 | if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) { |
56055d3a AA |
4348 | /* |
4349 | * put just found gap into cache to speed up | |
4350 | * subsequent requests | |
4351 | */ | |
d100eef2 ZL |
4352 | if ((flags & EXT4_GET_BLOCKS_NO_PUT_HOLE) == 0) |
4353 | ext4_ext_put_gap_in_cache(inode, path, map->m_lblk); | |
a86c6181 AT |
4354 | goto out2; |
4355 | } | |
4d33b1ef | 4356 | |
a86c6181 | 4357 | /* |
c2ea3fde | 4358 | * Okay, we need to do block allocation. |
63f57933 | 4359 | */ |
7b415bf6 | 4360 | map->m_flags &= ~EXT4_MAP_FROM_CLUSTER; |
4d33b1ef | 4361 | newex.ee_block = cpu_to_le32(map->m_lblk); |
d0abafac | 4362 | cluster_offset = EXT4_LBLK_COFF(sbi, map->m_lblk); |
4d33b1ef TT |
4363 | |
4364 | /* | |
4365 | * If we are doing bigalloc, check to see if the extent returned | |
4366 | * by ext4_ext_find_extent() implies a cluster we can use. | |
4367 | */ | |
4368 | if (cluster_offset && ex && | |
d8990240 | 4369 | get_implied_cluster_alloc(inode->i_sb, map, ex, path)) { |
4d33b1ef TT |
4370 | ar.len = allocated = map->m_len; |
4371 | newblock = map->m_pblk; | |
7b415bf6 | 4372 | map->m_flags |= EXT4_MAP_FROM_CLUSTER; |
4d33b1ef TT |
4373 | goto got_allocated_blocks; |
4374 | } | |
a86c6181 | 4375 | |
c9de560d | 4376 | /* find neighbour allocated blocks */ |
e35fd660 | 4377 | ar.lleft = map->m_lblk; |
c9de560d AT |
4378 | err = ext4_ext_search_left(inode, path, &ar.lleft, &ar.pleft); |
4379 | if (err) | |
4380 | goto out2; | |
e35fd660 | 4381 | ar.lright = map->m_lblk; |
4d33b1ef TT |
4382 | ex2 = NULL; |
4383 | err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright, &ex2); | |
c9de560d AT |
4384 | if (err) |
4385 | goto out2; | |
25d14f98 | 4386 | |
4d33b1ef TT |
4387 | /* Check if the extent after searching to the right implies a |
4388 | * cluster we can use. */ | |
4389 | if ((sbi->s_cluster_ratio > 1) && ex2 && | |
d8990240 | 4390 | get_implied_cluster_alloc(inode->i_sb, map, ex2, path)) { |
4d33b1ef TT |
4391 | ar.len = allocated = map->m_len; |
4392 | newblock = map->m_pblk; | |
7b415bf6 | 4393 | map->m_flags |= EXT4_MAP_FROM_CLUSTER; |
4d33b1ef TT |
4394 | goto got_allocated_blocks; |
4395 | } | |
4396 | ||
749269fa AA |
4397 | /* |
4398 | * See if request is beyond maximum number of blocks we can have in | |
4399 | * a single extent. For an initialized extent this limit is | |
556615dc LC |
4400 | * EXT_INIT_MAX_LEN and for an unwritten extent this limit is |
4401 | * EXT_UNWRITTEN_MAX_LEN. | |
749269fa | 4402 | */ |
e35fd660 | 4403 | if (map->m_len > EXT_INIT_MAX_LEN && |
556615dc | 4404 | !(flags & EXT4_GET_BLOCKS_UNWRIT_EXT)) |
e35fd660 | 4405 | map->m_len = EXT_INIT_MAX_LEN; |
556615dc LC |
4406 | else if (map->m_len > EXT_UNWRITTEN_MAX_LEN && |
4407 | (flags & EXT4_GET_BLOCKS_UNWRIT_EXT)) | |
4408 | map->m_len = EXT_UNWRITTEN_MAX_LEN; | |
749269fa | 4409 | |
e35fd660 | 4410 | /* Check if we can really insert (m_lblk)::(m_lblk + m_len) extent */ |
e35fd660 | 4411 | newex.ee_len = cpu_to_le16(map->m_len); |
4d33b1ef | 4412 | err = ext4_ext_check_overlap(sbi, inode, &newex, path); |
25d14f98 | 4413 | if (err) |
b939e376 | 4414 | allocated = ext4_ext_get_actual_len(&newex); |
25d14f98 | 4415 | else |
e35fd660 | 4416 | allocated = map->m_len; |
c9de560d AT |
4417 | |
4418 | /* allocate new block */ | |
4419 | ar.inode = inode; | |
e35fd660 TT |
4420 | ar.goal = ext4_ext_find_goal(inode, path, map->m_lblk); |
4421 | ar.logical = map->m_lblk; | |
4d33b1ef TT |
4422 | /* |
4423 | * We calculate the offset from the beginning of the cluster | |
4424 | * for the logical block number, since when we allocate a | |
4425 | * physical cluster, the physical block should start at the | |
4426 | * same offset from the beginning of the cluster. This is | |
4427 | * needed so that future calls to get_implied_cluster_alloc() | |
4428 | * work correctly. | |
4429 | */ | |
f5a44db5 | 4430 | offset = EXT4_LBLK_COFF(sbi, map->m_lblk); |
4d33b1ef TT |
4431 | ar.len = EXT4_NUM_B2C(sbi, offset+allocated); |
4432 | ar.goal -= offset; | |
4433 | ar.logical -= offset; | |
c9de560d AT |
4434 | if (S_ISREG(inode->i_mode)) |
4435 | ar.flags = EXT4_MB_HINT_DATA; | |
4436 | else | |
4437 | /* disable in-core preallocation for non-regular files */ | |
4438 | ar.flags = 0; | |
556b27ab VH |
4439 | if (flags & EXT4_GET_BLOCKS_NO_NORMALIZE) |
4440 | ar.flags |= EXT4_MB_HINT_NOPREALLOC; | |
c9de560d | 4441 | newblock = ext4_mb_new_blocks(handle, &ar, &err); |
a86c6181 AT |
4442 | if (!newblock) |
4443 | goto out2; | |
84fe3bef | 4444 | ext_debug("allocate new block: goal %llu, found %llu/%u\n", |
498e5f24 | 4445 | ar.goal, newblock, allocated); |
4d33b1ef | 4446 | free_on_err = 1; |
7b415bf6 | 4447 | allocated_clusters = ar.len; |
4d33b1ef TT |
4448 | ar.len = EXT4_C2B(sbi, ar.len) - offset; |
4449 | if (ar.len > allocated) | |
4450 | ar.len = allocated; | |
a86c6181 | 4451 | |
4d33b1ef | 4452 | got_allocated_blocks: |
a86c6181 | 4453 | /* try to insert new extent into found leaf and return */ |
4d33b1ef | 4454 | ext4_ext_store_pblock(&newex, newblock + offset); |
c9de560d | 4455 | newex.ee_len = cpu_to_le16(ar.len); |
556615dc LC |
4456 | /* Mark unwritten */ |
4457 | if (flags & EXT4_GET_BLOCKS_UNWRIT_EXT){ | |
4458 | ext4_ext_mark_unwritten(&newex); | |
a25a4e1a | 4459 | map->m_flags |= EXT4_MAP_UNWRITTEN; |
8d5d02e6 | 4460 | /* |
744692dc | 4461 | * io_end structure was created for every IO write to an |
556615dc | 4462 | * unwritten extent. To avoid unnecessary conversion, |
744692dc | 4463 | * here we flag the IO that really needs the conversion. |
5f524950 | 4464 | * For non asycn direct IO case, flag the inode state |
25985edc | 4465 | * that we need to perform conversion when IO is done. |
8d5d02e6 | 4466 | */ |
c8b459f4 | 4467 | if (flags & EXT4_GET_BLOCKS_PRE_IO) |
82e54229 | 4468 | set_unwritten = 1; |
8d5d02e6 | 4469 | } |
c8d46e41 | 4470 | |
a4e5d88b DM |
4471 | err = 0; |
4472 | if ((flags & EXT4_GET_BLOCKS_KEEP_SIZE) == 0) | |
4473 | err = check_eofblocks_fl(handle, inode, map->m_lblk, | |
4474 | path, ar.len); | |
575a1d4b | 4475 | if (!err) |
dfe50809 | 4476 | err = ext4_ext_insert_extent(handle, inode, &path, |
575a1d4b | 4477 | &newex, flags); |
82e54229 DM |
4478 | |
4479 | if (!err && set_unwritten) { | |
4480 | if (io) | |
4481 | ext4_set_io_unwritten_flag(inode, io); | |
4482 | else | |
4483 | ext4_set_inode_state(inode, | |
4484 | EXT4_STATE_DIO_UNWRITTEN); | |
4485 | } | |
4486 | ||
4d33b1ef | 4487 | if (err && free_on_err) { |
7132de74 MP |
4488 | int fb_flags = flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE ? |
4489 | EXT4_FREE_BLOCKS_NO_QUOT_UPDATE : 0; | |
315054f0 | 4490 | /* free data blocks we just allocated */ |
c9de560d AT |
4491 | /* not a good idea to call discard here directly, |
4492 | * but otherwise we'd need to call it every free() */ | |
c2ea3fde | 4493 | ext4_discard_preallocations(inode); |
c8e15130 TT |
4494 | ext4_free_blocks(handle, inode, NULL, newblock, |
4495 | EXT4_C2B(sbi, allocated_clusters), fb_flags); | |
a86c6181 | 4496 | goto out2; |
315054f0 | 4497 | } |
a86c6181 | 4498 | |
a86c6181 | 4499 | /* previous routine could use block we allocated */ |
bf89d16f | 4500 | newblock = ext4_ext_pblock(&newex); |
b939e376 | 4501 | allocated = ext4_ext_get_actual_len(&newex); |
e35fd660 TT |
4502 | if (allocated > map->m_len) |
4503 | allocated = map->m_len; | |
4504 | map->m_flags |= EXT4_MAP_NEW; | |
a86c6181 | 4505 | |
5f634d06 AK |
4506 | /* |
4507 | * Update reserved blocks/metadata blocks after successful | |
4508 | * block allocation which had been deferred till now. | |
4509 | */ | |
7b415bf6 | 4510 | if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) { |
81fdbb4a | 4511 | unsigned int reserved_clusters; |
7b415bf6 | 4512 | /* |
81fdbb4a | 4513 | * Check how many clusters we had reserved this allocated range |
7b415bf6 AK |
4514 | */ |
4515 | reserved_clusters = get_reserved_cluster_alloc(inode, | |
4516 | map->m_lblk, allocated); | |
4517 | if (map->m_flags & EXT4_MAP_FROM_CLUSTER) { | |
4518 | if (reserved_clusters) { | |
4519 | /* | |
4520 | * We have clusters reserved for this range. | |
4521 | * But since we are not doing actual allocation | |
4522 | * and are simply using blocks from previously | |
4523 | * allocated cluster, we should release the | |
4524 | * reservation and not claim quota. | |
4525 | */ | |
4526 | ext4_da_update_reserve_space(inode, | |
4527 | reserved_clusters, 0); | |
4528 | } | |
4529 | } else { | |
4530 | BUG_ON(allocated_clusters < reserved_clusters); | |
7b415bf6 | 4531 | if (reserved_clusters < allocated_clusters) { |
5356f261 | 4532 | struct ext4_inode_info *ei = EXT4_I(inode); |
7b415bf6 AK |
4533 | int reservation = allocated_clusters - |
4534 | reserved_clusters; | |
4535 | /* | |
4536 | * It seems we claimed few clusters outside of | |
4537 | * the range of this allocation. We should give | |
4538 | * it back to the reservation pool. This can | |
4539 | * happen in the following case: | |
4540 | * | |
4541 | * * Suppose s_cluster_ratio is 4 (i.e., each | |
4542 | * cluster has 4 blocks. Thus, the clusters | |
4543 | * are [0-3],[4-7],[8-11]... | |
4544 | * * First comes delayed allocation write for | |
4545 | * logical blocks 10 & 11. Since there were no | |
4546 | * previous delayed allocated blocks in the | |
4547 | * range [8-11], we would reserve 1 cluster | |
4548 | * for this write. | |
4549 | * * Next comes write for logical blocks 3 to 8. | |
4550 | * In this case, we will reserve 2 clusters | |
4551 | * (for [0-3] and [4-7]; and not for [8-11] as | |
4552 | * that range has a delayed allocated blocks. | |
4553 | * Thus total reserved clusters now becomes 3. | |
4554 | * * Now, during the delayed allocation writeout | |
4555 | * time, we will first write blocks [3-8] and | |
4556 | * allocate 3 clusters for writing these | |
4557 | * blocks. Also, we would claim all these | |
4558 | * three clusters above. | |
4559 | * * Now when we come here to writeout the | |
4560 | * blocks [10-11], we would expect to claim | |
4561 | * the reservation of 1 cluster we had made | |
4562 | * (and we would claim it since there are no | |
4563 | * more delayed allocated blocks in the range | |
4564 | * [8-11]. But our reserved cluster count had | |
4565 | * already gone to 0. | |
4566 | * | |
4567 | * Thus, at the step 4 above when we determine | |
4568 | * that there are still some unwritten delayed | |
4569 | * allocated blocks outside of our current | |
4570 | * block range, we should increment the | |
4571 | * reserved clusters count so that when the | |
4572 | * remaining blocks finally gets written, we | |
4573 | * could claim them. | |
4574 | */ | |
5356f261 AK |
4575 | dquot_reserve_block(inode, |
4576 | EXT4_C2B(sbi, reservation)); | |
4577 | spin_lock(&ei->i_block_reservation_lock); | |
4578 | ei->i_reserved_data_blocks += reservation; | |
4579 | spin_unlock(&ei->i_block_reservation_lock); | |
7b415bf6 | 4580 | } |
232ec872 LC |
4581 | /* |
4582 | * We will claim quota for all newly allocated blocks. | |
4583 | * We're updating the reserved space *after* the | |
4584 | * correction above so we do not accidentally free | |
4585 | * all the metadata reservation because we might | |
4586 | * actually need it later on. | |
4587 | */ | |
4588 | ext4_da_update_reserve_space(inode, allocated_clusters, | |
4589 | 1); | |
7b415bf6 AK |
4590 | } |
4591 | } | |
5f634d06 | 4592 | |
b436b9be JK |
4593 | /* |
4594 | * Cache the extent and update transaction to commit on fdatasync only | |
556615dc | 4595 | * when it is _not_ an unwritten extent. |
b436b9be | 4596 | */ |
556615dc | 4597 | if ((flags & EXT4_GET_BLOCKS_UNWRIT_EXT) == 0) |
b436b9be | 4598 | ext4_update_inode_fsync_trans(handle, inode, 1); |
69eb33dc | 4599 | else |
b436b9be | 4600 | ext4_update_inode_fsync_trans(handle, inode, 0); |
a86c6181 | 4601 | out: |
e35fd660 TT |
4602 | if (allocated > map->m_len) |
4603 | allocated = map->m_len; | |
a86c6181 | 4604 | ext4_ext_show_leaf(inode, path); |
e35fd660 TT |
4605 | map->m_flags |= EXT4_MAP_MAPPED; |
4606 | map->m_pblk = newblock; | |
4607 | map->m_len = allocated; | |
a86c6181 | 4608 | out2: |
b7ea89ad TT |
4609 | ext4_ext_drop_refs(path); |
4610 | kfree(path); | |
e861304b | 4611 | |
63b99968 TT |
4612 | trace_ext4_ext_map_blocks_exit(inode, flags, map, |
4613 | err ? err : allocated); | |
4614 | ext4_es_lru_add(inode); | |
7877191c | 4615 | return err ? err : allocated; |
a86c6181 AT |
4616 | } |
4617 | ||
819c4920 | 4618 | void ext4_ext_truncate(handle_t *handle, struct inode *inode) |
a86c6181 | 4619 | { |
a86c6181 | 4620 | struct super_block *sb = inode->i_sb; |
725d26d3 | 4621 | ext4_lblk_t last_block; |
a86c6181 AT |
4622 | int err = 0; |
4623 | ||
a86c6181 | 4624 | /* |
d0d856e8 RD |
4625 | * TODO: optimization is possible here. |
4626 | * Probably we need not scan at all, | |
4627 | * because page truncation is enough. | |
a86c6181 | 4628 | */ |
a86c6181 AT |
4629 | |
4630 | /* we have to know where to truncate from in crash case */ | |
4631 | EXT4_I(inode)->i_disksize = inode->i_size; | |
4632 | ext4_mark_inode_dirty(handle, inode); | |
4633 | ||
4634 | last_block = (inode->i_size + sb->s_blocksize - 1) | |
4635 | >> EXT4_BLOCK_SIZE_BITS(sb); | |
8acd5e9b | 4636 | retry: |
51865fda ZL |
4637 | err = ext4_es_remove_extent(inode, last_block, |
4638 | EXT_MAX_BLOCKS - last_block); | |
94eec0fc | 4639 | if (err == -ENOMEM) { |
8acd5e9b TT |
4640 | cond_resched(); |
4641 | congestion_wait(BLK_RW_ASYNC, HZ/50); | |
4642 | goto retry; | |
4643 | } | |
4644 | if (err) { | |
4645 | ext4_std_error(inode->i_sb, err); | |
4646 | return; | |
4647 | } | |
5f95d21f | 4648 | err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1); |
8acd5e9b | 4649 | ext4_std_error(inode->i_sb, err); |
a86c6181 AT |
4650 | } |
4651 | ||
0e8b6879 | 4652 | static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, |
c174e6d6 DM |
4653 | ext4_lblk_t len, loff_t new_size, |
4654 | int flags, int mode) | |
0e8b6879 LC |
4655 | { |
4656 | struct inode *inode = file_inode(file); | |
4657 | handle_t *handle; | |
4658 | int ret = 0; | |
4659 | int ret2 = 0; | |
4660 | int retries = 0; | |
4661 | struct ext4_map_blocks map; | |
4662 | unsigned int credits; | |
c174e6d6 | 4663 | loff_t epos; |
0e8b6879 LC |
4664 | |
4665 | map.m_lblk = offset; | |
c174e6d6 | 4666 | map.m_len = len; |
0e8b6879 LC |
4667 | /* |
4668 | * Don't normalize the request if it can fit in one extent so | |
4669 | * that it doesn't get unnecessarily split into multiple | |
4670 | * extents. | |
4671 | */ | |
556615dc | 4672 | if (len <= EXT_UNWRITTEN_MAX_LEN) |
0e8b6879 LC |
4673 | flags |= EXT4_GET_BLOCKS_NO_NORMALIZE; |
4674 | ||
4675 | /* | |
4676 | * credits to insert 1 extent into extent tree | |
4677 | */ | |
4678 | credits = ext4_chunk_trans_blocks(inode, len); | |
4679 | ||
4680 | retry: | |
c174e6d6 | 4681 | while (ret >= 0 && len) { |
0e8b6879 LC |
4682 | handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, |
4683 | credits); | |
4684 | if (IS_ERR(handle)) { | |
4685 | ret = PTR_ERR(handle); | |
4686 | break; | |
4687 | } | |
4688 | ret = ext4_map_blocks(handle, inode, &map, flags); | |
4689 | if (ret <= 0) { | |
4690 | ext4_debug("inode #%lu: block %u: len %u: " | |
4691 | "ext4_ext_map_blocks returned %d", | |
4692 | inode->i_ino, map.m_lblk, | |
4693 | map.m_len, ret); | |
4694 | ext4_mark_inode_dirty(handle, inode); | |
4695 | ret2 = ext4_journal_stop(handle); | |
4696 | break; | |
4697 | } | |
c174e6d6 DM |
4698 | map.m_lblk += ret; |
4699 | map.m_len = len = len - ret; | |
4700 | epos = (loff_t)map.m_lblk << inode->i_blkbits; | |
4701 | inode->i_ctime = ext4_current_time(inode); | |
4702 | if (new_size) { | |
4703 | if (epos > new_size) | |
4704 | epos = new_size; | |
4705 | if (ext4_update_inode_size(inode, epos) & 0x1) | |
4706 | inode->i_mtime = inode->i_ctime; | |
4707 | } else { | |
4708 | if (epos > inode->i_size) | |
4709 | ext4_set_inode_flag(inode, | |
4710 | EXT4_INODE_EOFBLOCKS); | |
4711 | } | |
4712 | ext4_mark_inode_dirty(handle, inode); | |
0e8b6879 LC |
4713 | ret2 = ext4_journal_stop(handle); |
4714 | if (ret2) | |
4715 | break; | |
4716 | } | |
4717 | if (ret == -ENOSPC && | |
4718 | ext4_should_retry_alloc(inode->i_sb, &retries)) { | |
4719 | ret = 0; | |
4720 | goto retry; | |
4721 | } | |
4722 | ||
4723 | return ret > 0 ? ret2 : ret; | |
4724 | } | |
4725 | ||
b8a86845 LC |
4726 | static long ext4_zero_range(struct file *file, loff_t offset, |
4727 | loff_t len, int mode) | |
4728 | { | |
4729 | struct inode *inode = file_inode(file); | |
4730 | handle_t *handle = NULL; | |
4731 | unsigned int max_blocks; | |
4732 | loff_t new_size = 0; | |
4733 | int ret = 0; | |
4734 | int flags; | |
69dc9536 | 4735 | int credits; |
c174e6d6 | 4736 | int partial_begin, partial_end; |
b8a86845 LC |
4737 | loff_t start, end; |
4738 | ext4_lblk_t lblk; | |
4739 | struct address_space *mapping = inode->i_mapping; | |
4740 | unsigned int blkbits = inode->i_blkbits; | |
4741 | ||
4742 | trace_ext4_zero_range(inode, offset, len, mode); | |
4743 | ||
6c5e73d3 | 4744 | if (!S_ISREG(inode->i_mode)) |
4745 | return -EINVAL; | |
4746 | ||
e1ee60fd NJ |
4747 | /* Call ext4_force_commit to flush all data in case of data=journal. */ |
4748 | if (ext4_should_journal_data(inode)) { | |
4749 | ret = ext4_force_commit(inode->i_sb); | |
4750 | if (ret) | |
4751 | return ret; | |
4752 | } | |
4753 | ||
b8a86845 LC |
4754 | /* |
4755 | * Write out all dirty pages to avoid race conditions | |
4756 | * Then release them. | |
4757 | */ | |
4758 | if (mapping->nrpages && mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) { | |
4759 | ret = filemap_write_and_wait_range(mapping, offset, | |
4760 | offset + len - 1); | |
4761 | if (ret) | |
4762 | return ret; | |
4763 | } | |
4764 | ||
4765 | /* | |
4766 | * Round up offset. This is not fallocate, we neet to zero out | |
4767 | * blocks, so convert interior block aligned part of the range to | |
4768 | * unwritten and possibly manually zero out unaligned parts of the | |
4769 | * range. | |
4770 | */ | |
4771 | start = round_up(offset, 1 << blkbits); | |
4772 | end = round_down((offset + len), 1 << blkbits); | |
4773 | ||
4774 | if (start < offset || end > offset + len) | |
4775 | return -EINVAL; | |
c174e6d6 DM |
4776 | partial_begin = offset & ((1 << blkbits) - 1); |
4777 | partial_end = (offset + len) & ((1 << blkbits) - 1); | |
b8a86845 LC |
4778 | |
4779 | lblk = start >> blkbits; | |
4780 | max_blocks = (end >> blkbits); | |
4781 | if (max_blocks < lblk) | |
4782 | max_blocks = 0; | |
4783 | else | |
4784 | max_blocks -= lblk; | |
4785 | ||
556615dc | 4786 | flags = EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT | |
713e8dde TT |
4787 | EXT4_GET_BLOCKS_CONVERT_UNWRITTEN | |
4788 | EXT4_EX_NOCACHE; | |
b8a86845 LC |
4789 | if (mode & FALLOC_FL_KEEP_SIZE) |
4790 | flags |= EXT4_GET_BLOCKS_KEEP_SIZE; | |
4791 | ||
4792 | mutex_lock(&inode->i_mutex); | |
4793 | ||
4794 | /* | |
4795 | * Indirect files do not support unwritten extnets | |
4796 | */ | |
4797 | if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { | |
4798 | ret = -EOPNOTSUPP; | |
4799 | goto out_mutex; | |
4800 | } | |
4801 | ||
4802 | if (!(mode & FALLOC_FL_KEEP_SIZE) && | |
4803 | offset + len > i_size_read(inode)) { | |
4804 | new_size = offset + len; | |
4805 | ret = inode_newsize_ok(inode, new_size); | |
4806 | if (ret) | |
4807 | goto out_mutex; | |
4808 | /* | |
4809 | * If we have a partial block after EOF we have to allocate | |
4810 | * the entire block. | |
4811 | */ | |
c174e6d6 | 4812 | if (partial_end) |
b8a86845 LC |
4813 | max_blocks += 1; |
4814 | } | |
4815 | ||
4816 | if (max_blocks > 0) { | |
4817 | ||
4818 | /* Now release the pages and zero block aligned part of pages*/ | |
4819 | truncate_pagecache_range(inode, start, end - 1); | |
c174e6d6 | 4820 | inode->i_mtime = inode->i_ctime = ext4_current_time(inode); |
b8a86845 LC |
4821 | |
4822 | /* Wait all existing dio workers, newcomers will block on i_mutex */ | |
4823 | ext4_inode_block_unlocked_dio(inode); | |
4824 | inode_dio_wait(inode); | |
4825 | ||
713e8dde TT |
4826 | ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size, |
4827 | flags, mode); | |
4828 | if (ret) | |
4829 | goto out_dio; | |
b8a86845 LC |
4830 | /* |
4831 | * Remove entire range from the extent status tree. | |
713e8dde TT |
4832 | * |
4833 | * ext4_es_remove_extent(inode, lblk, max_blocks) is | |
4834 | * NOT sufficient. I'm not sure why this is the case, | |
4835 | * but let's be conservative and remove the extent | |
4836 | * status tree for the entire inode. There should be | |
4837 | * no outstanding delalloc extents thanks to the | |
4838 | * filemap_write_and_wait_range() call above. | |
b8a86845 | 4839 | */ |
713e8dde | 4840 | ret = ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS); |
b8a86845 LC |
4841 | if (ret) |
4842 | goto out_dio; | |
4843 | } | |
c174e6d6 DM |
4844 | if (!partial_begin && !partial_end) |
4845 | goto out_dio; | |
4846 | ||
69dc9536 DM |
4847 | /* |
4848 | * In worst case we have to writeout two nonadjacent unwritten | |
4849 | * blocks and update the inode | |
4850 | */ | |
4851 | credits = (2 * ext4_ext_index_trans_blocks(inode, 2)) + 1; | |
4852 | if (ext4_should_journal_data(inode)) | |
4853 | credits += 2; | |
4854 | handle = ext4_journal_start(inode, EXT4_HT_MISC, credits); | |
b8a86845 LC |
4855 | if (IS_ERR(handle)) { |
4856 | ret = PTR_ERR(handle); | |
4857 | ext4_std_error(inode->i_sb, ret); | |
4858 | goto out_dio; | |
4859 | } | |
4860 | ||
4861 | inode->i_mtime = inode->i_ctime = ext4_current_time(inode); | |
e5b30416 | 4862 | if (new_size) { |
4631dbf6 | 4863 | ext4_update_inode_size(inode, new_size); |
e5b30416 | 4864 | } else { |
b8a86845 LC |
4865 | /* |
4866 | * Mark that we allocate beyond EOF so the subsequent truncate | |
4867 | * can proceed even if the new size is the same as i_size. | |
4868 | */ | |
4869 | if ((offset + len) > i_size_read(inode)) | |
4870 | ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS); | |
4871 | } | |
b8a86845 LC |
4872 | ext4_mark_inode_dirty(handle, inode); |
4873 | ||
4874 | /* Zero out partial block at the edges of the range */ | |
4875 | ret = ext4_zero_partial_blocks(handle, inode, offset, len); | |
4876 | ||
4877 | if (file->f_flags & O_SYNC) | |
4878 | ext4_handle_sync(handle); | |
4879 | ||
4880 | ext4_journal_stop(handle); | |
4881 | out_dio: | |
4882 | ext4_inode_resume_unlocked_dio(inode); | |
4883 | out_mutex: | |
4884 | mutex_unlock(&inode->i_mutex); | |
4885 | return ret; | |
4886 | } | |
4887 | ||
a2df2a63 | 4888 | /* |
2fe17c10 | 4889 | * preallocate space for a file. This implements ext4's fallocate file |
a2df2a63 AA |
4890 | * operation, which gets called from sys_fallocate system call. |
4891 | * For block-mapped files, posix_fallocate should fall back to the method | |
4892 | * of writing zeroes to the required new blocks (the same behavior which is | |
4893 | * expected for file systems which do not support fallocate() system call). | |
4894 | */ | |
2fe17c10 | 4895 | long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) |
a2df2a63 | 4896 | { |
496ad9aa | 4897 | struct inode *inode = file_inode(file); |
f282ac19 | 4898 | loff_t new_size = 0; |
498e5f24 | 4899 | unsigned int max_blocks; |
a2df2a63 | 4900 | int ret = 0; |
a4e5d88b | 4901 | int flags; |
0e8b6879 | 4902 | ext4_lblk_t lblk; |
0e8b6879 | 4903 | unsigned int blkbits = inode->i_blkbits; |
a2df2a63 | 4904 | |
a4bb6b64 | 4905 | /* Return error if mode is not supported */ |
9eb79482 | 4906 | if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE | |
b8a86845 | 4907 | FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE)) |
a4bb6b64 AH |
4908 | return -EOPNOTSUPP; |
4909 | ||
4910 | if (mode & FALLOC_FL_PUNCH_HOLE) | |
aeb2817a | 4911 | return ext4_punch_hole(inode, offset, len); |
a4bb6b64 | 4912 | |
0c8d414f TM |
4913 | ret = ext4_convert_inline_data(inode); |
4914 | if (ret) | |
4915 | return ret; | |
4916 | ||
8bad6fc8 ZL |
4917 | /* |
4918 | * currently supporting (pre)allocate mode for extent-based | |
4919 | * files _only_ | |
4920 | */ | |
4921 | if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) | |
4922 | return -EOPNOTSUPP; | |
4923 | ||
40c406c7 TT |
4924 | if (mode & FALLOC_FL_COLLAPSE_RANGE) |
4925 | return ext4_collapse_range(inode, offset, len); | |
4926 | ||
b8a86845 LC |
4927 | if (mode & FALLOC_FL_ZERO_RANGE) |
4928 | return ext4_zero_range(file, offset, len, mode); | |
4929 | ||
0562e0ba | 4930 | trace_ext4_fallocate_enter(inode, offset, len, mode); |
0e8b6879 | 4931 | lblk = offset >> blkbits; |
fd28784a AK |
4932 | /* |
4933 | * We can't just convert len to max_blocks because | |
4934 | * If blocksize = 4096 offset = 3072 and len = 2048 | |
4935 | */ | |
a2df2a63 | 4936 | max_blocks = (EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits) |
0e8b6879 LC |
4937 | - lblk; |
4938 | ||
556615dc | 4939 | flags = EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT; |
0e8b6879 LC |
4940 | if (mode & FALLOC_FL_KEEP_SIZE) |
4941 | flags |= EXT4_GET_BLOCKS_KEEP_SIZE; | |
4942 | ||
55bd725a | 4943 | mutex_lock(&inode->i_mutex); |
f282ac19 LC |
4944 | |
4945 | if (!(mode & FALLOC_FL_KEEP_SIZE) && | |
4946 | offset + len > i_size_read(inode)) { | |
4947 | new_size = offset + len; | |
4948 | ret = inode_newsize_ok(inode, new_size); | |
4949 | if (ret) | |
4950 | goto out; | |
6d19c42b | 4951 | } |
f282ac19 | 4952 | |
c174e6d6 DM |
4953 | ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size, |
4954 | flags, mode); | |
0e8b6879 LC |
4955 | if (ret) |
4956 | goto out; | |
f282ac19 | 4957 | |
c174e6d6 DM |
4958 | if (file->f_flags & O_SYNC && EXT4_SB(inode->i_sb)->s_journal) { |
4959 | ret = jbd2_complete_transaction(EXT4_SB(inode->i_sb)->s_journal, | |
4960 | EXT4_I(inode)->i_sync_tid); | |
f282ac19 | 4961 | } |
f282ac19 | 4962 | out: |
55bd725a | 4963 | mutex_unlock(&inode->i_mutex); |
0e8b6879 LC |
4964 | trace_ext4_fallocate_exit(inode, offset, max_blocks, ret); |
4965 | return ret; | |
a2df2a63 | 4966 | } |
6873fa0d | 4967 | |
0031462b MC |
4968 | /* |
4969 | * This function convert a range of blocks to written extents | |
4970 | * The caller of this function will pass the start offset and the size. | |
4971 | * all unwritten extents within this range will be converted to | |
4972 | * written extents. | |
4973 | * | |
4974 | * This function is called from the direct IO end io call back | |
4975 | * function, to convert the fallocated extents after IO is completed. | |
109f5565 | 4976 | * Returns 0 on success. |
0031462b | 4977 | */ |
6b523df4 JK |
4978 | int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode, |
4979 | loff_t offset, ssize_t len) | |
0031462b | 4980 | { |
0031462b MC |
4981 | unsigned int max_blocks; |
4982 | int ret = 0; | |
4983 | int ret2 = 0; | |
2ed88685 | 4984 | struct ext4_map_blocks map; |
0031462b MC |
4985 | unsigned int credits, blkbits = inode->i_blkbits; |
4986 | ||
2ed88685 | 4987 | map.m_lblk = offset >> blkbits; |
0031462b MC |
4988 | /* |
4989 | * We can't just convert len to max_blocks because | |
4990 | * If blocksize = 4096 offset = 3072 and len = 2048 | |
4991 | */ | |
2ed88685 TT |
4992 | max_blocks = ((EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits) - |
4993 | map.m_lblk); | |
0031462b | 4994 | /* |
6b523df4 JK |
4995 | * This is somewhat ugly but the idea is clear: When transaction is |
4996 | * reserved, everything goes into it. Otherwise we rather start several | |
4997 | * smaller transactions for conversion of each extent separately. | |
0031462b | 4998 | */ |
6b523df4 JK |
4999 | if (handle) { |
5000 | handle = ext4_journal_start_reserved(handle, | |
5001 | EXT4_HT_EXT_CONVERT); | |
5002 | if (IS_ERR(handle)) | |
5003 | return PTR_ERR(handle); | |
5004 | credits = 0; | |
5005 | } else { | |
5006 | /* | |
5007 | * credits to insert 1 extent into extent tree | |
5008 | */ | |
5009 | credits = ext4_chunk_trans_blocks(inode, max_blocks); | |
5010 | } | |
0031462b | 5011 | while (ret >= 0 && ret < max_blocks) { |
2ed88685 TT |
5012 | map.m_lblk += ret; |
5013 | map.m_len = (max_blocks -= ret); | |
6b523df4 JK |
5014 | if (credits) { |
5015 | handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, | |
5016 | credits); | |
5017 | if (IS_ERR(handle)) { | |
5018 | ret = PTR_ERR(handle); | |
5019 | break; | |
5020 | } | |
0031462b | 5021 | } |
2ed88685 | 5022 | ret = ext4_map_blocks(handle, inode, &map, |
c7064ef1 | 5023 | EXT4_GET_BLOCKS_IO_CONVERT_EXT); |
b06acd38 LC |
5024 | if (ret <= 0) |
5025 | ext4_warning(inode->i_sb, | |
5026 | "inode #%lu: block %u: len %u: " | |
5027 | "ext4_ext_map_blocks returned %d", | |
5028 | inode->i_ino, map.m_lblk, | |
5029 | map.m_len, ret); | |
0031462b | 5030 | ext4_mark_inode_dirty(handle, inode); |
6b523df4 JK |
5031 | if (credits) |
5032 | ret2 = ext4_journal_stop(handle); | |
5033 | if (ret <= 0 || ret2) | |
0031462b MC |
5034 | break; |
5035 | } | |
6b523df4 JK |
5036 | if (!credits) |
5037 | ret2 = ext4_journal_stop(handle); | |
0031462b MC |
5038 | return ret > 0 ? ret2 : ret; |
5039 | } | |
6d9c85eb | 5040 | |
6873fa0d | 5041 | /* |
69eb33dc ZL |
5042 | * If newes is not existing extent (newes->ec_pblk equals zero) find |
5043 | * delayed extent at start of newes and update newes accordingly and | |
91dd8c11 LC |
5044 | * return start of the next delayed extent. |
5045 | * | |
69eb33dc | 5046 | * If newes is existing extent (newes->ec_pblk is not equal zero) |
91dd8c11 | 5047 | * return start of next delayed extent or EXT_MAX_BLOCKS if no delayed |
69eb33dc | 5048 | * extent found. Leave newes unmodified. |
6873fa0d | 5049 | */ |
91dd8c11 | 5050 | static int ext4_find_delayed_extent(struct inode *inode, |
69eb33dc | 5051 | struct extent_status *newes) |
6873fa0d | 5052 | { |
b3aff3e3 | 5053 | struct extent_status es; |
be401363 | 5054 | ext4_lblk_t block, next_del; |
6873fa0d | 5055 | |
69eb33dc | 5056 | if (newes->es_pblk == 0) { |
e30b5dca YZ |
5057 | ext4_es_find_delayed_extent_range(inode, newes->es_lblk, |
5058 | newes->es_lblk + newes->es_len - 1, &es); | |
5059 | ||
6d9c85eb | 5060 | /* |
69eb33dc | 5061 | * No extent in extent-tree contains block @newes->es_pblk, |
6d9c85eb | 5062 | * then the block may stay in 1)a hole or 2)delayed-extent. |
6d9c85eb | 5063 | */ |
06b0c886 | 5064 | if (es.es_len == 0) |
b3aff3e3 | 5065 | /* A hole found. */ |
91dd8c11 | 5066 | return 0; |
b3aff3e3 | 5067 | |
69eb33dc | 5068 | if (es.es_lblk > newes->es_lblk) { |
b3aff3e3 | 5069 | /* A hole found. */ |
69eb33dc ZL |
5070 | newes->es_len = min(es.es_lblk - newes->es_lblk, |
5071 | newes->es_len); | |
91dd8c11 | 5072 | return 0; |
6873fa0d | 5073 | } |
6d9c85eb | 5074 | |
69eb33dc | 5075 | newes->es_len = es.es_lblk + es.es_len - newes->es_lblk; |
6873fa0d ES |
5076 | } |
5077 | ||
69eb33dc | 5078 | block = newes->es_lblk + newes->es_len; |
e30b5dca | 5079 | ext4_es_find_delayed_extent_range(inode, block, EXT_MAX_BLOCKS, &es); |
be401363 ZL |
5080 | if (es.es_len == 0) |
5081 | next_del = EXT_MAX_BLOCKS; | |
5082 | else | |
5083 | next_del = es.es_lblk; | |
5084 | ||
91dd8c11 | 5085 | return next_del; |
6873fa0d | 5086 | } |
6873fa0d ES |
5087 | /* fiemap flags we can handle specified here */ |
5088 | #define EXT4_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR) | |
5089 | ||
3a06d778 AK |
5090 | static int ext4_xattr_fiemap(struct inode *inode, |
5091 | struct fiemap_extent_info *fieinfo) | |
6873fa0d ES |
5092 | { |
5093 | __u64 physical = 0; | |
5094 | __u64 length; | |
5095 | __u32 flags = FIEMAP_EXTENT_LAST; | |
5096 | int blockbits = inode->i_sb->s_blocksize_bits; | |
5097 | int error = 0; | |
5098 | ||
5099 | /* in-inode? */ | |
19f5fb7a | 5100 | if (ext4_test_inode_state(inode, EXT4_STATE_XATTR)) { |
6873fa0d ES |
5101 | struct ext4_iloc iloc; |
5102 | int offset; /* offset of xattr in inode */ | |
5103 | ||
5104 | error = ext4_get_inode_loc(inode, &iloc); | |
5105 | if (error) | |
5106 | return error; | |
a60697f4 | 5107 | physical = (__u64)iloc.bh->b_blocknr << blockbits; |
6873fa0d ES |
5108 | offset = EXT4_GOOD_OLD_INODE_SIZE + |
5109 | EXT4_I(inode)->i_extra_isize; | |
5110 | physical += offset; | |
5111 | length = EXT4_SB(inode->i_sb)->s_inode_size - offset; | |
5112 | flags |= FIEMAP_EXTENT_DATA_INLINE; | |
fd2dd9fb | 5113 | brelse(iloc.bh); |
6873fa0d | 5114 | } else { /* external block */ |
a60697f4 | 5115 | physical = (__u64)EXT4_I(inode)->i_file_acl << blockbits; |
6873fa0d ES |
5116 | length = inode->i_sb->s_blocksize; |
5117 | } | |
5118 | ||
5119 | if (physical) | |
5120 | error = fiemap_fill_next_extent(fieinfo, 0, physical, | |
5121 | length, flags); | |
5122 | return (error < 0 ? error : 0); | |
5123 | } | |
5124 | ||
5125 | int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, | |
5126 | __u64 start, __u64 len) | |
5127 | { | |
5128 | ext4_lblk_t start_blk; | |
6873fa0d ES |
5129 | int error = 0; |
5130 | ||
94191985 TM |
5131 | if (ext4_has_inline_data(inode)) { |
5132 | int has_inline = 1; | |
5133 | ||
5134 | error = ext4_inline_data_fiemap(inode, fieinfo, &has_inline); | |
5135 | ||
5136 | if (has_inline) | |
5137 | return error; | |
5138 | } | |
5139 | ||
7869a4a6 TT |
5140 | if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) { |
5141 | error = ext4_ext_precache(inode); | |
5142 | if (error) | |
5143 | return error; | |
5144 | } | |
5145 | ||
6873fa0d | 5146 | /* fallback to generic here if not in extents fmt */ |
12e9b892 | 5147 | if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) |
6873fa0d ES |
5148 | return generic_block_fiemap(inode, fieinfo, start, len, |
5149 | ext4_get_block); | |
5150 | ||
5151 | if (fiemap_check_flags(fieinfo, EXT4_FIEMAP_FLAGS)) | |
5152 | return -EBADR; | |
5153 | ||
5154 | if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) { | |
5155 | error = ext4_xattr_fiemap(inode, fieinfo); | |
5156 | } else { | |
aca92ff6 LM |
5157 | ext4_lblk_t len_blks; |
5158 | __u64 last_blk; | |
5159 | ||
6873fa0d | 5160 | start_blk = start >> inode->i_sb->s_blocksize_bits; |
aca92ff6 | 5161 | last_blk = (start + len - 1) >> inode->i_sb->s_blocksize_bits; |
f17722f9 LC |
5162 | if (last_blk >= EXT_MAX_BLOCKS) |
5163 | last_blk = EXT_MAX_BLOCKS-1; | |
aca92ff6 | 5164 | len_blks = ((ext4_lblk_t) last_blk) - start_blk + 1; |
6873fa0d ES |
5165 | |
5166 | /* | |
91dd8c11 LC |
5167 | * Walk the extent tree gathering extent information |
5168 | * and pushing extents back to the user. | |
6873fa0d | 5169 | */ |
91dd8c11 LC |
5170 | error = ext4_fill_fiemap_extents(inode, start_blk, |
5171 | len_blks, fieinfo); | |
6873fa0d | 5172 | } |
107a7bd3 | 5173 | ext4_es_lru_add(inode); |
6873fa0d ES |
5174 | return error; |
5175 | } | |
9eb79482 NJ |
5176 | |
5177 | /* | |
5178 | * ext4_access_path: | |
5179 | * Function to access the path buffer for marking it dirty. | |
5180 | * It also checks if there are sufficient credits left in the journal handle | |
5181 | * to update path. | |
5182 | */ | |
5183 | static int | |
5184 | ext4_access_path(handle_t *handle, struct inode *inode, | |
5185 | struct ext4_ext_path *path) | |
5186 | { | |
5187 | int credits, err; | |
5188 | ||
5189 | if (!ext4_handle_valid(handle)) | |
5190 | return 0; | |
5191 | ||
5192 | /* | |
5193 | * Check if need to extend journal credits | |
5194 | * 3 for leaf, sb, and inode plus 2 (bmap and group | |
5195 | * descriptor) for each block group; assume two block | |
5196 | * groups | |
5197 | */ | |
5198 | if (handle->h_buffer_credits < 7) { | |
5199 | credits = ext4_writepage_trans_blocks(inode); | |
5200 | err = ext4_ext_truncate_extend_restart(handle, inode, credits); | |
5201 | /* EAGAIN is success */ | |
5202 | if (err && err != -EAGAIN) | |
5203 | return err; | |
5204 | } | |
5205 | ||
5206 | err = ext4_ext_get_access(handle, inode, path); | |
5207 | return err; | |
5208 | } | |
5209 | ||
5210 | /* | |
5211 | * ext4_ext_shift_path_extents: | |
5212 | * Shift the extents of a path structure lying between path[depth].p_ext | |
5213 | * and EXT_LAST_EXTENT(path[depth].p_hdr) downwards, by subtracting shift | |
5214 | * from starting block for each extent. | |
5215 | */ | |
5216 | static int | |
5217 | ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift, | |
5218 | struct inode *inode, handle_t *handle, | |
5219 | ext4_lblk_t *start) | |
5220 | { | |
5221 | int depth, err = 0; | |
5222 | struct ext4_extent *ex_start, *ex_last; | |
5223 | bool update = 0; | |
5224 | depth = path->p_depth; | |
5225 | ||
5226 | while (depth >= 0) { | |
5227 | if (depth == path->p_depth) { | |
5228 | ex_start = path[depth].p_ext; | |
5229 | if (!ex_start) | |
5230 | return -EIO; | |
5231 | ||
5232 | ex_last = EXT_LAST_EXTENT(path[depth].p_hdr); | |
5233 | if (!ex_last) | |
5234 | return -EIO; | |
5235 | ||
5236 | err = ext4_access_path(handle, inode, path + depth); | |
5237 | if (err) | |
5238 | goto out; | |
5239 | ||
5240 | if (ex_start == EXT_FIRST_EXTENT(path[depth].p_hdr)) | |
5241 | update = 1; | |
5242 | ||
847c6c42 | 5243 | *start = le32_to_cpu(ex_last->ee_block) + |
9eb79482 NJ |
5244 | ext4_ext_get_actual_len(ex_last); |
5245 | ||
5246 | while (ex_start <= ex_last) { | |
847c6c42 | 5247 | le32_add_cpu(&ex_start->ee_block, -shift); |
6dd834ef LC |
5248 | /* Try to merge to the left. */ |
5249 | if ((ex_start > | |
5250 | EXT_FIRST_EXTENT(path[depth].p_hdr)) && | |
5251 | ext4_ext_try_to_merge_right(inode, | |
5252 | path, ex_start - 1)) | |
5253 | ex_last--; | |
5254 | else | |
5255 | ex_start++; | |
9eb79482 NJ |
5256 | } |
5257 | err = ext4_ext_dirty(handle, inode, path + depth); | |
5258 | if (err) | |
5259 | goto out; | |
5260 | ||
5261 | if (--depth < 0 || !update) | |
5262 | break; | |
5263 | } | |
5264 | ||
5265 | /* Update index too */ | |
5266 | err = ext4_access_path(handle, inode, path + depth); | |
5267 | if (err) | |
5268 | goto out; | |
5269 | ||
847c6c42 | 5270 | le32_add_cpu(&path[depth].p_idx->ei_block, -shift); |
9eb79482 NJ |
5271 | err = ext4_ext_dirty(handle, inode, path + depth); |
5272 | if (err) | |
5273 | goto out; | |
5274 | ||
5275 | /* we are done if current index is not a starting index */ | |
5276 | if (path[depth].p_idx != EXT_FIRST_INDEX(path[depth].p_hdr)) | |
5277 | break; | |
5278 | ||
5279 | depth--; | |
5280 | } | |
5281 | ||
5282 | out: | |
5283 | return err; | |
5284 | } | |
5285 | ||
5286 | /* | |
5287 | * ext4_ext_shift_extents: | |
5288 | * All the extents which lies in the range from start to the last allocated | |
5289 | * block for the file are shifted downwards by shift blocks. | |
5290 | * On success, 0 is returned, error otherwise. | |
5291 | */ | |
5292 | static int | |
5293 | ext4_ext_shift_extents(struct inode *inode, handle_t *handle, | |
5294 | ext4_lblk_t start, ext4_lblk_t shift) | |
5295 | { | |
5296 | struct ext4_ext_path *path; | |
5297 | int ret = 0, depth; | |
5298 | struct ext4_extent *extent; | |
f8fb4f41 | 5299 | ext4_lblk_t stop_block; |
9eb79482 NJ |
5300 | ext4_lblk_t ex_start, ex_end; |
5301 | ||
5302 | /* Let path point to the last extent */ | |
5303 | path = ext4_ext_find_extent(inode, EXT_MAX_BLOCKS - 1, NULL, 0); | |
5304 | if (IS_ERR(path)) | |
5305 | return PTR_ERR(path); | |
5306 | ||
5307 | depth = path->p_depth; | |
5308 | extent = path[depth].p_ext; | |
ee4bd0d9 TT |
5309 | if (!extent) |
5310 | goto out; | |
9eb79482 | 5311 | |
847c6c42 ZL |
5312 | stop_block = le32_to_cpu(extent->ee_block) + |
5313 | ext4_ext_get_actual_len(extent); | |
9eb79482 NJ |
5314 | |
5315 | /* Nothing to shift, if hole is at the end of file */ | |
5316 | if (start >= stop_block) | |
ee4bd0d9 | 5317 | goto out; |
9eb79482 NJ |
5318 | |
5319 | /* | |
5320 | * Don't start shifting extents until we make sure the hole is big | |
5321 | * enough to accomodate the shift. | |
5322 | */ | |
ee4bd0d9 | 5323 | path = ext4_ext_find_extent(inode, start - 1, &path, 0); |
8dc79ec4 DM |
5324 | if (IS_ERR(path)) |
5325 | return PTR_ERR(path); | |
9eb79482 NJ |
5326 | depth = path->p_depth; |
5327 | extent = path[depth].p_ext; | |
8dc79ec4 DM |
5328 | if (extent) { |
5329 | ex_start = le32_to_cpu(extent->ee_block); | |
5330 | ex_end = le32_to_cpu(extent->ee_block) + | |
847c6c42 | 5331 | ext4_ext_get_actual_len(extent); |
8dc79ec4 DM |
5332 | } else { |
5333 | ex_start = 0; | |
5334 | ex_end = 0; | |
5335 | } | |
9eb79482 NJ |
5336 | |
5337 | if ((start == ex_start && shift > ex_start) || | |
5338 | (shift > start - ex_end)) | |
5339 | return -EINVAL; | |
5340 | ||
5341 | /* Its safe to start updating extents */ | |
5342 | while (start < stop_block) { | |
ee4bd0d9 | 5343 | path = ext4_ext_find_extent(inode, start, &path, 0); |
9eb79482 NJ |
5344 | if (IS_ERR(path)) |
5345 | return PTR_ERR(path); | |
5346 | depth = path->p_depth; | |
5347 | extent = path[depth].p_ext; | |
a18ed359 DM |
5348 | if (!extent) { |
5349 | EXT4_ERROR_INODE(inode, "unexpected hole at %lu", | |
5350 | (unsigned long) start); | |
5351 | return -EIO; | |
5352 | } | |
f8fb4f41 | 5353 | if (start > le32_to_cpu(extent->ee_block)) { |
9eb79482 | 5354 | /* Hole, move to the next extent */ |
f8fb4f41 DM |
5355 | if (extent < EXT_LAST_EXTENT(path[depth].p_hdr)) { |
5356 | path[depth].p_ext++; | |
5357 | } else { | |
5358 | start = ext4_ext_next_allocated_block(path); | |
f8fb4f41 | 5359 | continue; |
9eb79482 NJ |
5360 | } |
5361 | } | |
5362 | ret = ext4_ext_shift_path_extents(path, shift, inode, | |
5363 | handle, &start); | |
9eb79482 NJ |
5364 | if (ret) |
5365 | break; | |
5366 | } | |
ee4bd0d9 TT |
5367 | out: |
5368 | ext4_ext_drop_refs(path); | |
5369 | kfree(path); | |
9eb79482 NJ |
5370 | return ret; |
5371 | } | |
5372 | ||
5373 | /* | |
5374 | * ext4_collapse_range: | |
5375 | * This implements the fallocate's collapse range functionality for ext4 | |
5376 | * Returns: 0 and non-zero on error. | |
5377 | */ | |
5378 | int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) | |
5379 | { | |
5380 | struct super_block *sb = inode->i_sb; | |
5381 | ext4_lblk_t punch_start, punch_stop; | |
5382 | handle_t *handle; | |
5383 | unsigned int credits; | |
a8680e0d | 5384 | loff_t new_size, ioffset; |
9eb79482 NJ |
5385 | int ret; |
5386 | ||
9eb79482 | 5387 | /* Collapse range works only on fs block size aligned offsets. */ |
ee98fa3a NJ |
5388 | if (offset & (EXT4_CLUSTER_SIZE(sb) - 1) || |
5389 | len & (EXT4_CLUSTER_SIZE(sb) - 1)) | |
9eb79482 NJ |
5390 | return -EINVAL; |
5391 | ||
5392 | if (!S_ISREG(inode->i_mode)) | |
86f1ca38 | 5393 | return -EINVAL; |
9eb79482 NJ |
5394 | |
5395 | trace_ext4_collapse_range(inode, offset, len); | |
5396 | ||
5397 | punch_start = offset >> EXT4_BLOCK_SIZE_BITS(sb); | |
5398 | punch_stop = (offset + len) >> EXT4_BLOCK_SIZE_BITS(sb); | |
5399 | ||
1ce01c4a NJ |
5400 | /* Call ext4_force_commit to flush all data in case of data=journal. */ |
5401 | if (ext4_should_journal_data(inode)) { | |
5402 | ret = ext4_force_commit(inode->i_sb); | |
5403 | if (ret) | |
5404 | return ret; | |
5405 | } | |
5406 | ||
a8680e0d NJ |
5407 | /* |
5408 | * Need to round down offset to be aligned with page size boundary | |
5409 | * for page size > block size. | |
5410 | */ | |
5411 | ioffset = round_down(offset, PAGE_SIZE); | |
5412 | ||
9eb79482 | 5413 | /* Write out all dirty pages */ |
a8680e0d NJ |
5414 | ret = filemap_write_and_wait_range(inode->i_mapping, ioffset, |
5415 | LLONG_MAX); | |
9eb79482 NJ |
5416 | if (ret) |
5417 | return ret; | |
5418 | ||
5419 | /* Take mutex lock */ | |
5420 | mutex_lock(&inode->i_mutex); | |
5421 | ||
23fffa92 LC |
5422 | /* |
5423 | * There is no need to overlap collapse range with EOF, in which case | |
5424 | * it is effectively a truncate operation | |
5425 | */ | |
5426 | if (offset + len >= i_size_read(inode)) { | |
5427 | ret = -EINVAL; | |
5428 | goto out_mutex; | |
5429 | } | |
5430 | ||
9eb79482 NJ |
5431 | /* Currently just for extent based files */ |
5432 | if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { | |
5433 | ret = -EOPNOTSUPP; | |
5434 | goto out_mutex; | |
5435 | } | |
5436 | ||
a8680e0d | 5437 | truncate_pagecache(inode, ioffset); |
9eb79482 NJ |
5438 | |
5439 | /* Wait for existing dio to complete */ | |
5440 | ext4_inode_block_unlocked_dio(inode); | |
5441 | inode_dio_wait(inode); | |
5442 | ||
5443 | credits = ext4_writepage_trans_blocks(inode); | |
5444 | handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE, credits); | |
5445 | if (IS_ERR(handle)) { | |
5446 | ret = PTR_ERR(handle); | |
5447 | goto out_dio; | |
5448 | } | |
5449 | ||
5450 | down_write(&EXT4_I(inode)->i_data_sem); | |
5451 | ext4_discard_preallocations(inode); | |
5452 | ||
5453 | ret = ext4_es_remove_extent(inode, punch_start, | |
2c1d2328 | 5454 | EXT_MAX_BLOCKS - punch_start); |
9eb79482 NJ |
5455 | if (ret) { |
5456 | up_write(&EXT4_I(inode)->i_data_sem); | |
5457 | goto out_stop; | |
5458 | } | |
5459 | ||
5460 | ret = ext4_ext_remove_space(inode, punch_start, punch_stop - 1); | |
5461 | if (ret) { | |
5462 | up_write(&EXT4_I(inode)->i_data_sem); | |
5463 | goto out_stop; | |
5464 | } | |
ef24f6c2 | 5465 | ext4_discard_preallocations(inode); |
9eb79482 NJ |
5466 | |
5467 | ret = ext4_ext_shift_extents(inode, handle, punch_stop, | |
5468 | punch_stop - punch_start); | |
5469 | if (ret) { | |
5470 | up_write(&EXT4_I(inode)->i_data_sem); | |
5471 | goto out_stop; | |
5472 | } | |
5473 | ||
5474 | new_size = i_size_read(inode) - len; | |
9337d5d3 | 5475 | i_size_write(inode, new_size); |
9eb79482 NJ |
5476 | EXT4_I(inode)->i_disksize = new_size; |
5477 | ||
9eb79482 NJ |
5478 | up_write(&EXT4_I(inode)->i_data_sem); |
5479 | if (IS_SYNC(inode)) | |
5480 | ext4_handle_sync(handle); | |
5481 | inode->i_mtime = inode->i_ctime = ext4_current_time(inode); | |
5482 | ext4_mark_inode_dirty(handle, inode); | |
5483 | ||
5484 | out_stop: | |
5485 | ext4_journal_stop(handle); | |
5486 | out_dio: | |
5487 | ext4_inode_resume_unlocked_dio(inode); | |
5488 | out_mutex: | |
5489 | mutex_unlock(&inode->i_mutex); | |
5490 | return ret; | |
5491 | } | |
fcf6b1b7 DM |
5492 | |
5493 | /** | |
5494 | * ext4_swap_extents - Swap extents between two inodes | |
5495 | * | |
5496 | * @inode1: First inode | |
5497 | * @inode2: Second inode | |
5498 | * @lblk1: Start block for first inode | |
5499 | * @lblk2: Start block for second inode | |
5500 | * @count: Number of blocks to swap | |
5501 | * @mark_unwritten: Mark second inode's extents as unwritten after swap | |
5502 | * @erp: Pointer to save error value | |
5503 | * | |
5504 | * This helper routine does exactly what is promise "swap extents". All other | |
5505 | * stuff such as page-cache locking consistency, bh mapping consistency or | |
5506 | * extent's data copying must be performed by caller. | |
5507 | * Locking: | |
5508 | * i_mutex is held for both inodes | |
5509 | * i_data_sem is locked for write for both inodes | |
5510 | * Assumptions: | |
5511 | * All pages from requested range are locked for both inodes | |
5512 | */ | |
5513 | int | |
5514 | ext4_swap_extents(handle_t *handle, struct inode *inode1, | |
5515 | struct inode *inode2, ext4_lblk_t lblk1, ext4_lblk_t lblk2, | |
5516 | ext4_lblk_t count, int unwritten, int *erp) | |
5517 | { | |
5518 | struct ext4_ext_path *path1 = NULL; | |
5519 | struct ext4_ext_path *path2 = NULL; | |
5520 | int replaced_count = 0; | |
5521 | ||
5522 | BUG_ON(!rwsem_is_locked(&EXT4_I(inode1)->i_data_sem)); | |
5523 | BUG_ON(!rwsem_is_locked(&EXT4_I(inode2)->i_data_sem)); | |
5524 | BUG_ON(!mutex_is_locked(&inode1->i_mutex)); | |
5525 | BUG_ON(!mutex_is_locked(&inode1->i_mutex)); | |
5526 | ||
5527 | *erp = ext4_es_remove_extent(inode1, lblk1, count); | |
19008f6d | 5528 | if (unlikely(*erp)) |
fcf6b1b7 DM |
5529 | return 0; |
5530 | *erp = ext4_es_remove_extent(inode2, lblk2, count); | |
19008f6d | 5531 | if (unlikely(*erp)) |
fcf6b1b7 DM |
5532 | return 0; |
5533 | ||
5534 | while (count) { | |
5535 | struct ext4_extent *ex1, *ex2, tmp_ex; | |
5536 | ext4_lblk_t e1_blk, e2_blk; | |
5537 | int e1_len, e2_len, len; | |
5538 | int split = 0; | |
5539 | ||
5540 | path1 = ext4_ext_find_extent(inode1, lblk1, NULL, EXT4_EX_NOCACHE); | |
19008f6d | 5541 | if (unlikely(IS_ERR(path1))) { |
fcf6b1b7 | 5542 | *erp = PTR_ERR(path1); |
19008f6d TT |
5543 | path1 = NULL; |
5544 | finish: | |
5545 | count = 0; | |
5546 | goto repeat; | |
fcf6b1b7 DM |
5547 | } |
5548 | path2 = ext4_ext_find_extent(inode2, lblk2, NULL, EXT4_EX_NOCACHE); | |
19008f6d | 5549 | if (unlikely(IS_ERR(path2))) { |
fcf6b1b7 | 5550 | *erp = PTR_ERR(path2); |
19008f6d TT |
5551 | path2 = NULL; |
5552 | goto finish; | |
fcf6b1b7 DM |
5553 | } |
5554 | ex1 = path1[path1->p_depth].p_ext; | |
5555 | ex2 = path2[path2->p_depth].p_ext; | |
5556 | /* Do we have somthing to swap ? */ | |
5557 | if (unlikely(!ex2 || !ex1)) | |
19008f6d | 5558 | goto finish; |
fcf6b1b7 DM |
5559 | |
5560 | e1_blk = le32_to_cpu(ex1->ee_block); | |
5561 | e2_blk = le32_to_cpu(ex2->ee_block); | |
5562 | e1_len = ext4_ext_get_actual_len(ex1); | |
5563 | e2_len = ext4_ext_get_actual_len(ex2); | |
5564 | ||
5565 | /* Hole handling */ | |
5566 | if (!in_range(lblk1, e1_blk, e1_len) || | |
5567 | !in_range(lblk2, e2_blk, e2_len)) { | |
5568 | ext4_lblk_t next1, next2; | |
5569 | ||
5570 | /* if hole after extent, then go to next extent */ | |
5571 | next1 = ext4_ext_next_allocated_block(path1); | |
5572 | next2 = ext4_ext_next_allocated_block(path2); | |
5573 | /* If hole before extent, then shift to that extent */ | |
5574 | if (e1_blk > lblk1) | |
5575 | next1 = e1_blk; | |
5576 | if (e2_blk > lblk2) | |
5577 | next2 = e1_blk; | |
5578 | /* Do we have something to swap */ | |
5579 | if (next1 == EXT_MAX_BLOCKS || next2 == EXT_MAX_BLOCKS) | |
19008f6d | 5580 | goto finish; |
fcf6b1b7 DM |
5581 | /* Move to the rightest boundary */ |
5582 | len = next1 - lblk1; | |
5583 | if (len < next2 - lblk2) | |
5584 | len = next2 - lblk2; | |
5585 | if (len > count) | |
5586 | len = count; | |
5587 | lblk1 += len; | |
5588 | lblk2 += len; | |
5589 | count -= len; | |
5590 | goto repeat; | |
5591 | } | |
5592 | ||
5593 | /* Prepare left boundary */ | |
5594 | if (e1_blk < lblk1) { | |
5595 | split = 1; | |
5596 | *erp = ext4_force_split_extent_at(handle, inode1, | |
dfe50809 | 5597 | &path1, lblk1, 0); |
19008f6d TT |
5598 | if (unlikely(*erp)) |
5599 | goto finish; | |
fcf6b1b7 DM |
5600 | } |
5601 | if (e2_blk < lblk2) { | |
5602 | split = 1; | |
5603 | *erp = ext4_force_split_extent_at(handle, inode2, | |
dfe50809 | 5604 | &path2, lblk2, 0); |
19008f6d TT |
5605 | if (unlikely(*erp)) |
5606 | goto finish; | |
fcf6b1b7 | 5607 | } |
dfe50809 | 5608 | /* ext4_split_extent_at() may result in leaf extent split, |
fcf6b1b7 DM |
5609 | * path must to be revalidated. */ |
5610 | if (split) | |
5611 | goto repeat; | |
5612 | ||
5613 | /* Prepare right boundary */ | |
5614 | len = count; | |
5615 | if (len > e1_blk + e1_len - lblk1) | |
5616 | len = e1_blk + e1_len - lblk1; | |
5617 | if (len > e2_blk + e2_len - lblk2) | |
5618 | len = e2_blk + e2_len - lblk2; | |
5619 | ||
5620 | if (len != e1_len) { | |
5621 | split = 1; | |
5622 | *erp = ext4_force_split_extent_at(handle, inode1, | |
dfe50809 | 5623 | &path1, lblk1 + len, 0); |
19008f6d TT |
5624 | if (unlikely(*erp)) |
5625 | goto finish; | |
fcf6b1b7 DM |
5626 | } |
5627 | if (len != e2_len) { | |
5628 | split = 1; | |
5629 | *erp = ext4_force_split_extent_at(handle, inode2, | |
dfe50809 | 5630 | &path2, lblk2 + len, 0); |
fcf6b1b7 | 5631 | if (*erp) |
19008f6d | 5632 | goto finish; |
fcf6b1b7 | 5633 | } |
dfe50809 | 5634 | /* ext4_split_extent_at() may result in leaf extent split, |
fcf6b1b7 DM |
5635 | * path must to be revalidated. */ |
5636 | if (split) | |
5637 | goto repeat; | |
5638 | ||
5639 | BUG_ON(e2_len != e1_len); | |
5640 | *erp = ext4_ext_get_access(handle, inode1, path1 + path1->p_depth); | |
19008f6d TT |
5641 | if (unlikely(*erp)) |
5642 | goto finish; | |
fcf6b1b7 | 5643 | *erp = ext4_ext_get_access(handle, inode2, path2 + path2->p_depth); |
19008f6d TT |
5644 | if (unlikely(*erp)) |
5645 | goto finish; | |
fcf6b1b7 DM |
5646 | |
5647 | /* Both extents are fully inside boundaries. Swap it now */ | |
5648 | tmp_ex = *ex1; | |
5649 | ext4_ext_store_pblock(ex1, ext4_ext_pblock(ex2)); | |
5650 | ext4_ext_store_pblock(ex2, ext4_ext_pblock(&tmp_ex)); | |
5651 | ex1->ee_len = cpu_to_le16(e2_len); | |
5652 | ex2->ee_len = cpu_to_le16(e1_len); | |
5653 | if (unwritten) | |
5654 | ext4_ext_mark_unwritten(ex2); | |
5655 | if (ext4_ext_is_unwritten(&tmp_ex)) | |
5656 | ext4_ext_mark_unwritten(ex1); | |
5657 | ||
5658 | ext4_ext_try_to_merge(handle, inode2, path2, ex2); | |
5659 | ext4_ext_try_to_merge(handle, inode1, path1, ex1); | |
5660 | *erp = ext4_ext_dirty(handle, inode2, path2 + | |
5661 | path2->p_depth); | |
19008f6d TT |
5662 | if (unlikely(*erp)) |
5663 | goto finish; | |
fcf6b1b7 DM |
5664 | *erp = ext4_ext_dirty(handle, inode1, path1 + |
5665 | path1->p_depth); | |
5666 | /* | |
5667 | * Looks scarry ah..? second inode already points to new blocks, | |
5668 | * and it was successfully dirtied. But luckily error may happen | |
5669 | * only due to journal error, so full transaction will be | |
5670 | * aborted anyway. | |
5671 | */ | |
19008f6d TT |
5672 | if (unlikely(*erp)) |
5673 | goto finish; | |
fcf6b1b7 DM |
5674 | lblk1 += len; |
5675 | lblk2 += len; | |
5676 | replaced_count += len; | |
5677 | count -= len; | |
5678 | ||
5679 | repeat: | |
b7ea89ad TT |
5680 | ext4_ext_drop_refs(path1); |
5681 | kfree(path1); | |
5682 | ext4_ext_drop_refs(path2); | |
5683 | kfree(path2); | |
5684 | path1 = path2 = NULL; | |
fcf6b1b7 | 5685 | } |
fcf6b1b7 DM |
5686 | return replaced_count; |
5687 | } |