Blame view
fs/ext4/file.c
23 KB
b24413180 License cleanup: ... |
1 |
// SPDX-License-Identifier: GPL-2.0 |
ac27a0ec1 [PATCH] ext4: ini... |
2 |
/* |
617ba13b3 [PATCH] ext4: ren... |
3 |
* linux/fs/ext4/file.c |
ac27a0ec1 [PATCH] ext4: ini... |
4 5 6 7 8 9 10 11 12 13 14 15 |
* * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) * Laboratoire MASI - Institut Blaise Pascal * Universite Pierre et Marie Curie (Paris VI) * * from * * linux/fs/minix/file.c * * Copyright (C) 1991, 1992 Linus Torvalds * |
617ba13b3 [PATCH] ext4: ren... |
16 |
* ext4 fs regular file handling primitives |
ac27a0ec1 [PATCH] ext4: ini... |
17 18 19 20 21 22 23 |
* * 64-bit file support on 64-bit platforms by Jakub Jelinek * (jj@sunsite.ms.mff.cuni.cz) */ #include <linux/time.h> #include <linux/fs.h> |
545052e9e ext4: Switch to i... |
24 |
#include <linux/iomap.h> |
bc0b0d6d6 ext4: update the ... |
25 26 |
#include <linux/mount.h> #include <linux/path.h> |
c94c2acf8 dax: move DAX-rel... |
27 |
#include <linux/dax.h> |
871a29315 dquot: cleanup dq... |
28 |
#include <linux/quotaops.h> |
c8c0df241 ext4: introduce l... |
29 |
#include <linux/pagevec.h> |
e2e40f2c1 fs: move struct k... |
30 |
#include <linux/uio.h> |
b8a6176c2 ext4: Support for... |
31 |
#include <linux/mman.h> |
378f32bab ext4: introduce d... |
32 |
#include <linux/backing-dev.h> |
3dcf54515 ext4: move header... |
33 34 |
#include "ext4.h" #include "ext4_jbd2.h" |
ac27a0ec1 [PATCH] ext4: ini... |
35 36 |
#include "xattr.h" #include "acl.h" |
569342dc2 ext4: move inode ... |
37 |
#include "truncate.h" |
ac27a0ec1 [PATCH] ext4: ini... |
38 |
|
141f59b91 ANDROID: ext4, f2... |
39 |
static bool ext4_dio_supported(struct kiocb *iocb, struct iov_iter *iter) |
b1b4705d5 ext4: introduce d... |
40 |
{ |
141f59b91 ANDROID: ext4, f2... |
41 |
struct inode *inode = file_inode(iocb->ki_filp); |
8d6c90c9d ANDROID: fscrypt:... |
42 43 |
if (!fscrypt_dio_supported(iocb, iter)) return false; |
b1b4705d5 ext4: introduce d... |
44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
if (fsverity_active(inode)) return false; if (ext4_should_journal_data(inode)) return false; if (ext4_has_inline_data(inode)) return false; return true; } static ssize_t ext4_dio_read_iter(struct kiocb *iocb, struct iov_iter *to) { ssize_t ret; struct inode *inode = file_inode(iocb->ki_filp); if (iocb->ki_flags & IOCB_NOWAIT) { if (!inode_trylock_shared(inode)) return -EAGAIN; } else { inode_lock_shared(inode); } |
141f59b91 ANDROID: ext4, f2... |
64 |
if (!ext4_dio_supported(iocb, to)) { |
b1b4705d5 ext4: introduce d... |
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 |
inode_unlock_shared(inode); /* * Fallback to buffered I/O if the operation being performed on * the inode is not supported by direct I/O. The IOCB_DIRECT * flag needs to be cleared here in order to ensure that the * direct I/O path within generic_file_read_iter() is not * taken. */ iocb->ki_flags &= ~IOCB_DIRECT; return generic_file_read_iter(iocb, to); } ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL, is_sync_kiocb(iocb)); inode_unlock_shared(inode); file_accessed(iocb->ki_filp); return ret; } |
364443cbc ext4: convert DAX... |
84 85 86 87 88 |
#ifdef CONFIG_FS_DAX static ssize_t ext4_dax_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct inode *inode = file_inode(iocb->ki_filp); ssize_t ret; |
f629afe33 ext4: fix ext4_da... |
89 90 |
if (iocb->ki_flags & IOCB_NOWAIT) { if (!inode_trylock_shared(inode)) |
728fbc0e1 ext4: nowait aio ... |
91 |
return -EAGAIN; |
f629afe33 ext4: fix ext4_da... |
92 |
} else { |
728fbc0e1 ext4: nowait aio ... |
93 94 |
inode_lock_shared(inode); } |
364443cbc ext4: convert DAX... |
95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 |
/* * Recheck under inode lock - at this point we are sure it cannot * change anymore */ if (!IS_DAX(inode)) { inode_unlock_shared(inode); /* Fallback to buffered IO in case we cannot support DAX */ return generic_file_read_iter(iocb, to); } ret = dax_iomap_rw(iocb, to, &ext4_iomap_ops); inode_unlock_shared(inode); file_accessed(iocb->ki_filp); return ret; } #endif static ssize_t ext4_file_read_iter(struct kiocb *iocb, struct iov_iter *to) { |
b1b4705d5 ext4: introduce d... |
114 115 116 |
struct inode *inode = file_inode(iocb->ki_filp); if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) |
0db1ff222 ext4: add shutdow... |
117 |
return -EIO; |
364443cbc ext4: convert DAX... |
118 119 120 121 |
if (!iov_iter_count(to)) return 0; /* skip atime */ #ifdef CONFIG_FS_DAX |
b1b4705d5 ext4: introduce d... |
122 |
if (IS_DAX(inode)) |
364443cbc ext4: convert DAX... |
123 124 |
return ext4_dax_read_iter(iocb, to); #endif |
b1b4705d5 ext4: introduce d... |
125 126 |
if (iocb->ki_flags & IOCB_DIRECT) return ext4_dio_read_iter(iocb, to); |
364443cbc ext4: convert DAX... |
127 128 |
return generic_file_read_iter(iocb, to); } |
ac27a0ec1 [PATCH] ext4: ini... |
129 130 |
/* * Called when an inode is released. Note that this is different |
617ba13b3 [PATCH] ext4: ren... |
131 |
* from ext4_file_open: open gets called at every open, but release |
ac27a0ec1 [PATCH] ext4: ini... |
132 133 |
* gets called only when /all/ the files are closed. */ |
af5bc92dd ext4: Fix whitesp... |
134 |
static int ext4_release_file(struct inode *inode, struct file *filp) |
ac27a0ec1 [PATCH] ext4: ini... |
135 |
{ |
19f5fb7ad ext4: Use bitops ... |
136 |
if (ext4_test_inode_state(inode, EXT4_STATE_DA_ALLOC_CLOSE)) { |
7d8f9f7d1 ext4: Automatical... |
137 |
ext4_alloc_da_blocks(inode); |
19f5fb7ad ext4: Use bitops ... |
138 |
ext4_clear_inode_state(inode, EXT4_STATE_DA_ALLOC_CLOSE); |
7d8f9f7d1 ext4: Automatical... |
139 |
} |
ac27a0ec1 [PATCH] ext4: ini... |
140 141 |
/* if we are the last writer on the inode, drop the block reservation */ if ((filp->f_mode & FMODE_WRITE) && |
d6014301b ext4: Fix discard... |
142 |
(atomic_read(&inode->i_writecount) == 1) && |
e030a2881 ext4: fix coding ... |
143 |
!EXT4_I(inode)->i_reserved_data_blocks) { |
0e855ac8b ext4: Convert tru... |
144 |
down_write(&EXT4_I(inode)->i_data_sem); |
27bc446e2 ext4: limit the l... |
145 |
ext4_discard_preallocations(inode, 0); |
0e855ac8b ext4: Convert tru... |
146 |
up_write(&EXT4_I(inode)->i_data_sem); |
ac27a0ec1 [PATCH] ext4: ini... |
147 148 |
} if (is_dx(inode) && filp->private_data) |
617ba13b3 [PATCH] ext4: ren... |
149 |
ext4_htree_free_dir_info(filp->private_data); |
ac27a0ec1 [PATCH] ext4: ini... |
150 151 152 |
return 0; } |
e9e3bcecf ext4: serialize u... |
153 154 155 156 157 158 159 160 161 |
/* * This tests whether the IO in question is block-aligned or not. * Ext4 utilizes unwritten extents when hole-filling during direct IO, and they * are converted to written only after the IO is complete. Until they are * mapped, these blocks appear as holes, so dio_zero_block() will assume that * it needs to zero out portions of the start and/or end block. If 2 AIO * threads are at work on the same unwritten block, they must be synchronized * or one thread will zero the other's data, causing corruption. */ |
aa9714d0e ext4: Start with ... |
162 163 |
static bool ext4_unaligned_io(struct inode *inode, struct iov_iter *from, loff_t pos) |
e9e3bcecf ext4: serialize u... |
164 165 |
{ struct super_block *sb = inode->i_sb; |
aa9714d0e ext4: Start with ... |
166 |
unsigned long blockmask = sb->s_blocksize - 1; |
e9e3bcecf ext4: serialize u... |
167 |
|
9b884164d convert ext4 to -... |
168 |
if ((pos | iov_iter_alignment(from)) & blockmask) |
aa9714d0e ext4: Start with ... |
169 |
return true; |
e9e3bcecf ext4: serialize u... |
170 |
|
aa9714d0e ext4: Start with ... |
171 172 173 174 175 176 177 178 179 180 |
return false; } static bool ext4_extending_io(struct inode *inode, loff_t offset, size_t len) { if (offset + len > i_size_read(inode) || offset + len > EXT4_I(inode)->i_disksize) return true; return false; |
e9e3bcecf ext4: serialize u... |
181 |
} |
213bcd9cc ext4: factor out ... |
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 |
/* Is IO overwriting allocated and initialized blocks? */ static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len) { struct ext4_map_blocks map; unsigned int blkbits = inode->i_blkbits; int err, blklen; if (pos + len > i_size_read(inode)) return false; map.m_lblk = pos >> blkbits; map.m_len = EXT4_MAX_BLOCKS(len, pos, blkbits); blklen = map.m_len; err = ext4_map_blocks(NULL, inode, &map, 0); /* * 'err==len' means that all of the blocks have been preallocated, * regardless of whether they have been initialized or not. To exclude * unwritten extents, we need to check m_flags. */ return err == blklen && (map.m_flags & EXT4_MAP_MAPPED); } |
aa9714d0e ext4: Start with ... |
204 205 |
static ssize_t ext4_generic_write_checks(struct kiocb *iocb, struct iov_iter *from) |
213bcd9cc ext4: factor out ... |
206 207 208 |
{ struct inode *inode = file_inode(iocb->ki_filp); ssize_t ret; |
378f32bab ext4: introduce d... |
209 210 |
if (unlikely(IS_IMMUTABLE(inode))) return -EPERM; |
213bcd9cc ext4: factor out ... |
211 212 213 |
ret = generic_write_checks(iocb, from); if (ret <= 0) return ret; |
02b016ca7 ext4: enforce the... |
214 |
|
213bcd9cc ext4: factor out ... |
215 216 217 218 219 220 221 222 223 224 225 |
/* * If we have encountered a bitmap-format file, the size limit * is smaller than s_maxbytes, which is for extent-mapped files. */ if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); if (iocb->ki_pos >= sbi->s_bitmap_maxbytes) return -EFBIG; iov_iter_truncate(from, sbi->s_bitmap_maxbytes - iocb->ki_pos); } |
378f32bab ext4: introduce d... |
226 |
|
aa9714d0e ext4: Start with ... |
227 228 229 230 231 232 233 234 235 236 |
return iov_iter_count(from); } static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from) { ssize_t ret, count; count = ext4_generic_write_checks(iocb, from); if (count <= 0) return count; |
378f32bab ext4: introduce d... |
237 238 239 |
ret = file_modified(iocb->ki_filp); if (ret) return ret; |
aa9714d0e ext4: Start with ... |
240 |
return count; |
213bcd9cc ext4: factor out ... |
241 |
} |
378f32bab ext4: introduce d... |
242 243 244 245 246 247 248 249 |
static ssize_t ext4_buffered_write_iter(struct kiocb *iocb, struct iov_iter *from) { ssize_t ret; struct inode *inode = file_inode(iocb->ki_filp); if (iocb->ki_flags & IOCB_NOWAIT) return -EOPNOTSUPP; |
aa75f4d3d ext4: main fast-c... |
250 |
ext4_fc_start_update(inode); |
378f32bab ext4: introduce d... |
251 252 253 254 255 256 257 258 259 260 261 |
inode_lock(inode); ret = ext4_write_checks(iocb, from); if (ret <= 0) goto out; current->backing_dev_info = inode_to_bdi(inode); ret = generic_perform_write(iocb->ki_filp, from, iocb->ki_pos); current->backing_dev_info = NULL; out: inode_unlock(inode); |
aa75f4d3d ext4: main fast-c... |
262 |
ext4_fc_stop_update(inode); |
378f32bab ext4: introduce d... |
263 264 265 266 267 268 269 |
if (likely(ret > 0)) { iocb->ki_pos += ret; ret = generic_write_sync(iocb, ret); } return ret; } |
569342dc2 ext4: move inode ... |
270 271 272 273 274 275 276 |
static ssize_t ext4_handle_inode_extension(struct inode *inode, loff_t offset, ssize_t written, size_t count) { handle_t *handle; bool truncate = false; u8 blkbits = inode->i_blkbits; ext4_lblk_t written_blk, end_blk; |
4209ae12b ext4: handle ext4... |
277 |
int ret; |
569342dc2 ext4: move inode ... |
278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 |
/* * Note that EXT4_I(inode)->i_disksize can get extended up to * inode->i_size while the I/O was running due to writeback of delalloc * blocks. But, the code in ext4_iomap_alloc() is careful to use * zeroed/unwritten extents if this is possible; thus we won't leave * uninitialized blocks in a file even if we didn't succeed in writing * as much as we intended. */ WARN_ON_ONCE(i_size_read(inode) < EXT4_I(inode)->i_disksize); if (offset + count <= EXT4_I(inode)->i_disksize) { /* * We need to ensure that the inode is removed from the orphan * list if it has been added prematurely, due to writeback of * delalloc blocks. */ if (!list_empty(&EXT4_I(inode)->i_orphan) && inode->i_nlink) { handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); if (IS_ERR(handle)) { ext4_orphan_del(NULL, inode); return PTR_ERR(handle); } ext4_orphan_del(handle, inode); ext4_journal_stop(handle); } return written; } if (written < 0) goto truncate; handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); if (IS_ERR(handle)) { written = PTR_ERR(handle); goto truncate; } |
4209ae12b ext4: handle ext4... |
317 318 319 320 321 322 323 324 |
if (ext4_update_inode_size(inode, offset + written)) { ret = ext4_mark_inode_dirty(handle, inode); if (unlikely(ret)) { written = ret; ext4_journal_stop(handle); goto truncate; } } |
569342dc2 ext4: move inode ... |
325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 |
/* * We may need to truncate allocated but not written blocks beyond EOF. */ written_blk = ALIGN(offset + written, 1 << blkbits); end_blk = ALIGN(offset + count, 1 << blkbits); if (written_blk < end_blk && ext4_can_truncate(inode)) truncate = true; /* * Remove the inode from the orphan list if it has been extended and * everything went OK. */ if (!truncate && inode->i_nlink) ext4_orphan_del(handle, inode); ext4_journal_stop(handle); if (truncate) { truncate: ext4_truncate_failed_write(inode); /* * If the truncate operation failed early, then the inode may * still be on the orphan list. In that case, we need to try * remove the inode from the in-memory linked list. */ if (inode->i_nlink) ext4_orphan_del(NULL, inode); } return written; } |
378f32bab ext4: introduce d... |
356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 |
static int ext4_dio_write_end_io(struct kiocb *iocb, ssize_t size, int error, unsigned int flags) { loff_t offset = iocb->ki_pos; struct inode *inode = file_inode(iocb->ki_filp); if (error) return error; if (size && flags & IOMAP_DIO_UNWRITTEN) return ext4_convert_unwritten_extents(NULL, inode, offset, size); return 0; } static const struct iomap_dio_ops ext4_dio_write_ops = { .end_io = ext4_dio_write_end_io, }; |
aa9714d0e ext4: Start with ... |
375 376 377 378 379 380 381 382 383 384 385 |
/* * The intention here is to start with shared lock acquired then see if any * condition requires an exclusive inode lock. If yes, then we restart the * whole operation by releasing the shared lock and acquiring exclusive lock. * * - For unaligned_io we never take shared lock as it may cause data corruption * when two unaligned IO tries to modify the same block e.g. while zeroing. * * - For extending writes case we don't take the shared lock, since it requires * updating inode i_disksize and/or orphan handling with exclusive lock. * |
bc6385dab ext4: Move to sha... |
386 387 |
* - shared locking will only be true mostly with overwrites. Otherwise we will * switch to exclusive i_rwsem lock. |
aa9714d0e ext4: Start with ... |
388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 |
*/ static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from, bool *ilock_shared, bool *extend) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); loff_t offset; size_t count; ssize_t ret; restart: ret = ext4_generic_write_checks(iocb, from); if (ret <= 0) goto out; offset = iocb->ki_pos; count = ret; if (ext4_extending_io(inode, offset, count)) *extend = true; /* * Determine whether the IO operation will overwrite allocated |
bc6385dab ext4: Move to sha... |
409 |
* and initialized blocks. |
aa9714d0e ext4: Start with ... |
410 411 412 413 |
* We need exclusive i_rwsem for changing security info * in file_modified(). */ if (*ilock_shared && (!IS_NOSEC(inode) || *extend || |
aa9714d0e ext4: Start with ... |
414 |
!ext4_overwrite_io(inode, offset, count))) { |
0b3171b6d ext4: do not bloc... |
415 416 417 418 |
if (iocb->ki_flags & IOCB_NOWAIT) { ret = -EAGAIN; goto out; } |
aa9714d0e ext4: Start with ... |
419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 |
inode_unlock_shared(inode); *ilock_shared = false; inode_lock(inode); goto restart; } ret = file_modified(file); if (ret < 0) goto out; return count; out: if (*ilock_shared) inode_unlock_shared(inode); else inode_unlock(inode); return ret; } |
378f32bab ext4: introduce d... |
437 438 439 |
static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) { ssize_t ret; |
378f32bab ext4: introduce d... |
440 441 |
handle_t *handle; struct inode *inode = file_inode(iocb->ki_filp); |
aa9714d0e ext4: Start with ... |
442 443 |
loff_t offset = iocb->ki_pos; size_t count = iov_iter_count(from); |
8cd115bdd ext4: Optimize ex... |
444 |
const struct iomap_ops *iomap_ops = &ext4_iomap_ops; |
aa9714d0e ext4: Start with ... |
445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 |
bool extend = false, unaligned_io = false; bool ilock_shared = true; /* * We initially start with shared inode lock unless it is * unaligned IO which needs exclusive lock anyways. */ if (ext4_unaligned_io(inode, from, offset)) { unaligned_io = true; ilock_shared = false; } /* * Quick check here without any i_rwsem lock to see if it is extending * IO. A more reliable check is done in ext4_dio_write_checks() with * proper locking in place. */ if (offset + count > i_size_read(inode)) ilock_shared = false; |
378f32bab ext4: introduce d... |
463 464 |
if (iocb->ki_flags & IOCB_NOWAIT) { |
aa9714d0e ext4: Start with ... |
465 466 467 468 469 470 471 |
if (ilock_shared) { if (!inode_trylock_shared(inode)) return -EAGAIN; } else { if (!inode_trylock(inode)) return -EAGAIN; } |
378f32bab ext4: introduce d... |
472 |
} else { |
aa9714d0e ext4: Start with ... |
473 474 475 476 |
if (ilock_shared) inode_lock_shared(inode); else inode_lock(inode); |
378f32bab ext4: introduce d... |
477 |
} |
aa9714d0e ext4: Start with ... |
478 |
/* Fallback to buffered I/O if the inode does not support direct I/O. */ |
141f59b91 ANDROID: ext4, f2... |
479 |
if (!ext4_dio_supported(iocb, from)) { |
aa9714d0e ext4: Start with ... |
480 481 482 483 |
if (ilock_shared) inode_unlock_shared(inode); else inode_unlock(inode); |
378f32bab ext4: introduce d... |
484 485 |
return ext4_buffered_write_iter(iocb, from); } |
aa9714d0e ext4: Start with ... |
486 487 |
ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend); if (ret <= 0) |
378f32bab ext4: introduce d... |
488 |
return ret; |
378f32bab ext4: introduce d... |
489 |
|
6e014c621 ext4: don't block... |
490 491 492 493 494 |
/* if we're going to block and IOCB_NOWAIT is set, return -EAGAIN */ if ((iocb->ki_flags & IOCB_NOWAIT) && (unaligned_io || extend)) { ret = -EAGAIN; goto out; } |
378f32bab ext4: introduce d... |
495 |
offset = iocb->ki_pos; |
aa9714d0e ext4: Start with ... |
496 |
count = ret; |
378f32bab ext4: introduce d... |
497 498 |
/* |
aa9714d0e ext4: Start with ... |
499 500 501 502 503 504 505 506 |
* Unaligned direct IO must be serialized among each other as zeroing * of partial blocks of two competing unaligned IOs can result in data * corruption. * * So we make sure we don't allow any unaligned IO in flight. * For IOs where we need not wait (like unaligned non-AIO DIO), * below inode_dio_wait() may anyway become a no-op, since we start * with exclusive lock. |
378f32bab ext4: introduce d... |
507 |
*/ |
aa9714d0e ext4: Start with ... |
508 509 |
if (unaligned_io) inode_dio_wait(inode); |
378f32bab ext4: introduce d... |
510 |
|
aa9714d0e ext4: Start with ... |
511 |
if (extend) { |
378f32bab ext4: introduce d... |
512 513 514 515 516 |
handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; } |
aa75f4d3d ext4: main fast-c... |
517 |
ext4_fc_start_update(inode); |
378f32bab ext4: introduce d... |
518 |
ret = ext4_orphan_add(handle, inode); |
aa75f4d3d ext4: main fast-c... |
519 |
ext4_fc_stop_update(inode); |
378f32bab ext4: introduce d... |
520 521 522 523 |
if (ret) { ext4_journal_stop(handle); goto out; } |
378f32bab ext4: introduce d... |
524 525 |
ext4_journal_stop(handle); } |
8cd115bdd ext4: Optimize ex... |
526 527 528 |
if (ilock_shared) iomap_ops = &ext4_iomap_overwrite_ops; ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops, |
aa9714d0e ext4: Start with ... |
529 |
is_sync_kiocb(iocb) || unaligned_io || extend); |
60263d588 iomap: fall back ... |
530 531 |
if (ret == -ENOTBLK) ret = 0; |
378f32bab ext4: introduce d... |
532 533 534 535 536 |
if (extend) ret = ext4_handle_inode_extension(inode, offset, ret, count); out: |
aa9714d0e ext4: Start with ... |
537 |
if (ilock_shared) |
378f32bab ext4: introduce d... |
538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 |
inode_unlock_shared(inode); else inode_unlock(inode); if (ret >= 0 && iov_iter_count(from)) { ssize_t err; loff_t endbyte; offset = iocb->ki_pos; err = ext4_buffered_write_iter(iocb, from); if (err < 0) return err; /* * We need to ensure that the pages within the page cache for * the range covered by this I/O are written to disk and * invalidated. This is in attempt to preserve the expected * direct I/O semantics in the case we fallback to buffered I/O * to complete off the I/O request. */ ret += err; endbyte = offset + err - 1; err = filemap_write_and_wait_range(iocb->ki_filp->f_mapping, offset, endbyte); if (!err) invalidate_mapping_pages(iocb->ki_filp->f_mapping, offset >> PAGE_SHIFT, endbyte >> PAGE_SHIFT); } return ret; } |
776722e85 ext4: DAX iomap w... |
570 571 572 573 |
#ifdef CONFIG_FS_DAX static ssize_t ext4_dax_write_iter(struct kiocb *iocb, struct iov_iter *from) { |
776722e85 ext4: DAX iomap w... |
574 |
ssize_t ret; |
569342dc2 ext4: move inode ... |
575 576 |
size_t count; loff_t offset; |
0b9f230b9 ext4: move inode ... |
577 578 |
handle_t *handle; bool extend = false; |
569342dc2 ext4: move inode ... |
579 |
struct inode *inode = file_inode(iocb->ki_filp); |
776722e85 ext4: DAX iomap w... |
580 |
|
f629afe33 ext4: fix ext4_da... |
581 582 |
if (iocb->ki_flags & IOCB_NOWAIT) { if (!inode_trylock(inode)) |
728fbc0e1 ext4: nowait aio ... |
583 |
return -EAGAIN; |
f629afe33 ext4: fix ext4_da... |
584 |
} else { |
728fbc0e1 ext4: nowait aio ... |
585 586 |
inode_lock(inode); } |
378f32bab ext4: introduce d... |
587 |
|
776722e85 ext4: DAX iomap w... |
588 589 590 |
ret = ext4_write_checks(iocb, from); if (ret <= 0) goto out; |
776722e85 ext4: DAX iomap w... |
591 |
|
569342dc2 ext4: move inode ... |
592 593 |
offset = iocb->ki_pos; count = iov_iter_count(from); |
0b9f230b9 ext4: move inode ... |
594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 |
if (offset + count > EXT4_I(inode)->i_disksize) { handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; } ret = ext4_orphan_add(handle, inode); if (ret) { ext4_journal_stop(handle); goto out; } extend = true; ext4_journal_stop(handle); } |
776722e85 ext4: DAX iomap w... |
611 |
ret = dax_iomap_rw(iocb, from, &ext4_iomap_ops); |
0b9f230b9 ext4: move inode ... |
612 613 614 |
if (extend) ret = ext4_handle_inode_extension(inode, offset, ret, count); |
776722e85 ext4: DAX iomap w... |
615 |
out: |
ff5462e39 ext4: fix DAX wri... |
616 |
inode_unlock(inode); |
776722e85 ext4: DAX iomap w... |
617 618 619 620 621 |
if (ret > 0) ret = generic_write_sync(iocb, ret); return ret; } #endif |
ac27a0ec1 [PATCH] ext4: ini... |
622 |
static ssize_t |
9b884164d convert ext4 to -... |
623 |
ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from) |
ac27a0ec1 [PATCH] ext4: ini... |
624 |
{ |
8ad2850f4 ext4: move ext4_f... |
625 |
struct inode *inode = file_inode(iocb->ki_filp); |
7608e6104 ext4: inline gene... |
626 |
|
0db1ff222 ext4: add shutdow... |
627 628 |
if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; |
776722e85 ext4: DAX iomap w... |
629 630 631 632 |
#ifdef CONFIG_FS_DAX if (IS_DAX(inode)) return ext4_dax_write_iter(iocb, from); #endif |
378f32bab ext4: introduce d... |
633 634 |
if (iocb->ki_flags & IOCB_DIRECT) return ext4_dio_write_iter(iocb, from); |
aa75f4d3d ext4: main fast-c... |
635 636 |
else return ext4_buffered_write_iter(iocb, from); |
ac27a0ec1 [PATCH] ext4: ini... |
637 |
} |
923ae0ff9 ext4: add DAX fun... |
638 |
#ifdef CONFIG_FS_DAX |
71fe98996 fs: ext4: add new... |
639 |
static vm_fault_t ext4_dax_huge_fault(struct vm_fault *vmf, |
c791ace1e mm: replace FAULT... |
640 |
enum page_entry_size pe_size) |
923ae0ff9 ext4: add DAX fun... |
641 |
{ |
71fe98996 fs: ext4: add new... |
642 643 |
int error = 0; vm_fault_t result; |
224464231 ext4: fix ENOSPC ... |
644 |
int retries = 0; |
fb26a1cbe ext4: return to s... |
645 |
handle_t *handle = NULL; |
11bac8000 mm, fs: reduce fa... |
646 |
struct inode *inode = file_inode(vmf->vma->vm_file); |
ea3d7209c ext4: fix races b... |
647 |
struct super_block *sb = inode->i_sb; |
fd96b8da6 ext4: fix fault h... |
648 649 650 651 652 653 654 655 656 657 658 659 660 661 |
/* * We have to distinguish real writes from writes which will result in a * COW page; COW writes should *not* poke the journal (the file will not * be changed). Doing so would cause unintended failures when mounted * read-only. * * We check for VM_SHARED rather than vmf->cow_page since the latter is * unset for pe_size != PE_SIZE_PTE (i.e. only in do_cow_fault); for * other sizes, dax_iomap_fault will handle splitting / fallback so that * we eventually come back with a COW page. */ bool write = (vmf->flags & FAULT_FLAG_WRITE) && (vmf->vma->vm_flags & VM_SHARED); |
b8a6176c2 ext4: Support for... |
662 |
pfn_t pfn; |
01a33b4ac ext4: start trans... |
663 664 665 |
if (write) { sb_start_pagefault(sb); |
11bac8000 mm, fs: reduce fa... |
666 |
file_update_time(vmf->vma->vm_file); |
fb26a1cbe ext4: return to s... |
667 |
down_read(&EXT4_I(inode)->i_mmap_sem); |
224464231 ext4: fix ENOSPC ... |
668 |
retry: |
fb26a1cbe ext4: return to s... |
669 670 |
handle = ext4_journal_start_sb(sb, EXT4_HT_WRITE_PAGE, EXT4_DATA_TRANS_BLOCKS(sb)); |
497f6926d ext4: Simplify er... |
671 672 673 674 675 |
if (IS_ERR(handle)) { up_read(&EXT4_I(inode)->i_mmap_sem); sb_end_pagefault(sb); return VM_FAULT_SIGBUS; } |
fb26a1cbe ext4: return to s... |
676 677 |
} else { down_read(&EXT4_I(inode)->i_mmap_sem); |
1db175428 ext4: Simplify DA... |
678 |
} |
224464231 ext4: fix ENOSPC ... |
679 |
result = dax_iomap_fault(vmf, pe_size, &pfn, &error, &ext4_iomap_ops); |
fb26a1cbe ext4: return to s... |
680 |
if (write) { |
497f6926d ext4: Simplify er... |
681 |
ext4_journal_stop(handle); |
224464231 ext4: fix ENOSPC ... |
682 683 684 685 |
if ((result & VM_FAULT_ERROR) && error == -ENOSPC && ext4_should_retry_alloc(sb, &retries)) goto retry; |
b8a6176c2 ext4: Support for... |
686 687 688 |
/* Handling synchronous page fault? */ if (result & VM_FAULT_NEEDDSYNC) result = dax_finish_sync_fault(vmf, pe_size, pfn); |
fb26a1cbe ext4: return to s... |
689 |
up_read(&EXT4_I(inode)->i_mmap_sem); |
01a33b4ac ext4: start trans... |
690 |
sb_end_pagefault(sb); |
fb26a1cbe ext4: return to s... |
691 692 693 |
} else { up_read(&EXT4_I(inode)->i_mmap_sem); } |
01a33b4ac ext4: start trans... |
694 695 |
return result; |
923ae0ff9 ext4: add DAX fun... |
696 |
} |
71fe98996 fs: ext4: add new... |
697 |
static vm_fault_t ext4_dax_fault(struct vm_fault *vmf) |
c791ace1e mm: replace FAULT... |
698 699 700 |
{ return ext4_dax_huge_fault(vmf, PE_SIZE_PTE); } |
923ae0ff9 ext4: add DAX fun... |
701 702 |
static const struct vm_operations_struct ext4_dax_vm_ops = { .fault = ext4_dax_fault, |
c791ace1e mm: replace FAULT... |
703 |
.huge_fault = ext4_dax_huge_fault, |
1e9d180ba ext2, ext4: fix i... |
704 |
.page_mkwrite = ext4_dax_fault, |
91d25ba8a dax: use common 4... |
705 |
.pfn_mkwrite = ext4_dax_fault, |
923ae0ff9 ext4: add DAX fun... |
706 707 708 709 |
}; #else #define ext4_dax_vm_ops ext4_file_vm_ops #endif |
f0f37e2f7 const: mark struc... |
710 |
static const struct vm_operations_struct ext4_file_vm_ops = { |
ea3d7209c ext4: fix races b... |
711 |
.fault = ext4_filemap_fault, |
f1820361f mm: implement ->m... |
712 |
.map_pages = filemap_map_pages, |
2e9ee8503 ext4: Use page_mk... |
713 714 715 716 717 |
.page_mkwrite = ext4_page_mkwrite, }; static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) { |
c9c7429c2 ext4 crypto: impl... |
718 |
struct inode *inode = file->f_mapping->host; |
e46bfc3f0 ext4: disable map... |
719 720 |
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); struct dax_device *dax_dev = sbi->s_daxdev; |
c9c7429c2 ext4 crypto: impl... |
721 |
|
e46bfc3f0 ext4: disable map... |
722 |
if (unlikely(ext4_forced_shutdown(sbi))) |
0db1ff222 ext4: add shutdow... |
723 |
return -EIO; |
b8a6176c2 ext4: Support for... |
724 |
/* |
e46bfc3f0 ext4: disable map... |
725 726 |
* We don't support synchronous mappings for non-DAX files and * for DAX files if underneath dax_device is not synchronous. |
b8a6176c2 ext4: Support for... |
727 |
*/ |
e46bfc3f0 ext4: disable map... |
728 |
if (!daxdev_mapping_supported(vma, dax_dev)) |
b8a6176c2 ext4: Support for... |
729 |
return -EOPNOTSUPP; |
2e9ee8503 ext4: Use page_mk... |
730 |
file_accessed(file); |
923ae0ff9 ext4: add DAX fun... |
731 732 |
if (IS_DAX(file_inode(file))) { vma->vm_ops = &ext4_dax_vm_ops; |
e1fb4a086 dax: remove VM_MI... |
733 |
vma->vm_flags |= VM_HUGEPAGE; |
923ae0ff9 ext4: add DAX fun... |
734 735 736 |
} else { vma->vm_ops = &ext4_file_vm_ops; } |
2e9ee8503 ext4: Use page_mk... |
737 738 |
return 0; } |
833a95088 ext4: factor out ... |
739 740 |
static int ext4_sample_last_mounted(struct super_block *sb, struct vfsmount *mnt) |
bc0b0d6d6 ext4: update the ... |
741 |
{ |
833a95088 ext4: factor out ... |
742 |
struct ext4_sb_info *sbi = EXT4_SB(sb); |
bc0b0d6d6 ext4: update the ... |
743 744 |
struct path path; char buf[64], *cp; |
833a95088 ext4: factor out ... |
745 746 |
handle_t *handle; int err; |
9b5f6c9b8 ext4: make s_moun... |
747 |
if (likely(ext4_test_mount_flag(sb, EXT4_MF_MNTDIR_SAMPLED))) |
833a95088 ext4: factor out ... |
748 |
return 0; |
db6516a5e ext4: do not upda... |
749 |
if (sb_rdonly(sb) || !sb_start_intwrite_trylock(sb)) |
833a95088 ext4: factor out ... |
750 |
return 0; |
9b5f6c9b8 ext4: make s_moun... |
751 |
ext4_set_mount_flag(sb, EXT4_MF_MNTDIR_SAMPLED); |
833a95088 ext4: factor out ... |
752 753 754 755 756 757 758 759 760 761 |
/* * Sample where the filesystem has been mounted and * store it in the superblock for sysadmin convenience * when trying to sort through large numbers of block * devices or filesystem images. */ memset(buf, 0, sizeof(buf)); path.mnt = mnt; path.dentry = mnt->mnt_root; cp = d_path(&path, buf, sizeof(buf)); |
db6516a5e ext4: do not upda... |
762 |
err = 0; |
833a95088 ext4: factor out ... |
763 |
if (IS_ERR(cp)) |
db6516a5e ext4: do not upda... |
764 |
goto out; |
833a95088 ext4: factor out ... |
765 766 |
handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1); |
db6516a5e ext4: do not upda... |
767 |
err = PTR_ERR(handle); |
833a95088 ext4: factor out ... |
768 |
if (IS_ERR(handle)) |
db6516a5e ext4: do not upda... |
769 |
goto out; |
833a95088 ext4: factor out ... |
770 771 772 |
BUFFER_TRACE(sbi->s_sbh, "get_write_access"); err = ext4_journal_get_write_access(handle, sbi->s_sbh); if (err) |
db6516a5e ext4: do not upda... |
773 |
goto out_journal; |
85958f60e ext4: don't leak ... |
774 |
strncpy(sbi->s_es->s_last_mounted, cp, |
833a95088 ext4: factor out ... |
775 776 |
sizeof(sbi->s_es->s_last_mounted)); ext4_handle_dirty_super(handle, sb); |
db6516a5e ext4: do not upda... |
777 |
out_journal: |
833a95088 ext4: factor out ... |
778 |
ext4_journal_stop(handle); |
db6516a5e ext4: do not upda... |
779 780 |
out: sb_end_intwrite(sb); |
833a95088 ext4: factor out ... |
781 782 |
return err; } |
e030a2881 ext4: fix coding ... |
783 |
static int ext4_file_open(struct inode *inode, struct file *filp) |
833a95088 ext4: factor out ... |
784 |
{ |
c9c7429c2 ext4 crypto: impl... |
785 |
int ret; |
bc0b0d6d6 ext4: update the ... |
786 |
|
0db1ff222 ext4: add shutdow... |
787 788 |
if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; |
833a95088 ext4: factor out ... |
789 790 791 |
ret = ext4_sample_last_mounted(inode->i_sb, filp->f_path.mnt); if (ret) return ret; |
9dd78d8c9 ext4: use dget_pa... |
792 |
|
09a5c31c9 ext4: switch to f... |
793 794 |
ret = fscrypt_file_open(inode, filp); if (ret) |
c93d8f885 ext4: add basic f... |
795 796 797 798 |
return ret; ret = fsverity_file_open(inode, filp); if (ret) |
09a5c31c9 ext4: switch to f... |
799 |
return ret; |
8aefcd557 ext4: dynamically... |
800 801 802 803 |
/* * Set up the jbd2_inode if we are opening the inode for * writing and the journal is present */ |
a361293f5 jbd2: Fix oops in... |
804 |
if (filp->f_mode & FMODE_WRITE) { |
c9c7429c2 ext4 crypto: impl... |
805 |
ret = ext4_inode_attach_jinode(inode); |
a361293f5 jbd2: Fix oops in... |
806 807 |
if (ret < 0) return ret; |
8aefcd557 ext4: dynamically... |
808 |
} |
728fbc0e1 ext4: nowait aio ... |
809 |
|
766ef1e10 ext4: flag as sup... |
810 |
filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC; |
abdd438b2 ext4 crypto: hand... |
811 |
return dquot_file_open(inode, filp); |
bc0b0d6d6 ext4: update the ... |
812 |
} |
e0d10bfa9 ext4: improve lls... |
813 |
/* |
ec7268ce2 ext4: use core vf... |
814 815 816 |
* ext4_llseek() handles both block-mapped and extent-mapped maxbytes values * by calling generic_file_llseek_size() with the appropriate maxbytes * value for each. |
e0d10bfa9 ext4: improve lls... |
817 |
*/ |
965c8e59c lseek: the "whenc... |
818 |
loff_t ext4_llseek(struct file *file, loff_t offset, int whence) |
e0d10bfa9 ext4: improve lls... |
819 820 821 822 823 824 825 826 |
{ struct inode *inode = file->f_mapping->host; loff_t maxbytes; if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) maxbytes = EXT4_SB(inode->i_sb)->s_bitmap_maxbytes; else maxbytes = inode->i_sb->s_maxbytes; |
e0d10bfa9 ext4: improve lls... |
827 |
|
965c8e59c lseek: the "whenc... |
828 |
switch (whence) { |
545052e9e ext4: Switch to i... |
829 |
default: |
965c8e59c lseek: the "whenc... |
830 |
return generic_file_llseek_size(file, offset, whence, |
c8c0df241 ext4: introduce l... |
831 |
maxbytes, i_size_read(inode)); |
c8c0df241 ext4: introduce l... |
832 |
case SEEK_HOLE: |
545052e9e ext4: Switch to i... |
833 |
inode_lock_shared(inode); |
09edf4d38 ext4: introduce n... |
834 835 |
offset = iomap_seek_hole(inode, offset, &ext4_iomap_report_ops); |
545052e9e ext4: Switch to i... |
836 837 838 839 |
inode_unlock_shared(inode); break; case SEEK_DATA: inode_lock_shared(inode); |
09edf4d38 ext4: introduce n... |
840 841 |
offset = iomap_seek_data(inode, offset, &ext4_iomap_report_ops); |
545052e9e ext4: Switch to i... |
842 843 |
inode_unlock_shared(inode); break; |
c8c0df241 ext4: introduce l... |
844 |
} |
545052e9e ext4: Switch to i... |
845 846 847 |
if (offset < 0) return offset; return vfs_setpos(file, offset, maxbytes); |
e0d10bfa9 ext4: improve lls... |
848 |
} |
617ba13b3 [PATCH] ext4: ren... |
849 |
const struct file_operations ext4_file_operations = { |
e0d10bfa9 ext4: improve lls... |
850 |
.llseek = ext4_llseek, |
364443cbc ext4: convert DAX... |
851 |
.read_iter = ext4_file_read_iter, |
9b884164d convert ext4 to -... |
852 |
.write_iter = ext4_file_write_iter, |
72f9da1d5 ext4: start to su... |
853 |
.iopoll = iomap_dio_iopoll, |
5cdd7b2d7 Convert ext4 to u... |
854 |
.unlocked_ioctl = ext4_ioctl, |
ac27a0ec1 [PATCH] ext4: ini... |
855 |
#ifdef CONFIG_COMPAT |
617ba13b3 [PATCH] ext4: ren... |
856 |
.compat_ioctl = ext4_compat_ioctl, |
ac27a0ec1 [PATCH] ext4: ini... |
857 |
#endif |
2e9ee8503 ext4: Use page_mk... |
858 |
.mmap = ext4_file_mmap, |
b8a6176c2 ext4: Support for... |
859 |
.mmap_supported_flags = MAP_SYNC, |
bc0b0d6d6 ext4: update the ... |
860 |
.open = ext4_file_open, |
617ba13b3 [PATCH] ext4: ren... |
861 862 |
.release = ext4_release_file, .fsync = ext4_sync_file, |
dbe6ec815 ext2/4, xfs: call... |
863 |
.get_unmapped_area = thp_get_unmapped_area, |
ac27a0ec1 [PATCH] ext4: ini... |
864 |
.splice_read = generic_file_splice_read, |
8d0207652 ->splice_write() ... |
865 |
.splice_write = iter_file_splice_write, |
2fe17c107 fallocate should ... |
866 |
.fallocate = ext4_fallocate, |
ac27a0ec1 [PATCH] ext4: ini... |
867 |
}; |
754661f14 [PATCH] mark stru... |
868 |
const struct inode_operations ext4_file_inode_operations = { |
617ba13b3 [PATCH] ext4: ren... |
869 |
.setattr = ext4_setattr, |
99652ea56 ext4: Add statx s... |
870 |
.getattr = ext4_file_getattr, |
617ba13b3 [PATCH] ext4: ren... |
871 |
.listxattr = ext4_listxattr, |
4e34e719e fs: take the ACL ... |
872 |
.get_acl = ext4_get_acl, |
64e178a71 ext2/3/4: use gen... |
873 |
.set_acl = ext4_set_acl, |
6873fa0de Hook ext4 to the ... |
874 |
.fiemap = ext4_fiemap, |
ac27a0ec1 [PATCH] ext4: ini... |
875 |
}; |