Blame view
mm/filemap.c
64.5 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 |
/* * linux/mm/filemap.c * * Copyright (C) 1994-1999 Linus Torvalds */ /* * This file handles the generic file mmap semantics used by * most "normal" filesystems (but you don't /have/ to use this: * the NFS filesystem used to do this differently, for example) */ |
1da177e4c
|
12 13 14 15 |
#include <linux/module.h> #include <linux/slab.h> #include <linux/compiler.h> #include <linux/fs.h> |
c22ce143d
|
16 |
#include <linux/uaccess.h> |
1da177e4c
|
17 |
#include <linux/aio.h> |
c59ede7b7
|
18 |
#include <linux/capability.h> |
1da177e4c
|
19 20 21 22 23 24 25 26 27 |
#include <linux/kernel_stat.h> #include <linux/mm.h> #include <linux/swap.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/file.h> #include <linux/uio.h> #include <linux/hash.h> #include <linux/writeback.h> |
53253383f
|
28 |
#include <linux/backing-dev.h> |
1da177e4c
|
29 30 31 32 |
#include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/security.h> #include <linux/syscalls.h> |
44110fe38
|
33 |
#include <linux/cpuset.h> |
2f718ffc1
|
34 |
#include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */ |
8a9f3ccd2
|
35 |
#include <linux/memcontrol.h> |
4f98a2fee
|
36 |
#include <linux/mm_inline.h> /* for page_is_file_cache() */ |
0f8053a50
|
37 |
#include "internal.h" |
1da177e4c
|
38 |
/* |
1da177e4c
|
39 40 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba
|
41 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c
|
42 |
|
1da177e4c
|
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
#include <asm/mman.h> /* * Shared mappings implemented 30.11.1994. It's not fully working yet, * though. * * Shared mappings now work. 15.8.1995 Bruno. * * finished 'unifying' the page and buffer cache and SMP-threaded the * page-cache, 21.05.1999, Ingo Molnar <mingo@redhat.com> * * SMP-threaded pagemap-LRU 1999, Andrea Arcangeli <andrea@suse.de> */ /* * Lock ordering: * |
25d9e2d15
|
60 |
* ->i_mmap_lock (truncate_pagecache) |
1da177e4c
|
61 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919
|
62 63 |
* ->swap_lock (exclusive_swap_page, others) * ->mapping->tree_lock |
1da177e4c
|
64 |
* |
1b1dcc1b5
|
65 |
* ->i_mutex |
1da177e4c
|
66 67 68 69 |
* ->i_mmap_lock (truncate->unmap_mapping_range) * * ->mmap_sem * ->i_mmap_lock |
b8072f099
|
70 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
1da177e4c
|
71 72 73 74 75 |
* ->mapping->tree_lock (arch-dependent flush_dcache_mmap_lock) * * ->mmap_sem * ->lock_page (access_process_vm) * |
82591e6ea
|
76 77 |
* ->i_mutex (generic_file_buffered_write) * ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c
|
78 |
* |
1b1dcc1b5
|
79 |
* ->i_mutex |
1da177e4c
|
80 81 82 83 84 85 86 87 88 89 |
* ->i_alloc_sem (various) * * ->inode_lock * ->sb_lock (fs/fs-writeback.c) * ->mapping->tree_lock (__sync_single_inode) * * ->i_mmap_lock * ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099
|
90 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c
|
91 |
* |
b8072f099
|
92 |
* ->page_table_lock or pte_lock |
5d337b919
|
93 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c
|
94 95 96 |
* ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) |
053837fce
|
97 |
* ->zone.lru_lock (check_pte_range->isolate_lru_page) |
1da177e4c
|
98 99 100 101 102 103 104 105 |
* ->private_lock (page_remove_rmap->set_page_dirty) * ->tree_lock (page_remove_rmap->set_page_dirty) * ->inode_lock (page_remove_rmap->set_page_dirty) * ->inode_lock (zap_pte_range->set_page_dirty) * ->private_lock (zap_pte_range->__set_page_dirty_buffers) * * ->task->proc_lock * ->dcache_lock (proc_pid_lookup) |
6a46079cf
|
106 107 108 109 |
* * (code doesn't rely on that order, so you could switch it around) * ->tasklist_lock (memory_failure, collect_procs_ao) * ->i_mmap_lock |
1da177e4c
|
110 111 112 113 114 |
*/ /* * Remove a page from the page cache and free it. Caller has to make * sure the page is locked and that nobody else uses it - or that usage |
19fd62312
|
115 |
* is safe. The caller must hold the mapping's tree_lock. |
1da177e4c
|
116 117 118 119 120 121 122 123 |
*/ void __remove_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; radix_tree_delete(&mapping->page_tree, page->index); page->mapping = NULL; mapping->nrpages--; |
347ce434d
|
124 |
__dec_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac
|
125 126 |
if (PageSwapBacked(page)) __dec_zone_page_state(page, NR_SHMEM); |
45426812d
|
127 |
BUG_ON(page_mapped(page)); |
3a6927906
|
128 129 130 131 132 133 134 135 136 137 138 139 |
/* * Some filesystems seem to re-dirty the page even after * the VM has canceled the dirty bit (eg ext3 journaling). * * Fix it up by doing a final dirty accounting check after * having removed the page entirely. */ if (PageDirty(page) && mapping_cap_account_dirty(mapping)) { dec_zone_page_state(page, NR_FILE_DIRTY); dec_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); } |
1da177e4c
|
140 141 142 143 144 |
} void remove_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; |
cd7619d6b
|
145 |
BUG_ON(!PageLocked(page)); |
1da177e4c
|
146 |
|
19fd62312
|
147 |
spin_lock_irq(&mapping->tree_lock); |
1da177e4c
|
148 |
__remove_from_page_cache(page); |
19fd62312
|
149 |
spin_unlock_irq(&mapping->tree_lock); |
e767e0561
|
150 |
mem_cgroup_uncharge_cache_page(page); |
1da177e4c
|
151 152 153 154 155 156 |
} static int sync_page(void *word) { struct address_space *mapping; struct page *page; |
07808b74e
|
157 |
page = container_of((unsigned long *)word, struct page, flags); |
1da177e4c
|
158 159 |
/* |
dd1d5afca
|
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 |
* page_mapping() is being called without PG_locked held. * Some knowledge of the state and use of the page is used to * reduce the requirements down to a memory barrier. * The danger here is of a stale page_mapping() return value * indicating a struct address_space different from the one it's * associated with when it is associated with one. * After smp_mb(), it's either the correct page_mapping() for * the page, or an old page_mapping() and the page's own * page_mapping() has gone NULL. * The ->sync_page() address_space operation must tolerate * page_mapping() going NULL. By an amazing coincidence, * this comes about because none of the users of the page * in the ->sync_page() methods make essential use of the * page_mapping(), merely passing the page down to the backing * device's unplug functions when it's non-NULL, which in turn |
4c21e2f24
|
175 |
* ignore it for all cases but swap, where only page_private(page) is |
dd1d5afca
|
176 177 178 |
* of interest. When page_mapping() does go NULL, the entire * call stack gracefully ignores the page and returns. * -- wli |
1da177e4c
|
179 180 181 182 183 184 185 186 |
*/ smp_mb(); mapping = page_mapping(page); if (mapping && mapping->a_ops && mapping->a_ops->sync_page) mapping->a_ops->sync_page(page); io_schedule(); return 0; } |
2687a3569
|
187 188 189 190 191 |
static int sync_page_killable(void *word) { sync_page(word); return fatal_signal_pending(current) ? -EINTR : 0; } |
1da177e4c
|
192 |
/** |
485bb99b4
|
193 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b
|
194 195 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03
|
196 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b
|
197 |
* @sync_mode: enable synchronous operation |
1da177e4c
|
198 |
* |
485bb99b4
|
199 200 201 |
* Start writeback against all of a mapping's dirty pages that lie * within the byte offsets <start, end> inclusive. * |
1da177e4c
|
202 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4
|
203 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c
|
204 205 206 |
* these two operations is that if a dirty page/buffer is encountered, it must * be waited upon, and not just skipped over. */ |
ebcf28e1c
|
207 208 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c
|
209 210 211 212 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd
|
213 |
.nr_to_write = LONG_MAX, |
111ebb6e6
|
214 215 |
.range_start = start, .range_end = end, |
1da177e4c
|
216 217 218 219 220 221 222 223 224 225 226 227 |
}; if (!mapping_cap_writeback_dirty(mapping)) return 0; ret = do_writepages(mapping, &wbc); return ret; } static inline int __filemap_fdatawrite(struct address_space *mapping, int sync_mode) { |
111ebb6e6
|
228 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c
|
229 230 231 232 233 234 235 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf
|
236 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c
|
237 |
loff_t end) |
1da177e4c
|
238 239 240 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf
|
241 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c
|
242 |
|
485bb99b4
|
243 244 245 246 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c
|
247 248 249 250 251 252 253 254 |
* This is a mostly non-blocking flush. Not suitable for data-integrity * purposes - I/O may not be started against all dirty pages. */ int filemap_flush(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_NONE); } EXPORT_SYMBOL(filemap_flush); |
485bb99b4
|
255 |
/** |
94004ed72
|
256 257 258 259 |
* filemap_fdatawait_range - wait for writeback to complete * @mapping: address space structure to wait for * @start_byte: offset in bytes where the range starts * @end_byte: offset in bytes where the range ends (inclusive) |
485bb99b4
|
260 |
* |
94004ed72
|
261 262 |
* Walk the list of under-writeback pages of the given address space * in the given range and wait for all of them. |
1da177e4c
|
263 |
*/ |
94004ed72
|
264 265 |
int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) |
1da177e4c
|
266 |
{ |
94004ed72
|
267 268 |
pgoff_t index = start_byte >> PAGE_CACHE_SHIFT; pgoff_t end = end_byte >> PAGE_CACHE_SHIFT; |
1da177e4c
|
269 270 271 |
struct pagevec pvec; int nr_pages; int ret = 0; |
1da177e4c
|
272 |
|
94004ed72
|
273 |
if (end_byte < start_byte) |
1da177e4c
|
274 275 276 |
return 0; pagevec_init(&pvec, 0); |
1da177e4c
|
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 |
while ((index <= end) && (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, PAGECACHE_TAG_WRITEBACK, min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1)) != 0) { unsigned i; for (i = 0; i < nr_pages; i++) { struct page *page = pvec.pages[i]; /* until radix tree lookup accepts end_index */ if (page->index > end) continue; wait_on_page_writeback(page); if (PageError(page)) ret = -EIO; } pagevec_release(&pvec); cond_resched(); } /* Check for outstanding write errors */ if (test_and_clear_bit(AS_ENOSPC, &mapping->flags)) ret = -ENOSPC; if (test_and_clear_bit(AS_EIO, &mapping->flags)) ret = -EIO; return ret; } |
d3bccb6f4
|
306 307 308 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
485bb99b4
|
309 |
* filemap_fdatawait - wait for all under-writeback pages to complete |
1da177e4c
|
310 |
* @mapping: address space structure to wait for |
485bb99b4
|
311 312 313 |
* * Walk the list of under-writeback pages of the given address space * and wait for all of them. |
1da177e4c
|
314 315 316 317 318 319 320 |
*/ int filemap_fdatawait(struct address_space *mapping) { loff_t i_size = i_size_read(mapping->host); if (i_size == 0) return 0; |
94004ed72
|
321 |
return filemap_fdatawait_range(mapping, 0, i_size - 1); |
1da177e4c
|
322 323 324 325 326 |
} EXPORT_SYMBOL(filemap_fdatawait); int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982
|
327 |
int err = 0; |
1da177e4c
|
328 329 |
if (mapping->nrpages) { |
28fd12982
|
330 331 332 333 334 335 336 337 338 339 340 341 |
err = filemap_fdatawrite(mapping); /* * Even if the above returned error, the pages may be * written partially (e.g. -ENOSPC), so we wait for it. * But the -EIO is special case, it may indicate the worst * thing (e.g. bug) happened, so we avoid waiting for it. */ if (err != -EIO) { int err2 = filemap_fdatawait(mapping); if (!err) err = err2; } |
1da177e4c
|
342 |
} |
28fd12982
|
343 |
return err; |
1da177e4c
|
344 |
} |
28fd12982
|
345 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c
|
346 |
|
485bb99b4
|
347 348 349 350 351 352 |
/** * filemap_write_and_wait_range - write out & wait on a file range * @mapping: the address_space for the pages * @lstart: offset in bytes where the range starts * @lend: offset in bytes where the range ends (inclusive) * |
469eb4d03
|
353 354 355 356 357 |
* Write out and wait upon file offsets lstart->lend, inclusive. * * Note that `lend' is inclusive (describes the last byte to be written) so * that this function can be used to write to the very end-of-file (end = -1). */ |
1da177e4c
|
358 359 360 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982
|
361 |
int err = 0; |
1da177e4c
|
362 363 |
if (mapping->nrpages) { |
28fd12982
|
364 365 366 367 |
err = __filemap_fdatawrite_range(mapping, lstart, lend, WB_SYNC_ALL); /* See comment of filemap_write_and_wait() */ if (err != -EIO) { |
94004ed72
|
368 369 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982
|
370 371 372 |
if (!err) err = err2; } |
1da177e4c
|
373 |
} |
28fd12982
|
374 |
return err; |
1da177e4c
|
375 |
} |
f69955855
|
376 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c
|
377 |
|
485bb99b4
|
378 |
/** |
e286781d5
|
379 |
* add_to_page_cache_locked - add a locked page to the pagecache |
485bb99b4
|
380 381 382 383 384 |
* @page: page to add * @mapping: the page's address_space * @offset: page index * @gfp_mask: page allocation mode * |
e286781d5
|
385 |
* This function is used to add a page to the pagecache. It must be locked. |
1da177e4c
|
386 387 |
* This function does not add the page to the LRU. The caller must do that. */ |
e286781d5
|
388 |
int add_to_page_cache_locked(struct page *page, struct address_space *mapping, |
6daa0e286
|
389 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c
|
390 |
{ |
e286781d5
|
391 392 393 394 395 |
int error; VM_BUG_ON(!PageLocked(page)); error = mem_cgroup_cache_charge(page, current->mm, |
2c26fdd70
|
396 |
gfp_mask & GFP_RECLAIM_MASK); |
35c754d79
|
397 398 |
if (error) goto out; |
1da177e4c
|
399 |
|
35c754d79
|
400 |
error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM); |
1da177e4c
|
401 |
if (error == 0) { |
e286781d5
|
402 403 404 |
page_cache_get(page); page->mapping = mapping; page->index = offset; |
19fd62312
|
405 |
spin_lock_irq(&mapping->tree_lock); |
1da177e4c
|
406 |
error = radix_tree_insert(&mapping->page_tree, offset, page); |
e286781d5
|
407 |
if (likely(!error)) { |
1da177e4c
|
408 |
mapping->nrpages++; |
347ce434d
|
409 |
__inc_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac
|
410 411 |
if (PageSwapBacked(page)) __inc_zone_page_state(page, NR_SHMEM); |
e767e0561
|
412 |
spin_unlock_irq(&mapping->tree_lock); |
e286781d5
|
413 414 |
} else { page->mapping = NULL; |
e767e0561
|
415 |
spin_unlock_irq(&mapping->tree_lock); |
69029cd55
|
416 |
mem_cgroup_uncharge_cache_page(page); |
e286781d5
|
417 418 |
page_cache_release(page); } |
1da177e4c
|
419 |
radix_tree_preload_end(); |
35c754d79
|
420 |
} else |
69029cd55
|
421 |
mem_cgroup_uncharge_cache_page(page); |
8a9f3ccd2
|
422 |
out: |
1da177e4c
|
423 424 |
return error; } |
e286781d5
|
425 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c
|
426 427 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286
|
428 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c
|
429 |
{ |
4f98a2fee
|
430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 |
int ret; /* * Splice_read and readahead add shmem/tmpfs pages into the page cache * before shmem_readpage has a chance to mark them as SwapBacked: they * need to go on the active_anon lru below, and mem_cgroup_cache_charge * (called in add_to_page_cache) needs to know where they're going too. */ if (mapping_cap_swap_backed(mapping)) SetPageSwapBacked(page); ret = add_to_page_cache(page, mapping, offset, gfp_mask); if (ret == 0) { if (page_is_file_cache(page)) lru_cache_add_file(page); else lru_cache_add_active_anon(page); } |
1da177e4c
|
448 449 |
return ret; } |
18bc0bbd1
|
450 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c
|
451 |
|
44110fe38
|
452 |
#ifdef CONFIG_NUMA |
2ae88149a
|
453 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38
|
454 455 456 |
{ if (cpuset_do_page_mem_spread()) { int n = cpuset_mem_spread_node(); |
6484eb3e2
|
457 |
return alloc_pages_exact_node(n, gfp, 0); |
44110fe38
|
458 |
} |
2ae88149a
|
459 |
return alloc_pages(gfp, 0); |
44110fe38
|
460 |
} |
2ae88149a
|
461 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38
|
462 |
#endif |
db37648cd
|
463 464 465 466 467 |
static int __sleep_on_page_lock(void *word) { io_schedule(); return 0; } |
1da177e4c
|
468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 |
/* * In order to wait for pages to become available there must be * waitqueues associated with pages. By using a hash table of * waitqueues where the bucket discipline is to maintain all * waiters on the same queue and wake all when any of the pages * become available, and for the woken contexts to check to be * sure the appropriate page became available, this saves space * at a cost of "thundering herd" phenomena during rare hash * collisions. */ static wait_queue_head_t *page_waitqueue(struct page *page) { const struct zone *zone = page_zone(page); return &zone->wait_table[hash_ptr(page, zone->wait_table_bits)]; } static inline void wake_up_page(struct page *page, int bit) { __wake_up_bit(page_waitqueue(page), &page->flags, bit); } |
920c7a5d0
|
489 |
void wait_on_page_bit(struct page *page, int bit_nr) |
1da177e4c
|
490 491 492 493 494 495 496 497 498 499 |
{ DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); if (test_bit(bit_nr, &page->flags)) __wait_on_bit(page_waitqueue(page), &wait, sync_page, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(wait_on_page_bit); /** |
385e1ca5f
|
500 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc
|
501 502 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f
|
503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 |
* * Add an arbitrary @waiter to the wait queue for the nominated @page. */ void add_page_wait_queue(struct page *page, wait_queue_t *waiter) { wait_queue_head_t *q = page_waitqueue(page); unsigned long flags; spin_lock_irqsave(&q->lock, flags); __add_wait_queue(q, waiter); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL_GPL(add_page_wait_queue); /** |
485bb99b4
|
518 |
* unlock_page - unlock a locked page |
1da177e4c
|
519 520 521 522 523 524 525 |
* @page: the page * * Unlocks the page and wakes up sleepers in ___wait_on_page_locked(). * Also wakes sleepers in wait_on_page_writeback() because the wakeup * mechananism between PageLocked pages and PageWriteback pages is shared. * But that's OK - sleepers in wait_on_page_writeback() just go back to sleep. * |
8413ac9d8
|
526 527 |
* The mb is necessary to enforce ordering between the clear_bit and the read * of the waitqueue (to avoid SMP races with a parallel wait_on_page_locked()). |
1da177e4c
|
528 |
*/ |
920c7a5d0
|
529 |
void unlock_page(struct page *page) |
1da177e4c
|
530 |
{ |
8413ac9d8
|
531 532 533 |
VM_BUG_ON(!PageLocked(page)); clear_bit_unlock(PG_locked, &page->flags); smp_mb__after_clear_bit(); |
1da177e4c
|
534 535 536 |
wake_up_page(page, PG_locked); } EXPORT_SYMBOL(unlock_page); |
485bb99b4
|
537 538 539 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c
|
540 541 542 |
*/ void end_page_writeback(struct page *page) { |
ac6aadb24
|
543 544 545 546 547 |
if (TestClearPageReclaim(page)) rotate_reclaimable_page(page); if (!test_clear_page_writeback(page)) BUG(); |
1da177e4c
|
548 549 550 551 |
smp_mb__after_clear_bit(); wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
485bb99b4
|
552 553 554 |
/** * __lock_page - get a lock on the page, assuming we need to sleep to get it * @page: the page to lock |
1da177e4c
|
555 |
* |
485bb99b4
|
556 |
* Ugly. Running sync_page() in state TASK_UNINTERRUPTIBLE is scary. If some |
1da177e4c
|
557 558 559 560 |
* random driver's requestfn sets TASK_RUNNING, we could busywait. However * chances are that on the second loop, the block layer's plug list is empty, * so sync_page() will then return in state TASK_UNINTERRUPTIBLE. */ |
920c7a5d0
|
561 |
void __lock_page(struct page *page) |
1da177e4c
|
562 563 564 565 566 567 568 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); __wait_on_bit_lock(page_waitqueue(page), &wait, sync_page, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(__lock_page); |
b5606c2d4
|
569 |
int __lock_page_killable(struct page *page) |
2687a3569
|
570 571 572 573 574 575 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); return __wait_on_bit_lock(page_waitqueue(page), &wait, sync_page_killable, TASK_KILLABLE); } |
18bc0bbd1
|
576 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569
|
577 |
|
7682486b3
|
578 579 580 581 |
/** * __lock_page_nosync - get a lock on the page, without calling sync_page() * @page: the page to lock * |
db37648cd
|
582 583 584 |
* Variant of lock_page that does not require the caller to hold a reference * on the page's mapping. */ |
920c7a5d0
|
585 |
void __lock_page_nosync(struct page *page) |
db37648cd
|
586 587 588 589 590 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); __wait_on_bit_lock(page_waitqueue(page), &wait, __sleep_on_page_lock, TASK_UNINTERRUPTIBLE); } |
485bb99b4
|
591 592 593 594 595 |
/** * find_get_page - find and get a page reference * @mapping: the address_space to search * @offset: the page index * |
da6052f7b
|
596 597 |
* Is there a pagecache struct page at the given (mapping, offset) tuple? * If yes, increment its refcount and return it; if no, return NULL. |
1da177e4c
|
598 |
*/ |
a60637c85
|
599 |
struct page *find_get_page(struct address_space *mapping, pgoff_t offset) |
1da177e4c
|
600 |
{ |
a60637c85
|
601 |
void **pagep; |
1da177e4c
|
602 |
struct page *page; |
a60637c85
|
603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 |
rcu_read_lock(); repeat: page = NULL; pagep = radix_tree_lookup_slot(&mapping->page_tree, offset); if (pagep) { page = radix_tree_deref_slot(pagep); if (unlikely(!page || page == RADIX_TREE_RETRY)) goto repeat; if (!page_cache_get_speculative(page)) goto repeat; /* * Has the page moved? * This is part of the lockless pagecache protocol. See * include/linux/pagemap.h for details. */ if (unlikely(page != *pagep)) { page_cache_release(page); goto repeat; } } rcu_read_unlock(); |
1da177e4c
|
626 627 |
return page; } |
1da177e4c
|
628 |
EXPORT_SYMBOL(find_get_page); |
485bb99b4
|
629 |
/** |
1da177e4c
|
630 |
* find_lock_page - locate, pin and lock a pagecache page |
67be2dd1b
|
631 632 |
* @mapping: the address_space to search * @offset: the page index |
1da177e4c
|
633 634 635 636 637 638 |
* * Locates the desired pagecache page, locks it, increments its reference * count and returns its address. * * Returns zero if the page was not present. find_lock_page() may sleep. */ |
a60637c85
|
639 |
struct page *find_lock_page(struct address_space *mapping, pgoff_t offset) |
1da177e4c
|
640 641 |
{ struct page *page; |
1da177e4c
|
642 |
repeat: |
a60637c85
|
643 |
page = find_get_page(mapping, offset); |
1da177e4c
|
644 |
if (page) { |
a60637c85
|
645 646 647 648 649 650 |
lock_page(page); /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); page_cache_release(page); goto repeat; |
1da177e4c
|
651 |
} |
a60637c85
|
652 |
VM_BUG_ON(page->index != offset); |
1da177e4c
|
653 |
} |
1da177e4c
|
654 655 |
return page; } |
1da177e4c
|
656 657 658 659 |
EXPORT_SYMBOL(find_lock_page); /** * find_or_create_page - locate or add a pagecache page |
67be2dd1b
|
660 661 662 |
* @mapping: the page's address_space * @index: the page's index into the mapping * @gfp_mask: page allocation mode |
1da177e4c
|
663 664 665 666 667 668 669 670 671 672 673 674 675 |
* * Locates a page in the pagecache. If the page is not present, a new page * is allocated using @gfp_mask and is added to the pagecache and to the VM's * LRU list. The returned page is locked and has its reference count * incremented. * * find_or_create_page() may sleep, even if @gfp_flags specifies an atomic * allocation! * * find_or_create_page() returns the desired page's address, or zero on * memory exhaustion. */ struct page *find_or_create_page(struct address_space *mapping, |
57f6b96c0
|
676 |
pgoff_t index, gfp_t gfp_mask) |
1da177e4c
|
677 |
{ |
eb2be1893
|
678 |
struct page *page; |
1da177e4c
|
679 680 681 682 |
int err; repeat: page = find_lock_page(mapping, index); if (!page) { |
eb2be1893
|
683 684 685 |
page = __page_cache_alloc(gfp_mask); if (!page) return NULL; |
67d58ac47
|
686 687 688 689 690 691 692 693 |
/* * We want a regular kernel memory (not highmem or DMA etc) * allocation for the radix tree nodes, but we need to honour * the context-specific requirements the caller has asked for. * GFP_RECLAIM_MASK collects those requirements. */ err = add_to_page_cache_lru(page, mapping, index, (gfp_mask & GFP_RECLAIM_MASK)); |
eb2be1893
|
694 695 696 697 698 |
if (unlikely(err)) { page_cache_release(page); page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c
|
699 |
} |
1da177e4c
|
700 |
} |
1da177e4c
|
701 702 |
return page; } |
1da177e4c
|
703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 |
EXPORT_SYMBOL(find_or_create_page); /** * find_get_pages - gang pagecache lookup * @mapping: The address_space to search * @start: The starting page index * @nr_pages: The maximum number of pages * @pages: Where the resulting pages are placed * * find_get_pages() will search for and return a group of up to * @nr_pages pages in the mapping. The pages are placed at @pages. * find_get_pages() takes a reference against the returned pages. * * The search returns a group of mapping-contiguous pages with ascending * indexes. There may be holes in the indices due to not-present pages. * * find_get_pages() returns the number of pages which were found. */ unsigned find_get_pages(struct address_space *mapping, pgoff_t start, unsigned int nr_pages, struct page **pages) { unsigned int i; unsigned int ret; |
a60637c85
|
726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 |
unsigned int nr_found; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, (void ***)pages, start, nr_pages); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: page = radix_tree_deref_slot((void **)pages[i]); if (unlikely(!page)) continue; /* * this can only trigger if nr_found == 1, making livelock * a non issue. */ if (unlikely(page == RADIX_TREE_RETRY)) goto restart; if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *((void **)pages[i]))) { page_cache_release(page); goto repeat; } |
1da177e4c
|
754 |
|
a60637c85
|
755 756 757 758 |
pages[ret] = page; ret++; } rcu_read_unlock(); |
1da177e4c
|
759 760 |
return ret; } |
ebf43500e
|
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 |
/** * find_get_pages_contig - gang contiguous pagecache lookup * @mapping: The address_space to search * @index: The starting page index * @nr_pages: The maximum number of pages * @pages: Where the resulting pages are placed * * find_get_pages_contig() works exactly like find_get_pages(), except * that the returned number of pages are guaranteed to be contiguous. * * find_get_pages_contig() returns the number of pages which were found. */ unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index, unsigned int nr_pages, struct page **pages) { unsigned int i; unsigned int ret; |
a60637c85
|
778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 |
unsigned int nr_found; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, (void ***)pages, index, nr_pages); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: page = radix_tree_deref_slot((void **)pages[i]); if (unlikely(!page)) continue; /* * this can only trigger if nr_found == 1, making livelock * a non issue. */ if (unlikely(page == RADIX_TREE_RETRY)) goto restart; |
ebf43500e
|
797 |
|
a60637c85
|
798 |
if (page->mapping == NULL || page->index != index) |
ebf43500e
|
799 |
break; |
a60637c85
|
800 801 802 803 804 805 806 807 808 809 810 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *((void **)pages[i]))) { page_cache_release(page); goto repeat; } pages[ret] = page; ret++; |
ebf43500e
|
811 812 |
index++; } |
a60637c85
|
813 814 |
rcu_read_unlock(); return ret; |
ebf43500e
|
815 |
} |
ef71c15c4
|
816 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e
|
817 |
|
485bb99b4
|
818 819 820 821 822 823 824 825 |
/** * find_get_pages_tag - find and return pages that match @tag * @mapping: the address_space to search * @index: the starting page index * @tag: the tag index * @nr_pages: the maximum number of pages * @pages: where the resulting pages are placed * |
1da177e4c
|
826 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4
|
827 |
* @tag. We update @index to index the next page for the traversal. |
1da177e4c
|
828 829 830 831 832 833 |
*/ unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, int tag, unsigned int nr_pages, struct page **pages) { unsigned int i; unsigned int ret; |
a60637c85
|
834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 |
unsigned int nr_found; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_tag_slot(&mapping->page_tree, (void ***)pages, *index, nr_pages, tag); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: page = radix_tree_deref_slot((void **)pages[i]); if (unlikely(!page)) continue; /* * this can only trigger if nr_found == 1, making livelock * a non issue. */ if (unlikely(page == RADIX_TREE_RETRY)) goto restart; if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *((void **)pages[i]))) { page_cache_release(page); goto repeat; } pages[ret] = page; ret++; } rcu_read_unlock(); |
1da177e4c
|
867 |
|
1da177e4c
|
868 869 |
if (ret) *index = pages[ret - 1]->index + 1; |
a60637c85
|
870 |
|
1da177e4c
|
871 872 |
return ret; } |
ef71c15c4
|
873 |
EXPORT_SYMBOL(find_get_pages_tag); |
1da177e4c
|
874 |
|
485bb99b4
|
875 876 877 878 879 |
/** * grab_cache_page_nowait - returns locked page at given index in given cache * @mapping: target address_space * @index: the page index * |
72fd4a35a
|
880 |
* Same as grab_cache_page(), but do not wait if the page is unavailable. |
1da177e4c
|
881 882 883 884 885 886 887 888 |
* This is intended for speculative data generators, where the data can * be regenerated if the page couldn't be grabbed. This routine should * be safe to call while holding the lock for another page. * * Clear __GFP_FS when allocating the page to avoid recursion into the fs * and deadlock against the caller's locked page. */ struct page * |
57f6b96c0
|
889 |
grab_cache_page_nowait(struct address_space *mapping, pgoff_t index) |
1da177e4c
|
890 891 |
{ struct page *page = find_get_page(mapping, index); |
1da177e4c
|
892 893 |
if (page) { |
529ae9aaa
|
894 |
if (trylock_page(page)) |
1da177e4c
|
895 896 897 898 |
return page; page_cache_release(page); return NULL; } |
2ae88149a
|
899 |
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); |
67d58ac47
|
900 |
if (page && add_to_page_cache_lru(page, mapping, index, GFP_NOFS)) { |
1da177e4c
|
901 902 903 904 905 |
page_cache_release(page); page = NULL; } return page; } |
1da177e4c
|
906 |
EXPORT_SYMBOL(grab_cache_page_nowait); |
76d42bd96
|
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 |
/* * CD/DVDs are error prone. When a medium error occurs, the driver may fail * a _large_ part of the i/o request. Imagine the worst scenario: * * ---R__________________________________________B__________ * ^ reading here ^ bad block(assume 4k) * * read(R) => miss => readahead(R...B) => media error => frustrating retries * => failing the whole request => read(R) => read(R+1) => * readahead(R+1...B+1) => bang => read(R+2) => read(R+3) => * readahead(R+3...B+2) => bang => read(R+3) => read(R+4) => * readahead(R+4...B+3) => bang => read(R+4) => read(R+5) => ...... * * It is going insane. Fix it by quickly scaling down the readahead size. */ static void shrink_readahead_size_eio(struct file *filp, struct file_ra_state *ra) { |
76d42bd96
|
925 |
ra->ra_pages /= 4; |
76d42bd96
|
926 |
} |
485bb99b4
|
927 |
/** |
36e789144
|
928 |
* do_generic_file_read - generic file read routine |
485bb99b4
|
929 930 931 932 933 |
* @filp: the file to read * @ppos: current file position * @desc: read_descriptor * @actor: read method * |
1da177e4c
|
934 |
* This is a generic file read routine, and uses the |
485bb99b4
|
935 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c
|
936 937 938 |
* * This is really ugly. But the goto's actually try to clarify some * of the logic when it comes to error handling etc. |
1da177e4c
|
939 |
*/ |
36e789144
|
940 941 |
static void do_generic_file_read(struct file *filp, loff_t *ppos, read_descriptor_t *desc, read_actor_t actor) |
1da177e4c
|
942 |
{ |
36e789144
|
943 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c
|
944 |
struct inode *inode = mapping->host; |
36e789144
|
945 |
struct file_ra_state *ra = &filp->f_ra; |
57f6b96c0
|
946 947 948 949 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372
|
950 |
unsigned int prev_offset; |
1da177e4c
|
951 |
int error; |
1da177e4c
|
952 |
|
1da177e4c
|
953 |
index = *ppos >> PAGE_CACHE_SHIFT; |
7ff81078d
|
954 955 |
prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); |
1da177e4c
|
956 957 |
last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; offset = *ppos & ~PAGE_CACHE_MASK; |
1da177e4c
|
958 959 |
for (;;) { struct page *page; |
57f6b96c0
|
960 |
pgoff_t end_index; |
a32ea1e1f
|
961 |
loff_t isize; |
1da177e4c
|
962 |
unsigned long nr, ret; |
1da177e4c
|
963 |
cond_resched(); |
1da177e4c
|
964 965 |
find_page: page = find_get_page(mapping, index); |
3ea89ee86
|
966 |
if (!page) { |
cf914a7d6
|
967 |
page_cache_sync_readahead(mapping, |
7ff81078d
|
968 |
ra, filp, |
3ea89ee86
|
969 970 971 972 973 974 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6
|
975 |
page_cache_async_readahead(mapping, |
7ff81078d
|
976 |
ra, filp, page, |
3ea89ee86
|
977 |
index, last_index - index); |
1da177e4c
|
978 |
} |
8ab22b9ab
|
979 980 981 982 |
if (!PageUptodate(page)) { if (inode->i_blkbits == PAGE_CACHE_SHIFT || !mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; |
529ae9aaa
|
983 |
if (!trylock_page(page)) |
8ab22b9ab
|
984 985 986 987 988 989 |
goto page_not_up_to_date; if (!mapping->a_ops->is_partially_uptodate(page, desc, offset)) goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c
|
990 |
page_ok: |
a32ea1e1f
|
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 |
/* * i_size must be checked after we know the page is Uptodate. * * Checking i_size after the check allows us to calculate * the correct value for "nr", which means the zero-filled * part of the page is not copied back to userspace (unless * another truncate extends the file - this is desired though). */ isize = i_size_read(inode); end_index = (isize - 1) >> PAGE_CACHE_SHIFT; if (unlikely(!isize || index > end_index)) { page_cache_release(page); goto out; } /* nr is the maximum number of bytes to copy from this page */ nr = PAGE_CACHE_SIZE; if (index == end_index) { nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1; if (nr <= offset) { page_cache_release(page); goto out; } } nr = nr - offset; |
1da177e4c
|
1017 1018 1019 1020 1021 1022 1023 1024 1025 |
/* If users can be writing to this page using arbitrary * virtual addresses, take care about potential aliasing * before reading the page on the kernel side. */ if (mapping_writably_mapped(mapping)) flush_dcache_page(page); /* |
ec0f16372
|
1026 1027 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c
|
1028 |
*/ |
ec0f16372
|
1029 |
if (prev_index != index || offset != prev_offset) |
1da177e4c
|
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 |
mark_page_accessed(page); prev_index = index; /* * Ok, we have the page, and it's up-to-date, so * now we can copy it to user space... * * The actor routine returns how many bytes were actually used.. * NOTE! This may not be the same as how much of a user buffer * we filled up (we may be padding etc), so we can only update * "pos" here (the actor routine has to update the user buffer * pointers and the remaining count). */ ret = actor(desc, page, offset, nr); offset += ret; index += offset >> PAGE_CACHE_SHIFT; offset &= ~PAGE_CACHE_MASK; |
6ce745ed3
|
1047 |
prev_offset = offset; |
1da177e4c
|
1048 1049 1050 1051 1052 1053 1054 1055 |
page_cache_release(page); if (ret == nr && desc->count) continue; goto out; page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235
|
1056 1057 1058 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c
|
1059 |
|
8ab22b9ab
|
1060 |
page_not_up_to_date_locked: |
da6052f7b
|
1061 |
/* Did it get truncated before we got the lock? */ |
1da177e4c
|
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 |
if (!page->mapping) { unlock_page(page); page_cache_release(page); continue; } /* Did somebody else fill it already? */ if (PageUptodate(page)) { unlock_page(page); goto page_ok; } readpage: /* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7
|
1077 1078 1079 1080 1081 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { page_cache_release(page); goto find_page; } |
1da177e4c
|
1082 |
goto readpage_error; |
994fc28c7
|
1083 |
} |
1da177e4c
|
1084 1085 |
if (!PageUptodate(page)) { |
854623235
|
1086 1087 1088 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c
|
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* * invalidate_inode_pages got it */ unlock_page(page); page_cache_release(page); goto find_page; } unlock_page(page); |
7ff81078d
|
1099 |
shrink_readahead_size_eio(filp, ra); |
854623235
|
1100 1101 |
error = -EIO; goto readpage_error; |
1da177e4c
|
1102 1103 1104 |
} unlock_page(page); } |
1da177e4c
|
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 |
goto page_ok; readpage_error: /* UHHUH! A synchronous read error occurred. Report it */ desc->error = error; page_cache_release(page); goto out; no_cached_page: /* * Ok, it wasn't cached, so we need to create a new * page.. */ |
eb2be1893
|
1118 1119 1120 1121 |
page = page_cache_alloc_cold(mapping); if (!page) { desc->error = -ENOMEM; goto out; |
1da177e4c
|
1122 |
} |
eb2be1893
|
1123 |
error = add_to_page_cache_lru(page, mapping, |
1da177e4c
|
1124 1125 |
index, GFP_KERNEL); if (error) { |
eb2be1893
|
1126 |
page_cache_release(page); |
1da177e4c
|
1127 1128 1129 1130 1131 |
if (error == -EEXIST) goto find_page; desc->error = error; goto out; } |
1da177e4c
|
1132 1133 1134 1135 |
goto readpage; } out: |
7ff81078d
|
1136 1137 1138 |
ra->prev_pos = prev_index; ra->prev_pos <<= PAGE_CACHE_SHIFT; ra->prev_pos |= prev_offset; |
1da177e4c
|
1139 |
|
f4e6b498d
|
1140 |
*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; |
0c6aa2639
|
1141 |
file_accessed(filp); |
1da177e4c
|
1142 |
} |
1da177e4c
|
1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 |
int file_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset, unsigned long size) { char *kaddr; unsigned long left, count = desc->count; if (size > count) size = count; /* * Faults on the destination of a read are common, so do it before * taking the kmap. */ if (!fault_in_pages_writeable(desc->arg.buf, size)) { kaddr = kmap_atomic(page, KM_USER0); left = __copy_to_user_inatomic(desc->arg.buf, kaddr + offset, size); kunmap_atomic(kaddr, KM_USER0); if (left == 0) goto success; } /* Do it the slow way */ kaddr = kmap(page); left = __copy_to_user(desc->arg.buf, kaddr + offset, size); kunmap(page); if (left) { size -= left; desc->error = -EFAULT; } success: desc->count = count - size; desc->written += size; desc->arg.buf += size; return size; } |
0ceb33143
|
1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 |
/* * Performs necessary checks before doing a write * @iov: io vector request * @nr_segs: number of segments in the iovec * @count: number of bytes to write * @access_flags: type of access: %VERIFY_READ or %VERIFY_WRITE * * Adjust number of segments and amount of bytes to write (nr_segs should be * properly initialized first). Returns appropriate error code that caller * should return or zero in case that write should be allowed. */ int generic_segment_checks(const struct iovec *iov, unsigned long *nr_segs, size_t *count, int access_flags) { unsigned long seg; size_t cnt = 0; for (seg = 0; seg < *nr_segs; seg++) { const struct iovec *iv = &iov[seg]; /* * If any segment has a negative length, or the cumulative * length ever wraps negative then return -EINVAL. */ cnt += iv->iov_len; if (unlikely((ssize_t)(cnt|iv->iov_len) < 0)) return -EINVAL; if (access_ok(access_flags, iv->iov_base, iv->iov_len)) continue; if (seg == 0) return -EFAULT; *nr_segs = seg; cnt -= iv->iov_len; /* This segment is no good */ break; } *count = cnt; return 0; } EXPORT_SYMBOL(generic_segment_checks); |
485bb99b4
|
1219 |
/** |
b2abacf3a
|
1220 |
* generic_file_aio_read - generic filesystem read routine |
485bb99b4
|
1221 1222 1223 |
* @iocb: kernel I/O control block * @iov: io vector request * @nr_segs: number of segments in the iovec |
b2abacf3a
|
1224 |
* @pos: current file position |
485bb99b4
|
1225 |
* |
1da177e4c
|
1226 1227 1228 1229 |
* This is the "read()" routine for all filesystems * that can use the page cache directly. */ ssize_t |
543ade1fc
|
1230 1231 |
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c
|
1232 1233 1234 1235 1236 |
{ struct file *filp = iocb->ki_filp; ssize_t retval; unsigned long seg; size_t count; |
543ade1fc
|
1237 |
loff_t *ppos = &iocb->ki_pos; |
1da177e4c
|
1238 1239 |
count = 0; |
0ceb33143
|
1240 1241 1242 |
retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; |
1da177e4c
|
1243 1244 1245 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { |
543ade1fc
|
1246 |
loff_t size; |
1da177e4c
|
1247 1248 1249 1250 1251 |
struct address_space *mapping; struct inode *inode; mapping = filp->f_mapping; inode = mapping->host; |
1da177e4c
|
1252 1253 1254 1255 |
if (!count) goto out; /* skip atime */ size = i_size_read(inode); if (pos < size) { |
48b47c561
|
1256 1257 |
retval = filemap_write_and_wait_range(mapping, pos, pos + iov_length(iov, nr_segs) - 1); |
a969e903a
|
1258 1259 1260 1261 |
if (!retval) { retval = mapping->a_ops->direct_IO(READ, iocb, iov, pos, nr_segs); } |
1da177e4c
|
1262 1263 |
if (retval > 0) *ppos = pos + retval; |
11fa977ec
|
1264 1265 1266 1267 |
if (retval) { file_accessed(filp); goto out; } |
0e0bcae3b
|
1268 |
} |
1da177e4c
|
1269 |
} |
11fa977ec
|
1270 1271 |
for (seg = 0; seg < nr_segs; seg++) { read_descriptor_t desc; |
1da177e4c
|
1272 |
|
11fa977ec
|
1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 |
desc.written = 0; desc.arg.buf = iov[seg].iov_base; desc.count = iov[seg].iov_len; if (desc.count == 0) continue; desc.error = 0; do_generic_file_read(filp, ppos, &desc, file_read_actor); retval += desc.written; if (desc.error) { retval = retval ?: desc.error; break; |
1da177e4c
|
1284 |
} |
11fa977ec
|
1285 1286 |
if (desc.count > 0) break; |
1da177e4c
|
1287 1288 1289 1290 |
} out: return retval; } |
1da177e4c
|
1291 |
EXPORT_SYMBOL(generic_file_aio_read); |
1da177e4c
|
1292 1293 |
static ssize_t do_readahead(struct address_space *mapping, struct file *filp, |
57f6b96c0
|
1294 |
pgoff_t index, unsigned long nr) |
1da177e4c
|
1295 1296 1297 |
{ if (!mapping || !mapping->a_ops || !mapping->a_ops->readpage) return -EINVAL; |
f7e839dd3
|
1298 |
force_page_cache_readahead(mapping, filp, index, nr); |
1da177e4c
|
1299 1300 |
return 0; } |
6673e0c3f
|
1301 |
SYSCALL_DEFINE(readahead)(int fd, loff_t offset, size_t count) |
1da177e4c
|
1302 1303 1304 1305 1306 1307 1308 1309 1310 |
{ ssize_t ret; struct file *file; ret = -EBADF; file = fget(fd); if (file) { if (file->f_mode & FMODE_READ) { struct address_space *mapping = file->f_mapping; |
57f6b96c0
|
1311 1312 |
pgoff_t start = offset >> PAGE_CACHE_SHIFT; pgoff_t end = (offset + count - 1) >> PAGE_CACHE_SHIFT; |
1da177e4c
|
1313 1314 1315 1316 1317 1318 1319 |
unsigned long len = end - start + 1; ret = do_readahead(mapping, file, start, len); } fput(file); } return ret; } |
6673e0c3f
|
1320 1321 1322 1323 1324 1325 1326 |
#ifdef CONFIG_HAVE_SYSCALL_WRAPPERS asmlinkage long SyS_readahead(long fd, loff_t offset, long count) { return SYSC_readahead((int) fd, offset, (size_t) count); } SYSCALL_ALIAS(sys_readahead, SyS_readahead); #endif |
1da177e4c
|
1327 1328 |
#ifdef CONFIG_MMU |
485bb99b4
|
1329 1330 1331 1332 1333 |
/** * page_cache_read - adds requested page to the page cache if not already there * @file: file to read * @offset: page index * |
1da177e4c
|
1334 1335 1336 |
* This adds the requested page to the page cache if it isn't already there, * and schedules an I/O to read in its contents from disk. */ |
920c7a5d0
|
1337 |
static int page_cache_read(struct file *file, pgoff_t offset) |
1da177e4c
|
1338 1339 1340 |
{ struct address_space *mapping = file->f_mapping; struct page *page; |
994fc28c7
|
1341 |
int ret; |
1da177e4c
|
1342 |
|
994fc28c7
|
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 |
do { page = page_cache_alloc_cold(mapping); if (!page) return -ENOMEM; ret = add_to_page_cache_lru(page, mapping, offset, GFP_KERNEL); if (ret == 0) ret = mapping->a_ops->readpage(file, page); else if (ret == -EEXIST) ret = 0; /* losing race to add is OK */ |
1da177e4c
|
1353 |
|
1da177e4c
|
1354 |
page_cache_release(page); |
1da177e4c
|
1355 |
|
994fc28c7
|
1356 1357 1358 |
} while (ret == AOP_TRUNCATED_PAGE); return ret; |
1da177e4c
|
1359 1360 1361 |
} #define MMAP_LOTSAMISS (100) |
ef00e08e2
|
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 |
/* * Synchronous readahead happens when we don't even find * a page in the page cache at all. */ static void do_sync_mmap_readahead(struct vm_area_struct *vma, struct file_ra_state *ra, struct file *file, pgoff_t offset) { unsigned long ra_pages; struct address_space *mapping = file->f_mapping; /* If we don't want any read-ahead, don't bother */ if (VM_RandomReadHint(vma)) return; |
70ac23cfa
|
1377 1378 |
if (VM_SequentialReadHint(vma) || offset - 1 == (ra->prev_pos >> PAGE_CACHE_SHIFT)) { |
7ffc59b4d
|
1379 1380 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
ef00e08e2
|
1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 |
return; } if (ra->mmap_miss < INT_MAX) ra->mmap_miss++; /* * Do we miss much more than hit in this file? If so, * stop bothering with read-ahead. It will only hurt. */ if (ra->mmap_miss > MMAP_LOTSAMISS) return; |
d30a11004
|
1393 1394 1395 |
/* * mmap read-around */ |
ef00e08e2
|
1396 1397 |
ra_pages = max_sane_readahead(ra->ra_pages); if (ra_pages) { |
d30a11004
|
1398 1399 1400 1401 |
ra->start = max_t(long, 0, offset - ra_pages/2); ra->size = ra_pages; ra->async_size = 0; ra_submit(ra, mapping, file); |
ef00e08e2
|
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 |
} } /* * Asynchronous readahead happens when we find the page and PG_readahead, * so we want to possibly extend the readahead further.. */ static void do_async_mmap_readahead(struct vm_area_struct *vma, struct file_ra_state *ra, struct file *file, struct page *page, pgoff_t offset) { struct address_space *mapping = file->f_mapping; /* If we don't want any read-ahead, don't bother */ if (VM_RandomReadHint(vma)) return; if (ra->mmap_miss > 0) ra->mmap_miss--; if (PageReadahead(page)) |
2fad6f5de
|
1423 1424 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
ef00e08e2
|
1425 |
} |
485bb99b4
|
1426 |
/** |
54cb8821d
|
1427 |
* filemap_fault - read in file data for page fault handling |
d0217ac04
|
1428 1429 |
* @vma: vma in which the fault was taken * @vmf: struct vm_fault containing details of the fault |
485bb99b4
|
1430 |
* |
54cb8821d
|
1431 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c
|
1432 1433 1434 1435 1436 1437 |
* mapped memory region to read in file data during a page fault. * * The goto's are kind of ugly, but this streamlines the normal case of having * it in the page cache, and handles the special cases reasonably without * having a lot of duplicated code. */ |
d0217ac04
|
1438 |
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
1da177e4c
|
1439 1440 |
{ int error; |
54cb8821d
|
1441 |
struct file *file = vma->vm_file; |
1da177e4c
|
1442 1443 1444 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2
|
1445 |
pgoff_t offset = vmf->pgoff; |
1da177e4c
|
1446 |
struct page *page; |
2004dc8ee
|
1447 |
pgoff_t size; |
83c54070e
|
1448 |
int ret = 0; |
1da177e4c
|
1449 |
|
1da177e4c
|
1450 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2
|
1451 |
if (offset >= size) |
5307cc1aa
|
1452 |
return VM_FAULT_SIGBUS; |
1da177e4c
|
1453 |
|
1da177e4c
|
1454 |
/* |
1da177e4c
|
1455 1456 |
* Do we have something in the page cache already? */ |
ef00e08e2
|
1457 1458 |
page = find_get_page(mapping, offset); if (likely(page)) { |
1da177e4c
|
1459 |
/* |
ef00e08e2
|
1460 1461 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c
|
1462 |
*/ |
ef00e08e2
|
1463 1464 |
do_async_mmap_readahead(vma, ra, file, page, offset); lock_page(page); |
1da177e4c
|
1465 |
|
ef00e08e2
|
1466 1467 1468 1469 1470 |
/* Did it get truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); put_page(page); goto no_cached_page; |
1da177e4c
|
1471 |
} |
ef00e08e2
|
1472 1473 1474 1475 1476 1477 1478 |
} else { /* No page in the page cache at all */ do_sync_mmap_readahead(vma, ra, file, offset); count_vm_event(PGMAJFAULT); ret = VM_FAULT_MAJOR; retry_find: page = find_lock_page(mapping, offset); |
1da177e4c
|
1479 1480 1481 |
if (!page) goto no_cached_page; } |
1da177e4c
|
1482 |
/* |
d00806b18
|
1483 1484 |
* We have a locked page in the page cache, now we need to check * that it's up-to-date. If not, it is going to be due to an error. |
1da177e4c
|
1485 |
*/ |
d00806b18
|
1486 |
if (unlikely(!PageUptodate(page))) |
1da177e4c
|
1487 |
goto page_not_uptodate; |
ef00e08e2
|
1488 1489 1490 1491 |
/* * Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
d00806b18
|
1492 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2
|
1493 |
if (unlikely(offset >= size)) { |
d00806b18
|
1494 |
unlock_page(page); |
745ad48e8
|
1495 |
page_cache_release(page); |
5307cc1aa
|
1496 |
return VM_FAULT_SIGBUS; |
d00806b18
|
1497 |
} |
ef00e08e2
|
1498 |
ra->prev_pos = (loff_t)offset << PAGE_CACHE_SHIFT; |
d0217ac04
|
1499 |
vmf->page = page; |
83c54070e
|
1500 |
return ret | VM_FAULT_LOCKED; |
1da177e4c
|
1501 |
|
1da177e4c
|
1502 1503 1504 1505 1506 |
no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ |
ef00e08e2
|
1507 |
error = page_cache_read(file, offset); |
1da177e4c
|
1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 |
/* * The page we want has now been added to the page cache. * In the unlikely event that someone removed it in the * meantime, we'll just come back here and read it again. */ if (error >= 0) goto retry_find; /* * An error return from page_cache_read can result if the * system is low on memory, or a problem occurs while trying * to schedule I/O. */ if (error == -ENOMEM) |
d0217ac04
|
1523 1524 |
return VM_FAULT_OOM; return VM_FAULT_SIGBUS; |
1da177e4c
|
1525 1526 |
page_not_uptodate: |
1da177e4c
|
1527 1528 1529 1530 1531 1532 |
/* * Umm, take care of errors if the page isn't up-to-date. * Try to re-read it _once_. We do this synchronously, * because there really aren't any performance issues here * and we need to check for errors. */ |
1da177e4c
|
1533 |
ClearPageError(page); |
994fc28c7
|
1534 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e
|
1535 1536 1537 1538 1539 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
d00806b18
|
1540 1541 1542 |
page_cache_release(page); if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7
|
1543 |
goto retry_find; |
1da177e4c
|
1544 |
|
d00806b18
|
1545 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96
|
1546 |
shrink_readahead_size_eio(file, ra); |
d0217ac04
|
1547 |
return VM_FAULT_SIGBUS; |
54cb8821d
|
1548 1549 |
} EXPORT_SYMBOL(filemap_fault); |
f0f37e2f7
|
1550 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d
|
1551 |
.fault = filemap_fault, |
1da177e4c
|
1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 |
}; /* This is used for a general mmap of a disk file */ int generic_file_mmap(struct file * file, struct vm_area_struct * vma) { struct address_space *mapping = file->f_mapping; if (!mapping->a_ops->readpage) return -ENOEXEC; file_accessed(file); vma->vm_ops = &generic_file_vm_ops; |
d0217ac04
|
1564 |
vma->vm_flags |= VM_CAN_NONLINEAR; |
1da177e4c
|
1565 1566 |
return 0; } |
1da177e4c
|
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 |
/* * This is for filesystems which do not implement ->writepage. */ int generic_file_readonly_mmap(struct file *file, struct vm_area_struct *vma) { if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE)) return -EINVAL; return generic_file_mmap(file, vma); } #else int generic_file_mmap(struct file * file, struct vm_area_struct * vma) { return -ENOSYS; } int generic_file_readonly_mmap(struct file * file, struct vm_area_struct * vma) { return -ENOSYS; } #endif /* CONFIG_MMU */ EXPORT_SYMBOL(generic_file_mmap); EXPORT_SYMBOL(generic_file_readonly_mmap); |
6fe6900e1
|
1590 |
static struct page *__read_cache_page(struct address_space *mapping, |
57f6b96c0
|
1591 |
pgoff_t index, |
1da177e4c
|
1592 1593 1594 |
int (*filler)(void *,struct page*), void *data) { |
eb2be1893
|
1595 |
struct page *page; |
1da177e4c
|
1596 1597 1598 1599 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
eb2be1893
|
1600 1601 1602 1603 1604 1605 1606 1607 |
page = page_cache_alloc_cold(mapping); if (!page) return ERR_PTR(-ENOMEM); err = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL); if (unlikely(err)) { page_cache_release(page); if (err == -EEXIST) goto repeat; |
1da177e4c
|
1608 |
/* Presumably ENOMEM for radix tree node */ |
1da177e4c
|
1609 1610 |
return ERR_PTR(err); } |
1da177e4c
|
1611 1612 1613 1614 1615 1616 |
err = filler(data, page); if (err < 0) { page_cache_release(page); page = ERR_PTR(err); } } |
1da177e4c
|
1617 1618 |
return page; } |
7682486b3
|
1619 1620 1621 1622 1623 1624 1625 |
/** * read_cache_page_async - read into page cache, fill it if needed * @mapping: the page's address_space * @index: the page index * @filler: function to perform the read * @data: destination for read data * |
6fe6900e1
|
1626 1627 |
* Same as read_cache_page, but don't wait for page to become unlocked * after submitting it to the filler. |
7682486b3
|
1628 1629 1630 1631 1632 |
* * Read into the page cache. If a page already exists, and PageUptodate() is * not set, try to fill the page but don't wait for it to become unlocked. * * If the page does not get brought uptodate, return -EIO. |
1da177e4c
|
1633 |
*/ |
6fe6900e1
|
1634 |
struct page *read_cache_page_async(struct address_space *mapping, |
57f6b96c0
|
1635 |
pgoff_t index, |
1da177e4c
|
1636 1637 1638 1639 1640 1641 1642 1643 1644 |
int (*filler)(void *,struct page*), void *data) { struct page *page; int err; retry: page = __read_cache_page(mapping, index, filler, data); if (IS_ERR(page)) |
c855ff371
|
1645 |
return page; |
1da177e4c
|
1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 |
if (PageUptodate(page)) goto out; lock_page(page); if (!page->mapping) { unlock_page(page); page_cache_release(page); goto retry; } if (PageUptodate(page)) { unlock_page(page); goto out; } err = filler(data, page); if (err < 0) { page_cache_release(page); |
c855ff371
|
1662 |
return ERR_PTR(err); |
1da177e4c
|
1663 |
} |
c855ff371
|
1664 |
out: |
6fe6900e1
|
1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 |
mark_page_accessed(page); return page; } EXPORT_SYMBOL(read_cache_page_async); /** * read_cache_page - read into page cache, fill it if needed * @mapping: the page's address_space * @index: the page index * @filler: function to perform the read * @data: destination for read data * * Read into the page cache. If a page already exists, and PageUptodate() is * not set, try to fill the page then wait for it to become unlocked. * * If the page does not get brought uptodate, return -EIO. */ struct page *read_cache_page(struct address_space *mapping, |
57f6b96c0
|
1683 |
pgoff_t index, |
6fe6900e1
|
1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 |
int (*filler)(void *,struct page*), void *data) { struct page *page; page = read_cache_page_async(mapping, index, filler, data); if (IS_ERR(page)) goto out; wait_on_page_locked(page); if (!PageUptodate(page)) { page_cache_release(page); page = ERR_PTR(-EIO); } out: |
1da177e4c
|
1698 1699 |
return page; } |
1da177e4c
|
1700 1701 1702 |
EXPORT_SYMBOL(read_cache_page); /* |
1da177e4c
|
1703 1704 1705 1706 1707 |
* The logic we want is * * if suid or (sgid and xgrp) * remove privs */ |
01de85e05
|
1708 |
int should_remove_suid(struct dentry *dentry) |
1da177e4c
|
1709 1710 1711 |
{ mode_t mode = dentry->d_inode->i_mode; int kill = 0; |
1da177e4c
|
1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 |
/* suid always must be killed */ if (unlikely(mode & S_ISUID)) kill = ATTR_KILL_SUID; /* * sgid without any exec bits is just a mandatory locking mark; leave * it alone. If some exec bits are set, it's a real sgid; kill it. */ if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) kill |= ATTR_KILL_SGID; |
7f5ff766a
|
1723 |
if (unlikely(kill && !capable(CAP_FSETID) && S_ISREG(mode))) |
01de85e05
|
1724 |
return kill; |
1da177e4c
|
1725 |
|
01de85e05
|
1726 1727 |
return 0; } |
d23a147bb
|
1728 |
EXPORT_SYMBOL(should_remove_suid); |
01de85e05
|
1729 |
|
7f3d4ee10
|
1730 |
static int __remove_suid(struct dentry *dentry, int kill) |
01de85e05
|
1731 1732 1733 1734 1735 1736 |
{ struct iattr newattrs; newattrs.ia_valid = ATTR_FORCE | kill; return notify_change(dentry, &newattrs); } |
2f1936b87
|
1737 |
int file_remove_suid(struct file *file) |
01de85e05
|
1738 |
{ |
2f1936b87
|
1739 |
struct dentry *dentry = file->f_path.dentry; |
b53767719
|
1740 1741 1742 |
int killsuid = should_remove_suid(dentry); int killpriv = security_inode_need_killpriv(dentry); int error = 0; |
01de85e05
|
1743 |
|
b53767719
|
1744 1745 1746 1747 1748 1749 |
if (killpriv < 0) return killpriv; if (killpriv) error = security_inode_killpriv(dentry); if (!error && killsuid) error = __remove_suid(dentry, killsuid); |
01de85e05
|
1750 |
|
b53767719
|
1751 |
return error; |
1da177e4c
|
1752 |
} |
2f1936b87
|
1753 |
EXPORT_SYMBOL(file_remove_suid); |
1da177e4c
|
1754 |
|
2f718ffc1
|
1755 |
static size_t __iovec_copy_from_user_inatomic(char *vaddr, |
1da177e4c
|
1756 1757 |
const struct iovec *iov, size_t base, size_t bytes) { |
f18005369
|
1758 |
size_t copied = 0, left = 0; |
1da177e4c
|
1759 1760 1761 1762 1763 1764 |
while (bytes) { char __user *buf = iov->iov_base + base; int copy = min(bytes, iov->iov_len - base); base = 0; |
f18005369
|
1765 |
left = __copy_from_user_inatomic(vaddr, buf, copy); |
1da177e4c
|
1766 1767 1768 1769 |
copied += copy; bytes -= copy; vaddr += copy; iov++; |
01408c493
|
1770 |
if (unlikely(left)) |
1da177e4c
|
1771 |
break; |
1da177e4c
|
1772 1773 1774 1775 1776 |
} return copied - left; } /* |
2f718ffc1
|
1777 |
* Copy as much as we can into the page and return the number of bytes which |
af901ca18
|
1778 |
* were successfully copied. If a fault is encountered then return the number of |
2f718ffc1
|
1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 |
* bytes which were copied. */ size_t iov_iter_copy_from_user_atomic(struct page *page, struct iov_iter *i, unsigned long offset, size_t bytes) { char *kaddr; size_t copied; BUG_ON(!in_atomic()); kaddr = kmap_atomic(page, KM_USER0); if (likely(i->nr_segs == 1)) { int left; char __user *buf = i->iov->iov_base + i->iov_offset; |
f18005369
|
1792 |
left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); |
2f718ffc1
|
1793 1794 1795 1796 1797 1798 1799 1800 1801 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } kunmap_atomic(kaddr, KM_USER0); return copied; } |
89e107877
|
1802 |
EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); |
2f718ffc1
|
1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 |
/* * This has the same sideeffects and return value as * iov_iter_copy_from_user_atomic(). * The difference is that it attempts to resolve faults. * Page must not be locked. */ size_t iov_iter_copy_from_user(struct page *page, struct iov_iter *i, unsigned long offset, size_t bytes) { char *kaddr; size_t copied; kaddr = kmap(page); if (likely(i->nr_segs == 1)) { int left; char __user *buf = i->iov->iov_base + i->iov_offset; |
f18005369
|
1820 |
left = __copy_from_user(kaddr + offset, buf, bytes); |
2f718ffc1
|
1821 1822 1823 1824 1825 1826 1827 1828 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } kunmap(page); return copied; } |
89e107877
|
1829 |
EXPORT_SYMBOL(iov_iter_copy_from_user); |
2f718ffc1
|
1830 |
|
f7009264c
|
1831 |
void iov_iter_advance(struct iov_iter *i, size_t bytes) |
2f718ffc1
|
1832 |
{ |
f7009264c
|
1833 |
BUG_ON(i->count < bytes); |
2f718ffc1
|
1834 1835 |
if (likely(i->nr_segs == 1)) { i->iov_offset += bytes; |
f7009264c
|
1836 |
i->count -= bytes; |
2f718ffc1
|
1837 1838 1839 |
} else { const struct iovec *iov = i->iov; size_t base = i->iov_offset; |
124d3b704
|
1840 1841 |
/* * The !iov->iov_len check ensures we skip over unlikely |
f7009264c
|
1842 |
* zero-length segments (without overruning the iovec). |
124d3b704
|
1843 |
*/ |
94ad374a0
|
1844 |
while (bytes || unlikely(i->count && !iov->iov_len)) { |
f7009264c
|
1845 |
int copy; |
2f718ffc1
|
1846 |
|
f7009264c
|
1847 1848 1849 |
copy = min(bytes, iov->iov_len - base); BUG_ON(!i->count || i->count < copy); i->count -= copy; |
2f718ffc1
|
1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 |
bytes -= copy; base += copy; if (iov->iov_len == base) { iov++; base = 0; } } i->iov = iov; i->iov_offset = base; } } |
89e107877
|
1861 |
EXPORT_SYMBOL(iov_iter_advance); |
2f718ffc1
|
1862 |
|
afddba49d
|
1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 |
/* * Fault in the first iovec of the given iov_iter, to a maximum length * of bytes. Returns 0 on success, or non-zero if the memory could not be * accessed (ie. because it is an invalid address). * * writev-intensive code may want this to prefault several iovecs -- that * would be possible (callers must not rely on the fact that _only_ the * first iovec will be faulted with the current implementation). */ int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) |
2f718ffc1
|
1873 |
{ |
2f718ffc1
|
1874 |
char __user *buf = i->iov->iov_base + i->iov_offset; |
afddba49d
|
1875 1876 |
bytes = min(bytes, i->iov->iov_len - i->iov_offset); return fault_in_pages_readable(buf, bytes); |
2f718ffc1
|
1877 |
} |
89e107877
|
1878 |
EXPORT_SYMBOL(iov_iter_fault_in_readable); |
2f718ffc1
|
1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 |
/* * Return the count of just the current iov_iter segment. */ size_t iov_iter_single_seg_count(struct iov_iter *i) { const struct iovec *iov = i->iov; if (i->nr_segs == 1) return i->count; else return min(i->count, iov->iov_len - i->iov_offset); } |
89e107877
|
1891 |
EXPORT_SYMBOL(iov_iter_single_seg_count); |
2f718ffc1
|
1892 1893 |
/* |
1da177e4c
|
1894 1895 |
* Performs necessary checks before doing a write * |
485bb99b4
|
1896 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c
|
1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 |
* Returns appropriate error code that caller should return or * zero in case that write should be allowed. */ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk) { struct inode *inode = file->f_mapping->host; unsigned long limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; if (unlikely(*pos < 0)) return -EINVAL; |
1da177e4c
|
1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 |
if (!isblk) { /* FIXME: this is for backwards compatibility with 2.4 */ if (file->f_flags & O_APPEND) *pos = i_size_read(inode); if (limit != RLIM_INFINITY) { if (*pos >= limit) { send_sig(SIGXFSZ, current, 0); return -EFBIG; } if (*count > limit - (typeof(limit))*pos) { *count = limit - (typeof(limit))*pos; } } } /* * LFS rule */ if (unlikely(*pos + *count > MAX_NON_LFS && !(file->f_flags & O_LARGEFILE))) { if (*pos >= MAX_NON_LFS) { |
1da177e4c
|
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 |
return -EFBIG; } if (*count > MAX_NON_LFS - (unsigned long)*pos) { *count = MAX_NON_LFS - (unsigned long)*pos; } } /* * Are we about to exceed the fs block limit ? * * If we have written data it becomes a short write. If we have * exceeded without writing data we send a signal and return EFBIG. * Linus frestrict idea will clean these up nicely.. */ if (likely(!isblk)) { if (unlikely(*pos >= inode->i_sb->s_maxbytes)) { if (*count || *pos > inode->i_sb->s_maxbytes) { |
1da177e4c
|
1946 1947 1948 1949 1950 1951 1952 1953 |
return -EFBIG; } /* zero-length writes at ->s_maxbytes are OK */ } if (unlikely(*pos + *count > inode->i_sb->s_maxbytes)) *count = inode->i_sb->s_maxbytes - *pos; } else { |
9361401eb
|
1954 |
#ifdef CONFIG_BLOCK |
1da177e4c
|
1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 |
loff_t isize; if (bdev_read_only(I_BDEV(inode))) return -EPERM; isize = i_size_read(inode); if (*pos >= isize) { if (*count || *pos > isize) return -ENOSPC; } if (*pos + *count > isize) *count = isize - *pos; |
9361401eb
|
1966 1967 1968 |
#else return -EPERM; #endif |
1da177e4c
|
1969 1970 1971 1972 |
} return 0; } EXPORT_SYMBOL(generic_write_checks); |
afddba49d
|
1973 1974 1975 1976 1977 |
int pagecache_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { const struct address_space_operations *aops = mapping->a_ops; |
4e02ed4b4
|
1978 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d
|
1979 |
pagep, fsdata); |
afddba49d
|
1980 1981 1982 1983 1984 1985 1986 1987 |
} EXPORT_SYMBOL(pagecache_write_begin); int pagecache_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata) { const struct address_space_operations *aops = mapping->a_ops; |
afddba49d
|
1988 |
|
4e02ed4b4
|
1989 1990 |
mark_page_accessed(page); return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d
|
1991 1992 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c
|
1993 1994 1995 1996 1997 1998 1999 2000 2001 |
ssize_t generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, unsigned long *nr_segs, loff_t pos, loff_t *ppos, size_t count, size_t ocount) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; ssize_t written; |
a969e903a
|
2002 2003 |
size_t write_len; pgoff_t end; |
1da177e4c
|
2004 2005 2006 |
if (count != ocount) *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); |
a969e903a
|
2007 2008 |
write_len = iov_length(iov, *nr_segs); end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; |
a969e903a
|
2009 |
|
48b47c561
|
2010 |
written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); |
a969e903a
|
2011 2012 2013 2014 2015 2016 2017 |
if (written) goto out; /* * After a write we want buffered reads to be sure to go to disk to get * the new data. We invalidate clean cached page from the region we're * about to write. We do this *before* the write so that we can return |
6ccfa806a
|
2018 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a
|
2019 2020 2021 2022 |
*/ if (mapping->nrpages) { written = invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); |
6ccfa806a
|
2023 2024 2025 2026 2027 2028 2029 |
/* * If a page can not be invalidated, return 0 to fall back * to buffered write. */ if (written) { if (written == -EBUSY) return 0; |
a969e903a
|
2030 |
goto out; |
6ccfa806a
|
2031 |
} |
a969e903a
|
2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 |
} written = mapping->a_ops->direct_IO(WRITE, iocb, iov, pos, *nr_segs); /* * Finally, try again to invalidate clean pages which might have been * cached by non-direct readahead, or faulted in by get_user_pages() * if the source of the write was an mmap'ed region of the file * we're writing. Either one is a pretty crazy thing to do, * so we don't support it 100%. If this invalidation * fails, tough, the write still worked... */ if (mapping->nrpages) { invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); } |
1da177e4c
|
2048 2049 2050 2051 2052 2053 2054 2055 |
if (written > 0) { loff_t end = pos + written; if (end > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, end); mark_inode_dirty(inode); } *ppos = end; } |
a969e903a
|
2056 |
out: |
1da177e4c
|
2057 2058 2059 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893
|
2060 2061 2062 2063 |
/* * Find or create a page at the given pagecache position. Return the locked * page. This function is specifically for buffered writes. */ |
54566b2c1
|
2064 2065 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893
|
2066 2067 2068 |
{ int status; struct page *page; |
54566b2c1
|
2069 2070 2071 |
gfp_t gfp_notmask = 0; if (flags & AOP_FLAG_NOFS) gfp_notmask = __GFP_FS; |
eb2be1893
|
2072 2073 2074 2075 |
repeat: page = find_lock_page(mapping, index); if (likely(page)) return page; |
54566b2c1
|
2076 |
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~gfp_notmask); |
eb2be1893
|
2077 2078 |
if (!page) return NULL; |
54566b2c1
|
2079 2080 |
status = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL & ~gfp_notmask); |
eb2be1893
|
2081 2082 2083 2084 2085 2086 2087 2088 |
if (unlikely(status)) { page_cache_release(page); if (status == -EEXIST) goto repeat; return NULL; } return page; } |
54566b2c1
|
2089 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893
|
2090 |
|
afddba49d
|
2091 2092 2093 2094 2095 2096 2097 |
static ssize_t generic_perform_write(struct file *file, struct iov_iter *i, loff_t pos) { struct address_space *mapping = file->f_mapping; const struct address_space_operations *a_ops = mapping->a_ops; long status = 0; ssize_t written = 0; |
674b892ed
|
2098 2099 2100 2101 2102 2103 2104 |
unsigned int flags = 0; /* * Copies from kernel address space cannot fail (NFSD is a big user). */ if (segment_eq(get_fs(), KERNEL_DS)) flags |= AOP_FLAG_UNINTERRUPTIBLE; |
afddba49d
|
2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 |
do { struct page *page; pgoff_t index; /* Pagecache index for current page */ unsigned long offset; /* Offset into pagecache page */ unsigned long bytes; /* Bytes to write to page */ size_t copied; /* Bytes copied from user */ void *fsdata; offset = (pos & (PAGE_CACHE_SIZE - 1)); index = pos >> PAGE_CACHE_SHIFT; bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_count(i)); again: /* * Bring in the user page that we will copy from _first_. * Otherwise there's a nasty deadlock on copying from the * same page as we're writing to, without it being marked * up-to-date. * * Not only is this an optimisation, but it is also required * to check that the address is actually valid, when atomic * usercopies are used, below. */ if (unlikely(iov_iter_fault_in_readable(i, bytes))) { status = -EFAULT; break; } |
674b892ed
|
2135 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d
|
2136 2137 2138 2139 2140 2141 2142 2143 |
&page, &fsdata); if (unlikely(status)) break; pagefault_disable(); copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); pagefault_enable(); flush_dcache_page(page); |
c8236db9c
|
2144 |
mark_page_accessed(page); |
afddba49d
|
2145 2146 2147 2148 2149 2150 2151 |
status = a_ops->write_end(file, mapping, pos, bytes, copied, page, fsdata); if (unlikely(status < 0)) break; copied = status; cond_resched(); |
124d3b704
|
2152 |
iov_iter_advance(i, copied); |
afddba49d
|
2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 |
if (unlikely(copied == 0)) { /* * If we were unable to copy any data at all, we must * fall back to a single segment length write. * * If we didn't fallback here, we could livelock * because not all segments in the iov can be copied at * once without a pagefault. */ bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_single_seg_count(i)); goto again; } |
afddba49d
|
2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); } while (iov_iter_count(i)); return written ? written : status; } ssize_t generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos, loff_t *ppos, size_t count, ssize_t written) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; |
afddba49d
|
2183 2184 2185 2186 |
ssize_t status; struct iov_iter i; iov_iter_init(&i, iov, nr_segs, count, written); |
4e02ed4b4
|
2187 |
status = generic_perform_write(file, &i, pos); |
1da177e4c
|
2188 |
|
1da177e4c
|
2189 |
if (likely(status >= 0)) { |
afddba49d
|
2190 2191 |
written += status; *ppos = pos + status; |
1da177e4c
|
2192 2193 2194 2195 2196 2197 2198 2199 |
} /* * If we get here for O_DIRECT writes then we must have fallen through * to buffered writes (block instantiation inside i_size). So we sync * the file data here, to try to honour O_DIRECT expectations. */ if (unlikely(file->f_flags & O_DIRECT) && written) |
48b47c561
|
2200 2201 |
status = filemap_write_and_wait_range(mapping, pos, pos + written - 1); |
1da177e4c
|
2202 |
|
1da177e4c
|
2203 2204 2205 |
return written ? written : status; } EXPORT_SYMBOL(generic_file_buffered_write); |
e4dd9de3c
|
2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 |
/** * __generic_file_aio_write - write data to a file * @iocb: IO state structure (file, offset, etc.) * @iov: vector with data to write * @nr_segs: number of segments in the vector * @ppos: position where to write * * This function does all the work needed for actually writing data to a * file. It does all basic checks, removes SUID from the file, updates * modification times and calls proper subroutines depending on whether we * do direct IO or a standard buffered write. * * It expects i_mutex to be grabbed unless we work on a block device or similar * object which does not need locking at all. * * This function does *not* take care of syncing data in case of O_SYNC write. * A caller has to handle it. This is mainly due to the fact that we want to * avoid syncing under i_mutex. */ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t *ppos) |
1da177e4c
|
2227 2228 |
{ struct file *file = iocb->ki_filp; |
fb5527e68
|
2229 |
struct address_space * mapping = file->f_mapping; |
1da177e4c
|
2230 2231 2232 |
size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; |
1da177e4c
|
2233 2234 2235 2236 2237 |
loff_t pos; ssize_t written; ssize_t err; ocount = 0; |
0ceb33143
|
2238 2239 2240 |
err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (err) return err; |
1da177e4c
|
2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 |
count = ocount; pos = *ppos; vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); /* We can write back this queue in page reclaim */ current->backing_dev_info = mapping->backing_dev_info; written = 0; err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); if (err) goto out; if (count == 0) goto out; |
2f1936b87
|
2257 |
err = file_remove_suid(file); |
1da177e4c
|
2258 2259 |
if (err) goto out; |
870f48179
|
2260 |
file_update_time(file); |
1da177e4c
|
2261 2262 2263 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { |
fb5527e68
|
2264 2265 2266 2267 2268 |
loff_t endbyte; ssize_t written_buffered; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, ppos, count, ocount); |
1da177e4c
|
2269 2270 2271 2272 2273 2274 2275 2276 |
if (written < 0 || written == count) goto out; /* * direct-io write to a hole: fall through to buffered I/O * for completing the rest of the request. */ pos += written; count -= written; |
fb5527e68
|
2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 |
written_buffered = generic_file_buffered_write(iocb, iov, nr_segs, pos, ppos, count, written); /* * If generic_file_buffered_write() retuned a synchronous error * then we want to return the number of bytes which were * direct-written, or the error code if that was zero. Note * that this differs from normal direct-io semantics, which * will return -EFOO even if some bytes were written. */ if (written_buffered < 0) { err = written_buffered; goto out; } |
1da177e4c
|
2291 |
|
fb5527e68
|
2292 2293 2294 2295 2296 2297 |
/* * We need to ensure that the page cache pages are written to * disk and invalidated to preserve the expected O_DIRECT * semantics. */ endbyte = pos + written_buffered - written - 1; |
ef51c9762
|
2298 2299 2300 2301 |
err = do_sync_mapping_range(file->f_mapping, pos, endbyte, SYNC_FILE_RANGE_WAIT_BEFORE| SYNC_FILE_RANGE_WRITE| SYNC_FILE_RANGE_WAIT_AFTER); |
fb5527e68
|
2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 |
if (err == 0) { written = written_buffered; invalidate_mapping_pages(mapping, pos >> PAGE_CACHE_SHIFT, endbyte >> PAGE_CACHE_SHIFT); } else { /* * We don't know how much we wrote, so just return * the number of bytes which were direct-written */ } } else { written = generic_file_buffered_write(iocb, iov, nr_segs, pos, ppos, count, written); } |
1da177e4c
|
2317 2318 2319 2320 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
e4dd9de3c
|
2321 |
EXPORT_SYMBOL(__generic_file_aio_write); |
e4dd9de3c
|
2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 |
/** * generic_file_aio_write - write data to a file * @iocb: IO state structure * @iov: vector with data to write * @nr_segs: number of segments in the vector * @pos: position in file where to write * * This is a wrapper around __generic_file_aio_write() to be used by most * filesystems. It takes care of syncing the file in case of O_SYNC file * and acquires i_mutex as needed. */ |
027445c37
|
2333 2334 |
ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c
|
2335 2336 |
{ struct file *file = iocb->ki_filp; |
148f948ba
|
2337 |
struct inode *inode = file->f_mapping->host; |
1da177e4c
|
2338 |
ssize_t ret; |
1da177e4c
|
2339 2340 |
BUG_ON(iocb->ki_pos != pos); |
1b1dcc1b5
|
2341 |
mutex_lock(&inode->i_mutex); |
e4dd9de3c
|
2342 |
ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); |
1b1dcc1b5
|
2343 |
mutex_unlock(&inode->i_mutex); |
1da177e4c
|
2344 |
|
148f948ba
|
2345 |
if (ret > 0 || ret == -EIOCBQUEUED) { |
1da177e4c
|
2346 |
ssize_t err; |
148f948ba
|
2347 |
err = generic_write_sync(file, pos, ret); |
c7b50db21
|
2348 |
if (err < 0 && ret > 0) |
1da177e4c
|
2349 2350 2351 2352 2353 |
ret = err; } return ret; } EXPORT_SYMBOL(generic_file_aio_write); |
cf9a2ae8d
|
2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 |
/** * try_to_release_page() - release old fs-specific metadata on a page * * @page: the page which the kernel is trying to free * @gfp_mask: memory allocation flags (and I/O mode) * * The address_space is to try to release any data against the page * (presumably at page->private). If the release was successful, return `1'. * Otherwise return zero. * |
266cf658e
|
2364 2365 2366 |
* This may also be called if PG_fscache is set on a page, indicating that the * page is known to the local caching routines. * |
cf9a2ae8d
|
2367 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
3f31fddfa
|
2368 |
* this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS). |
cf9a2ae8d
|
2369 |
* |
cf9a2ae8d
|
2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 |
*/ int try_to_release_page(struct page *page, gfp_t gfp_mask) { struct address_space * const mapping = page->mapping; BUG_ON(!PageLocked(page)); if (PageWriteback(page)) return 0; if (mapping && mapping->a_ops->releasepage) return mapping->a_ops->releasepage(page, gfp_mask); return try_to_free_buffers(page); } EXPORT_SYMBOL(try_to_release_page); |