Blame view
mm/filemap.c
67.4 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) */ |
b95f1b31b
|
12 |
#include <linux/export.h> |
1da177e4c
|
13 14 |
#include <linux/compiler.h> #include <linux/fs.h> |
c22ce143d
|
15 |
#include <linux/uaccess.h> |
1da177e4c
|
16 |
#include <linux/aio.h> |
c59ede7b7
|
17 |
#include <linux/capability.h> |
1da177e4c
|
18 |
#include <linux/kernel_stat.h> |
5a0e3ad6a
|
19 |
#include <linux/gfp.h> |
1da177e4c
|
20 21 22 23 24 25 26 27 |
#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 |
#include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/security.h> |
44110fe38
|
32 |
#include <linux/cpuset.h> |
2f718ffc1
|
33 |
#include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */ |
8a9f3ccd2
|
34 |
#include <linux/memcontrol.h> |
c515e1fd3
|
35 |
#include <linux/cleancache.h> |
0f8053a50
|
36 |
#include "internal.h" |
1da177e4c
|
37 |
/* |
1da177e4c
|
38 39 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba
|
40 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c
|
41 |
|
1da177e4c
|
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 |
#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: * |
3d48ae45e
|
59 |
* ->i_mmap_mutex (truncate_pagecache) |
1da177e4c
|
60 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919
|
61 62 |
* ->swap_lock (exclusive_swap_page, others) * ->mapping->tree_lock |
1da177e4c
|
63 |
* |
1b1dcc1b5
|
64 |
* ->i_mutex |
3d48ae45e
|
65 |
* ->i_mmap_mutex (truncate->unmap_mapping_range) |
1da177e4c
|
66 67 |
* * ->mmap_sem |
3d48ae45e
|
68 |
* ->i_mmap_mutex |
b8072f099
|
69 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
1da177e4c
|
70 71 72 73 74 |
* ->mapping->tree_lock (arch-dependent flush_dcache_mmap_lock) * * ->mmap_sem * ->lock_page (access_process_vm) * |
82591e6ea
|
75 76 |
* ->i_mutex (generic_file_buffered_write) * ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c
|
77 |
* |
f758eeabe
|
78 |
* bdi->wb.list_lock |
a66979aba
|
79 |
* sb_lock (fs/fs-writeback.c) |
1da177e4c
|
80 81 |
* ->mapping->tree_lock (__sync_single_inode) * |
3d48ae45e
|
82 |
* ->i_mmap_mutex |
1da177e4c
|
83 84 85 |
* ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099
|
86 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c
|
87 |
* |
b8072f099
|
88 |
* ->page_table_lock or pte_lock |
5d337b919
|
89 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c
|
90 91 92 |
* ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) |
053837fce
|
93 |
* ->zone.lru_lock (check_pte_range->isolate_lru_page) |
1da177e4c
|
94 95 |
* ->private_lock (page_remove_rmap->set_page_dirty) * ->tree_lock (page_remove_rmap->set_page_dirty) |
f758eeabe
|
96 |
* bdi.wb->list_lock (page_remove_rmap->set_page_dirty) |
250df6ed2
|
97 |
* ->inode->i_lock (page_remove_rmap->set_page_dirty) |
f758eeabe
|
98 |
* bdi.wb->list_lock (zap_pte_range->set_page_dirty) |
250df6ed2
|
99 |
* ->inode->i_lock (zap_pte_range->set_page_dirty) |
1da177e4c
|
100 101 |
* ->private_lock (zap_pte_range->__set_page_dirty_buffers) * |
9a3c531df
|
102 103 |
* ->i_mmap_mutex * ->tasklist_lock (memory_failure, collect_procs_ao) |
1da177e4c
|
104 105 106 |
*/ /* |
e64a782fe
|
107 |
* Delete a page from the page cache and free it. Caller has to make |
1da177e4c
|
108 |
* sure the page is locked and that nobody else uses it - or that usage |
19fd62312
|
109 |
* is safe. The caller must hold the mapping's tree_lock. |
1da177e4c
|
110 |
*/ |
e64a782fe
|
111 |
void __delete_from_page_cache(struct page *page) |
1da177e4c
|
112 113 |
{ struct address_space *mapping = page->mapping; |
c515e1fd3
|
114 115 116 117 118 119 120 121 |
/* * if we're uptodate, flush out into the cleancache, otherwise * invalidate any existing cleancache entries. We can't leave * stale data around in the cleancache once our page is gone */ if (PageUptodate(page) && PageMappedToDisk(page)) cleancache_put_page(page); else |
3167760f8
|
122 |
cleancache_invalidate_page(mapping, page); |
c515e1fd3
|
123 |
|
1da177e4c
|
124 125 |
radix_tree_delete(&mapping->page_tree, page->index); page->mapping = NULL; |
b85e0effd
|
126 |
/* Leave page->index set: truncation lookup relies upon it */ |
1da177e4c
|
127 |
mapping->nrpages--; |
347ce434d
|
128 |
__dec_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac
|
129 130 |
if (PageSwapBacked(page)) __dec_zone_page_state(page, NR_SHMEM); |
45426812d
|
131 |
BUG_ON(page_mapped(page)); |
3a6927906
|
132 133 134 135 136 137 138 139 140 141 142 143 |
/* * 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
|
144 |
} |
702cfbf93
|
145 146 147 148 149 150 151 152 153 |
/** * delete_from_page_cache - delete page from page cache * @page: the page which the kernel is trying to remove from page cache * * This must be called only on pages that have been verified to be in the page * cache and locked. It will never put the page into the free list, the caller * has a reference on the page. */ void delete_from_page_cache(struct page *page) |
1da177e4c
|
154 155 |
{ struct address_space *mapping = page->mapping; |
6072d13c4
|
156 |
void (*freepage)(struct page *); |
1da177e4c
|
157 |
|
cd7619d6b
|
158 |
BUG_ON(!PageLocked(page)); |
1da177e4c
|
159 |
|
6072d13c4
|
160 |
freepage = mapping->a_ops->freepage; |
19fd62312
|
161 |
spin_lock_irq(&mapping->tree_lock); |
e64a782fe
|
162 |
__delete_from_page_cache(page); |
19fd62312
|
163 |
spin_unlock_irq(&mapping->tree_lock); |
e767e0561
|
164 |
mem_cgroup_uncharge_cache_page(page); |
6072d13c4
|
165 166 167 |
if (freepage) freepage(page); |
97cecb5a2
|
168 169 170 |
page_cache_release(page); } EXPORT_SYMBOL(delete_from_page_cache); |
7eaceacca
|
171 |
static int sleep_on_page(void *word) |
1da177e4c
|
172 |
{ |
1da177e4c
|
173 174 175 |
io_schedule(); return 0; } |
7eaceacca
|
176 |
static int sleep_on_page_killable(void *word) |
2687a3569
|
177 |
{ |
7eaceacca
|
178 |
sleep_on_page(word); |
2687a3569
|
179 180 |
return fatal_signal_pending(current) ? -EINTR : 0; } |
1da177e4c
|
181 |
/** |
485bb99b4
|
182 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b
|
183 184 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03
|
185 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b
|
186 |
* @sync_mode: enable synchronous operation |
1da177e4c
|
187 |
* |
485bb99b4
|
188 189 190 |
* Start writeback against all of a mapping's dirty pages that lie * within the byte offsets <start, end> inclusive. * |
1da177e4c
|
191 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4
|
192 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c
|
193 194 195 |
* these two operations is that if a dirty page/buffer is encountered, it must * be waited upon, and not just skipped over. */ |
ebcf28e1c
|
196 197 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c
|
198 199 200 201 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd
|
202 |
.nr_to_write = LONG_MAX, |
111ebb6e6
|
203 204 |
.range_start = start, .range_end = end, |
1da177e4c
|
205 206 207 208 209 210 211 212 213 214 215 216 |
}; 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
|
217 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c
|
218 219 220 221 222 223 224 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf
|
225 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c
|
226 |
loff_t end) |
1da177e4c
|
227 228 229 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf
|
230 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c
|
231 |
|
485bb99b4
|
232 233 234 235 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c
|
236 237 238 239 240 241 242 243 |
* 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
|
244 |
/** |
94004ed72
|
245 246 247 248 |
* 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
|
249 |
* |
94004ed72
|
250 251 |
* Walk the list of under-writeback pages of the given address space * in the given range and wait for all of them. |
1da177e4c
|
252 |
*/ |
94004ed72
|
253 254 |
int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) |
1da177e4c
|
255 |
{ |
94004ed72
|
256 257 |
pgoff_t index = start_byte >> PAGE_CACHE_SHIFT; pgoff_t end = end_byte >> PAGE_CACHE_SHIFT; |
1da177e4c
|
258 259 260 |
struct pagevec pvec; int nr_pages; int ret = 0; |
1da177e4c
|
261 |
|
94004ed72
|
262 |
if (end_byte < start_byte) |
1da177e4c
|
263 264 265 |
return 0; pagevec_init(&pvec, 0); |
1da177e4c
|
266 267 268 269 270 271 272 273 274 275 276 277 278 279 |
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); |
212260aa0
|
280 |
if (TestClearPageError(page)) |
1da177e4c
|
281 282 283 284 285 286 287 288 289 290 291 292 293 294 |
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
|
295 296 297 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
485bb99b4
|
298 |
* filemap_fdatawait - wait for all under-writeback pages to complete |
1da177e4c
|
299 |
* @mapping: address space structure to wait for |
485bb99b4
|
300 301 302 |
* * Walk the list of under-writeback pages of the given address space * and wait for all of them. |
1da177e4c
|
303 304 305 306 307 308 309 |
*/ int filemap_fdatawait(struct address_space *mapping) { loff_t i_size = i_size_read(mapping->host); if (i_size == 0) return 0; |
94004ed72
|
310 |
return filemap_fdatawait_range(mapping, 0, i_size - 1); |
1da177e4c
|
311 312 313 314 315 |
} EXPORT_SYMBOL(filemap_fdatawait); int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982
|
316 |
int err = 0; |
1da177e4c
|
317 318 |
if (mapping->nrpages) { |
28fd12982
|
319 320 321 322 323 324 325 326 327 328 329 330 |
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
|
331 |
} |
28fd12982
|
332 |
return err; |
1da177e4c
|
333 |
} |
28fd12982
|
334 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c
|
335 |
|
485bb99b4
|
336 337 338 339 340 341 |
/** * 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
|
342 343 344 345 346 |
* 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
|
347 348 349 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982
|
350 |
int err = 0; |
1da177e4c
|
351 352 |
if (mapping->nrpages) { |
28fd12982
|
353 354 355 356 |
err = __filemap_fdatawrite_range(mapping, lstart, lend, WB_SYNC_ALL); /* See comment of filemap_write_and_wait() */ if (err != -EIO) { |
94004ed72
|
357 358 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982
|
359 360 361 |
if (!err) err = err2; } |
1da177e4c
|
362 |
} |
28fd12982
|
363 |
return err; |
1da177e4c
|
364 |
} |
f69955855
|
365 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c
|
366 |
|
485bb99b4
|
367 |
/** |
ef6a3c631
|
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 |
* replace_page_cache_page - replace a pagecache page with a new one * @old: page to be replaced * @new: page to replace with * @gfp_mask: allocation mode * * This function replaces a page in the pagecache with a new one. On * success it acquires the pagecache reference for the new page and * drops it for the old page. Both the old and new pages must be * locked. This function does not add the new page to the LRU, the * caller must do that. * * The remove + add is atomic. The only way this function can fail is * memory allocation failure. */ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) { int error; |
ef6a3c631
|
385 386 387 388 |
VM_BUG_ON(!PageLocked(old)); VM_BUG_ON(!PageLocked(new)); VM_BUG_ON(new->mapping); |
ef6a3c631
|
389 390 391 392 393 394 395 396 397 398 399 400 401 |
error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM); if (!error) { struct address_space *mapping = old->mapping; void (*freepage)(struct page *); pgoff_t offset = old->index; freepage = mapping->a_ops->freepage; page_cache_get(new); new->mapping = mapping; new->index = offset; spin_lock_irq(&mapping->tree_lock); |
e64a782fe
|
402 |
__delete_from_page_cache(old); |
ef6a3c631
|
403 404 405 406 407 408 409 |
error = radix_tree_insert(&mapping->page_tree, offset, new); BUG_ON(error); mapping->nrpages++; __inc_zone_page_state(new, NR_FILE_PAGES); if (PageSwapBacked(new)) __inc_zone_page_state(new, NR_SHMEM); spin_unlock_irq(&mapping->tree_lock); |
ab936cbcd
|
410 411 |
/* mem_cgroup codes must not be called under tree_lock */ mem_cgroup_replace_page_cache(old, new); |
ef6a3c631
|
412 413 414 415 |
radix_tree_preload_end(); if (freepage) freepage(old); page_cache_release(old); |
ef6a3c631
|
416 417 418 419 420 421 422 |
} return error; } EXPORT_SYMBOL_GPL(replace_page_cache_page); /** |
e286781d5
|
423 |
* add_to_page_cache_locked - add a locked page to the pagecache |
485bb99b4
|
424 425 426 427 428 |
* @page: page to add * @mapping: the page's address_space * @offset: page index * @gfp_mask: page allocation mode * |
e286781d5
|
429 |
* This function is used to add a page to the pagecache. It must be locked. |
1da177e4c
|
430 431 |
* This function does not add the page to the LRU. The caller must do that. */ |
e286781d5
|
432 |
int add_to_page_cache_locked(struct page *page, struct address_space *mapping, |
6daa0e286
|
433 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c
|
434 |
{ |
e286781d5
|
435 436 437 |
int error; VM_BUG_ON(!PageLocked(page)); |
31475dd61
|
438 |
VM_BUG_ON(PageSwapBacked(page)); |
e286781d5
|
439 440 |
error = mem_cgroup_cache_charge(page, current->mm, |
2c26fdd70
|
441 |
gfp_mask & GFP_RECLAIM_MASK); |
35c754d79
|
442 443 |
if (error) goto out; |
1da177e4c
|
444 |
|
35c754d79
|
445 |
error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM); |
1da177e4c
|
446 |
if (error == 0) { |
e286781d5
|
447 448 449 |
page_cache_get(page); page->mapping = mapping; page->index = offset; |
19fd62312
|
450 |
spin_lock_irq(&mapping->tree_lock); |
1da177e4c
|
451 |
error = radix_tree_insert(&mapping->page_tree, offset, page); |
e286781d5
|
452 |
if (likely(!error)) { |
1da177e4c
|
453 |
mapping->nrpages++; |
347ce434d
|
454 |
__inc_zone_page_state(page, NR_FILE_PAGES); |
e767e0561
|
455 |
spin_unlock_irq(&mapping->tree_lock); |
e286781d5
|
456 457 |
} else { page->mapping = NULL; |
b85e0effd
|
458 |
/* Leave page->index set: truncation relies upon it */ |
e767e0561
|
459 |
spin_unlock_irq(&mapping->tree_lock); |
69029cd55
|
460 |
mem_cgroup_uncharge_cache_page(page); |
e286781d5
|
461 462 |
page_cache_release(page); } |
1da177e4c
|
463 |
radix_tree_preload_end(); |
35c754d79
|
464 |
} else |
69029cd55
|
465 |
mem_cgroup_uncharge_cache_page(page); |
8a9f3ccd2
|
466 |
out: |
1da177e4c
|
467 468 |
return error; } |
e286781d5
|
469 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c
|
470 471 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286
|
472 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c
|
473 |
{ |
4f98a2fee
|
474 |
int ret; |
4f98a2fee
|
475 |
ret = add_to_page_cache(page, mapping, offset, gfp_mask); |
31475dd61
|
476 477 |
if (ret == 0) lru_cache_add_file(page); |
1da177e4c
|
478 479 |
return ret; } |
18bc0bbd1
|
480 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c
|
481 |
|
44110fe38
|
482 |
#ifdef CONFIG_NUMA |
2ae88149a
|
483 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38
|
484 |
{ |
c0ff7453b
|
485 486 |
int n; struct page *page; |
44110fe38
|
487 |
if (cpuset_do_page_mem_spread()) { |
cc9a6c877
|
488 489 490 491 492 493 |
unsigned int cpuset_mems_cookie; do { cpuset_mems_cookie = get_mems_allowed(); n = cpuset_mem_spread_node(); page = alloc_pages_exact_node(n, gfp, 0); } while (!put_mems_allowed(cpuset_mems_cookie) && !page); |
c0ff7453b
|
494 |
return page; |
44110fe38
|
495 |
} |
2ae88149a
|
496 |
return alloc_pages(gfp, 0); |
44110fe38
|
497 |
} |
2ae88149a
|
498 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38
|
499 |
#endif |
1da177e4c
|
500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 |
/* * 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
|
521 |
void wait_on_page_bit(struct page *page, int bit_nr) |
1da177e4c
|
522 523 524 525 |
{ DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); if (test_bit(bit_nr, &page->flags)) |
7eaceacca
|
526 |
__wait_on_bit(page_waitqueue(page), &wait, sleep_on_page, |
1da177e4c
|
527 528 529 |
TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(wait_on_page_bit); |
f62e00cc3
|
530 531 532 533 534 535 536 537 538 539 |
int wait_on_page_bit_killable(struct page *page, int bit_nr) { DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); if (!test_bit(bit_nr, &page->flags)) return 0; return __wait_on_bit(page_waitqueue(page), &wait, sleep_on_page_killable, TASK_KILLABLE); } |
1da177e4c
|
540 |
/** |
385e1ca5f
|
541 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc
|
542 543 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f
|
544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 |
* * 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
|
559 |
* unlock_page - unlock a locked page |
1da177e4c
|
560 561 562 563 564 565 566 |
* @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
|
567 568 |
* 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
|
569 |
*/ |
920c7a5d0
|
570 |
void unlock_page(struct page *page) |
1da177e4c
|
571 |
{ |
8413ac9d8
|
572 573 574 |
VM_BUG_ON(!PageLocked(page)); clear_bit_unlock(PG_locked, &page->flags); smp_mb__after_clear_bit(); |
1da177e4c
|
575 576 577 |
wake_up_page(page, PG_locked); } EXPORT_SYMBOL(unlock_page); |
485bb99b4
|
578 579 580 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c
|
581 582 583 |
*/ void end_page_writeback(struct page *page) { |
ac6aadb24
|
584 585 586 587 588 |
if (TestClearPageReclaim(page)) rotate_reclaimable_page(page); if (!test_clear_page_writeback(page)) BUG(); |
1da177e4c
|
589 590 591 592 |
smp_mb__after_clear_bit(); wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
485bb99b4
|
593 594 595 |
/** * __lock_page - get a lock on the page, assuming we need to sleep to get it * @page: the page to lock |
1da177e4c
|
596 |
*/ |
920c7a5d0
|
597 |
void __lock_page(struct page *page) |
1da177e4c
|
598 599 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); |
7eaceacca
|
600 |
__wait_on_bit_lock(page_waitqueue(page), &wait, sleep_on_page, |
1da177e4c
|
601 602 603 |
TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(__lock_page); |
b5606c2d4
|
604 |
int __lock_page_killable(struct page *page) |
2687a3569
|
605 606 607 608 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); return __wait_on_bit_lock(page_waitqueue(page), &wait, |
7eaceacca
|
609 |
sleep_on_page_killable, TASK_KILLABLE); |
2687a3569
|
610 |
} |
18bc0bbd1
|
611 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569
|
612 |
|
d065bd810
|
613 614 615 |
int __lock_page_or_retry(struct page *page, struct mm_struct *mm, unsigned int flags) { |
37b23e052
|
616 617 618 619 620 621 622 623 624 625 626 627 |
if (flags & FAULT_FLAG_ALLOW_RETRY) { /* * CAUTION! In this case, mmap_sem is not released * even though return 0. */ if (flags & FAULT_FLAG_RETRY_NOWAIT) return 0; up_read(&mm->mmap_sem); if (flags & FAULT_FLAG_KILLABLE) wait_on_page_locked_killable(page); else |
318b275fb
|
628 |
wait_on_page_locked(page); |
d065bd810
|
629 |
return 0; |
37b23e052
|
630 631 632 633 634 635 636 637 638 639 640 641 |
} else { if (flags & FAULT_FLAG_KILLABLE) { int ret; ret = __lock_page_killable(page); if (ret) { up_read(&mm->mmap_sem); return 0; } } else __lock_page(page); return 1; |
d065bd810
|
642 643 |
} } |
485bb99b4
|
644 645 646 647 648 |
/** * find_get_page - find and get a page reference * @mapping: the address_space to search * @offset: the page index * |
da6052f7b
|
649 650 |
* 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
|
651 |
*/ |
a60637c85
|
652 |
struct page *find_get_page(struct address_space *mapping, pgoff_t offset) |
1da177e4c
|
653 |
{ |
a60637c85
|
654 |
void **pagep; |
1da177e4c
|
655 |
struct page *page; |
a60637c85
|
656 657 658 659 660 661 |
rcu_read_lock(); repeat: page = NULL; pagep = radix_tree_lookup_slot(&mapping->page_tree, offset); if (pagep) { page = radix_tree_deref_slot(pagep); |
27d20fddc
|
662 663 |
if (unlikely(!page)) goto out; |
a2c16d6cb
|
664 |
if (radix_tree_exception(page)) { |
8079b1c85
|
665 666 667 668 669 670 671 672 |
if (radix_tree_deref_retry(page)) goto repeat; /* * Otherwise, shmem/tmpfs must be storing a swap entry * here as an exceptional entry: so return it without * attempting to raise page count. */ goto out; |
a2c16d6cb
|
673 |
} |
a60637c85
|
674 675 676 677 678 679 680 681 682 683 684 685 686 |
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; } } |
27d20fddc
|
687 |
out: |
a60637c85
|
688 |
rcu_read_unlock(); |
1da177e4c
|
689 690 |
return page; } |
1da177e4c
|
691 |
EXPORT_SYMBOL(find_get_page); |
485bb99b4
|
692 |
/** |
1da177e4c
|
693 |
* find_lock_page - locate, pin and lock a pagecache page |
67be2dd1b
|
694 695 |
* @mapping: the address_space to search * @offset: the page index |
1da177e4c
|
696 697 698 699 700 701 |
* * 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
|
702 |
struct page *find_lock_page(struct address_space *mapping, pgoff_t offset) |
1da177e4c
|
703 704 |
{ struct page *page; |
1da177e4c
|
705 |
repeat: |
a60637c85
|
706 |
page = find_get_page(mapping, offset); |
a2c16d6cb
|
707 |
if (page && !radix_tree_exception(page)) { |
a60637c85
|
708 709 710 711 712 713 |
lock_page(page); /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); page_cache_release(page); goto repeat; |
1da177e4c
|
714 |
} |
a60637c85
|
715 |
VM_BUG_ON(page->index != offset); |
1da177e4c
|
716 |
} |
1da177e4c
|
717 718 |
return page; } |
1da177e4c
|
719 720 721 722 |
EXPORT_SYMBOL(find_lock_page); /** * find_or_create_page - locate or add a pagecache page |
67be2dd1b
|
723 724 725 |
* @mapping: the page's address_space * @index: the page's index into the mapping * @gfp_mask: page allocation mode |
1da177e4c
|
726 727 728 729 730 731 732 733 734 735 736 737 738 |
* * 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
|
739 |
pgoff_t index, gfp_t gfp_mask) |
1da177e4c
|
740 |
{ |
eb2be1893
|
741 |
struct page *page; |
1da177e4c
|
742 743 744 745 |
int err; repeat: page = find_lock_page(mapping, index); if (!page) { |
eb2be1893
|
746 747 748 |
page = __page_cache_alloc(gfp_mask); if (!page) return NULL; |
67d58ac47
|
749 750 751 752 753 754 755 756 |
/* * 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
|
757 758 759 760 761 |
if (unlikely(err)) { page_cache_release(page); page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c
|
762 |
} |
1da177e4c
|
763 |
} |
1da177e4c
|
764 765 |
return page; } |
1da177e4c
|
766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 |
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) { |
0fc9d1040
|
787 788 789 790 791 792 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85
|
793 794 795 |
rcu_read_lock(); restart: |
0fc9d1040
|
796 |
radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, start) { |
a60637c85
|
797 798 |
struct page *page; repeat: |
0fc9d1040
|
799 |
page = radix_tree_deref_slot(slot); |
a60637c85
|
800 801 |
if (unlikely(!page)) continue; |
9d8aa4ea8
|
802 |
|
a2c16d6cb
|
803 |
if (radix_tree_exception(page)) { |
8079b1c85
|
804 805 806 807 808 809 |
if (radix_tree_deref_retry(page)) { /* * Transient condition which can only trigger * when entry at index 0 moves out of or back * to root: none yet gotten, safe to restart. */ |
0fc9d1040
|
810 |
WARN_ON(iter.index); |
8079b1c85
|
811 812 |
goto restart; } |
a2c16d6cb
|
813 |
/* |
8079b1c85
|
814 815 816 |
* Otherwise, shmem/tmpfs must be storing a swap entry * here as an exceptional entry: so skip over it - * we only reach this from invalidate_mapping_pages(). |
a2c16d6cb
|
817 |
*/ |
8079b1c85
|
818 |
continue; |
27d20fddc
|
819 |
} |
a60637c85
|
820 821 822 823 824 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040
|
825 |
if (unlikely(page != *slot)) { |
a60637c85
|
826 827 828 |
page_cache_release(page); goto repeat; } |
1da177e4c
|
829 |
|
a60637c85
|
830 |
pages[ret] = page; |
0fc9d1040
|
831 832 |
if (++ret == nr_pages) break; |
a60637c85
|
833 |
} |
5b280c0cc
|
834 |
|
a60637c85
|
835 |
rcu_read_unlock(); |
1da177e4c
|
836 837 |
return ret; } |
ebf43500e
|
838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 |
/** * 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) { |
0fc9d1040
|
853 854 855 856 857 858 |
struct radix_tree_iter iter; void **slot; unsigned int ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85
|
859 860 861 |
rcu_read_lock(); restart: |
0fc9d1040
|
862 |
radix_tree_for_each_contig(slot, &mapping->page_tree, &iter, index) { |
a60637c85
|
863 864 |
struct page *page; repeat: |
0fc9d1040
|
865 866 |
page = radix_tree_deref_slot(slot); /* The hole, there no reason to continue */ |
a60637c85
|
867 |
if (unlikely(!page)) |
0fc9d1040
|
868 |
break; |
9d8aa4ea8
|
869 |
|
a2c16d6cb
|
870 |
if (radix_tree_exception(page)) { |
8079b1c85
|
871 872 873 874 875 876 877 878 |
if (radix_tree_deref_retry(page)) { /* * Transient condition which can only trigger * when entry at index 0 moves out of or back * to root: none yet gotten, safe to restart. */ goto restart; } |
a2c16d6cb
|
879 |
/* |
8079b1c85
|
880 881 882 |
* Otherwise, shmem/tmpfs must be storing a swap entry * here as an exceptional entry: so stop looking for * contiguous pages. |
a2c16d6cb
|
883 |
*/ |
8079b1c85
|
884 |
break; |
a2c16d6cb
|
885 |
} |
ebf43500e
|
886 |
|
a60637c85
|
887 888 889 890 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040
|
891 |
if (unlikely(page != *slot)) { |
a60637c85
|
892 893 894 |
page_cache_release(page); goto repeat; } |
9cbb4cb21
|
895 896 897 898 899 |
/* * must check mapping and index after taking the ref. * otherwise we can get both false positives and false * negatives, which is just confusing to the caller. */ |
0fc9d1040
|
900 |
if (page->mapping == NULL || page->index != iter.index) { |
9cbb4cb21
|
901 902 903 |
page_cache_release(page); break; } |
a60637c85
|
904 |
pages[ret] = page; |
0fc9d1040
|
905 906 |
if (++ret == nr_pages) break; |
ebf43500e
|
907 |
} |
a60637c85
|
908 909 |
rcu_read_unlock(); return ret; |
ebf43500e
|
910 |
} |
ef71c15c4
|
911 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e
|
912 |
|
485bb99b4
|
913 914 915 916 917 918 919 920 |
/** * 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
|
921 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4
|
922 |
* @tag. We update @index to index the next page for the traversal. |
1da177e4c
|
923 924 925 926 |
*/ unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, int tag, unsigned int nr_pages, struct page **pages) { |
0fc9d1040
|
927 928 929 930 931 932 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85
|
933 934 935 |
rcu_read_lock(); restart: |
0fc9d1040
|
936 937 |
radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, *index, tag) { |
a60637c85
|
938 939 |
struct page *page; repeat: |
0fc9d1040
|
940 |
page = radix_tree_deref_slot(slot); |
a60637c85
|
941 942 |
if (unlikely(!page)) continue; |
9d8aa4ea8
|
943 |
|
a2c16d6cb
|
944 |
if (radix_tree_exception(page)) { |
8079b1c85
|
945 946 947 948 949 950 951 952 |
if (radix_tree_deref_retry(page)) { /* * Transient condition which can only trigger * when entry at index 0 moves out of or back * to root: none yet gotten, safe to restart. */ goto restart; } |
a2c16d6cb
|
953 |
/* |
8079b1c85
|
954 955 |
* This function is never used on a shmem/tmpfs * mapping, so a swap entry won't be found here. |
a2c16d6cb
|
956 |
*/ |
8079b1c85
|
957 |
BUG(); |
a2c16d6cb
|
958 |
} |
a60637c85
|
959 960 961 962 963 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040
|
964 |
if (unlikely(page != *slot)) { |
a60637c85
|
965 966 967 968 969 |
page_cache_release(page); goto repeat; } pages[ret] = page; |
0fc9d1040
|
970 971 |
if (++ret == nr_pages) break; |
a60637c85
|
972 |
} |
5b280c0cc
|
973 |
|
a60637c85
|
974 |
rcu_read_unlock(); |
1da177e4c
|
975 |
|
1da177e4c
|
976 977 |
if (ret) *index = pages[ret - 1]->index + 1; |
a60637c85
|
978 |
|
1da177e4c
|
979 980 |
return ret; } |
ef71c15c4
|
981 |
EXPORT_SYMBOL(find_get_pages_tag); |
1da177e4c
|
982 |
|
485bb99b4
|
983 984 985 986 987 |
/** * grab_cache_page_nowait - returns locked page at given index in given cache * @mapping: target address_space * @index: the page index * |
72fd4a35a
|
988 |
* Same as grab_cache_page(), but do not wait if the page is unavailable. |
1da177e4c
|
989 990 991 992 993 994 995 996 |
* 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
|
997 |
grab_cache_page_nowait(struct address_space *mapping, pgoff_t index) |
1da177e4c
|
998 999 |
{ struct page *page = find_get_page(mapping, index); |
1da177e4c
|
1000 1001 |
if (page) { |
529ae9aaa
|
1002 |
if (trylock_page(page)) |
1da177e4c
|
1003 1004 1005 1006 |
return page; page_cache_release(page); return NULL; } |
2ae88149a
|
1007 |
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); |
67d58ac47
|
1008 |
if (page && add_to_page_cache_lru(page, mapping, index, GFP_NOFS)) { |
1da177e4c
|
1009 1010 1011 1012 1013 |
page_cache_release(page); page = NULL; } return page; } |
1da177e4c
|
1014 |
EXPORT_SYMBOL(grab_cache_page_nowait); |
76d42bd96
|
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 |
/* * 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
|
1033 |
ra->ra_pages /= 4; |
76d42bd96
|
1034 |
} |
485bb99b4
|
1035 |
/** |
36e789144
|
1036 |
* do_generic_file_read - generic file read routine |
485bb99b4
|
1037 1038 1039 1040 1041 |
* @filp: the file to read * @ppos: current file position * @desc: read_descriptor * @actor: read method * |
1da177e4c
|
1042 |
* This is a generic file read routine, and uses the |
485bb99b4
|
1043 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c
|
1044 1045 1046 |
* * This is really ugly. But the goto's actually try to clarify some * of the logic when it comes to error handling etc. |
1da177e4c
|
1047 |
*/ |
36e789144
|
1048 1049 |
static void do_generic_file_read(struct file *filp, loff_t *ppos, read_descriptor_t *desc, read_actor_t actor) |
1da177e4c
|
1050 |
{ |
36e789144
|
1051 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c
|
1052 |
struct inode *inode = mapping->host; |
36e789144
|
1053 |
struct file_ra_state *ra = &filp->f_ra; |
57f6b96c0
|
1054 1055 1056 1057 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372
|
1058 |
unsigned int prev_offset; |
1da177e4c
|
1059 |
int error; |
1da177e4c
|
1060 |
|
1da177e4c
|
1061 |
index = *ppos >> PAGE_CACHE_SHIFT; |
7ff81078d
|
1062 1063 |
prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); |
1da177e4c
|
1064 1065 |
last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; offset = *ppos & ~PAGE_CACHE_MASK; |
1da177e4c
|
1066 1067 |
for (;;) { struct page *page; |
57f6b96c0
|
1068 |
pgoff_t end_index; |
a32ea1e1f
|
1069 |
loff_t isize; |
1da177e4c
|
1070 |
unsigned long nr, ret; |
1da177e4c
|
1071 |
cond_resched(); |
1da177e4c
|
1072 1073 |
find_page: page = find_get_page(mapping, index); |
3ea89ee86
|
1074 |
if (!page) { |
cf914a7d6
|
1075 |
page_cache_sync_readahead(mapping, |
7ff81078d
|
1076 |
ra, filp, |
3ea89ee86
|
1077 1078 1079 1080 1081 1082 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6
|
1083 |
page_cache_async_readahead(mapping, |
7ff81078d
|
1084 |
ra, filp, page, |
3ea89ee86
|
1085 |
index, last_index - index); |
1da177e4c
|
1086 |
} |
8ab22b9ab
|
1087 1088 1089 1090 |
if (!PageUptodate(page)) { if (inode->i_blkbits == PAGE_CACHE_SHIFT || !mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; |
529ae9aaa
|
1091 |
if (!trylock_page(page)) |
8ab22b9ab
|
1092 |
goto page_not_up_to_date; |
8d056cb96
|
1093 1094 1095 |
/* Did it get truncated before we got the lock? */ if (!page->mapping) goto page_not_up_to_date_locked; |
8ab22b9ab
|
1096 1097 1098 1099 1100 |
if (!mapping->a_ops->is_partially_uptodate(page, desc, offset)) goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c
|
1101 |
page_ok: |
a32ea1e1f
|
1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 |
/* * 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
|
1128 1129 1130 1131 1132 1133 1134 1135 1136 |
/* 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
|
1137 1138 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c
|
1139 |
*/ |
ec0f16372
|
1140 |
if (prev_index != index || offset != prev_offset) |
1da177e4c
|
1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 |
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
|
1158 |
prev_offset = offset; |
1da177e4c
|
1159 1160 1161 1162 1163 1164 1165 1166 |
page_cache_release(page); if (ret == nr && desc->count) continue; goto out; page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235
|
1167 1168 1169 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c
|
1170 |
|
8ab22b9ab
|
1171 |
page_not_up_to_date_locked: |
da6052f7b
|
1172 |
/* Did it get truncated before we got the lock? */ |
1da177e4c
|
1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 |
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: |
91803b499
|
1186 1187 1188 1189 1190 1191 |
/* * A previous I/O error may have been due to temporary * failures, eg. multipath errors. * PG_error will be set again if readpage fails. */ ClearPageError(page); |
1da177e4c
|
1192 1193 |
/* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7
|
1194 1195 1196 1197 1198 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { page_cache_release(page); goto find_page; } |
1da177e4c
|
1199 |
goto readpage_error; |
994fc28c7
|
1200 |
} |
1da177e4c
|
1201 1202 |
if (!PageUptodate(page)) { |
854623235
|
1203 1204 1205 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c
|
1206 1207 1208 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* |
2ecdc82ef
|
1209 |
* invalidate_mapping_pages got it |
1da177e4c
|
1210 1211 1212 1213 1214 1215 |
*/ unlock_page(page); page_cache_release(page); goto find_page; } unlock_page(page); |
7ff81078d
|
1216 |
shrink_readahead_size_eio(filp, ra); |
854623235
|
1217 1218 |
error = -EIO; goto readpage_error; |
1da177e4c
|
1219 1220 1221 |
} unlock_page(page); } |
1da177e4c
|
1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 |
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
|
1235 1236 1237 1238 |
page = page_cache_alloc_cold(mapping); if (!page) { desc->error = -ENOMEM; goto out; |
1da177e4c
|
1239 |
} |
eb2be1893
|
1240 |
error = add_to_page_cache_lru(page, mapping, |
1da177e4c
|
1241 1242 |
index, GFP_KERNEL); if (error) { |
eb2be1893
|
1243 |
page_cache_release(page); |
1da177e4c
|
1244 1245 1246 1247 1248 |
if (error == -EEXIST) goto find_page; desc->error = error; goto out; } |
1da177e4c
|
1249 1250 1251 1252 |
goto readpage; } out: |
7ff81078d
|
1253 1254 1255 |
ra->prev_pos = prev_index; ra->prev_pos <<= PAGE_CACHE_SHIFT; ra->prev_pos |= prev_offset; |
1da177e4c
|
1256 |
|
f4e6b498d
|
1257 |
*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; |
0c6aa2639
|
1258 |
file_accessed(filp); |
1da177e4c
|
1259 |
} |
1da177e4c
|
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 |
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)) { |
9b04c5fec
|
1275 |
kaddr = kmap_atomic(page); |
1da177e4c
|
1276 1277 |
left = __copy_to_user_inatomic(desc->arg.buf, kaddr + offset, size); |
9b04c5fec
|
1278 |
kunmap_atomic(kaddr); |
1da177e4c
|
1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 |
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
|
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 |
/* * 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
|
1336 |
/** |
b2abacf3a
|
1337 |
* generic_file_aio_read - generic filesystem read routine |
485bb99b4
|
1338 1339 1340 |
* @iocb: kernel I/O control block * @iov: io vector request * @nr_segs: number of segments in the iovec |
b2abacf3a
|
1341 |
* @pos: current file position |
485bb99b4
|
1342 |
* |
1da177e4c
|
1343 1344 1345 1346 |
* This is the "read()" routine for all filesystems * that can use the page cache directly. */ ssize_t |
543ade1fc
|
1347 1348 |
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c
|
1349 1350 1351 |
{ struct file *filp = iocb->ki_filp; ssize_t retval; |
66f998f61
|
1352 |
unsigned long seg = 0; |
1da177e4c
|
1353 |
size_t count; |
543ade1fc
|
1354 |
loff_t *ppos = &iocb->ki_pos; |
1da177e4c
|
1355 1356 |
count = 0; |
0ceb33143
|
1357 1358 1359 |
retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; |
1da177e4c
|
1360 1361 1362 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { |
543ade1fc
|
1363 |
loff_t size; |
1da177e4c
|
1364 1365 1366 1367 1368 |
struct address_space *mapping; struct inode *inode; mapping = filp->f_mapping; inode = mapping->host; |
1da177e4c
|
1369 1370 1371 1372 |
if (!count) goto out; /* skip atime */ size = i_size_read(inode); if (pos < size) { |
48b47c561
|
1373 1374 |
retval = filemap_write_and_wait_range(mapping, pos, pos + iov_length(iov, nr_segs) - 1); |
a969e903a
|
1375 |
if (!retval) { |
3deaa7190
|
1376 1377 1378 |
struct blk_plug plug; blk_start_plug(&plug); |
a969e903a
|
1379 1380 |
retval = mapping->a_ops->direct_IO(READ, iocb, iov, pos, nr_segs); |
3deaa7190
|
1381 |
blk_finish_plug(&plug); |
a969e903a
|
1382 |
} |
66f998f61
|
1383 |
if (retval > 0) { |
1da177e4c
|
1384 |
*ppos = pos + retval; |
66f998f61
|
1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 |
count -= retval; } /* * Btrfs can have a short DIO read if we encounter * compressed extents, so if there was an error, or if * we've already read everything we wanted to, or if * there was a short read because we hit EOF, go ahead * and return. Otherwise fallthrough to buffered io for * the rest of the read. */ if (retval < 0 || !count || *ppos >= size) { |
11fa977ec
|
1397 1398 1399 |
file_accessed(filp); goto out; } |
0e0bcae3b
|
1400 |
} |
1da177e4c
|
1401 |
} |
66f998f61
|
1402 |
count = retval; |
11fa977ec
|
1403 1404 |
for (seg = 0; seg < nr_segs; seg++) { read_descriptor_t desc; |
66f998f61
|
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 |
loff_t offset = 0; /* * If we did a short DIO read we need to skip the section of the * iov that we've already read data into. */ if (count) { if (count > iov[seg].iov_len) { count -= iov[seg].iov_len; continue; } offset = count; count = 0; } |
1da177e4c
|
1419 |
|
11fa977ec
|
1420 |
desc.written = 0; |
66f998f61
|
1421 1422 |
desc.arg.buf = iov[seg].iov_base + offset; desc.count = iov[seg].iov_len - offset; |
11fa977ec
|
1423 1424 1425 1426 1427 1428 1429 1430 |
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
|
1431 |
} |
11fa977ec
|
1432 1433 |
if (desc.count > 0) break; |
1da177e4c
|
1434 1435 1436 1437 |
} out: return retval; } |
1da177e4c
|
1438 |
EXPORT_SYMBOL(generic_file_aio_read); |
1da177e4c
|
1439 |
#ifdef CONFIG_MMU |
485bb99b4
|
1440 1441 1442 1443 1444 |
/** * page_cache_read - adds requested page to the page cache if not already there * @file: file to read * @offset: page index * |
1da177e4c
|
1445 1446 1447 |
* 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
|
1448 |
static int page_cache_read(struct file *file, pgoff_t offset) |
1da177e4c
|
1449 1450 1451 |
{ struct address_space *mapping = file->f_mapping; struct page *page; |
994fc28c7
|
1452 |
int ret; |
1da177e4c
|
1453 |
|
994fc28c7
|
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 |
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
|
1464 |
|
1da177e4c
|
1465 |
page_cache_release(page); |
1da177e4c
|
1466 |
|
994fc28c7
|
1467 1468 1469 |
} while (ret == AOP_TRUNCATED_PAGE); return ret; |
1da177e4c
|
1470 1471 1472 |
} #define MMAP_LOTSAMISS (100) |
ef00e08e2
|
1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 |
/* * 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; |
275b12bf5
|
1488 1489 |
if (!ra->ra_pages) return; |
ef00e08e2
|
1490 |
|
2cbea1d3a
|
1491 |
if (VM_SequentialReadHint(vma)) { |
7ffc59b4d
|
1492 1493 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
ef00e08e2
|
1494 1495 |
return; } |
207d04baa
|
1496 1497 |
/* Avoid banging the cache line if not needed */ if (ra->mmap_miss < MMAP_LOTSAMISS * 10) |
ef00e08e2
|
1498 1499 1500 1501 1502 1503 1504 1505 |
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
|
1506 1507 1508 |
/* * mmap read-around */ |
ef00e08e2
|
1509 |
ra_pages = max_sane_readahead(ra->ra_pages); |
275b12bf5
|
1510 1511 |
ra->start = max_t(long, 0, offset - ra_pages / 2); ra->size = ra_pages; |
2cbea1d3a
|
1512 |
ra->async_size = ra_pages / 4; |
275b12bf5
|
1513 |
ra_submit(ra, mapping, file); |
ef00e08e2
|
1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 |
} /* * 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
|
1534 1535 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
ef00e08e2
|
1536 |
} |
485bb99b4
|
1537 |
/** |
54cb8821d
|
1538 |
* filemap_fault - read in file data for page fault handling |
d0217ac04
|
1539 1540 |
* @vma: vma in which the fault was taken * @vmf: struct vm_fault containing details of the fault |
485bb99b4
|
1541 |
* |
54cb8821d
|
1542 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c
|
1543 1544 1545 1546 1547 1548 |
* 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
|
1549 |
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
1da177e4c
|
1550 1551 |
{ int error; |
54cb8821d
|
1552 |
struct file *file = vma->vm_file; |
1da177e4c
|
1553 1554 1555 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2
|
1556 |
pgoff_t offset = vmf->pgoff; |
1da177e4c
|
1557 |
struct page *page; |
2004dc8ee
|
1558 |
pgoff_t size; |
83c54070e
|
1559 |
int ret = 0; |
1da177e4c
|
1560 |
|
1da177e4c
|
1561 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2
|
1562 |
if (offset >= size) |
5307cc1aa
|
1563 |
return VM_FAULT_SIGBUS; |
1da177e4c
|
1564 |
|
1da177e4c
|
1565 |
/* |
1da177e4c
|
1566 1567 |
* Do we have something in the page cache already? */ |
ef00e08e2
|
1568 1569 |
page = find_get_page(mapping, offset); if (likely(page)) { |
1da177e4c
|
1570 |
/* |
ef00e08e2
|
1571 1572 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c
|
1573 |
*/ |
ef00e08e2
|
1574 |
do_async_mmap_readahead(vma, ra, file, page, offset); |
ef00e08e2
|
1575 1576 1577 1578 |
} else { /* No page in the page cache at all */ do_sync_mmap_readahead(vma, ra, file, offset); count_vm_event(PGMAJFAULT); |
456f998ec
|
1579 |
mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT); |
ef00e08e2
|
1580 1581 |
ret = VM_FAULT_MAJOR; retry_find: |
b522c94da
|
1582 |
page = find_get_page(mapping, offset); |
1da177e4c
|
1583 1584 1585 |
if (!page) goto no_cached_page; } |
d88c0922f
|
1586 1587 |
if (!lock_page_or_retry(page, vma->vm_mm, vmf->flags)) { page_cache_release(page); |
d065bd810
|
1588 |
return ret | VM_FAULT_RETRY; |
d88c0922f
|
1589 |
} |
b522c94da
|
1590 1591 1592 1593 1594 1595 1596 1597 |
/* Did it get truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); put_page(page); goto retry_find; } VM_BUG_ON(page->index != offset); |
1da177e4c
|
1598 |
/* |
d00806b18
|
1599 1600 |
* 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
|
1601 |
*/ |
d00806b18
|
1602 |
if (unlikely(!PageUptodate(page))) |
1da177e4c
|
1603 |
goto page_not_uptodate; |
ef00e08e2
|
1604 1605 1606 1607 |
/* * Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
d00806b18
|
1608 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2
|
1609 |
if (unlikely(offset >= size)) { |
d00806b18
|
1610 |
unlock_page(page); |
745ad48e8
|
1611 |
page_cache_release(page); |
5307cc1aa
|
1612 |
return VM_FAULT_SIGBUS; |
d00806b18
|
1613 |
} |
d0217ac04
|
1614 |
vmf->page = page; |
83c54070e
|
1615 |
return ret | VM_FAULT_LOCKED; |
1da177e4c
|
1616 |
|
1da177e4c
|
1617 1618 1619 1620 1621 |
no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ |
ef00e08e2
|
1622 |
error = page_cache_read(file, offset); |
1da177e4c
|
1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 |
/* * 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
|
1638 1639 |
return VM_FAULT_OOM; return VM_FAULT_SIGBUS; |
1da177e4c
|
1640 1641 |
page_not_uptodate: |
1da177e4c
|
1642 1643 1644 1645 1646 1647 |
/* * 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
|
1648 |
ClearPageError(page); |
994fc28c7
|
1649 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e
|
1650 1651 1652 1653 1654 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
d00806b18
|
1655 1656 1657 |
page_cache_release(page); if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7
|
1658 |
goto retry_find; |
1da177e4c
|
1659 |
|
d00806b18
|
1660 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96
|
1661 |
shrink_readahead_size_eio(file, ra); |
d0217ac04
|
1662 |
return VM_FAULT_SIGBUS; |
54cb8821d
|
1663 1664 |
} EXPORT_SYMBOL(filemap_fault); |
4fcf1c620
|
1665 1666 1667 1668 1669 |
int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) { struct page *page = vmf->page; struct inode *inode = vma->vm_file->f_path.dentry->d_inode; int ret = VM_FAULT_LOCKED; |
14da92001
|
1670 |
sb_start_pagefault(inode->i_sb); |
4fcf1c620
|
1671 1672 1673 1674 1675 1676 1677 |
file_update_time(vma->vm_file); lock_page(page); if (page->mapping != inode->i_mapping) { unlock_page(page); ret = VM_FAULT_NOPAGE; goto out; } |
14da92001
|
1678 1679 1680 1681 1682 1683 |
/* * We mark the page dirty already here so that when freeze is in * progress, we are guaranteed that writeback during freezing will * see the dirty page and writeprotect it again. */ set_page_dirty(page); |
4fcf1c620
|
1684 |
out: |
14da92001
|
1685 |
sb_end_pagefault(inode->i_sb); |
4fcf1c620
|
1686 1687 1688 |
return ret; } EXPORT_SYMBOL(filemap_page_mkwrite); |
f0f37e2f7
|
1689 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d
|
1690 |
.fault = filemap_fault, |
4fcf1c620
|
1691 |
.page_mkwrite = filemap_page_mkwrite, |
1da177e4c
|
1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 |
}; /* 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
|
1704 |
vma->vm_flags |= VM_CAN_NONLINEAR; |
1da177e4c
|
1705 1706 |
return 0; } |
1da177e4c
|
1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 |
/* * 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
|
1730 |
static struct page *__read_cache_page(struct address_space *mapping, |
57f6b96c0
|
1731 |
pgoff_t index, |
5e5358e7c
|
1732 |
int (*filler)(void *, struct page *), |
0531b2aac
|
1733 1734 |
void *data, gfp_t gfp) |
1da177e4c
|
1735 |
{ |
eb2be1893
|
1736 |
struct page *page; |
1da177e4c
|
1737 1738 1739 1740 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
0531b2aac
|
1741 |
page = __page_cache_alloc(gfp | __GFP_COLD); |
eb2be1893
|
1742 1743 |
if (!page) return ERR_PTR(-ENOMEM); |
e6f67b8c0
|
1744 |
err = add_to_page_cache_lru(page, mapping, index, gfp); |
eb2be1893
|
1745 1746 1747 1748 |
if (unlikely(err)) { page_cache_release(page); if (err == -EEXIST) goto repeat; |
1da177e4c
|
1749 |
/* Presumably ENOMEM for radix tree node */ |
1da177e4c
|
1750 1751 |
return ERR_PTR(err); } |
1da177e4c
|
1752 1753 1754 1755 1756 1757 |
err = filler(data, page); if (err < 0) { page_cache_release(page); page = ERR_PTR(err); } } |
1da177e4c
|
1758 1759 |
return page; } |
0531b2aac
|
1760 |
static struct page *do_read_cache_page(struct address_space *mapping, |
57f6b96c0
|
1761 |
pgoff_t index, |
5e5358e7c
|
1762 |
int (*filler)(void *, struct page *), |
0531b2aac
|
1763 1764 |
void *data, gfp_t gfp) |
1da177e4c
|
1765 1766 1767 1768 1769 |
{ struct page *page; int err; retry: |
0531b2aac
|
1770 |
page = __read_cache_page(mapping, index, filler, data, gfp); |
1da177e4c
|
1771 |
if (IS_ERR(page)) |
c855ff371
|
1772 |
return page; |
1da177e4c
|
1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 |
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
|
1789 |
return ERR_PTR(err); |
1da177e4c
|
1790 |
} |
c855ff371
|
1791 |
out: |
6fe6900e1
|
1792 1793 1794 |
mark_page_accessed(page); return page; } |
0531b2aac
|
1795 1796 1797 1798 1799 1800 |
/** * 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 |
5e5358e7c
|
1801 |
* @data: first arg to filler(data, page) function, often left as NULL |
0531b2aac
|
1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 |
* * Same as read_cache_page, but don't wait for page to become unlocked * after submitting it to the filler. * * 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. */ struct page *read_cache_page_async(struct address_space *mapping, pgoff_t index, |
5e5358e7c
|
1813 |
int (*filler)(void *, struct page *), |
0531b2aac
|
1814 1815 1816 1817 |
void *data) { return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); } |
6fe6900e1
|
1818 |
EXPORT_SYMBOL(read_cache_page_async); |
0531b2aac
|
1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 |
static struct page *wait_on_page_read(struct page *page) { if (!IS_ERR(page)) { wait_on_page_locked(page); if (!PageUptodate(page)) { page_cache_release(page); page = ERR_PTR(-EIO); } } return page; } /** * read_cache_page_gfp - read into page cache, using specified page allocation flags. * @mapping: the page's address_space * @index: the page index * @gfp: the page allocator flags to use if allocating * * This is the same as "read_mapping_page(mapping, index, NULL)", but with |
e6f67b8c0
|
1838 |
* any new page allocations done using the specified allocation flags. |
0531b2aac
|
1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 |
* * If the page does not get brought uptodate, return -EIO. */ struct page *read_cache_page_gfp(struct address_space *mapping, pgoff_t index, gfp_t gfp) { filler_t *filler = (filler_t *)mapping->a_ops->readpage; return wait_on_page_read(do_read_cache_page(mapping, index, filler, NULL, gfp)); } EXPORT_SYMBOL(read_cache_page_gfp); |
6fe6900e1
|
1851 1852 1853 1854 1855 |
/** * 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 |
5e5358e7c
|
1856 |
* @data: first arg to filler(data, page) function, often left as NULL |
6fe6900e1
|
1857 1858 1859 1860 1861 1862 1863 |
* * 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
|
1864 |
pgoff_t index, |
5e5358e7c
|
1865 |
int (*filler)(void *, struct page *), |
6fe6900e1
|
1866 1867 |
void *data) { |
0531b2aac
|
1868 |
return wait_on_page_read(read_cache_page_async(mapping, index, filler, data)); |
1da177e4c
|
1869 |
} |
1da177e4c
|
1870 |
EXPORT_SYMBOL(read_cache_page); |
2f718ffc1
|
1871 |
static size_t __iovec_copy_from_user_inatomic(char *vaddr, |
1da177e4c
|
1872 1873 |
const struct iovec *iov, size_t base, size_t bytes) { |
f18005369
|
1874 |
size_t copied = 0, left = 0; |
1da177e4c
|
1875 1876 1877 1878 1879 1880 |
while (bytes) { char __user *buf = iov->iov_base + base; int copy = min(bytes, iov->iov_len - base); base = 0; |
f18005369
|
1881 |
left = __copy_from_user_inatomic(vaddr, buf, copy); |
1da177e4c
|
1882 1883 1884 1885 |
copied += copy; bytes -= copy; vaddr += copy; iov++; |
01408c493
|
1886 |
if (unlikely(left)) |
1da177e4c
|
1887 |
break; |
1da177e4c
|
1888 1889 1890 1891 1892 |
} return copied - left; } /* |
2f718ffc1
|
1893 |
* Copy as much as we can into the page and return the number of bytes which |
af901ca18
|
1894 |
* were successfully copied. If a fault is encountered then return the number of |
2f718ffc1
|
1895 1896 1897 1898 1899 1900 1901 1902 1903 |
* 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()); |
9b04c5fec
|
1904 |
kaddr = kmap_atomic(page); |
2f718ffc1
|
1905 1906 1907 |
if (likely(i->nr_segs == 1)) { int left; char __user *buf = i->iov->iov_base + i->iov_offset; |
f18005369
|
1908 |
left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); |
2f718ffc1
|
1909 1910 1911 1912 1913 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } |
9b04c5fec
|
1914 |
kunmap_atomic(kaddr); |
2f718ffc1
|
1915 1916 1917 |
return copied; } |
89e107877
|
1918 |
EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); |
2f718ffc1
|
1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 |
/* * 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
|
1936 |
left = __copy_from_user(kaddr + offset, buf, bytes); |
2f718ffc1
|
1937 1938 1939 1940 1941 1942 1943 1944 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } kunmap(page); return copied; } |
89e107877
|
1945 |
EXPORT_SYMBOL(iov_iter_copy_from_user); |
2f718ffc1
|
1946 |
|
f7009264c
|
1947 |
void iov_iter_advance(struct iov_iter *i, size_t bytes) |
2f718ffc1
|
1948 |
{ |
f7009264c
|
1949 |
BUG_ON(i->count < bytes); |
2f718ffc1
|
1950 1951 |
if (likely(i->nr_segs == 1)) { i->iov_offset += bytes; |
f7009264c
|
1952 |
i->count -= bytes; |
2f718ffc1
|
1953 1954 1955 |
} else { const struct iovec *iov = i->iov; size_t base = i->iov_offset; |
39be79c16
|
1956 |
unsigned long nr_segs = i->nr_segs; |
2f718ffc1
|
1957 |
|
124d3b704
|
1958 1959 |
/* * The !iov->iov_len check ensures we skip over unlikely |
f7009264c
|
1960 |
* zero-length segments (without overruning the iovec). |
124d3b704
|
1961 |
*/ |
94ad374a0
|
1962 |
while (bytes || unlikely(i->count && !iov->iov_len)) { |
f7009264c
|
1963 |
int copy; |
2f718ffc1
|
1964 |
|
f7009264c
|
1965 1966 1967 |
copy = min(bytes, iov->iov_len - base); BUG_ON(!i->count || i->count < copy); i->count -= copy; |
2f718ffc1
|
1968 1969 1970 1971 |
bytes -= copy; base += copy; if (iov->iov_len == base) { iov++; |
39be79c16
|
1972 |
nr_segs--; |
2f718ffc1
|
1973 1974 1975 1976 1977 |
base = 0; } } i->iov = iov; i->iov_offset = base; |
39be79c16
|
1978 |
i->nr_segs = nr_segs; |
2f718ffc1
|
1979 1980 |
} } |
89e107877
|
1981 |
EXPORT_SYMBOL(iov_iter_advance); |
2f718ffc1
|
1982 |
|
afddba49d
|
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 |
/* * 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
|
1993 |
{ |
2f718ffc1
|
1994 |
char __user *buf = i->iov->iov_base + i->iov_offset; |
afddba49d
|
1995 1996 |
bytes = min(bytes, i->iov->iov_len - i->iov_offset); return fault_in_pages_readable(buf, bytes); |
2f718ffc1
|
1997 |
} |
89e107877
|
1998 |
EXPORT_SYMBOL(iov_iter_fault_in_readable); |
2f718ffc1
|
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 |
/* * 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
|
2011 |
EXPORT_SYMBOL(iov_iter_single_seg_count); |
2f718ffc1
|
2012 2013 |
/* |
1da177e4c
|
2014 2015 |
* Performs necessary checks before doing a write * |
485bb99b4
|
2016 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c
|
2017 2018 2019 2020 2021 2022 |
* 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; |
59e99e5b9
|
2023 |
unsigned long limit = rlimit(RLIMIT_FSIZE); |
1da177e4c
|
2024 2025 2026 |
if (unlikely(*pos < 0)) return -EINVAL; |
1da177e4c
|
2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 |
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
|
2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 |
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
|
2066 2067 2068 2069 2070 2071 2072 2073 |
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
|
2074 |
#ifdef CONFIG_BLOCK |
1da177e4c
|
2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 |
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
|
2086 2087 2088 |
#else return -EPERM; #endif |
1da177e4c
|
2089 2090 2091 2092 |
} return 0; } EXPORT_SYMBOL(generic_write_checks); |
afddba49d
|
2093 2094 2095 2096 2097 |
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
|
2098 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d
|
2099 |
pagep, fsdata); |
afddba49d
|
2100 2101 2102 2103 2104 2105 2106 2107 |
} 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
|
2108 |
|
4e02ed4b4
|
2109 2110 |
mark_page_accessed(page); return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d
|
2111 2112 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c
|
2113 2114 2115 2116 2117 2118 2119 2120 2121 |
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
|
2122 2123 |
size_t write_len; pgoff_t end; |
1da177e4c
|
2124 2125 2126 |
if (count != ocount) *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); |
a969e903a
|
2127 2128 |
write_len = iov_length(iov, *nr_segs); end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; |
a969e903a
|
2129 |
|
48b47c561
|
2130 |
written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); |
a969e903a
|
2131 2132 2133 2134 2135 2136 2137 |
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
|
2138 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a
|
2139 2140 2141 2142 |
*/ if (mapping->nrpages) { written = invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); |
6ccfa806a
|
2143 2144 2145 2146 2147 2148 2149 |
/* * If a page can not be invalidated, return 0 to fall back * to buffered write. */ if (written) { if (written == -EBUSY) return 0; |
a969e903a
|
2150 |
goto out; |
6ccfa806a
|
2151 |
} |
a969e903a
|
2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 |
} 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
|
2168 |
if (written > 0) { |
0116651c8
|
2169 2170 2171 |
pos += written; if (pos > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, pos); |
1da177e4c
|
2172 2173 |
mark_inode_dirty(inode); } |
0116651c8
|
2174 |
*ppos = pos; |
1da177e4c
|
2175 |
} |
a969e903a
|
2176 |
out: |
1da177e4c
|
2177 2178 2179 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893
|
2180 2181 2182 2183 |
/* * Find or create a page at the given pagecache position. Return the locked * page. This function is specifically for buffered writes. */ |
54566b2c1
|
2184 2185 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893
|
2186 2187 |
{ int status; |
0faa70cb0
|
2188 |
gfp_t gfp_mask; |
eb2be1893
|
2189 |
struct page *page; |
54566b2c1
|
2190 |
gfp_t gfp_notmask = 0; |
0faa70cb0
|
2191 |
|
1010bb1b8
|
2192 2193 2194 |
gfp_mask = mapping_gfp_mask(mapping); if (mapping_cap_account_dirty(mapping)) gfp_mask |= __GFP_WRITE; |
54566b2c1
|
2195 2196 |
if (flags & AOP_FLAG_NOFS) gfp_notmask = __GFP_FS; |
eb2be1893
|
2197 2198 |
repeat: page = find_lock_page(mapping, index); |
c585a2678
|
2199 |
if (page) |
3d08bcc88
|
2200 |
goto found; |
eb2be1893
|
2201 |
|
0faa70cb0
|
2202 |
page = __page_cache_alloc(gfp_mask & ~gfp_notmask); |
eb2be1893
|
2203 2204 |
if (!page) return NULL; |
54566b2c1
|
2205 2206 |
status = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL & ~gfp_notmask); |
eb2be1893
|
2207 2208 2209 2210 2211 2212 |
if (unlikely(status)) { page_cache_release(page); if (status == -EEXIST) goto repeat; return NULL; } |
3d08bcc88
|
2213 2214 |
found: wait_on_page_writeback(page); |
eb2be1893
|
2215 2216 |
return page; } |
54566b2c1
|
2217 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893
|
2218 |
|
afddba49d
|
2219 2220 2221 2222 2223 2224 2225 |
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
|
2226 2227 2228 2229 2230 2231 2232 |
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
|
2233 2234 2235 |
do { struct page *page; |
afddba49d
|
2236 2237 2238 2239 2240 2241 |
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)); |
afddba49d
|
2242 2243 2244 2245 |
bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_count(i)); again: |
afddba49d
|
2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 |
/* * 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
|
2260 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d
|
2261 2262 2263 |
&page, &fsdata); if (unlikely(status)) break; |
931e80e4b
|
2264 2265 |
if (mapping_writably_mapped(mapping)) flush_dcache_page(page); |
afddba49d
|
2266 2267 2268 2269 |
pagefault_disable(); copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); pagefault_enable(); flush_dcache_page(page); |
c8236db9c
|
2270 |
mark_page_accessed(page); |
afddba49d
|
2271 2272 2273 2274 2275 2276 2277 |
status = a_ops->write_end(file, mapping, pos, bytes, copied, page, fsdata); if (unlikely(status < 0)) break; copied = status; cond_resched(); |
124d3b704
|
2278 |
iov_iter_advance(i, copied); |
afddba49d
|
2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 |
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
|
2292 2293 2294 2295 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); |
a50527b19
|
2296 2297 2298 2299 |
if (fatal_signal_pending(current)) { status = -EINTR; break; } |
afddba49d
|
2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 |
} 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; |
afddba49d
|
2311 2312 2313 2314 |
ssize_t status; struct iov_iter i; iov_iter_init(&i, iov, nr_segs, count, written); |
4e02ed4b4
|
2315 |
status = generic_perform_write(file, &i, pos); |
1da177e4c
|
2316 |
|
1da177e4c
|
2317 |
if (likely(status >= 0)) { |
afddba49d
|
2318 2319 |
written += status; *ppos = pos + status; |
1da177e4c
|
2320 2321 |
} |
1da177e4c
|
2322 2323 2324 |
return written ? written : status; } EXPORT_SYMBOL(generic_file_buffered_write); |
e4dd9de3c
|
2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 |
/** * __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
|
2346 2347 |
{ struct file *file = iocb->ki_filp; |
fb5527e68
|
2348 |
struct address_space * mapping = file->f_mapping; |
1da177e4c
|
2349 2350 2351 |
size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; |
1da177e4c
|
2352 2353 2354 2355 2356 |
loff_t pos; ssize_t written; ssize_t err; ocount = 0; |
0ceb33143
|
2357 2358 2359 |
err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (err) return err; |
1da177e4c
|
2360 2361 2362 |
count = ocount; pos = *ppos; |
1da177e4c
|
2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 |
/* 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
|
2373 |
err = file_remove_suid(file); |
1da177e4c
|
2374 2375 |
if (err) goto out; |
c3b2da314
|
2376 2377 2378 |
err = file_update_time(file); if (err) goto out; |
1da177e4c
|
2379 2380 2381 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { |
fb5527e68
|
2382 2383 2384 2385 2386 |
loff_t endbyte; ssize_t written_buffered; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, ppos, count, ocount); |
1da177e4c
|
2387 2388 2389 2390 2391 2392 2393 2394 |
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
|
2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 |
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
|
2409 |
|
fb5527e68
|
2410 2411 2412 2413 2414 2415 |
/* * 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; |
c05c4edd8
|
2416 |
err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); |
fb5527e68
|
2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 |
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
|
2432 2433 2434 2435 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
e4dd9de3c
|
2436 |
EXPORT_SYMBOL(__generic_file_aio_write); |
e4dd9de3c
|
2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 |
/** * 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
|
2448 2449 |
ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c
|
2450 2451 |
{ struct file *file = iocb->ki_filp; |
148f948ba
|
2452 |
struct inode *inode = file->f_mapping->host; |
55602dd66
|
2453 |
struct blk_plug plug; |
1da177e4c
|
2454 |
ssize_t ret; |
1da177e4c
|
2455 2456 |
BUG_ON(iocb->ki_pos != pos); |
14da92001
|
2457 |
sb_start_write(inode->i_sb); |
1b1dcc1b5
|
2458 |
mutex_lock(&inode->i_mutex); |
55602dd66
|
2459 |
blk_start_plug(&plug); |
e4dd9de3c
|
2460 |
ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); |
1b1dcc1b5
|
2461 |
mutex_unlock(&inode->i_mutex); |
1da177e4c
|
2462 |
|
148f948ba
|
2463 |
if (ret > 0 || ret == -EIOCBQUEUED) { |
1da177e4c
|
2464 |
ssize_t err; |
148f948ba
|
2465 |
err = generic_write_sync(file, pos, ret); |
c7b50db21
|
2466 |
if (err < 0 && ret > 0) |
1da177e4c
|
2467 2468 |
ret = err; } |
55602dd66
|
2469 |
blk_finish_plug(&plug); |
14da92001
|
2470 |
sb_end_write(inode->i_sb); |
1da177e4c
|
2471 2472 2473 |
return ret; } EXPORT_SYMBOL(generic_file_aio_write); |
cf9a2ae8d
|
2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 |
/** * 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
|
2484 2485 2486 |
* 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
|
2487 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
3f31fddfa
|
2488 |
* this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS). |
cf9a2ae8d
|
2489 |
* |
cf9a2ae8d
|
2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 |
*/ 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); |