Blame view
mm/filemap.c
71.9 KB
1da177e4c Linux-2.6.12-rc2 |
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 mm: Map most file... |
12 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
13 14 |
#include <linux/compiler.h> #include <linux/fs.h> |
c22ce143d [PATCH] x86: cach... |
15 |
#include <linux/uaccess.h> |
1da177e4c Linux-2.6.12-rc2 |
16 |
#include <linux/aio.h> |
c59ede7b7 [PATCH] move capa... |
17 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
18 |
#include <linux/kernel_stat.h> |
5a0e3ad6a include cleanup: ... |
19 |
#include <linux/gfp.h> |
1da177e4c Linux-2.6.12-rc2 |
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 Include <linux/ba... |
28 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
29 30 31 |
#include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/security.h> |
44110fe38 [PATCH] cpuset me... |
32 |
#include <linux/cpuset.h> |
2f718ffc1 mm: buffered writ... |
33 |
#include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */ |
8a9f3ccd2 Memory controller... |
34 |
#include <linux/memcontrol.h> |
c515e1fd3 mm/fs: add hooks ... |
35 |
#include <linux/cleancache.h> |
0f8053a50 [PATCH] mm: make ... |
36 |
#include "internal.h" |
fe0bfaaff mm: trace filemap... |
37 38 |
#define CREATE_TRACE_POINTS #include <trace/events/filemap.h> |
1da177e4c Linux-2.6.12-rc2 |
39 |
/* |
1da177e4c Linux-2.6.12-rc2 |
40 41 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba vfs: Introduce ne... |
42 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c Linux-2.6.12-rc2 |
43 |
|
1da177e4c Linux-2.6.12-rc2 |
44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 |
#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 mm: Convert i_mma... |
61 |
* ->i_mmap_mutex (truncate_pagecache) |
1da177e4c Linux-2.6.12-rc2 |
62 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919 [PATCH] swap: swa... |
63 64 |
* ->swap_lock (exclusive_swap_page, others) * ->mapping->tree_lock |
1da177e4c Linux-2.6.12-rc2 |
65 |
* |
1b1dcc1b5 [PATCH] mutex sub... |
66 |
* ->i_mutex |
3d48ae45e mm: Convert i_mma... |
67 |
* ->i_mmap_mutex (truncate->unmap_mapping_range) |
1da177e4c Linux-2.6.12-rc2 |
68 69 |
* * ->mmap_sem |
3d48ae45e mm: Convert i_mma... |
70 |
* ->i_mmap_mutex |
b8072f099 [PATCH] mm: updat... |
71 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
1da177e4c Linux-2.6.12-rc2 |
72 73 74 75 76 |
* ->mapping->tree_lock (arch-dependent flush_dcache_mmap_lock) * * ->mmap_sem * ->lock_page (access_process_vm) * |
82591e6ea [PATCH] mm: more ... |
77 78 |
* ->i_mutex (generic_file_buffered_write) * ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c Linux-2.6.12-rc2 |
79 |
* |
f758eeabe writeback: split ... |
80 |
* bdi->wb.list_lock |
a66979aba fs: move i_wb_lis... |
81 |
* sb_lock (fs/fs-writeback.c) |
1da177e4c Linux-2.6.12-rc2 |
82 83 |
* ->mapping->tree_lock (__sync_single_inode) * |
3d48ae45e mm: Convert i_mma... |
84 |
* ->i_mmap_mutex |
1da177e4c Linux-2.6.12-rc2 |
85 86 87 |
* ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099 [PATCH] mm: updat... |
88 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c Linux-2.6.12-rc2 |
89 |
* |
b8072f099 [PATCH] mm: updat... |
90 |
* ->page_table_lock or pte_lock |
5d337b919 [PATCH] swap: swa... |
91 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c Linux-2.6.12-rc2 |
92 93 94 |
* ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) |
053837fce [PATCH] mm: migra... |
95 |
* ->zone.lru_lock (check_pte_range->isolate_lru_page) |
1da177e4c Linux-2.6.12-rc2 |
96 97 |
* ->private_lock (page_remove_rmap->set_page_dirty) * ->tree_lock (page_remove_rmap->set_page_dirty) |
f758eeabe writeback: split ... |
98 |
* bdi.wb->list_lock (page_remove_rmap->set_page_dirty) |
250df6ed2 fs: protect inode... |
99 |
* ->inode->i_lock (page_remove_rmap->set_page_dirty) |
f758eeabe writeback: split ... |
100 |
* bdi.wb->list_lock (zap_pte_range->set_page_dirty) |
250df6ed2 fs: protect inode... |
101 |
* ->inode->i_lock (zap_pte_range->set_page_dirty) |
1da177e4c Linux-2.6.12-rc2 |
102 103 |
* ->private_lock (zap_pte_range->__set_page_dirty_buffers) * |
9a3c531df mm: update stale ... |
104 105 |
* ->i_mmap_mutex * ->tasklist_lock (memory_failure, collect_procs_ao) |
1da177e4c Linux-2.6.12-rc2 |
106 107 108 |
*/ /* |
e64a782fe mm: change __remo... |
109 |
* Delete a page from the page cache and free it. Caller has to make |
1da177e4c Linux-2.6.12-rc2 |
110 |
* sure the page is locked and that nobody else uses it - or that usage |
19fd62312 mm: spinlock tree... |
111 |
* is safe. The caller must hold the mapping's tree_lock. |
1da177e4c Linux-2.6.12-rc2 |
112 |
*/ |
e64a782fe mm: change __remo... |
113 |
void __delete_from_page_cache(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
114 115 |
{ struct address_space *mapping = page->mapping; |
fe0bfaaff mm: trace filemap... |
116 |
trace_mm_filemap_delete_from_page_cache(page); |
c515e1fd3 mm/fs: add hooks ... |
117 118 119 120 121 122 123 124 |
/* * 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 mm: cleancache: s... |
125 |
cleancache_invalidate_page(mapping, page); |
c515e1fd3 mm/fs: add hooks ... |
126 |
|
1da177e4c Linux-2.6.12-rc2 |
127 128 |
radix_tree_delete(&mapping->page_tree, page->index); page->mapping = NULL; |
b85e0effd mm: consistent tr... |
129 |
/* Leave page->index set: truncation lookup relies upon it */ |
1da177e4c Linux-2.6.12-rc2 |
130 |
mapping->nrpages--; |
347ce434d [PATCH] zoned vm ... |
131 |
__dec_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac mm: oom analysis:... |
132 133 |
if (PageSwapBacked(page)) __dec_zone_page_state(page, NR_SHMEM); |
45426812d mm: debug check f... |
134 |
BUG_ON(page_mapped(page)); |
3a6927906 Do dirty page acc... |
135 136 137 138 139 140 141 142 143 144 145 146 |
/* * 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 Linux-2.6.12-rc2 |
147 |
} |
702cfbf93 mm: goodbye remov... |
148 149 150 151 152 153 154 155 156 |
/** * 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 Linux-2.6.12-rc2 |
157 158 |
{ struct address_space *mapping = page->mapping; |
6072d13c4 Call the filesyst... |
159 |
void (*freepage)(struct page *); |
1da177e4c Linux-2.6.12-rc2 |
160 |
|
cd7619d6b [PATCH] Extermina... |
161 |
BUG_ON(!PageLocked(page)); |
1da177e4c Linux-2.6.12-rc2 |
162 |
|
6072d13c4 Call the filesyst... |
163 |
freepage = mapping->a_ops->freepage; |
19fd62312 mm: spinlock tree... |
164 |
spin_lock_irq(&mapping->tree_lock); |
e64a782fe mm: change __remo... |
165 |
__delete_from_page_cache(page); |
19fd62312 mm: spinlock tree... |
166 |
spin_unlock_irq(&mapping->tree_lock); |
e767e0561 memcg: fix deadlo... |
167 |
mem_cgroup_uncharge_cache_page(page); |
6072d13c4 Call the filesyst... |
168 169 170 |
if (freepage) freepage(page); |
97cecb5a2 mm: introduce del... |
171 172 173 |
page_cache_release(page); } EXPORT_SYMBOL(delete_from_page_cache); |
7eaceacca block: remove per... |
174 |
static int sleep_on_page(void *word) |
1da177e4c Linux-2.6.12-rc2 |
175 |
{ |
1da177e4c Linux-2.6.12-rc2 |
176 177 178 |
io_schedule(); return 0; } |
7eaceacca block: remove per... |
179 |
static int sleep_on_page_killable(void *word) |
2687a3569 Add lock_page_kil... |
180 |
{ |
7eaceacca block: remove per... |
181 |
sleep_on_page(word); |
2687a3569 Add lock_page_kil... |
182 183 |
return fatal_signal_pending(current) ? -EINTR : 0; } |
865ffef37 fs: fix fsync() e... |
184 185 186 187 |
static int filemap_check_errors(struct address_space *mapping) { int ret = 0; /* Check for outstanding write errors */ |
5a6ad555b mm/filemap.c: avo... |
188 189 |
if (test_bit(AS_ENOSPC, &mapping->flags) && test_and_clear_bit(AS_ENOSPC, &mapping->flags)) |
865ffef37 fs: fix fsync() e... |
190 |
ret = -ENOSPC; |
5a6ad555b mm/filemap.c: avo... |
191 192 |
if (test_bit(AS_EIO, &mapping->flags) && test_and_clear_bit(AS_EIO, &mapping->flags)) |
865ffef37 fs: fix fsync() e... |
193 194 195 |
ret = -EIO; return ret; } |
1da177e4c Linux-2.6.12-rc2 |
196 |
/** |
485bb99b4 [PATCH] kernel-do... |
197 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b [PATCH] DocBook: ... |
198 199 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03 [PATCH] filemap_f... |
200 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b [PATCH] DocBook: ... |
201 |
* @sync_mode: enable synchronous operation |
1da177e4c Linux-2.6.12-rc2 |
202 |
* |
485bb99b4 [PATCH] kernel-do... |
203 204 205 |
* Start writeback against all of a mapping's dirty pages that lie * within the byte offsets <start, end> inclusive. * |
1da177e4c Linux-2.6.12-rc2 |
206 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4 [PATCH] kernel-do... |
207 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c Linux-2.6.12-rc2 |
208 209 210 |
* these two operations is that if a dirty page/buffer is encountered, it must * be waited upon, and not just skipped over. */ |
ebcf28e1c [PATCH] fadvise()... |
211 212 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c Linux-2.6.12-rc2 |
213 214 215 216 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd mm: write_cache_p... |
217 |
.nr_to_write = LONG_MAX, |
111ebb6e6 [PATCH] writeback... |
218 219 |
.range_start = start, .range_end = end, |
1da177e4c Linux-2.6.12-rc2 |
220 221 222 223 224 225 226 227 228 229 230 231 |
}; 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 [PATCH] writeback... |
232 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c Linux-2.6.12-rc2 |
233 234 235 236 237 238 239 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf vfs: export filem... |
240 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c [PATCH] fadvise()... |
241 |
loff_t end) |
1da177e4c Linux-2.6.12-rc2 |
242 243 244 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf vfs: export filem... |
245 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c Linux-2.6.12-rc2 |
246 |
|
485bb99b4 [PATCH] kernel-do... |
247 248 249 250 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c Linux-2.6.12-rc2 |
251 252 253 254 255 256 257 258 |
* 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 [PATCH] kernel-do... |
259 |
/** |
94004ed72 kill wait_on_page... |
260 261 262 263 |
* 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 [PATCH] kernel-do... |
264 |
* |
94004ed72 kill wait_on_page... |
265 266 |
* Walk the list of under-writeback pages of the given address space * in the given range and wait for all of them. |
1da177e4c Linux-2.6.12-rc2 |
267 |
*/ |
94004ed72 kill wait_on_page... |
268 269 |
int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) |
1da177e4c Linux-2.6.12-rc2 |
270 |
{ |
94004ed72 kill wait_on_page... |
271 272 |
pgoff_t index = start_byte >> PAGE_CACHE_SHIFT; pgoff_t end = end_byte >> PAGE_CACHE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
273 274 |
struct pagevec pvec; int nr_pages; |
865ffef37 fs: fix fsync() e... |
275 |
int ret2, ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
276 |
|
94004ed72 kill wait_on_page... |
277 |
if (end_byte < start_byte) |
865ffef37 fs: fix fsync() e... |
278 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
279 280 |
pagevec_init(&pvec, 0); |
1da177e4c Linux-2.6.12-rc2 |
281 282 283 284 285 286 287 288 289 290 291 292 293 294 |
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 mm: clear PageErr... |
295 |
if (TestClearPageError(page)) |
1da177e4c Linux-2.6.12-rc2 |
296 297 298 299 300 |
ret = -EIO; } pagevec_release(&pvec); cond_resched(); } |
865ffef37 fs: fix fsync() e... |
301 302 303 304 |
out: ret2 = filemap_check_errors(mapping); if (!ret) ret = ret2; |
1da177e4c Linux-2.6.12-rc2 |
305 306 307 |
return ret; } |
d3bccb6f4 vfs: Introduce fi... |
308 309 310 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
485bb99b4 [PATCH] kernel-do... |
311 |
* filemap_fdatawait - wait for all under-writeback pages to complete |
1da177e4c Linux-2.6.12-rc2 |
312 |
* @mapping: address space structure to wait for |
485bb99b4 [PATCH] kernel-do... |
313 314 315 |
* * Walk the list of under-writeback pages of the given address space * and wait for all of them. |
1da177e4c Linux-2.6.12-rc2 |
316 317 318 319 320 321 322 |
*/ int filemap_fdatawait(struct address_space *mapping) { loff_t i_size = i_size_read(mapping->host); if (i_size == 0) return 0; |
94004ed72 kill wait_on_page... |
323 |
return filemap_fdatawait_range(mapping, 0, i_size - 1); |
1da177e4c Linux-2.6.12-rc2 |
324 325 326 327 328 |
} EXPORT_SYMBOL(filemap_fdatawait); int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982 [PATCH] Fix and a... |
329 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
330 331 |
if (mapping->nrpages) { |
28fd12982 [PATCH] Fix and a... |
332 333 334 335 336 337 338 339 340 341 342 343 |
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; } |
865ffef37 fs: fix fsync() e... |
344 345 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
346 |
} |
28fd12982 [PATCH] Fix and a... |
347 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
348 |
} |
28fd12982 [PATCH] Fix and a... |
349 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c Linux-2.6.12-rc2 |
350 |
|
485bb99b4 [PATCH] kernel-do... |
351 352 353 354 355 356 |
/** * 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 [PATCH] filemap_f... |
357 358 359 360 361 |
* 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 Linux-2.6.12-rc2 |
362 363 364 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982 [PATCH] Fix and a... |
365 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
366 367 |
if (mapping->nrpages) { |
28fd12982 [PATCH] Fix and a... |
368 369 370 371 |
err = __filemap_fdatawrite_range(mapping, lstart, lend, WB_SYNC_ALL); /* See comment of filemap_write_and_wait() */ if (err != -EIO) { |
94004ed72 kill wait_on_page... |
372 373 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982 [PATCH] Fix and a... |
374 375 376 |
if (!err) err = err2; } |
865ffef37 fs: fix fsync() e... |
377 378 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
379 |
} |
28fd12982 [PATCH] Fix and a... |
380 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
381 |
} |
f69955855 Export filemap_wr... |
382 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c Linux-2.6.12-rc2 |
383 |
|
485bb99b4 [PATCH] kernel-do... |
384 |
/** |
ef6a3c631 mm: add replace_p... |
385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 |
* 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 mm: add replace_p... |
402 403 404 405 |
VM_BUG_ON(!PageLocked(old)); VM_BUG_ON(!PageLocked(new)); VM_BUG_ON(new->mapping); |
ef6a3c631 mm: add replace_p... |
406 407 408 409 410 411 412 413 414 415 416 417 418 |
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 mm: change __remo... |
419 |
__delete_from_page_cache(old); |
ef6a3c631 mm: add replace_p... |
420 421 422 423 424 425 426 |
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 memcg: add mem_cg... |
427 428 |
/* mem_cgroup codes must not be called under tree_lock */ mem_cgroup_replace_page_cache(old, new); |
ef6a3c631 mm: add replace_p... |
429 430 431 432 |
radix_tree_preload_end(); if (freepage) freepage(old); page_cache_release(old); |
ef6a3c631 mm: add replace_p... |
433 434 435 436 437 |
} return error; } EXPORT_SYMBOL_GPL(replace_page_cache_page); |
e714f0cf0 mm + fs: prepare ... |
438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 |
static int page_cache_tree_insert(struct address_space *mapping, struct page *page) { void **slot; int error; slot = radix_tree_lookup_slot(&mapping->page_tree, page->index); if (slot) { void *p; p = radix_tree_deref_slot_protected(slot, &mapping->tree_lock); if (!radix_tree_exceptional_entry(p)) return -EEXIST; radix_tree_replace_slot(slot, page); mapping->nrpages++; return 0; } error = radix_tree_insert(&mapping->page_tree, page->index, page); if (!error) mapping->nrpages++; return error; } |
ef6a3c631 mm: add replace_p... |
460 |
/** |
e286781d5 mm: speculative p... |
461 |
* add_to_page_cache_locked - add a locked page to the pagecache |
485bb99b4 [PATCH] kernel-do... |
462 463 464 465 466 |
* @page: page to add * @mapping: the page's address_space * @offset: page index * @gfp_mask: page allocation mode * |
e286781d5 mm: speculative p... |
467 |
* This function is used to add a page to the pagecache. It must be locked. |
1da177e4c Linux-2.6.12-rc2 |
468 469 |
* This function does not add the page to the LRU. The caller must do that. */ |
e286781d5 mm: speculative p... |
470 |
int add_to_page_cache_locked(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
471 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
472 |
{ |
e286781d5 mm: speculative p... |
473 474 475 |
int error; VM_BUG_ON(!PageLocked(page)); |
31475dd61 mm: a few small u... |
476 |
VM_BUG_ON(PageSwapBacked(page)); |
e286781d5 mm: speculative p... |
477 478 |
error = mem_cgroup_cache_charge(page, current->mm, |
2c26fdd70 memcg: revert gfp... |
479 |
gfp_mask & GFP_RECLAIM_MASK); |
35c754d79 memory controller... |
480 |
if (error) |
66a0c8ee3 mm: cleanup add_t... |
481 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
482 |
|
5e4c0d974 lib/radix-tree.c:... |
483 |
error = radix_tree_maybe_preload(gfp_mask & ~__GFP_HIGHMEM); |
66a0c8ee3 mm: cleanup add_t... |
484 |
if (error) { |
69029cd55 memcg: remove ref... |
485 |
mem_cgroup_uncharge_cache_page(page); |
66a0c8ee3 mm: cleanup add_t... |
486 487 488 489 490 491 492 493 |
return error; } page_cache_get(page); page->mapping = mapping; page->index = offset; spin_lock_irq(&mapping->tree_lock); |
e714f0cf0 mm + fs: prepare ... |
494 |
error = page_cache_tree_insert(mapping, page); |
66a0c8ee3 mm: cleanup add_t... |
495 496 497 |
radix_tree_preload_end(); if (unlikely(error)) goto err_insert; |
66a0c8ee3 mm: cleanup add_t... |
498 499 500 501 502 503 504 505 506 507 |
__inc_zone_page_state(page, NR_FILE_PAGES); spin_unlock_irq(&mapping->tree_lock); trace_mm_filemap_add_to_page_cache(page); return 0; err_insert: page->mapping = NULL; /* Leave page->index set: truncation relies upon it */ spin_unlock_irq(&mapping->tree_lock); mem_cgroup_uncharge_cache_page(page); page_cache_release(page); |
1da177e4c Linux-2.6.12-rc2 |
508 509 |
return error; } |
e286781d5 mm: speculative p... |
510 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c Linux-2.6.12-rc2 |
511 512 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
513 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
514 |
{ |
4f98a2fee vmscan: split LRU... |
515 |
int ret; |
4f98a2fee vmscan: split LRU... |
516 |
ret = add_to_page_cache(page, mapping, offset, gfp_mask); |
31475dd61 mm: a few small u... |
517 518 |
if (ret == 0) lru_cache_add_file(page); |
1da177e4c Linux-2.6.12-rc2 |
519 520 |
return ret; } |
18bc0bbd1 Staging: pohmelfs... |
521 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c Linux-2.6.12-rc2 |
522 |
|
44110fe38 [PATCH] cpuset me... |
523 |
#ifdef CONFIG_NUMA |
2ae88149a [PATCH] mm: clean... |
524 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38 [PATCH] cpuset me... |
525 |
{ |
c0ff7453b cpuset,mm: fix no... |
526 527 |
int n; struct page *page; |
44110fe38 [PATCH] cpuset me... |
528 |
if (cpuset_do_page_mem_spread()) { |
cc9a6c877 cpuset: mm: reduc... |
529 530 |
unsigned int cpuset_mems_cookie; do { |
337c9823c mm: optimize put_... |
531 |
cpuset_mems_cookie = read_mems_allowed_begin(); |
cc9a6c877 cpuset: mm: reduc... |
532 533 |
n = cpuset_mem_spread_node(); page = alloc_pages_exact_node(n, gfp, 0); |
337c9823c mm: optimize put_... |
534 |
} while (!page && read_mems_allowed_retry(cpuset_mems_cookie)); |
cc9a6c877 cpuset: mm: reduc... |
535 |
|
c0ff7453b cpuset,mm: fix no... |
536 |
return page; |
44110fe38 [PATCH] cpuset me... |
537 |
} |
2ae88149a [PATCH] mm: clean... |
538 |
return alloc_pages(gfp, 0); |
44110fe38 [PATCH] cpuset me... |
539 |
} |
2ae88149a [PATCH] mm: clean... |
540 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38 [PATCH] cpuset me... |
541 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 |
/* * 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 mm: remove fastca... |
563 |
void wait_on_page_bit(struct page *page, int bit_nr) |
1da177e4c Linux-2.6.12-rc2 |
564 565 566 567 |
{ DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); if (test_bit(bit_nr, &page->flags)) |
7eaceacca block: remove per... |
568 |
__wait_on_bit(page_waitqueue(page), &wait, sleep_on_page, |
1da177e4c Linux-2.6.12-rc2 |
569 570 571 |
TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(wait_on_page_bit); |
f62e00cc3 mm: introduce wai... |
572 573 574 575 576 577 578 579 580 581 |
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 Linux-2.6.12-rc2 |
582 |
/** |
385e1ca5f CacheFiles: Permi... |
583 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc filemap: fix kern... |
584 585 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f CacheFiles: Permi... |
586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 |
* * 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 [PATCH] kernel-do... |
601 |
* unlock_page - unlock a locked page |
1da177e4c Linux-2.6.12-rc2 |
602 603 604 605 606 607 608 |
* @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 mm: page lock use... |
609 610 |
* 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 Linux-2.6.12-rc2 |
611 |
*/ |
920c7a5d0 mm: remove fastca... |
612 |
void unlock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
613 |
{ |
8413ac9d8 mm: page lock use... |
614 615 616 |
VM_BUG_ON(!PageLocked(page)); clear_bit_unlock(PG_locked, &page->flags); smp_mb__after_clear_bit(); |
1da177e4c Linux-2.6.12-rc2 |
617 618 619 |
wake_up_page(page, PG_locked); } EXPORT_SYMBOL(unlock_page); |
485bb99b4 [PATCH] kernel-do... |
620 621 622 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c Linux-2.6.12-rc2 |
623 624 625 |
*/ void end_page_writeback(struct page *page) { |
9057c212b mm: avoid unneces... |
626 627 628 629 630 631 632 633 634 |
/* * TestClearPageReclaim could be used here but it is an atomic * operation and overkill in this particular case. Failing to * shuffle a page marked for immediate reclaim is too mild to * justify taking an atomic operation penalty at the end of * ever page writeback. */ if (PageReclaim(page)) { ClearPageReclaim(page); |
ac6aadb24 mm: rotate_reclai... |
635 |
rotate_reclaimable_page(page); |
9057c212b mm: avoid unneces... |
636 |
} |
ac6aadb24 mm: rotate_reclai... |
637 638 639 |
if (!test_clear_page_writeback(page)) BUG(); |
1da177e4c Linux-2.6.12-rc2 |
640 641 642 643 |
smp_mb__after_clear_bit(); wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
485bb99b4 [PATCH] kernel-do... |
644 645 646 |
/** * __lock_page - get a lock on the page, assuming we need to sleep to get it * @page: the page to lock |
1da177e4c Linux-2.6.12-rc2 |
647 |
*/ |
920c7a5d0 mm: remove fastca... |
648 |
void __lock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
649 650 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); |
7eaceacca block: remove per... |
651 |
__wait_on_bit_lock(page_waitqueue(page), &wait, sleep_on_page, |
1da177e4c Linux-2.6.12-rc2 |
652 653 654 |
TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(__lock_page); |
b5606c2d4 remove final fast... |
655 |
int __lock_page_killable(struct page *page) |
2687a3569 Add lock_page_kil... |
656 657 658 659 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); return __wait_on_bit_lock(page_waitqueue(page), &wait, |
7eaceacca block: remove per... |
660 |
sleep_on_page_killable, TASK_KILLABLE); |
2687a3569 Add lock_page_kil... |
661 |
} |
18bc0bbd1 Staging: pohmelfs... |
662 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569 Add lock_page_kil... |
663 |
|
d065bd810 mm: retry page fa... |
664 665 666 |
int __lock_page_or_retry(struct page *page, struct mm_struct *mm, unsigned int flags) { |
37b23e052 x86,mm: make page... |
667 668 669 670 671 672 673 674 675 676 677 678 |
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 mm: allow GUP to ... |
679 |
wait_on_page_locked(page); |
d065bd810 mm: retry page fa... |
680 |
return 0; |
37b23e052 x86,mm: make page... |
681 682 683 684 685 686 687 688 689 690 691 692 |
} 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 mm: retry page fa... |
693 694 |
} } |
485bb99b4 [PATCH] kernel-do... |
695 |
/** |
3721b4212 mm: filemap: move... |
696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 |
* page_cache_next_hole - find the next hole (not-present entry) * @mapping: mapping * @index: index * @max_scan: maximum range to search * * Search the set [index, min(index+max_scan-1, MAX_INDEX)] for the * lowest indexed hole. * * Returns: the index of the hole if found, otherwise returns an index * outside of the set specified (in which case 'return - index >= * max_scan' will be true). In rare cases of index wrap-around, 0 will * be returned. * * page_cache_next_hole may be called under rcu_read_lock. However, * like radix_tree_gang_lookup, this will not atomically search a * snapshot of the tree at a single point in time. For example, if a * hole is created at index 5, then subsequently a hole is created at * index 10, page_cache_next_hole covering both indexes may return 10 * if called under rcu_read_lock. */ pgoff_t page_cache_next_hole(struct address_space *mapping, pgoff_t index, unsigned long max_scan) { unsigned long i; for (i = 0; i < max_scan; i++) { |
e714f0cf0 mm + fs: prepare ... |
722 723 724 725 |
struct page *page; page = radix_tree_lookup(&mapping->page_tree, index); if (!page || radix_tree_exceptional_entry(page)) |
3721b4212 mm: filemap: move... |
726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 |
break; index++; if (index == 0) break; } return index; } EXPORT_SYMBOL(page_cache_next_hole); /** * page_cache_prev_hole - find the prev hole (not-present entry) * @mapping: mapping * @index: index * @max_scan: maximum range to search * * Search backwards in the range [max(index-max_scan+1, 0), index] for * the first hole. * * Returns: the index of the hole if found, otherwise returns an index * outside of the set specified (in which case 'index - return >= * max_scan' will be true). In rare cases of wrap-around, ULONG_MAX * will be returned. * * page_cache_prev_hole may be called under rcu_read_lock. However, * like radix_tree_gang_lookup, this will not atomically search a * snapshot of the tree at a single point in time. For example, if a * hole is created at index 10, then subsequently a hole is created at * index 5, page_cache_prev_hole covering both indexes may return 5 if * called under rcu_read_lock. */ pgoff_t page_cache_prev_hole(struct address_space *mapping, pgoff_t index, unsigned long max_scan) { unsigned long i; for (i = 0; i < max_scan; i++) { |
e714f0cf0 mm + fs: prepare ... |
763 764 765 766 |
struct page *page; page = radix_tree_lookup(&mapping->page_tree, index); if (!page || radix_tree_exceptional_entry(page)) |
3721b4212 mm: filemap: move... |
767 768 769 770 771 772 773 774 775 776 777 |
break; index--; if (index == ULONG_MAX) break; } return index; } EXPORT_SYMBOL(page_cache_prev_hole); /** |
e714f0cf0 mm + fs: prepare ... |
778 |
* find_get_entry - find and get a page cache entry |
485bb99b4 [PATCH] kernel-do... |
779 |
* @mapping: the address_space to search |
e714f0cf0 mm + fs: prepare ... |
780 |
* @offset: the page cache index |
485bb99b4 [PATCH] kernel-do... |
781 |
* |
e714f0cf0 mm + fs: prepare ... |
782 783 784 785 786 787 788 |
* Looks up the page cache slot at @mapping & @offset. If there is a * page cache page, it is returned with an increased refcount. * * If the slot holds a shadow entry of a previously evicted page, it * is returned. * * Otherwise, %NULL is returned. |
1da177e4c Linux-2.6.12-rc2 |
789 |
*/ |
e714f0cf0 mm + fs: prepare ... |
790 |
struct page *find_get_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
791 |
{ |
a60637c85 mm: lockless page... |
792 |
void **pagep; |
1da177e4c Linux-2.6.12-rc2 |
793 |
struct page *page; |
a60637c85 mm: lockless page... |
794 795 796 797 798 799 |
rcu_read_lock(); repeat: page = NULL; pagep = radix_tree_lookup_slot(&mapping->page_tree, offset); if (pagep) { page = radix_tree_deref_slot(pagep); |
27d20fddc radix-tree: fix R... |
800 801 |
if (unlikely(!page)) goto out; |
a2c16d6cb mm: let swap use ... |
802 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
803 804 805 806 807 808 809 810 |
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 mm: let swap use ... |
811 |
} |
a60637c85 mm: lockless page... |
812 813 814 815 816 817 818 819 820 821 822 823 824 |
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 radix-tree: fix R... |
825 |
out: |
a60637c85 mm: lockless page... |
826 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
827 828 |
return page; } |
e714f0cf0 mm + fs: prepare ... |
829 |
EXPORT_SYMBOL(find_get_entry); |
1da177e4c Linux-2.6.12-rc2 |
830 |
|
485bb99b4 [PATCH] kernel-do... |
831 |
/** |
e714f0cf0 mm + fs: prepare ... |
832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 |
* find_lock_entry - locate, pin and lock a page cache entry * @mapping: the address_space to search * @offset: the page cache index * * Looks up the page cache slot at @mapping & @offset. If there is a * page cache page, it is returned locked and with an increased * refcount. * * If the slot holds a shadow entry of a previously evicted page, it * is returned. * * Otherwise, %NULL is returned. * * find_lock_entry() may sleep. */ struct page *find_lock_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
848 849 |
{ struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
850 |
repeat: |
e714f0cf0 mm + fs: prepare ... |
851 |
page = find_get_entry(mapping, offset); |
a2c16d6cb mm: let swap use ... |
852 |
if (page && !radix_tree_exception(page)) { |
a60637c85 mm: lockless page... |
853 854 855 856 857 858 |
lock_page(page); /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); page_cache_release(page); goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
859 |
} |
a60637c85 mm: lockless page... |
860 |
VM_BUG_ON(page->index != offset); |
1da177e4c Linux-2.6.12-rc2 |
861 |
} |
1da177e4c Linux-2.6.12-rc2 |
862 863 |
return page; } |
e714f0cf0 mm + fs: prepare ... |
864 865 866 |
EXPORT_SYMBOL(find_lock_entry); /** |
d618a27c7 mm: non-atomicall... |
867 |
* pagecache_get_page - find and get a page reference |
e714f0cf0 mm + fs: prepare ... |
868 869 |
* @mapping: the address_space to search * @offset: the page index |
d618a27c7 mm: non-atomicall... |
870 871 |
* @fgp_flags: PCG flags * @gfp_mask: gfp mask to use if a page is to be allocated |
e714f0cf0 mm + fs: prepare ... |
872 |
* |
d618a27c7 mm: non-atomicall... |
873 |
* Looks up the page cache slot at @mapping & @offset. |
1da177e4c Linux-2.6.12-rc2 |
874 |
* |
d618a27c7 mm: non-atomicall... |
875 |
* PCG flags modify how the page is returned |
e714f0cf0 mm + fs: prepare ... |
876 |
* |
d618a27c7 mm: non-atomicall... |
877 878 879 880 881 882 |
* FGP_ACCESSED: the page will be marked accessed * FGP_LOCK: Page is return locked * FGP_CREAT: If page is not present then a new page is allocated using * @gfp_mask and added to the page cache and the VM's LRU * list. The page is returned locked and with an increased * refcount. Otherwise, %NULL is returned. |
1da177e4c Linux-2.6.12-rc2 |
883 |
* |
d618a27c7 mm: non-atomicall... |
884 885 |
* If FGP_LOCK or FGP_CREAT are specified then the function may sleep even * if the GFP flags specified for FGP_CREAT are atomic. |
1da177e4c Linux-2.6.12-rc2 |
886 |
* |
d618a27c7 mm: non-atomicall... |
887 |
* If there is a page cache page, it is returned with an increased refcount. |
1da177e4c Linux-2.6.12-rc2 |
888 |
*/ |
d618a27c7 mm: non-atomicall... |
889 890 |
struct page *pagecache_get_page(struct address_space *mapping, pgoff_t offset, int fgp_flags, gfp_t cache_gfp_mask, gfp_t radix_gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
891 |
{ |
eb2be1893 mm: buffered writ... |
892 |
struct page *page; |
d618a27c7 mm: non-atomicall... |
893 |
|
1da177e4c Linux-2.6.12-rc2 |
894 |
repeat: |
d618a27c7 mm: non-atomicall... |
895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 |
page = find_get_entry(mapping, offset); if (radix_tree_exceptional_entry(page)) page = NULL; if (!page) goto no_page; if (fgp_flags & FGP_LOCK) { if (fgp_flags & FGP_NOWAIT) { if (!trylock_page(page)) { page_cache_release(page); return NULL; } } else { lock_page(page); } /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); page_cache_release(page); goto repeat; } VM_BUG_ON(page->index != offset); } if (page && (fgp_flags & FGP_ACCESSED)) mark_page_accessed(page); no_page: if (!page && (fgp_flags & FGP_CREAT)) { int err; if ((fgp_flags & FGP_WRITE) && mapping_cap_account_dirty(mapping)) cache_gfp_mask |= __GFP_WRITE; if (fgp_flags & FGP_NOFS) { cache_gfp_mask &= ~__GFP_FS; radix_gfp_mask &= ~__GFP_FS; } page = __page_cache_alloc(cache_gfp_mask); |
eb2be1893 mm: buffered writ... |
934 935 |
if (!page) return NULL; |
d618a27c7 mm: non-atomicall... |
936 937 938 939 940 941 942 943 944 |
if (WARN_ON_ONCE(!(fgp_flags & FGP_LOCK))) fgp_flags |= FGP_LOCK; /* Init accessed so avoit atomic mark_page_accessed later */ if (fgp_flags & FGP_ACCESSED) init_page_accessed(page); err = add_to_page_cache_lru(page, mapping, offset, radix_gfp_mask); |
eb2be1893 mm: buffered writ... |
945 946 947 948 949 |
if (unlikely(err)) { page_cache_release(page); page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
950 |
} |
1da177e4c Linux-2.6.12-rc2 |
951 |
} |
d618a27c7 mm: non-atomicall... |
952 |
|
1da177e4c Linux-2.6.12-rc2 |
953 954 |
return page; } |
d618a27c7 mm: non-atomicall... |
955 |
EXPORT_SYMBOL(pagecache_get_page); |
1da177e4c Linux-2.6.12-rc2 |
956 957 |
/** |
e714f0cf0 mm + fs: prepare ... |
958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 |
* find_get_entries - gang pagecache lookup * @mapping: The address_space to search * @start: The starting page cache index * @nr_entries: The maximum number of entries * @entries: Where the resulting entries are placed * @indices: The cache indices corresponding to the entries in @entries * * find_get_entries() will search for and return a group of up to * @nr_entries entries in the mapping. The entries are placed at * @entries. find_get_entries() takes a reference against any actual * pages it returns. * * The search returns a group of mapping-contiguous page cache entries * with ascending indexes. There may be holes in the indices due to * not-present pages. * * Any shadow entries of evicted pages are included in the returned * array. * * find_get_entries() returns the number of pages and shadow entries * which were found. */ unsigned find_get_entries(struct address_space *mapping, pgoff_t start, unsigned int nr_entries, struct page **entries, pgoff_t *indices) { void **slot; unsigned int ret = 0; struct radix_tree_iter iter; if (!nr_entries) return 0; rcu_read_lock(); restart: radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, start) { struct page *page; repeat: page = radix_tree_deref_slot(slot); if (unlikely(!page)) continue; if (radix_tree_exception(page)) { if (radix_tree_deref_retry(page)) goto restart; /* * Otherwise, we must be storing a swap entry * here as an exceptional entry: so return it * without attempting to raise page count. */ goto export; } if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *slot)) { page_cache_release(page); goto repeat; } export: indices[ret] = iter.index; entries[ret] = page; if (++ret == nr_entries) break; } rcu_read_unlock(); return ret; } /** |
1da177e4c Linux-2.6.12-rc2 |
1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 |
* 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 radix-tree: use i... |
1046 1047 1048 1049 1050 1051 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1052 1053 1054 |
rcu_read_lock(); restart: |
0fc9d1040 radix-tree: use i... |
1055 |
radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, start) { |
a60637c85 mm: lockless page... |
1056 1057 |
struct page *page; repeat: |
0fc9d1040 radix-tree: use i... |
1058 |
page = radix_tree_deref_slot(slot); |
a60637c85 mm: lockless page... |
1059 1060 |
if (unlikely(!page)) continue; |
9d8aa4ea8 mm: remove worryi... |
1061 |
|
a2c16d6cb mm: let swap use ... |
1062 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1063 1064 1065 1066 1067 1068 |
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 radix-tree: use i... |
1069 |
WARN_ON(iter.index); |
8079b1c85 mm: clarify the r... |
1070 1071 |
goto restart; } |
a2c16d6cb mm: let swap use ... |
1072 |
/* |
8079b1c85 mm: clarify the r... |
1073 1074 1075 |
* 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 mm: let swap use ... |
1076 |
*/ |
8079b1c85 mm: clarify the r... |
1077 |
continue; |
27d20fddc radix-tree: fix R... |
1078 |
} |
a60637c85 mm: lockless page... |
1079 1080 1081 1082 1083 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1084 |
if (unlikely(page != *slot)) { |
a60637c85 mm: lockless page... |
1085 1086 1087 |
page_cache_release(page); goto repeat; } |
1da177e4c Linux-2.6.12-rc2 |
1088 |
|
a60637c85 mm: lockless page... |
1089 |
pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1090 1091 |
if (++ret == nr_pages) break; |
a60637c85 mm: lockless page... |
1092 |
} |
5b280c0cc mm: don't return ... |
1093 |
|
a60637c85 mm: lockless page... |
1094 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1095 1096 |
return ret; } |
ebf43500e [PATCH] Add find_... |
1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 |
/** * 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 radix-tree: use i... |
1112 1113 1114 1115 1116 1117 |
struct radix_tree_iter iter; void **slot; unsigned int ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1118 1119 1120 |
rcu_read_lock(); restart: |
0fc9d1040 radix-tree: use i... |
1121 |
radix_tree_for_each_contig(slot, &mapping->page_tree, &iter, index) { |
a60637c85 mm: lockless page... |
1122 1123 |
struct page *page; repeat: |
0fc9d1040 radix-tree: use i... |
1124 1125 |
page = radix_tree_deref_slot(slot); /* The hole, there no reason to continue */ |
a60637c85 mm: lockless page... |
1126 |
if (unlikely(!page)) |
0fc9d1040 radix-tree: use i... |
1127 |
break; |
9d8aa4ea8 mm: remove worryi... |
1128 |
|
a2c16d6cb mm: let swap use ... |
1129 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1130 1131 1132 1133 1134 1135 1136 1137 |
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 mm: let swap use ... |
1138 |
/* |
8079b1c85 mm: clarify the r... |
1139 1140 1141 |
* Otherwise, shmem/tmpfs must be storing a swap entry * here as an exceptional entry: so stop looking for * contiguous pages. |
a2c16d6cb mm: let swap use ... |
1142 |
*/ |
8079b1c85 mm: clarify the r... |
1143 |
break; |
a2c16d6cb mm: let swap use ... |
1144 |
} |
ebf43500e [PATCH] Add find_... |
1145 |
|
a60637c85 mm: lockless page... |
1146 1147 1148 1149 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1150 |
if (unlikely(page != *slot)) { |
a60637c85 mm: lockless page... |
1151 1152 1153 |
page_cache_release(page); goto repeat; } |
9cbb4cb21 mm: find_get_page... |
1154 1155 1156 1157 1158 |
/* * 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 radix-tree: use i... |
1159 |
if (page->mapping == NULL || page->index != iter.index) { |
9cbb4cb21 mm: find_get_page... |
1160 1161 1162 |
page_cache_release(page); break; } |
a60637c85 mm: lockless page... |
1163 |
pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1164 1165 |
if (++ret == nr_pages) break; |
ebf43500e [PATCH] Add find_... |
1166 |
} |
a60637c85 mm: lockless page... |
1167 1168 |
rcu_read_unlock(); return ret; |
ebf43500e [PATCH] Add find_... |
1169 |
} |
ef71c15c4 AFS: export a cou... |
1170 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e [PATCH] Add find_... |
1171 |
|
485bb99b4 [PATCH] kernel-do... |
1172 1173 1174 1175 1176 1177 1178 1179 |
/** * 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 Linux-2.6.12-rc2 |
1180 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4 [PATCH] kernel-do... |
1181 |
* @tag. We update @index to index the next page for the traversal. |
1da177e4c Linux-2.6.12-rc2 |
1182 1183 1184 1185 |
*/ unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, int tag, unsigned int nr_pages, struct page **pages) { |
0fc9d1040 radix-tree: use i... |
1186 1187 1188 1189 1190 1191 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1192 1193 1194 |
rcu_read_lock(); restart: |
0fc9d1040 radix-tree: use i... |
1195 1196 |
radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, *index, tag) { |
a60637c85 mm: lockless page... |
1197 1198 |
struct page *page; repeat: |
0fc9d1040 radix-tree: use i... |
1199 |
page = radix_tree_deref_slot(slot); |
a60637c85 mm: lockless page... |
1200 1201 |
if (unlikely(!page)) continue; |
9d8aa4ea8 mm: remove worryi... |
1202 |
|
a2c16d6cb mm: let swap use ... |
1203 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1204 1205 1206 1207 1208 1209 1210 1211 |
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 mm: let swap use ... |
1212 |
/* |
8079b1c85 mm: clarify the r... |
1213 1214 |
* This function is never used on a shmem/tmpfs * mapping, so a swap entry won't be found here. |
a2c16d6cb mm: let swap use ... |
1215 |
*/ |
8079b1c85 mm: clarify the r... |
1216 |
BUG(); |
a2c16d6cb mm: let swap use ... |
1217 |
} |
a60637c85 mm: lockless page... |
1218 1219 1220 1221 1222 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1223 |
if (unlikely(page != *slot)) { |
a60637c85 mm: lockless page... |
1224 1225 1226 1227 1228 |
page_cache_release(page); goto repeat; } pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1229 1230 |
if (++ret == nr_pages) break; |
a60637c85 mm: lockless page... |
1231 |
} |
5b280c0cc mm: don't return ... |
1232 |
|
a60637c85 mm: lockless page... |
1233 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1234 |
|
1da177e4c Linux-2.6.12-rc2 |
1235 1236 |
if (ret) *index = pages[ret - 1]->index + 1; |
a60637c85 mm: lockless page... |
1237 |
|
1da177e4c Linux-2.6.12-rc2 |
1238 1239 |
return ret; } |
ef71c15c4 AFS: export a cou... |
1240 |
EXPORT_SYMBOL(find_get_pages_tag); |
1da177e4c Linux-2.6.12-rc2 |
1241 |
|
76d42bd96 [PATCH] readahead... |
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 |
/* * 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 [PATCH] readahead... |
1260 |
ra->ra_pages /= 4; |
76d42bd96 [PATCH] readahead... |
1261 |
} |
485bb99b4 [PATCH] kernel-do... |
1262 |
/** |
36e789144 kill do_generic_m... |
1263 |
* do_generic_file_read - generic file read routine |
485bb99b4 [PATCH] kernel-do... |
1264 1265 1266 1267 1268 |
* @filp: the file to read * @ppos: current file position * @desc: read_descriptor * @actor: read method * |
1da177e4c Linux-2.6.12-rc2 |
1269 |
* This is a generic file read routine, and uses the |
485bb99b4 [PATCH] kernel-do... |
1270 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c Linux-2.6.12-rc2 |
1271 1272 1273 |
* * This is really ugly. But the goto's actually try to clarify some * of the logic when it comes to error handling etc. |
1da177e4c Linux-2.6.12-rc2 |
1274 |
*/ |
36e789144 kill do_generic_m... |
1275 1276 |
static void do_generic_file_read(struct file *filp, loff_t *ppos, read_descriptor_t *desc, read_actor_t actor) |
1da177e4c Linux-2.6.12-rc2 |
1277 |
{ |
36e789144 kill do_generic_m... |
1278 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
1279 |
struct inode *inode = mapping->host; |
36e789144 kill do_generic_m... |
1280 |
struct file_ra_state *ra = &filp->f_ra; |
57f6b96c0 filemap: convert ... |
1281 1282 1283 1284 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372 readahead: improv... |
1285 |
unsigned int prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
1286 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
1287 |
|
1da177e4c Linux-2.6.12-rc2 |
1288 |
index = *ppos >> PAGE_CACHE_SHIFT; |
7ff81078d readahead: remove... |
1289 1290 |
prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); |
1da177e4c Linux-2.6.12-rc2 |
1291 1292 |
last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; offset = *ppos & ~PAGE_CACHE_MASK; |
1da177e4c Linux-2.6.12-rc2 |
1293 1294 |
for (;;) { struct page *page; |
57f6b96c0 filemap: convert ... |
1295 |
pgoff_t end_index; |
a32ea1e1f Fix read/truncate... |
1296 |
loff_t isize; |
1da177e4c Linux-2.6.12-rc2 |
1297 |
unsigned long nr, ret; |
1da177e4c Linux-2.6.12-rc2 |
1298 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
1299 1300 |
find_page: page = find_get_page(mapping, index); |
3ea89ee86 readahead: conver... |
1301 |
if (!page) { |
cf914a7d6 readahead: split ... |
1302 |
page_cache_sync_readahead(mapping, |
7ff81078d readahead: remove... |
1303 |
ra, filp, |
3ea89ee86 readahead: conver... |
1304 1305 1306 1307 1308 1309 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6 readahead: split ... |
1310 |
page_cache_async_readahead(mapping, |
7ff81078d readahead: remove... |
1311 |
ra, filp, page, |
3ea89ee86 readahead: conver... |
1312 |
index, last_index - index); |
1da177e4c Linux-2.6.12-rc2 |
1313 |
} |
8ab22b9ab vfs: pagecache us... |
1314 1315 1316 1317 |
if (!PageUptodate(page)) { if (inode->i_blkbits == PAGE_CACHE_SHIFT || !mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; |
529ae9aaa mm: rename page t... |
1318 |
if (!trylock_page(page)) |
8ab22b9ab vfs: pagecache us... |
1319 |
goto page_not_up_to_date; |
8d056cb96 mm/vfs: revalidat... |
1320 1321 1322 |
/* Did it get truncated before we got the lock? */ if (!page->mapping) goto page_not_up_to_date_locked; |
8ab22b9ab vfs: pagecache us... |
1323 1324 1325 1326 1327 |
if (!mapping->a_ops->is_partially_uptodate(page, desc, offset)) goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1328 |
page_ok: |
a32ea1e1f Fix read/truncate... |
1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 |
/* * 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 Linux-2.6.12-rc2 |
1355 1356 1357 1358 1359 1360 1361 1362 1363 |
/* 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 readahead: improv... |
1364 1365 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c Linux-2.6.12-rc2 |
1366 |
*/ |
ec0f16372 readahead: improv... |
1367 |
if (prev_index != index || offset != prev_offset) |
1da177e4c Linux-2.6.12-rc2 |
1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 |
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 readahead: code c... |
1385 |
prev_offset = offset; |
1da177e4c Linux-2.6.12-rc2 |
1386 1387 1388 1389 1390 1391 1392 1393 |
page_cache_release(page); if (ret == nr && desc->count) continue; goto out; page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235 do_generic_file_r... |
1394 1395 1396 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1397 |
|
8ab22b9ab vfs: pagecache us... |
1398 |
page_not_up_to_date_locked: |
da6052f7b [PATCH] update so... |
1399 |
/* Did it get truncated before we got the lock? */ |
1da177e4c Linux-2.6.12-rc2 |
1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 |
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 do_generic_file_r... |
1413 1414 1415 1416 1417 1418 |
/* * 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 Linux-2.6.12-rc2 |
1419 1420 |
/* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7 [PATCH] add AOP_T... |
1421 1422 1423 1424 1425 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { page_cache_release(page); goto find_page; } |
1da177e4c Linux-2.6.12-rc2 |
1426 |
goto readpage_error; |
994fc28c7 [PATCH] add AOP_T... |
1427 |
} |
1da177e4c Linux-2.6.12-rc2 |
1428 1429 |
if (!PageUptodate(page)) { |
854623235 do_generic_file_r... |
1430 1431 1432 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1433 1434 1435 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* |
2ecdc82ef kill unused inval... |
1436 |
* invalidate_mapping_pages got it |
1da177e4c Linux-2.6.12-rc2 |
1437 1438 1439 1440 1441 1442 |
*/ unlock_page(page); page_cache_release(page); goto find_page; } unlock_page(page); |
7ff81078d readahead: remove... |
1443 |
shrink_readahead_size_eio(filp, ra); |
854623235 do_generic_file_r... |
1444 1445 |
error = -EIO; goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1446 1447 1448 |
} unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 |
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 mm: buffered writ... |
1462 1463 1464 1465 |
page = page_cache_alloc_cold(mapping); if (!page) { desc->error = -ENOMEM; goto out; |
1da177e4c Linux-2.6.12-rc2 |
1466 |
} |
eb2be1893 mm: buffered writ... |
1467 |
error = add_to_page_cache_lru(page, mapping, |
1da177e4c Linux-2.6.12-rc2 |
1468 1469 |
index, GFP_KERNEL); if (error) { |
eb2be1893 mm: buffered writ... |
1470 |
page_cache_release(page); |
1da177e4c Linux-2.6.12-rc2 |
1471 1472 1473 1474 1475 |
if (error == -EEXIST) goto find_page; desc->error = error; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
1476 1477 1478 1479 |
goto readpage; } out: |
7ff81078d readahead: remove... |
1480 1481 1482 |
ra->prev_pos = prev_index; ra->prev_pos <<= PAGE_CACHE_SHIFT; ra->prev_pos |= prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
1483 |
|
f4e6b498d readahead: combin... |
1484 |
*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; |
0c6aa2639 mm: do_generic_fi... |
1485 |
file_accessed(filp); |
1da177e4c Linux-2.6.12-rc2 |
1486 |
} |
1da177e4c Linux-2.6.12-rc2 |
1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 |
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 mm: remove the se... |
1502 |
kaddr = kmap_atomic(page); |
1da177e4c Linux-2.6.12-rc2 |
1503 1504 |
left = __copy_to_user_inatomic(desc->arg.buf, kaddr + offset, size); |
9b04c5fec mm: remove the se... |
1505 |
kunmap_atomic(kaddr); |
1da177e4c Linux-2.6.12-rc2 |
1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 |
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 mm: move common s... |
1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 |
/* * 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 [PATCH] kernel-do... |
1563 |
/** |
b2abacf3a [PATCH] mm: fix i... |
1564 |
* generic_file_aio_read - generic filesystem read routine |
485bb99b4 [PATCH] kernel-do... |
1565 1566 1567 |
* @iocb: kernel I/O control block * @iov: io vector request * @nr_segs: number of segments in the iovec |
b2abacf3a [PATCH] mm: fix i... |
1568 |
* @pos: current file position |
485bb99b4 [PATCH] kernel-do... |
1569 |
* |
1da177e4c Linux-2.6.12-rc2 |
1570 1571 1572 1573 |
* This is the "read()" routine for all filesystems * that can use the page cache directly. */ ssize_t |
543ade1fc [PATCH] Streamlin... |
1574 1575 |
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c Linux-2.6.12-rc2 |
1576 1577 1578 |
{ struct file *filp = iocb->ki_filp; ssize_t retval; |
66f998f61 fs: allow short d... |
1579 |
unsigned long seg = 0; |
1da177e4c Linux-2.6.12-rc2 |
1580 |
size_t count; |
543ade1fc [PATCH] Streamlin... |
1581 |
loff_t *ppos = &iocb->ki_pos; |
1da177e4c Linux-2.6.12-rc2 |
1582 1583 |
count = 0; |
0ceb33143 mm: move common s... |
1584 1585 1586 |
retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; |
1da177e4c Linux-2.6.12-rc2 |
1587 1588 1589 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { |
543ade1fc [PATCH] Streamlin... |
1590 |
loff_t size; |
1da177e4c Linux-2.6.12-rc2 |
1591 1592 1593 1594 1595 |
struct address_space *mapping; struct inode *inode; mapping = filp->f_mapping; inode = mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
1596 1597 1598 1599 |
if (!count) goto out; /* skip atime */ size = i_size_read(inode); if (pos < size) { |
48b47c561 mm: direct IO sta... |
1600 1601 |
retval = filemap_write_and_wait_range(mapping, pos, pos + iov_length(iov, nr_segs) - 1); |
a969e903a kill generic_file... |
1602 1603 1604 1605 |
if (!retval) { retval = mapping->a_ops->direct_IO(READ, iocb, iov, pos, nr_segs); } |
66f998f61 fs: allow short d... |
1606 |
if (retval > 0) { |
1da177e4c Linux-2.6.12-rc2 |
1607 |
*ppos = pos + retval; |
66f998f61 fs: allow short d... |
1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 |
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 generic_file_aio_... |
1620 1621 1622 |
file_accessed(filp); goto out; } |
0e0bcae3b [GFS2] Fix direct... |
1623 |
} |
1da177e4c Linux-2.6.12-rc2 |
1624 |
} |
66f998f61 fs: allow short d... |
1625 |
count = retval; |
11fa977ec generic_file_aio_... |
1626 1627 |
for (seg = 0; seg < nr_segs; seg++) { read_descriptor_t desc; |
66f998f61 fs: allow short d... |
1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 |
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 Linux-2.6.12-rc2 |
1642 |
|
11fa977ec generic_file_aio_... |
1643 |
desc.written = 0; |
66f998f61 fs: allow short d... |
1644 1645 |
desc.arg.buf = iov[seg].iov_base + offset; desc.count = iov[seg].iov_len - offset; |
11fa977ec generic_file_aio_... |
1646 1647 1648 1649 1650 1651 1652 1653 |
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 Linux-2.6.12-rc2 |
1654 |
} |
11fa977ec generic_file_aio_... |
1655 1656 |
if (desc.count > 0) break; |
1da177e4c Linux-2.6.12-rc2 |
1657 1658 1659 1660 |
} out: return retval; } |
1da177e4c Linux-2.6.12-rc2 |
1661 |
EXPORT_SYMBOL(generic_file_aio_read); |
1da177e4c Linux-2.6.12-rc2 |
1662 |
#ifdef CONFIG_MMU |
485bb99b4 [PATCH] kernel-do... |
1663 1664 1665 1666 1667 |
/** * page_cache_read - adds requested page to the page cache if not already there * @file: file to read * @offset: page index * |
1da177e4c Linux-2.6.12-rc2 |
1668 1669 1670 |
* 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 mm: remove fastca... |
1671 |
static int page_cache_read(struct file *file, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1672 1673 1674 |
{ struct address_space *mapping = file->f_mapping; struct page *page; |
994fc28c7 [PATCH] add AOP_T... |
1675 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1676 |
|
994fc28c7 [PATCH] add AOP_T... |
1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 |
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 Linux-2.6.12-rc2 |
1687 |
|
1da177e4c Linux-2.6.12-rc2 |
1688 |
page_cache_release(page); |
1da177e4c Linux-2.6.12-rc2 |
1689 |
|
994fc28c7 [PATCH] add AOP_T... |
1690 1691 1692 |
} while (ret == AOP_TRUNCATED_PAGE); return ret; |
1da177e4c Linux-2.6.12-rc2 |
1693 1694 1695 |
} #define MMAP_LOTSAMISS (100) |
ef00e08e2 readahead: clean ... |
1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 |
/* * 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 */ |
64363aad5 mm: remove unused... |
1709 |
if (vma->vm_flags & VM_RAND_READ) |
ef00e08e2 readahead: clean ... |
1710 |
return; |
275b12bf5 readahead: return... |
1711 1712 |
if (!ra->ra_pages) return; |
ef00e08e2 readahead: clean ... |
1713 |
|
64363aad5 mm: remove unused... |
1714 |
if (vma->vm_flags & VM_SEQ_READ) { |
7ffc59b4d readahead: enforc... |
1715 1716 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
1717 1718 |
return; } |
207d04baa readahead: reduce... |
1719 1720 |
/* Avoid banging the cache line if not needed */ if (ra->mmap_miss < MMAP_LOTSAMISS * 10) |
ef00e08e2 readahead: clean ... |
1721 1722 1723 1724 1725 1726 1727 1728 |
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 readahead: record... |
1729 1730 1731 |
/* * mmap read-around */ |
ef00e08e2 readahead: clean ... |
1732 |
ra_pages = max_sane_readahead(ra->ra_pages); |
275b12bf5 readahead: return... |
1733 1734 |
ra->start = max_t(long, 0, offset - ra_pages / 2); ra->size = ra_pages; |
2cbea1d3a readahead: trigge... |
1735 |
ra->async_size = ra_pages / 4; |
275b12bf5 readahead: return... |
1736 |
ra_submit(ra, mapping, file); |
ef00e08e2 readahead: clean ... |
1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 |
} /* * 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 */ |
64363aad5 mm: remove unused... |
1752 |
if (vma->vm_flags & VM_RAND_READ) |
ef00e08e2 readahead: clean ... |
1753 1754 1755 1756 |
return; if (ra->mmap_miss > 0) ra->mmap_miss--; if (PageReadahead(page)) |
2fad6f5de readahead: enforc... |
1757 1758 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
1759 |
} |
485bb99b4 [PATCH] kernel-do... |
1760 |
/** |
54cb8821d mm: merge populat... |
1761 |
* filemap_fault - read in file data for page fault handling |
d0217ac04 mm: fault feedbac... |
1762 1763 |
* @vma: vma in which the fault was taken * @vmf: struct vm_fault containing details of the fault |
485bb99b4 [PATCH] kernel-do... |
1764 |
* |
54cb8821d mm: merge populat... |
1765 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c Linux-2.6.12-rc2 |
1766 1767 1768 1769 1770 1771 |
* 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 mm: fault feedbac... |
1772 |
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
1da177e4c Linux-2.6.12-rc2 |
1773 1774 |
{ int error; |
54cb8821d mm: merge populat... |
1775 |
struct file *file = vma->vm_file; |
1da177e4c Linux-2.6.12-rc2 |
1776 1777 1778 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2 readahead: clean ... |
1779 |
pgoff_t offset = vmf->pgoff; |
1da177e4c Linux-2.6.12-rc2 |
1780 |
struct page *page; |
2004dc8ee Use pgoff_t inste... |
1781 |
pgoff_t size; |
83c54070e mm: fault feedbac... |
1782 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1783 |
|
1da177e4c Linux-2.6.12-rc2 |
1784 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2 readahead: clean ... |
1785 |
if (offset >= size) |
5307cc1aa Remove broken ptr... |
1786 |
return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
1787 |
|
1da177e4c Linux-2.6.12-rc2 |
1788 |
/* |
494264208 mm: memcg: handle... |
1789 |
* Do we have something in the page cache already? |
1da177e4c Linux-2.6.12-rc2 |
1790 |
*/ |
ef00e08e2 readahead: clean ... |
1791 |
page = find_get_page(mapping, offset); |
45cac65b0 readahead: fault ... |
1792 |
if (likely(page) && !(vmf->flags & FAULT_FLAG_TRIED)) { |
1da177e4c Linux-2.6.12-rc2 |
1793 |
/* |
ef00e08e2 readahead: clean ... |
1794 1795 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c Linux-2.6.12-rc2 |
1796 |
*/ |
ef00e08e2 readahead: clean ... |
1797 |
do_async_mmap_readahead(vma, ra, file, page, offset); |
45cac65b0 readahead: fault ... |
1798 |
} else if (!page) { |
ef00e08e2 readahead: clean ... |
1799 1800 1801 |
/* No page in the page cache at all */ do_sync_mmap_readahead(vma, ra, file, offset); count_vm_event(PGMAJFAULT); |
456f998ec memcg: add the pa... |
1802 |
mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT); |
ef00e08e2 readahead: clean ... |
1803 1804 |
ret = VM_FAULT_MAJOR; retry_find: |
b522c94da mm: filemap_fault... |
1805 |
page = find_get_page(mapping, offset); |
1da177e4c Linux-2.6.12-rc2 |
1806 1807 1808 |
if (!page) goto no_cached_page; } |
d88c0922f Release page refe... |
1809 1810 |
if (!lock_page_or_retry(page, vma->vm_mm, vmf->flags)) { page_cache_release(page); |
d065bd810 mm: retry page fa... |
1811 |
return ret | VM_FAULT_RETRY; |
d88c0922f Release page refe... |
1812 |
} |
b522c94da mm: filemap_fault... |
1813 1814 1815 1816 1817 1818 1819 1820 |
/* 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 Linux-2.6.12-rc2 |
1821 |
/* |
d00806b18 mm: fix fault vs ... |
1822 1823 |
* 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 Linux-2.6.12-rc2 |
1824 |
*/ |
d00806b18 mm: fix fault vs ... |
1825 |
if (unlikely(!PageUptodate(page))) |
1da177e4c Linux-2.6.12-rc2 |
1826 |
goto page_not_uptodate; |
ef00e08e2 readahead: clean ... |
1827 1828 1829 1830 |
/* * Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
d00806b18 mm: fix fault vs ... |
1831 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2 readahead: clean ... |
1832 |
if (unlikely(offset >= size)) { |
d00806b18 mm: fix fault vs ... |
1833 |
unlock_page(page); |
745ad48e8 fix page release ... |
1834 |
page_cache_release(page); |
5307cc1aa Remove broken ptr... |
1835 |
return VM_FAULT_SIGBUS; |
d00806b18 mm: fix fault vs ... |
1836 |
} |
d0217ac04 mm: fault feedbac... |
1837 |
vmf->page = page; |
83c54070e mm: fault feedbac... |
1838 |
return ret | VM_FAULT_LOCKED; |
1da177e4c Linux-2.6.12-rc2 |
1839 |
|
1da177e4c Linux-2.6.12-rc2 |
1840 1841 1842 1843 1844 |
no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ |
ef00e08e2 readahead: clean ... |
1845 |
error = page_cache_read(file, offset); |
1da177e4c Linux-2.6.12-rc2 |
1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 |
/* * 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 mm: fault feedbac... |
1861 1862 |
return VM_FAULT_OOM; return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
1863 1864 |
page_not_uptodate: |
1da177e4c Linux-2.6.12-rc2 |
1865 1866 1867 1868 1869 1870 |
/* * 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 Linux-2.6.12-rc2 |
1871 |
ClearPageError(page); |
994fc28c7 [PATCH] add AOP_T... |
1872 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e mm: fix infinite ... |
1873 1874 1875 1876 1877 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
d00806b18 mm: fix fault vs ... |
1878 1879 1880 |
page_cache_release(page); if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7 [PATCH] add AOP_T... |
1881 |
goto retry_find; |
1da177e4c Linux-2.6.12-rc2 |
1882 |
|
d00806b18 mm: fix fault vs ... |
1883 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96 [PATCH] readahead... |
1884 |
shrink_readahead_size_eio(file, ra); |
d0217ac04 mm: fault feedbac... |
1885 |
return VM_FAULT_SIGBUS; |
54cb8821d mm: merge populat... |
1886 1887 |
} EXPORT_SYMBOL(filemap_fault); |
4fcf1c620 mm: Make default ... |
1888 1889 1890 |
int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) { struct page *page = vmf->page; |
496ad9aa8 new helper: file_... |
1891 |
struct inode *inode = file_inode(vma->vm_file); |
4fcf1c620 mm: Make default ... |
1892 |
int ret = VM_FAULT_LOCKED; |
14da92001 fs: Protect write... |
1893 |
sb_start_pagefault(inode->i_sb); |
4fcf1c620 mm: Make default ... |
1894 1895 1896 1897 1898 1899 1900 |
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 fs: Protect write... |
1901 1902 1903 1904 1905 1906 |
/* * 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); |
1d1d1a767 mm: only enforce ... |
1907 |
wait_for_stable_page(page); |
4fcf1c620 mm: Make default ... |
1908 |
out: |
14da92001 fs: Protect write... |
1909 |
sb_end_pagefault(inode->i_sb); |
4fcf1c620 mm: Make default ... |
1910 1911 1912 |
return ret; } EXPORT_SYMBOL(filemap_page_mkwrite); |
f0f37e2f7 const: mark struc... |
1913 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d mm: merge populat... |
1914 |
.fault = filemap_fault, |
4fcf1c620 mm: Make default ... |
1915 |
.page_mkwrite = filemap_page_mkwrite, |
0b173bc4d mm: kill vma flag... |
1916 |
.remap_pages = generic_file_remap_pages, |
1da177e4c Linux-2.6.12-rc2 |
1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 |
}; /* 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; return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 |
/* * 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); |
4fb08e5ab mm: remove read_c... |
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 |
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; } |
6fe6900e1 mm: make read_cac... |
1965 |
static struct page *__read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
1966 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
1967 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
1968 1969 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
1970 |
{ |
eb2be1893 mm: buffered writ... |
1971 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1972 1973 1974 1975 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
0531b2aac mm: add new 'read... |
1976 |
page = __page_cache_alloc(gfp | __GFP_COLD); |
eb2be1893 mm: buffered writ... |
1977 1978 |
if (!page) return ERR_PTR(-ENOMEM); |
e6f67b8c0 vfs: __read_cache... |
1979 |
err = add_to_page_cache_lru(page, mapping, index, gfp); |
eb2be1893 mm: buffered writ... |
1980 1981 1982 1983 |
if (unlikely(err)) { page_cache_release(page); if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1984 |
/* Presumably ENOMEM for radix tree node */ |
1da177e4c Linux-2.6.12-rc2 |
1985 1986 |
return ERR_PTR(err); } |
1da177e4c Linux-2.6.12-rc2 |
1987 1988 1989 1990 |
err = filler(data, page); if (err < 0) { page_cache_release(page); page = ERR_PTR(err); |
4fb08e5ab mm: remove read_c... |
1991 1992 |
} else { page = wait_on_page_read(page); |
1da177e4c Linux-2.6.12-rc2 |
1993 1994 |
} } |
1da177e4c Linux-2.6.12-rc2 |
1995 1996 |
return page; } |
0531b2aac mm: add new 'read... |
1997 |
static struct page *do_read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
1998 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
1999 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2000 2001 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
2002 2003 2004 2005 2006 |
{ struct page *page; int err; retry: |
0531b2aac mm: add new 'read... |
2007 |
page = __read_cache_page(mapping, index, filler, data, gfp); |
1da177e4c Linux-2.6.12-rc2 |
2008 |
if (IS_ERR(page)) |
c855ff371 Fix a bad error c... |
2009 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 |
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 Fix a bad error c... |
2026 |
return ERR_PTR(err); |
4fb08e5ab mm: remove read_c... |
2027 2028 2029 2030 |
} else { page = wait_on_page_read(page); if (IS_ERR(page)) return page; |
1da177e4c Linux-2.6.12-rc2 |
2031 |
} |
c855ff371 Fix a bad error c... |
2032 |
out: |
6fe6900e1 mm: make read_cac... |
2033 2034 2035 |
mark_page_accessed(page); return page; } |
0531b2aac mm: add new 'read... |
2036 2037 |
/** |
4fb08e5ab mm: remove read_c... |
2038 |
* read_cache_page - read into page cache, fill it if needed |
0531b2aac mm: add new 'read... |
2039 2040 2041 |
* @mapping: the page's address_space * @index: the page index * @filler: function to perform the read |
5e5358e7c mm: cleanup descr... |
2042 |
* @data: first arg to filler(data, page) function, often left as NULL |
0531b2aac mm: add new 'read... |
2043 |
* |
0531b2aac mm: add new 'read... |
2044 |
* Read into the page cache. If a page already exists, and PageUptodate() is |
4fb08e5ab mm: remove read_c... |
2045 |
* not set, try to fill the page and wait for it to become unlocked. |
0531b2aac mm: add new 'read... |
2046 2047 2048 |
* * If the page does not get brought uptodate, return -EIO. */ |
4fb08e5ab mm: remove read_c... |
2049 |
struct page *read_cache_page(struct address_space *mapping, |
0531b2aac mm: add new 'read... |
2050 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2051 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2052 2053 2054 2055 |
void *data) { return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); } |
4fb08e5ab mm: remove read_c... |
2056 |
EXPORT_SYMBOL(read_cache_page); |
0531b2aac mm: add new 'read... |
2057 2058 2059 2060 2061 2062 2063 2064 |
/** * 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 vfs: __read_cache... |
2065 |
* any new page allocations done using the specified allocation flags. |
0531b2aac mm: add new 'read... |
2066 2067 2068 2069 2070 2071 2072 2073 |
* * 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; |
4fb08e5ab mm: remove read_c... |
2074 |
return do_read_cache_page(mapping, index, filler, NULL, gfp); |
0531b2aac mm: add new 'read... |
2075 2076 |
} EXPORT_SYMBOL(read_cache_page_gfp); |
2f718ffc1 mm: buffered writ... |
2077 |
static size_t __iovec_copy_from_user_inatomic(char *vaddr, |
1da177e4c Linux-2.6.12-rc2 |
2078 2079 |
const struct iovec *iov, size_t base, size_t bytes) { |
f18005369 x86, mm: dont use... |
2080 |
size_t copied = 0, left = 0; |
1da177e4c Linux-2.6.12-rc2 |
2081 2082 2083 2084 2085 2086 |
while (bytes) { char __user *buf = iov->iov_base + base; int copy = min(bytes, iov->iov_len - base); base = 0; |
f18005369 x86, mm: dont use... |
2087 |
left = __copy_from_user_inatomic(vaddr, buf, copy); |
1da177e4c Linux-2.6.12-rc2 |
2088 2089 2090 2091 |
copied += copy; bytes -= copy; vaddr += copy; iov++; |
01408c493 [PATCH] Prepare f... |
2092 |
if (unlikely(left)) |
1da177e4c Linux-2.6.12-rc2 |
2093 |
break; |
1da177e4c Linux-2.6.12-rc2 |
2094 2095 2096 2097 2098 |
} return copied - left; } /* |
2f718ffc1 mm: buffered writ... |
2099 |
* Copy as much as we can into the page and return the number of bytes which |
af901ca18 tree-wide: fix as... |
2100 |
* were successfully copied. If a fault is encountered then return the number of |
2f718ffc1 mm: buffered writ... |
2101 2102 2103 2104 2105 2106 2107 |
* 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; |
9b04c5fec mm: remove the se... |
2108 |
kaddr = kmap_atomic(page); |
2f718ffc1 mm: buffered writ... |
2109 2110 2111 |
if (likely(i->nr_segs == 1)) { int left; char __user *buf = i->iov->iov_base + i->iov_offset; |
f18005369 x86, mm: dont use... |
2112 |
left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); |
2f718ffc1 mm: buffered writ... |
2113 2114 2115 2116 2117 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } |
9b04c5fec mm: remove the se... |
2118 |
kunmap_atomic(kaddr); |
2f718ffc1 mm: buffered writ... |
2119 2120 2121 |
return copied; } |
89e107877 fs: new cont helpers |
2122 |
EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); |
2f718ffc1 mm: buffered writ... |
2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 |
/* * 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 x86, mm: dont use... |
2140 |
left = __copy_from_user(kaddr + offset, buf, bytes); |
2f718ffc1 mm: buffered writ... |
2141 2142 2143 2144 2145 2146 2147 2148 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } kunmap(page); return copied; } |
89e107877 fs: new cont helpers |
2149 |
EXPORT_SYMBOL(iov_iter_copy_from_user); |
2f718ffc1 mm: buffered writ... |
2150 |
|
f7009264c iov_iter_advance(... |
2151 |
void iov_iter_advance(struct iov_iter *i, size_t bytes) |
2f718ffc1 mm: buffered writ... |
2152 |
{ |
f7009264c iov_iter_advance(... |
2153 |
BUG_ON(i->count < bytes); |
2f718ffc1 mm: buffered writ... |
2154 2155 |
if (likely(i->nr_segs == 1)) { i->iov_offset += bytes; |
f7009264c iov_iter_advance(... |
2156 |
i->count -= bytes; |
2f718ffc1 mm: buffered writ... |
2157 2158 2159 |
} else { const struct iovec *iov = i->iov; size_t base = i->iov_offset; |
39be79c16 vfs: iov_iter: ha... |
2160 |
unsigned long nr_segs = i->nr_segs; |
2f718ffc1 mm: buffered writ... |
2161 |
|
124d3b704 fix writev regres... |
2162 2163 |
/* * The !iov->iov_len check ensures we skip over unlikely |
f7009264c iov_iter_advance(... |
2164 |
* zero-length segments (without overruning the iovec). |
124d3b704 fix writev regres... |
2165 |
*/ |
94ad374a0 Fix off-by-one er... |
2166 |
while (bytes || unlikely(i->count && !iov->iov_len)) { |
f7009264c iov_iter_advance(... |
2167 |
int copy; |
2f718ffc1 mm: buffered writ... |
2168 |
|
f7009264c iov_iter_advance(... |
2169 2170 2171 |
copy = min(bytes, iov->iov_len - base); BUG_ON(!i->count || i->count < copy); i->count -= copy; |
2f718ffc1 mm: buffered writ... |
2172 2173 2174 2175 |
bytes -= copy; base += copy; if (iov->iov_len == base) { iov++; |
39be79c16 vfs: iov_iter: ha... |
2176 |
nr_segs--; |
2f718ffc1 mm: buffered writ... |
2177 2178 2179 2180 2181 |
base = 0; } } i->iov = iov; i->iov_offset = base; |
39be79c16 vfs: iov_iter: ha... |
2182 |
i->nr_segs = nr_segs; |
2f718ffc1 mm: buffered writ... |
2183 2184 |
} } |
89e107877 fs: new cont helpers |
2185 |
EXPORT_SYMBOL(iov_iter_advance); |
2f718ffc1 mm: buffered writ... |
2186 |
|
afddba49d fs: introduce wri... |
2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 |
/* * 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 mm: buffered writ... |
2197 |
{ |
2f718ffc1 mm: buffered writ... |
2198 |
char __user *buf = i->iov->iov_base + i->iov_offset; |
afddba49d fs: introduce wri... |
2199 2200 |
bytes = min(bytes, i->iov->iov_len - i->iov_offset); return fault_in_pages_readable(buf, bytes); |
2f718ffc1 mm: buffered writ... |
2201 |
} |
89e107877 fs: new cont helpers |
2202 |
EXPORT_SYMBOL(iov_iter_fault_in_readable); |
2f718ffc1 mm: buffered writ... |
2203 2204 2205 2206 |
/* * Return the count of just the current iov_iter segment. */ |
d28574e04 mm: minor cleanup... |
2207 |
size_t iov_iter_single_seg_count(const struct iov_iter *i) |
2f718ffc1 mm: buffered writ... |
2208 2209 2210 2211 2212 2213 2214 |
{ 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 fs: new cont helpers |
2215 |
EXPORT_SYMBOL(iov_iter_single_seg_count); |
2f718ffc1 mm: buffered writ... |
2216 2217 |
/* |
1da177e4c Linux-2.6.12-rc2 |
2218 2219 |
* Performs necessary checks before doing a write * |
485bb99b4 [PATCH] kernel-do... |
2220 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c Linux-2.6.12-rc2 |
2221 2222 2223 2224 2225 2226 |
* 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 mm: use rlimit he... |
2227 |
unsigned long limit = rlimit(RLIMIT_FSIZE); |
1da177e4c Linux-2.6.12-rc2 |
2228 2229 2230 |
if (unlikely(*pos < 0)) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 |
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 Linux-2.6.12-rc2 |
2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 |
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 Linux-2.6.12-rc2 |
2270 2271 2272 2273 2274 2275 2276 2277 |
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 [PATCH] BLOCK: Ma... |
2278 |
#ifdef CONFIG_BLOCK |
1da177e4c Linux-2.6.12-rc2 |
2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 |
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 [PATCH] BLOCK: Ma... |
2290 2291 2292 |
#else return -EPERM; #endif |
1da177e4c Linux-2.6.12-rc2 |
2293 2294 2295 2296 |
} return 0; } EXPORT_SYMBOL(generic_write_checks); |
afddba49d fs: introduce wri... |
2297 2298 2299 2300 2301 |
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 fs: remove prepar... |
2302 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d fs: introduce wri... |
2303 |
pagep, fsdata); |
afddba49d fs: introduce wri... |
2304 2305 2306 2307 2308 2309 2310 2311 |
} 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 fs: introduce wri... |
2312 |
|
4e02ed4b4 fs: remove prepar... |
2313 |
return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d fs: introduce wri... |
2314 2315 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c Linux-2.6.12-rc2 |
2316 2317 2318 2319 2320 2321 2322 2323 2324 |
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 kill generic_file... |
2325 2326 |
size_t write_len; pgoff_t end; |
1da177e4c Linux-2.6.12-rc2 |
2327 2328 2329 |
if (count != ocount) *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); |
a969e903a kill generic_file... |
2330 2331 |
write_len = iov_length(iov, *nr_segs); end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; |
a969e903a kill generic_file... |
2332 |
|
48b47c561 mm: direct IO sta... |
2333 |
written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); |
a969e903a kill generic_file... |
2334 2335 2336 2337 2338 2339 2340 |
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 VFS: fix dio writ... |
2341 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a kill generic_file... |
2342 2343 2344 2345 |
*/ if (mapping->nrpages) { written = invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); |
6ccfa806a VFS: fix dio writ... |
2346 2347 2348 2349 2350 2351 2352 |
/* * If a page can not be invalidated, return 0 to fall back * to buffered write. */ if (written) { if (written == -EBUSY) return 0; |
a969e903a kill generic_file... |
2353 |
goto out; |
6ccfa806a VFS: fix dio writ... |
2354 |
} |
a969e903a kill generic_file... |
2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 |
} 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 Linux-2.6.12-rc2 |
2371 |
if (written > 0) { |
0116651c8 mm: remove tempor... |
2372 2373 2374 |
pos += written; if (pos > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, pos); |
1da177e4c Linux-2.6.12-rc2 |
2375 2376 |
mark_inode_dirty(inode); } |
0116651c8 mm: remove tempor... |
2377 |
*ppos = pos; |
1da177e4c Linux-2.6.12-rc2 |
2378 |
} |
a969e903a kill generic_file... |
2379 |
out: |
1da177e4c Linux-2.6.12-rc2 |
2380 2381 2382 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893 mm: buffered writ... |
2383 2384 2385 2386 |
/* * Find or create a page at the given pagecache position. Return the locked * page. This function is specifically for buffered writes. */ |
54566b2c1 fs: symlink write... |
2387 2388 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893 mm: buffered writ... |
2389 |
{ |
eb2be1893 mm: buffered writ... |
2390 |
struct page *page; |
d618a27c7 mm: non-atomicall... |
2391 |
int fgp_flags = FGP_LOCK|FGP_ACCESSED|FGP_WRITE|FGP_CREAT; |
0faa70cb0 mm: filemap: pass... |
2392 |
|
54566b2c1 fs: symlink write... |
2393 |
if (flags & AOP_FLAG_NOFS) |
d618a27c7 mm: non-atomicall... |
2394 2395 2396 2397 2398 |
fgp_flags |= FGP_NOFS; page = pagecache_get_page(mapping, index, fgp_flags, mapping_gfp_mask(mapping), GFP_KERNEL); |
c585a2678 mm: remove likely... |
2399 |
if (page) |
d618a27c7 mm: non-atomicall... |
2400 |
wait_for_stable_page(page); |
eb2be1893 mm: buffered writ... |
2401 |
|
eb2be1893 mm: buffered writ... |
2402 2403 |
return page; } |
54566b2c1 fs: symlink write... |
2404 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893 mm: buffered writ... |
2405 |
|
afddba49d fs: introduce wri... |
2406 2407 2408 2409 2410 2411 2412 |
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 mm: restore KERNE... |
2413 2414 2415 2416 2417 2418 2419 |
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 fs: introduce wri... |
2420 2421 2422 |
do { struct page *page; |
afddba49d fs: introduce wri... |
2423 2424 2425 2426 2427 2428 |
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 fs: introduce wri... |
2429 2430 2431 2432 |
bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_count(i)); again: |
afddba49d fs: introduce wri... |
2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 |
/* * 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 mm: restore KERNE... |
2447 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d fs: introduce wri... |
2448 |
&page, &fsdata); |
d618a27c7 mm: non-atomicall... |
2449 |
if (unlikely(status < 0)) |
afddba49d fs: introduce wri... |
2450 |
break; |
931e80e4b mm: flush dcache ... |
2451 2452 |
if (mapping_writably_mapped(mapping)) flush_dcache_page(page); |
afddba49d fs: introduce wri... |
2453 |
copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); |
afddba49d fs: introduce wri... |
2454 2455 2456 2457 2458 2459 2460 2461 2462 |
flush_dcache_page(page); status = a_ops->write_end(file, mapping, pos, bytes, copied, page, fsdata); if (unlikely(status < 0)) break; copied = status; cond_resched(); |
124d3b704 fix writev regres... |
2463 |
iov_iter_advance(i, copied); |
afddba49d fs: introduce wri... |
2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 |
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 fs: introduce wri... |
2477 2478 2479 2480 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); |
a50527b19 fs: Make write(2)... |
2481 2482 2483 2484 |
if (fatal_signal_pending(current)) { status = -EINTR; break; } |
afddba49d fs: introduce wri... |
2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 |
} 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 fs: introduce wri... |
2496 2497 2498 2499 |
ssize_t status; struct iov_iter i; iov_iter_init(&i, iov, nr_segs, count, written); |
4e02ed4b4 fs: remove prepar... |
2500 |
status = generic_perform_write(file, &i, pos); |
1da177e4c Linux-2.6.12-rc2 |
2501 |
|
1da177e4c Linux-2.6.12-rc2 |
2502 |
if (likely(status >= 0)) { |
afddba49d fs: introduce wri... |
2503 2504 |
written += status; *ppos = pos + status; |
1da177e4c Linux-2.6.12-rc2 |
2505 2506 |
} |
1da177e4c Linux-2.6.12-rc2 |
2507 2508 2509 |
return written ? written : status; } EXPORT_SYMBOL(generic_file_buffered_write); |
e4dd9de3c vfs: Export __gen... |
2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 |
/** * __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 Linux-2.6.12-rc2 |
2531 2532 |
{ struct file *file = iocb->ki_filp; |
fb5527e68 [PATCH] direct-io... |
2533 |
struct address_space * mapping = file->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
2534 2535 2536 |
size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
2537 2538 2539 2540 2541 |
loff_t pos; ssize_t written; ssize_t err; ocount = 0; |
0ceb33143 mm: move common s... |
2542 2543 2544 |
err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (err) return err; |
1da177e4c Linux-2.6.12-rc2 |
2545 2546 2547 |
count = ocount; pos = *ppos; |
1da177e4c Linux-2.6.12-rc2 |
2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 |
/* 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 [patch 3/5] vfs: ... |
2558 |
err = file_remove_suid(file); |
1da177e4c Linux-2.6.12-rc2 |
2559 2560 |
if (err) goto out; |
c3b2da314 fs: introduce ino... |
2561 2562 2563 |
err = file_update_time(file); if (err) goto out; |
1da177e4c Linux-2.6.12-rc2 |
2564 2565 2566 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { |
fb5527e68 [PATCH] direct-io... |
2567 2568 2569 2570 2571 |
loff_t endbyte; ssize_t written_buffered; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, ppos, count, ocount); |
1da177e4c Linux-2.6.12-rc2 |
2572 2573 2574 2575 2576 2577 2578 2579 |
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 [PATCH] direct-io... |
2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 |
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 Linux-2.6.12-rc2 |
2594 |
|
fb5527e68 [PATCH] direct-io... |
2595 2596 2597 2598 2599 2600 |
/* * 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 direct I/O fallba... |
2601 |
err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); |
fb5527e68 [PATCH] direct-io... |
2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 |
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 Linux-2.6.12-rc2 |
2617 2618 2619 2620 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
e4dd9de3c vfs: Export __gen... |
2621 |
EXPORT_SYMBOL(__generic_file_aio_write); |
e4dd9de3c vfs: Export __gen... |
2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 |
/** * 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 [PATCH] Vectorize... |
2633 2634 |
ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c Linux-2.6.12-rc2 |
2635 2636 |
{ struct file *file = iocb->ki_filp; |
148f948ba vfs: Introduce ne... |
2637 |
struct inode *inode = file->f_mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
2638 |
ssize_t ret; |
1da177e4c Linux-2.6.12-rc2 |
2639 2640 |
BUG_ON(iocb->ki_pos != pos); |
1b1dcc1b5 [PATCH] mutex sub... |
2641 |
mutex_lock(&inode->i_mutex); |
e4dd9de3c vfs: Export __gen... |
2642 |
ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); |
1b1dcc1b5 [PATCH] mutex sub... |
2643 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2644 |
|
02afc27fa direct-io: Handle... |
2645 |
if (ret > 0) { |
1da177e4c Linux-2.6.12-rc2 |
2646 |
ssize_t err; |
148f948ba vfs: Introduce ne... |
2647 |
err = generic_write_sync(file, pos, ret); |
c7b50db21 vfs: Remove synci... |
2648 |
if (err < 0 && ret > 0) |
1da177e4c Linux-2.6.12-rc2 |
2649 2650 2651 2652 2653 |
ret = err; } return ret; } EXPORT_SYMBOL(generic_file_aio_write); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 |
/** * 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 FS-Cache: Recruit... |
2664 2665 2666 |
* 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 [PATCH] BLOCK: Mo... |
2667 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
3f31fddfa jbd: fix race bet... |
2668 |
* this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS). |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2669 |
* |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 |
*/ 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); |