Blame view
mm/filemap.c
72 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> |
f1820361f mm: implement ->m... |
36 |
#include <linux/rmap.h> |
0f8053a50 [PATCH] mm: make ... |
37 |
#include "internal.h" |
fe0bfaaff mm: trace filemap... |
38 39 |
#define CREATE_TRACE_POINTS #include <trace/events/filemap.h> |
1da177e4c Linux-2.6.12-rc2 |
40 |
/* |
1da177e4c Linux-2.6.12-rc2 |
41 42 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba vfs: Introduce ne... |
43 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c Linux-2.6.12-rc2 |
44 |
|
1da177e4c Linux-2.6.12-rc2 |
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 |
#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... |
62 |
* ->i_mmap_mutex (truncate_pagecache) |
1da177e4c Linux-2.6.12-rc2 |
63 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919 [PATCH] swap: swa... |
64 65 |
* ->swap_lock (exclusive_swap_page, others) * ->mapping->tree_lock |
1da177e4c Linux-2.6.12-rc2 |
66 |
* |
1b1dcc1b5 [PATCH] mutex sub... |
67 |
* ->i_mutex |
3d48ae45e mm: Convert i_mma... |
68 |
* ->i_mmap_mutex (truncate->unmap_mapping_range) |
1da177e4c Linux-2.6.12-rc2 |
69 70 |
* * ->mmap_sem |
3d48ae45e mm: Convert i_mma... |
71 |
* ->i_mmap_mutex |
b8072f099 [PATCH] mm: updat... |
72 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
1da177e4c Linux-2.6.12-rc2 |
73 74 75 76 77 |
* ->mapping->tree_lock (arch-dependent flush_dcache_mmap_lock) * * ->mmap_sem * ->lock_page (access_process_vm) * |
ccad23656 kill generic_file... |
78 |
* ->i_mutex (generic_perform_write) |
82591e6ea [PATCH] mm: more ... |
79 |
* ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c Linux-2.6.12-rc2 |
80 |
* |
f758eeabe writeback: split ... |
81 |
* bdi->wb.list_lock |
a66979aba fs: move i_wb_lis... |
82 |
* sb_lock (fs/fs-writeback.c) |
1da177e4c Linux-2.6.12-rc2 |
83 84 |
* ->mapping->tree_lock (__sync_single_inode) * |
3d48ae45e mm: Convert i_mma... |
85 |
* ->i_mmap_mutex |
1da177e4c Linux-2.6.12-rc2 |
86 87 88 |
* ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099 [PATCH] mm: updat... |
89 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c Linux-2.6.12-rc2 |
90 |
* |
b8072f099 [PATCH] mm: updat... |
91 |
* ->page_table_lock or pte_lock |
5d337b919 [PATCH] swap: swa... |
92 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c Linux-2.6.12-rc2 |
93 94 95 |
* ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) |
053837fce [PATCH] mm: migra... |
96 |
* ->zone.lru_lock (check_pte_range->isolate_lru_page) |
1da177e4c Linux-2.6.12-rc2 |
97 98 |
* ->private_lock (page_remove_rmap->set_page_dirty) * ->tree_lock (page_remove_rmap->set_page_dirty) |
f758eeabe writeback: split ... |
99 |
* bdi.wb->list_lock (page_remove_rmap->set_page_dirty) |
250df6ed2 fs: protect inode... |
100 |
* ->inode->i_lock (page_remove_rmap->set_page_dirty) |
f758eeabe writeback: split ... |
101 |
* bdi.wb->list_lock (zap_pte_range->set_page_dirty) |
250df6ed2 fs: protect inode... |
102 |
* ->inode->i_lock (zap_pte_range->set_page_dirty) |
1da177e4c Linux-2.6.12-rc2 |
103 104 |
* ->private_lock (zap_pte_range->__set_page_dirty_buffers) * |
9a3c531df mm: update stale ... |
105 106 |
* ->i_mmap_mutex * ->tasklist_lock (memory_failure, collect_procs_ao) |
1da177e4c Linux-2.6.12-rc2 |
107 |
*/ |
91b0abe36 mm + fs: store sh... |
108 109 110 |
static void page_cache_tree_delete(struct address_space *mapping, struct page *page, void *shadow) { |
449dd6984 mm: keep page cac... |
111 112 113 114 115 |
struct radix_tree_node *node; unsigned long index; unsigned int offset; unsigned int tag; void **slot; |
91b0abe36 mm + fs: store sh... |
116 |
|
449dd6984 mm: keep page cac... |
117 118 119 120 121 |
VM_BUG_ON(!PageLocked(page)); __radix_tree_lookup(&mapping->page_tree, page->index, &node, &slot); if (shadow) { |
91b0abe36 mm + fs: store sh... |
122 123 124 125 126 127 128 129 |
mapping->nrshadows++; /* * Make sure the nrshadows update is committed before * the nrpages update so that final truncate racing * with reclaim does not see both counters 0 at the * same time and miss a shadow entry. */ smp_wmb(); |
449dd6984 mm: keep page cac... |
130 |
} |
91b0abe36 mm + fs: store sh... |
131 |
mapping->nrpages--; |
449dd6984 mm: keep page cac... |
132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 |
if (!node) { /* Clear direct pointer tags in root node */ mapping->page_tree.gfp_mask &= __GFP_BITS_MASK; radix_tree_replace_slot(slot, shadow); return; } /* Clear tree tags for the removed page */ index = page->index; offset = index & RADIX_TREE_MAP_MASK; for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) { if (test_bit(offset, node->tags[tag])) radix_tree_tag_clear(&mapping->page_tree, index, tag); } /* Delete page, swap shadow entry */ radix_tree_replace_slot(slot, shadow); workingset_node_pages_dec(node); if (shadow) workingset_node_shadows_inc(node); else if (__radix_tree_delete_node(&mapping->page_tree, node)) return; /* * Track node that only contains shadow entries. * * Avoid acquiring the list_lru lock if already tracked. The * list_empty() test is safe as node->private_list is * protected by mapping->tree_lock. */ if (!workingset_node_pages(node) && list_empty(&node->private_list)) { node->private_data = mapping; list_lru_add(&workingset_shadow_nodes, &node->private_list); } |
91b0abe36 mm + fs: store sh... |
169 |
} |
1da177e4c Linux-2.6.12-rc2 |
170 |
/* |
e64a782fe mm: change __remo... |
171 |
* Delete a page from the page cache and free it. Caller has to make |
1da177e4c Linux-2.6.12-rc2 |
172 |
* sure the page is locked and that nobody else uses it - or that usage |
19fd62312 mm: spinlock tree... |
173 |
* is safe. The caller must hold the mapping's tree_lock. |
1da177e4c Linux-2.6.12-rc2 |
174 |
*/ |
91b0abe36 mm + fs: store sh... |
175 |
void __delete_from_page_cache(struct page *page, void *shadow) |
1da177e4c Linux-2.6.12-rc2 |
176 177 |
{ struct address_space *mapping = page->mapping; |
fe0bfaaff mm: trace filemap... |
178 |
trace_mm_filemap_delete_from_page_cache(page); |
c515e1fd3 mm/fs: add hooks ... |
179 180 181 182 183 184 185 186 |
/* * 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... |
187 |
cleancache_invalidate_page(mapping, page); |
c515e1fd3 mm/fs: add hooks ... |
188 |
|
91b0abe36 mm + fs: store sh... |
189 |
page_cache_tree_delete(mapping, page, shadow); |
1da177e4c Linux-2.6.12-rc2 |
190 |
page->mapping = NULL; |
b85e0effd mm: consistent tr... |
191 |
/* Leave page->index set: truncation lookup relies upon it */ |
91b0abe36 mm + fs: store sh... |
192 |
|
347ce434d [PATCH] zoned vm ... |
193 |
__dec_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac mm: oom analysis:... |
194 195 |
if (PageSwapBacked(page)) __dec_zone_page_state(page, NR_SHMEM); |
45426812d mm: debug check f... |
196 |
BUG_ON(page_mapped(page)); |
3a6927906 Do dirty page acc... |
197 198 199 200 201 202 203 204 205 206 207 208 |
/* * 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 |
209 |
} |
702cfbf93 mm: goodbye remov... |
210 211 212 213 214 215 216 217 218 |
/** * 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 |
219 220 |
{ struct address_space *mapping = page->mapping; |
6072d13c4 Call the filesyst... |
221 |
void (*freepage)(struct page *); |
1da177e4c Linux-2.6.12-rc2 |
222 |
|
cd7619d6b [PATCH] Extermina... |
223 |
BUG_ON(!PageLocked(page)); |
1da177e4c Linux-2.6.12-rc2 |
224 |
|
6072d13c4 Call the filesyst... |
225 |
freepage = mapping->a_ops->freepage; |
19fd62312 mm: spinlock tree... |
226 |
spin_lock_irq(&mapping->tree_lock); |
91b0abe36 mm + fs: store sh... |
227 |
__delete_from_page_cache(page, NULL); |
19fd62312 mm: spinlock tree... |
228 |
spin_unlock_irq(&mapping->tree_lock); |
e767e0561 memcg: fix deadlo... |
229 |
mem_cgroup_uncharge_cache_page(page); |
6072d13c4 Call the filesyst... |
230 231 232 |
if (freepage) freepage(page); |
97cecb5a2 mm: introduce del... |
233 234 235 |
page_cache_release(page); } EXPORT_SYMBOL(delete_from_page_cache); |
7eaceacca block: remove per... |
236 |
static int sleep_on_page(void *word) |
1da177e4c Linux-2.6.12-rc2 |
237 |
{ |
1da177e4c Linux-2.6.12-rc2 |
238 239 240 |
io_schedule(); return 0; } |
7eaceacca block: remove per... |
241 |
static int sleep_on_page_killable(void *word) |
2687a3569 Add lock_page_kil... |
242 |
{ |
7eaceacca block: remove per... |
243 |
sleep_on_page(word); |
2687a3569 Add lock_page_kil... |
244 245 |
return fatal_signal_pending(current) ? -EINTR : 0; } |
865ffef37 fs: fix fsync() e... |
246 247 248 249 250 251 252 253 254 255 |
static int filemap_check_errors(struct address_space *mapping) { int ret = 0; /* Check for outstanding write errors */ if (test_and_clear_bit(AS_ENOSPC, &mapping->flags)) ret = -ENOSPC; if (test_and_clear_bit(AS_EIO, &mapping->flags)) ret = -EIO; return ret; } |
1da177e4c Linux-2.6.12-rc2 |
256 |
/** |
485bb99b4 [PATCH] kernel-do... |
257 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b [PATCH] DocBook: ... |
258 259 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03 [PATCH] filemap_f... |
260 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b [PATCH] DocBook: ... |
261 |
* @sync_mode: enable synchronous operation |
1da177e4c Linux-2.6.12-rc2 |
262 |
* |
485bb99b4 [PATCH] kernel-do... |
263 264 265 |
* 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 |
266 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4 [PATCH] kernel-do... |
267 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c Linux-2.6.12-rc2 |
268 269 270 |
* 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()... |
271 272 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c Linux-2.6.12-rc2 |
273 274 275 276 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd mm: write_cache_p... |
277 |
.nr_to_write = LONG_MAX, |
111ebb6e6 [PATCH] writeback... |
278 279 |
.range_start = start, .range_end = end, |
1da177e4c Linux-2.6.12-rc2 |
280 281 282 283 284 285 286 287 288 289 290 291 |
}; 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... |
292 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c Linux-2.6.12-rc2 |
293 294 295 296 297 298 299 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf vfs: export filem... |
300 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c [PATCH] fadvise()... |
301 |
loff_t end) |
1da177e4c Linux-2.6.12-rc2 |
302 303 304 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf vfs: export filem... |
305 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c Linux-2.6.12-rc2 |
306 |
|
485bb99b4 [PATCH] kernel-do... |
307 308 309 310 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c Linux-2.6.12-rc2 |
311 312 313 314 315 316 317 318 |
* 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... |
319 |
/** |
94004ed72 kill wait_on_page... |
320 321 322 323 |
* 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... |
324 |
* |
94004ed72 kill wait_on_page... |
325 326 |
* 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 |
327 |
*/ |
94004ed72 kill wait_on_page... |
328 329 |
int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) |
1da177e4c Linux-2.6.12-rc2 |
330 |
{ |
94004ed72 kill wait_on_page... |
331 332 |
pgoff_t index = start_byte >> PAGE_CACHE_SHIFT; pgoff_t end = end_byte >> PAGE_CACHE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
333 334 |
struct pagevec pvec; int nr_pages; |
865ffef37 fs: fix fsync() e... |
335 |
int ret2, ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
336 |
|
94004ed72 kill wait_on_page... |
337 |
if (end_byte < start_byte) |
865ffef37 fs: fix fsync() e... |
338 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
339 340 |
pagevec_init(&pvec, 0); |
1da177e4c Linux-2.6.12-rc2 |
341 342 343 344 345 346 347 348 349 350 351 352 353 354 |
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... |
355 |
if (TestClearPageError(page)) |
1da177e4c Linux-2.6.12-rc2 |
356 357 358 359 360 |
ret = -EIO; } pagevec_release(&pvec); cond_resched(); } |
865ffef37 fs: fix fsync() e... |
361 362 363 364 |
out: ret2 = filemap_check_errors(mapping); if (!ret) ret = ret2; |
1da177e4c Linux-2.6.12-rc2 |
365 366 367 |
return ret; } |
d3bccb6f4 vfs: Introduce fi... |
368 369 370 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
485bb99b4 [PATCH] kernel-do... |
371 |
* filemap_fdatawait - wait for all under-writeback pages to complete |
1da177e4c Linux-2.6.12-rc2 |
372 |
* @mapping: address space structure to wait for |
485bb99b4 [PATCH] kernel-do... |
373 374 375 |
* * Walk the list of under-writeback pages of the given address space * and wait for all of them. |
1da177e4c Linux-2.6.12-rc2 |
376 377 378 379 380 381 382 |
*/ 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... |
383 |
return filemap_fdatawait_range(mapping, 0, i_size - 1); |
1da177e4c Linux-2.6.12-rc2 |
384 385 386 387 388 |
} EXPORT_SYMBOL(filemap_fdatawait); int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982 [PATCH] Fix and a... |
389 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
390 391 |
if (mapping->nrpages) { |
28fd12982 [PATCH] Fix and a... |
392 393 394 395 396 397 398 399 400 401 402 403 |
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... |
404 405 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
406 |
} |
28fd12982 [PATCH] Fix and a... |
407 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
408 |
} |
28fd12982 [PATCH] Fix and a... |
409 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c Linux-2.6.12-rc2 |
410 |
|
485bb99b4 [PATCH] kernel-do... |
411 412 413 414 415 416 |
/** * 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... |
417 418 419 420 421 |
* 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 |
422 423 424 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982 [PATCH] Fix and a... |
425 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
426 427 |
if (mapping->nrpages) { |
28fd12982 [PATCH] Fix and a... |
428 429 430 431 |
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... |
432 433 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982 [PATCH] Fix and a... |
434 435 436 |
if (!err) err = err2; } |
865ffef37 fs: fix fsync() e... |
437 438 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
439 |
} |
28fd12982 [PATCH] Fix and a... |
440 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
441 |
} |
f69955855 Export filemap_wr... |
442 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c Linux-2.6.12-rc2 |
443 |
|
485bb99b4 [PATCH] kernel-do... |
444 |
/** |
ef6a3c631 mm: add replace_p... |
445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 |
* 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... |
462 |
|
309381fea mm: dump page whe... |
463 464 465 |
VM_BUG_ON_PAGE(!PageLocked(old), old); VM_BUG_ON_PAGE(!PageLocked(new), new); VM_BUG_ON_PAGE(new->mapping, new); |
ef6a3c631 mm: add replace_p... |
466 |
|
ef6a3c631 mm: add replace_p... |
467 468 469 470 471 472 473 474 475 476 477 478 479 |
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); |
91b0abe36 mm + fs: store sh... |
480 |
__delete_from_page_cache(old, NULL); |
ef6a3c631 mm: add replace_p... |
481 482 483 484 485 486 487 |
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... |
488 489 |
/* mem_cgroup codes must not be called under tree_lock */ mem_cgroup_replace_page_cache(old, new); |
ef6a3c631 mm: add replace_p... |
490 491 492 493 |
radix_tree_preload_end(); if (freepage) freepage(old); page_cache_release(old); |
ef6a3c631 mm: add replace_p... |
494 495 496 497 498 |
} return error; } EXPORT_SYMBOL_GPL(replace_page_cache_page); |
0cd6144aa mm + fs: prepare ... |
499 |
static int page_cache_tree_insert(struct address_space *mapping, |
a528910e1 mm: thrash detect... |
500 |
struct page *page, void **shadowp) |
0cd6144aa mm + fs: prepare ... |
501 |
{ |
449dd6984 mm: keep page cac... |
502 |
struct radix_tree_node *node; |
0cd6144aa mm + fs: prepare ... |
503 504 |
void **slot; int error; |
449dd6984 mm: keep page cac... |
505 506 507 508 509 |
error = __radix_tree_create(&mapping->page_tree, page->index, &node, &slot); if (error) return error; if (*slot) { |
0cd6144aa mm + fs: prepare ... |
510 511 512 513 514 |
void *p; p = radix_tree_deref_slot_protected(slot, &mapping->tree_lock); if (!radix_tree_exceptional_entry(p)) return -EEXIST; |
a528910e1 mm: thrash detect... |
515 516 |
if (shadowp) *shadowp = p; |
449dd6984 mm: keep page cac... |
517 518 519 |
mapping->nrshadows--; if (node) workingset_node_shadows_dec(node); |
0cd6144aa mm + fs: prepare ... |
520 |
} |
449dd6984 mm: keep page cac... |
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 |
radix_tree_replace_slot(slot, page); mapping->nrpages++; if (node) { workingset_node_pages_inc(node); /* * Don't track node that contains actual pages. * * Avoid acquiring the list_lru lock if already * untracked. The list_empty() test is safe as * node->private_list is protected by * mapping->tree_lock. */ if (!list_empty(&node->private_list)) list_lru_del(&workingset_shadow_nodes, &node->private_list); } return 0; |
0cd6144aa mm + fs: prepare ... |
538 |
} |
a528910e1 mm: thrash detect... |
539 540 541 542 |
static int __add_to_page_cache_locked(struct page *page, struct address_space *mapping, pgoff_t offset, gfp_t gfp_mask, void **shadowp) |
1da177e4c Linux-2.6.12-rc2 |
543 |
{ |
e286781d5 mm: speculative p... |
544 |
int error; |
309381fea mm: dump page whe... |
545 546 |
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageSwapBacked(page), page); |
e286781d5 mm: speculative p... |
547 |
|
d715ae08f memcg: rename hig... |
548 |
error = mem_cgroup_charge_file(page, current->mm, |
2c26fdd70 memcg: revert gfp... |
549 |
gfp_mask & GFP_RECLAIM_MASK); |
35c754d79 memory controller... |
550 |
if (error) |
66a0c8ee3 mm: cleanup add_t... |
551 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
552 |
|
5e4c0d974 lib/radix-tree.c:... |
553 |
error = radix_tree_maybe_preload(gfp_mask & ~__GFP_HIGHMEM); |
66a0c8ee3 mm: cleanup add_t... |
554 |
if (error) { |
69029cd55 memcg: remove ref... |
555 |
mem_cgroup_uncharge_cache_page(page); |
66a0c8ee3 mm: cleanup add_t... |
556 557 558 559 560 561 562 563 |
return error; } page_cache_get(page); page->mapping = mapping; page->index = offset; spin_lock_irq(&mapping->tree_lock); |
a528910e1 mm: thrash detect... |
564 |
error = page_cache_tree_insert(mapping, page, shadowp); |
66a0c8ee3 mm: cleanup add_t... |
565 566 567 |
radix_tree_preload_end(); if (unlikely(error)) goto err_insert; |
66a0c8ee3 mm: cleanup add_t... |
568 569 570 571 572 573 574 575 576 577 |
__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 |
578 579 |
return error; } |
a528910e1 mm: thrash detect... |
580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 |
/** * add_to_page_cache_locked - add a locked page to the pagecache * @page: page to add * @mapping: the page's address_space * @offset: page index * @gfp_mask: page allocation mode * * This function is used to add a page to the pagecache. It must be locked. * This function does not add the page to the LRU. The caller must do that. */ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, pgoff_t offset, gfp_t gfp_mask) { return __add_to_page_cache_locked(page, mapping, offset, gfp_mask, NULL); } |
e286781d5 mm: speculative p... |
597 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c Linux-2.6.12-rc2 |
598 599 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
600 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
601 |
{ |
a528910e1 mm: thrash detect... |
602 |
void *shadow = NULL; |
4f98a2fee vmscan: split LRU... |
603 |
int ret; |
a528910e1 mm: thrash detect... |
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 |
__set_page_locked(page); ret = __add_to_page_cache_locked(page, mapping, offset, gfp_mask, &shadow); if (unlikely(ret)) __clear_page_locked(page); else { /* * The page might have been evicted from cache only * recently, in which case it should be activated like * any other repeatedly accessed page. */ if (shadow && workingset_refault(shadow)) { SetPageActive(page); workingset_activation(page); } else ClearPageActive(page); lru_cache_add(page); } |
1da177e4c Linux-2.6.12-rc2 |
622 623 |
return ret; } |
18bc0bbd1 Staging: pohmelfs... |
624 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c Linux-2.6.12-rc2 |
625 |
|
44110fe38 [PATCH] cpuset me... |
626 |
#ifdef CONFIG_NUMA |
2ae88149a [PATCH] mm: clean... |
627 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38 [PATCH] cpuset me... |
628 |
{ |
c0ff7453b cpuset,mm: fix no... |
629 630 |
int n; struct page *page; |
44110fe38 [PATCH] cpuset me... |
631 |
if (cpuset_do_page_mem_spread()) { |
cc9a6c877 cpuset: mm: reduc... |
632 633 |
unsigned int cpuset_mems_cookie; do { |
d26914d11 mm: optimize put_... |
634 |
cpuset_mems_cookie = read_mems_allowed_begin(); |
cc9a6c877 cpuset: mm: reduc... |
635 636 |
n = cpuset_mem_spread_node(); page = alloc_pages_exact_node(n, gfp, 0); |
d26914d11 mm: optimize put_... |
637 |
} while (!page && read_mems_allowed_retry(cpuset_mems_cookie)); |
cc9a6c877 cpuset: mm: reduc... |
638 |
|
c0ff7453b cpuset,mm: fix no... |
639 |
return page; |
44110fe38 [PATCH] cpuset me... |
640 |
} |
2ae88149a [PATCH] mm: clean... |
641 |
return alloc_pages(gfp, 0); |
44110fe38 [PATCH] cpuset me... |
642 |
} |
2ae88149a [PATCH] mm: clean... |
643 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38 [PATCH] cpuset me... |
644 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 |
/* * 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... |
666 |
void wait_on_page_bit(struct page *page, int bit_nr) |
1da177e4c Linux-2.6.12-rc2 |
667 668 669 670 |
{ DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); if (test_bit(bit_nr, &page->flags)) |
7eaceacca block: remove per... |
671 |
__wait_on_bit(page_waitqueue(page), &wait, sleep_on_page, |
1da177e4c Linux-2.6.12-rc2 |
672 673 674 |
TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(wait_on_page_bit); |
f62e00cc3 mm: introduce wai... |
675 676 677 678 679 680 681 682 683 684 |
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 |
685 |
/** |
385e1ca5f CacheFiles: Permi... |
686 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc filemap: fix kern... |
687 688 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f CacheFiles: Permi... |
689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 |
* * 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... |
704 |
* unlock_page - unlock a locked page |
1da177e4c Linux-2.6.12-rc2 |
705 706 707 708 709 710 711 |
* @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... |
712 713 |
* 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 |
714 |
*/ |
920c7a5d0 mm: remove fastca... |
715 |
void unlock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
716 |
{ |
309381fea mm: dump page whe... |
717 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
8413ac9d8 mm: page lock use... |
718 719 |
clear_bit_unlock(PG_locked, &page->flags); smp_mb__after_clear_bit(); |
1da177e4c Linux-2.6.12-rc2 |
720 721 722 |
wake_up_page(page, PG_locked); } EXPORT_SYMBOL(unlock_page); |
485bb99b4 [PATCH] kernel-do... |
723 724 725 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c Linux-2.6.12-rc2 |
726 727 728 |
*/ void end_page_writeback(struct page *page) { |
ac6aadb24 mm: rotate_reclai... |
729 730 731 732 733 |
if (TestClearPageReclaim(page)) rotate_reclaimable_page(page); if (!test_clear_page_writeback(page)) BUG(); |
1da177e4c Linux-2.6.12-rc2 |
734 735 736 737 |
smp_mb__after_clear_bit(); wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
485bb99b4 [PATCH] kernel-do... |
738 739 740 |
/** * __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 |
741 |
*/ |
920c7a5d0 mm: remove fastca... |
742 |
void __lock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
743 744 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); |
7eaceacca block: remove per... |
745 |
__wait_on_bit_lock(page_waitqueue(page), &wait, sleep_on_page, |
1da177e4c Linux-2.6.12-rc2 |
746 747 748 |
TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(__lock_page); |
b5606c2d4 remove final fast... |
749 |
int __lock_page_killable(struct page *page) |
2687a3569 Add lock_page_kil... |
750 751 752 753 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); return __wait_on_bit_lock(page_waitqueue(page), &wait, |
7eaceacca block: remove per... |
754 |
sleep_on_page_killable, TASK_KILLABLE); |
2687a3569 Add lock_page_kil... |
755 |
} |
18bc0bbd1 Staging: pohmelfs... |
756 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569 Add lock_page_kil... |
757 |
|
d065bd810 mm: retry page fa... |
758 759 760 |
int __lock_page_or_retry(struct page *page, struct mm_struct *mm, unsigned int flags) { |
37b23e052 x86,mm: make page... |
761 762 763 764 765 766 767 768 769 770 771 772 |
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 ... |
773 |
wait_on_page_locked(page); |
d065bd810 mm: retry page fa... |
774 |
return 0; |
37b23e052 x86,mm: make page... |
775 776 777 778 779 780 781 782 783 784 785 786 |
} 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... |
787 788 |
} } |
485bb99b4 [PATCH] kernel-do... |
789 |
/** |
e7b563bb2 mm: filemap: move... |
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 |
* 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++) { |
0cd6144aa mm + fs: prepare ... |
816 817 818 819 |
struct page *page; page = radix_tree_lookup(&mapping->page_tree, index); if (!page || radix_tree_exceptional_entry(page)) |
e7b563bb2 mm: filemap: move... |
820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 |
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++) { |
0cd6144aa mm + fs: prepare ... |
857 858 859 860 |
struct page *page; page = radix_tree_lookup(&mapping->page_tree, index); if (!page || radix_tree_exceptional_entry(page)) |
e7b563bb2 mm: filemap: move... |
861 862 863 864 865 866 867 868 869 870 871 |
break; index--; if (index == ULONG_MAX) break; } return index; } EXPORT_SYMBOL(page_cache_prev_hole); /** |
0cd6144aa mm + fs: prepare ... |
872 |
* find_get_entry - find and get a page cache entry |
485bb99b4 [PATCH] kernel-do... |
873 |
* @mapping: the address_space to search |
0cd6144aa mm + fs: prepare ... |
874 875 876 877 |
* @offset: the page cache index * * Looks up the page cache slot at @mapping & @offset. If there is a * page cache page, it is returned with an increased refcount. |
485bb99b4 [PATCH] kernel-do... |
878 |
* |
0cd6144aa mm + fs: prepare ... |
879 880 881 882 |
* 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 |
883 |
*/ |
0cd6144aa mm + fs: prepare ... |
884 |
struct page *find_get_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
885 |
{ |
a60637c85 mm: lockless page... |
886 |
void **pagep; |
1da177e4c Linux-2.6.12-rc2 |
887 |
struct page *page; |
a60637c85 mm: lockless page... |
888 889 890 891 892 893 |
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... |
894 895 |
if (unlikely(!page)) goto out; |
a2c16d6cb mm: let swap use ... |
896 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
897 898 899 900 901 902 903 904 |
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 ... |
905 |
} |
a60637c85 mm: lockless page... |
906 907 908 909 910 911 912 913 914 915 916 917 918 |
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... |
919 |
out: |
a60637c85 mm: lockless page... |
920 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
921 922 |
return page; } |
0cd6144aa mm + fs: prepare ... |
923 |
EXPORT_SYMBOL(find_get_entry); |
1da177e4c Linux-2.6.12-rc2 |
924 |
|
485bb99b4 [PATCH] kernel-do... |
925 |
/** |
0cd6144aa mm + fs: prepare ... |
926 |
* find_get_page - find and get a page reference |
67be2dd1b [PATCH] DocBook: ... |
927 928 |
* @mapping: the address_space to search * @offset: the page index |
1da177e4c Linux-2.6.12-rc2 |
929 |
* |
0cd6144aa mm + fs: prepare ... |
930 931 |
* Looks up the page cache slot at @mapping & @offset. If there is a * page cache page, it is returned with an increased refcount. |
1da177e4c Linux-2.6.12-rc2 |
932 |
* |
0cd6144aa mm + fs: prepare ... |
933 |
* Otherwise, %NULL is returned. |
1da177e4c Linux-2.6.12-rc2 |
934 |
*/ |
0cd6144aa mm + fs: prepare ... |
935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 |
struct page *find_get_page(struct address_space *mapping, pgoff_t offset) { struct page *page = find_get_entry(mapping, offset); if (radix_tree_exceptional_entry(page)) page = NULL; return page; } EXPORT_SYMBOL(find_get_page); /** * 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 |
962 963 |
{ struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
964 |
repeat: |
0cd6144aa mm + fs: prepare ... |
965 |
page = find_get_entry(mapping, offset); |
a2c16d6cb mm: let swap use ... |
966 |
if (page && !radix_tree_exception(page)) { |
a60637c85 mm: lockless page... |
967 968 969 970 971 972 |
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 |
973 |
} |
309381fea mm: dump page whe... |
974 |
VM_BUG_ON_PAGE(page->index != offset, page); |
1da177e4c Linux-2.6.12-rc2 |
975 |
} |
1da177e4c Linux-2.6.12-rc2 |
976 977 |
return page; } |
0cd6144aa mm + fs: prepare ... |
978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 |
EXPORT_SYMBOL(find_lock_entry); /** * find_lock_page - locate, pin and lock a pagecache page * @mapping: the address_space to search * @offset: the page 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. * * Otherwise, %NULL is returned. * * find_lock_page() may sleep. */ struct page *find_lock_page(struct address_space *mapping, pgoff_t offset) { struct page *page = find_lock_entry(mapping, offset); if (radix_tree_exceptional_entry(page)) page = NULL; return page; } |
1da177e4c Linux-2.6.12-rc2 |
1001 1002 1003 1004 |
EXPORT_SYMBOL(find_lock_page); /** * find_or_create_page - locate or add a pagecache page |
67be2dd1b [PATCH] DocBook: ... |
1005 1006 1007 |
* @mapping: the page's address_space * @index: the page's index into the mapping * @gfp_mask: page allocation mode |
1da177e4c Linux-2.6.12-rc2 |
1008 |
* |
0cd6144aa mm + fs: prepare ... |
1009 1010 1011 1012 1013 1014 1015 |
* 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 page is not present, 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. |
1da177e4c Linux-2.6.12-rc2 |
1016 |
* |
0cd6144aa mm + fs: prepare ... |
1017 |
* On memory exhaustion, %NULL is returned. |
1da177e4c Linux-2.6.12-rc2 |
1018 |
* |
0cd6144aa mm + fs: prepare ... |
1019 1020 |
* find_or_create_page() may sleep, even if @gfp_flags specifies an * atomic allocation! |
1da177e4c Linux-2.6.12-rc2 |
1021 1022 |
*/ struct page *find_or_create_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
1023 |
pgoff_t index, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
1024 |
{ |
eb2be1893 mm: buffered writ... |
1025 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1026 1027 1028 1029 |
int err; repeat: page = find_lock_page(mapping, index); if (!page) { |
eb2be1893 mm: buffered writ... |
1030 1031 1032 |
page = __page_cache_alloc(gfp_mask); if (!page) return NULL; |
67d58ac47 mm: pagecache gfp... |
1033 1034 1035 1036 1037 1038 1039 1040 |
/* * We want a regular kernel memory (not highmem or DMA etc) * allocation for the radix tree nodes, but we need to honour * the context-specific requirements the caller has asked for. * GFP_RECLAIM_MASK collects those requirements. */ err = add_to_page_cache_lru(page, mapping, index, (gfp_mask & GFP_RECLAIM_MASK)); |
eb2be1893 mm: buffered writ... |
1041 1042 1043 1044 1045 |
if (unlikely(err)) { page_cache_release(page); page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1046 |
} |
1da177e4c Linux-2.6.12-rc2 |
1047 |
} |
1da177e4c Linux-2.6.12-rc2 |
1048 1049 |
return page; } |
1da177e4c Linux-2.6.12-rc2 |
1050 1051 1052 |
EXPORT_SYMBOL(find_or_create_page); /** |
0cd6144aa mm + fs: prepare ... |
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 |
* 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 |
1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 |
* 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... |
1141 1142 1143 1144 1145 1146 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1147 1148 1149 |
rcu_read_lock(); restart: |
0fc9d1040 radix-tree: use i... |
1150 |
radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, start) { |
a60637c85 mm: lockless page... |
1151 1152 |
struct page *page; repeat: |
0fc9d1040 radix-tree: use i... |
1153 |
page = radix_tree_deref_slot(slot); |
a60637c85 mm: lockless page... |
1154 1155 |
if (unlikely(!page)) continue; |
9d8aa4ea8 mm: remove worryi... |
1156 |
|
a2c16d6cb mm: let swap use ... |
1157 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1158 1159 1160 1161 1162 1163 |
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... |
1164 |
WARN_ON(iter.index); |
8079b1c85 mm: clarify the r... |
1165 1166 |
goto restart; } |
a2c16d6cb mm: let swap use ... |
1167 |
/* |
8079b1c85 mm: clarify the r... |
1168 1169 1170 |
* 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 ... |
1171 |
*/ |
8079b1c85 mm: clarify the r... |
1172 |
continue; |
27d20fddc radix-tree: fix R... |
1173 |
} |
a60637c85 mm: lockless page... |
1174 1175 1176 1177 1178 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1179 |
if (unlikely(page != *slot)) { |
a60637c85 mm: lockless page... |
1180 1181 1182 |
page_cache_release(page); goto repeat; } |
1da177e4c Linux-2.6.12-rc2 |
1183 |
|
a60637c85 mm: lockless page... |
1184 |
pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1185 1186 |
if (++ret == nr_pages) break; |
a60637c85 mm: lockless page... |
1187 |
} |
5b280c0cc mm: don't return ... |
1188 |
|
a60637c85 mm: lockless page... |
1189 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1190 1191 |
return ret; } |
ebf43500e [PATCH] Add find_... |
1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 |
/** * 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... |
1207 1208 1209 1210 1211 1212 |
struct radix_tree_iter iter; void **slot; unsigned int ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1213 1214 1215 |
rcu_read_lock(); restart: |
0fc9d1040 radix-tree: use i... |
1216 |
radix_tree_for_each_contig(slot, &mapping->page_tree, &iter, index) { |
a60637c85 mm: lockless page... |
1217 1218 |
struct page *page; repeat: |
0fc9d1040 radix-tree: use i... |
1219 1220 |
page = radix_tree_deref_slot(slot); /* The hole, there no reason to continue */ |
a60637c85 mm: lockless page... |
1221 |
if (unlikely(!page)) |
0fc9d1040 radix-tree: use i... |
1222 |
break; |
9d8aa4ea8 mm: remove worryi... |
1223 |
|
a2c16d6cb mm: let swap use ... |
1224 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1225 1226 1227 1228 1229 1230 1231 1232 |
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 ... |
1233 |
/* |
8079b1c85 mm: clarify the r... |
1234 1235 1236 |
* 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 ... |
1237 |
*/ |
8079b1c85 mm: clarify the r... |
1238 |
break; |
a2c16d6cb mm: let swap use ... |
1239 |
} |
ebf43500e [PATCH] Add find_... |
1240 |
|
a60637c85 mm: lockless page... |
1241 1242 1243 1244 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1245 |
if (unlikely(page != *slot)) { |
a60637c85 mm: lockless page... |
1246 1247 1248 |
page_cache_release(page); goto repeat; } |
9cbb4cb21 mm: find_get_page... |
1249 1250 1251 1252 1253 |
/* * 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... |
1254 |
if (page->mapping == NULL || page->index != iter.index) { |
9cbb4cb21 mm: find_get_page... |
1255 1256 1257 |
page_cache_release(page); break; } |
a60637c85 mm: lockless page... |
1258 |
pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1259 1260 |
if (++ret == nr_pages) break; |
ebf43500e [PATCH] Add find_... |
1261 |
} |
a60637c85 mm: lockless page... |
1262 1263 |
rcu_read_unlock(); return ret; |
ebf43500e [PATCH] Add find_... |
1264 |
} |
ef71c15c4 AFS: export a cou... |
1265 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e [PATCH] Add find_... |
1266 |
|
485bb99b4 [PATCH] kernel-do... |
1267 1268 1269 1270 1271 1272 1273 1274 |
/** * 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 |
1275 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4 [PATCH] kernel-do... |
1276 |
* @tag. We update @index to index the next page for the traversal. |
1da177e4c Linux-2.6.12-rc2 |
1277 1278 1279 1280 |
*/ 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... |
1281 1282 1283 1284 1285 1286 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1287 1288 1289 |
rcu_read_lock(); restart: |
0fc9d1040 radix-tree: use i... |
1290 1291 |
radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, *index, tag) { |
a60637c85 mm: lockless page... |
1292 1293 |
struct page *page; repeat: |
0fc9d1040 radix-tree: use i... |
1294 |
page = radix_tree_deref_slot(slot); |
a60637c85 mm: lockless page... |
1295 1296 |
if (unlikely(!page)) continue; |
9d8aa4ea8 mm: remove worryi... |
1297 |
|
a2c16d6cb mm: let swap use ... |
1298 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1299 1300 1301 1302 1303 1304 1305 1306 |
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 ... |
1307 |
/* |
8079b1c85 mm: clarify the r... |
1308 1309 |
* This function is never used on a shmem/tmpfs * mapping, so a swap entry won't be found here. |
a2c16d6cb mm: let swap use ... |
1310 |
*/ |
8079b1c85 mm: clarify the r... |
1311 |
BUG(); |
a2c16d6cb mm: let swap use ... |
1312 |
} |
a60637c85 mm: lockless page... |
1313 1314 1315 1316 1317 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1318 |
if (unlikely(page != *slot)) { |
a60637c85 mm: lockless page... |
1319 1320 1321 1322 1323 |
page_cache_release(page); goto repeat; } pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1324 1325 |
if (++ret == nr_pages) break; |
a60637c85 mm: lockless page... |
1326 |
} |
5b280c0cc mm: don't return ... |
1327 |
|
a60637c85 mm: lockless page... |
1328 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1329 |
|
1da177e4c Linux-2.6.12-rc2 |
1330 1331 |
if (ret) *index = pages[ret - 1]->index + 1; |
a60637c85 mm: lockless page... |
1332 |
|
1da177e4c Linux-2.6.12-rc2 |
1333 1334 |
return ret; } |
ef71c15c4 AFS: export a cou... |
1335 |
EXPORT_SYMBOL(find_get_pages_tag); |
1da177e4c Linux-2.6.12-rc2 |
1336 |
|
485bb99b4 [PATCH] kernel-do... |
1337 1338 1339 1340 1341 |
/** * grab_cache_page_nowait - returns locked page at given index in given cache * @mapping: target address_space * @index: the page index * |
72fd4a35a [PATCH] Numerous ... |
1342 |
* Same as grab_cache_page(), but do not wait if the page is unavailable. |
1da177e4c Linux-2.6.12-rc2 |
1343 1344 1345 1346 1347 1348 1349 1350 |
* This is intended for speculative data generators, where the data can * be regenerated if the page couldn't be grabbed. This routine should * be safe to call while holding the lock for another page. * * Clear __GFP_FS when allocating the page to avoid recursion into the fs * and deadlock against the caller's locked page. */ struct page * |
57f6b96c0 filemap: convert ... |
1351 |
grab_cache_page_nowait(struct address_space *mapping, pgoff_t index) |
1da177e4c Linux-2.6.12-rc2 |
1352 1353 |
{ struct page *page = find_get_page(mapping, index); |
1da177e4c Linux-2.6.12-rc2 |
1354 1355 |
if (page) { |
529ae9aaa mm: rename page t... |
1356 |
if (trylock_page(page)) |
1da177e4c Linux-2.6.12-rc2 |
1357 1358 1359 1360 |
return page; page_cache_release(page); return NULL; } |
2ae88149a [PATCH] mm: clean... |
1361 |
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); |
67d58ac47 mm: pagecache gfp... |
1362 |
if (page && add_to_page_cache_lru(page, mapping, index, GFP_NOFS)) { |
1da177e4c Linux-2.6.12-rc2 |
1363 1364 1365 1366 1367 |
page_cache_release(page); page = NULL; } return page; } |
1da177e4c Linux-2.6.12-rc2 |
1368 |
EXPORT_SYMBOL(grab_cache_page_nowait); |
76d42bd96 [PATCH] readahead... |
1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 |
/* * 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... |
1387 |
ra->ra_pages /= 4; |
76d42bd96 [PATCH] readahead... |
1388 |
} |
485bb99b4 [PATCH] kernel-do... |
1389 |
/** |
36e789144 kill do_generic_m... |
1390 |
* do_generic_file_read - generic file read routine |
485bb99b4 [PATCH] kernel-do... |
1391 1392 |
* @filp: the file to read * @ppos: current file position |
6e58e79db introduce copy_pa... |
1393 1394 |
* @iter: data destination * @written: already copied |
485bb99b4 [PATCH] kernel-do... |
1395 |
* |
1da177e4c Linux-2.6.12-rc2 |
1396 |
* This is a generic file read routine, and uses the |
485bb99b4 [PATCH] kernel-do... |
1397 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c Linux-2.6.12-rc2 |
1398 1399 1400 |
* * 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 |
1401 |
*/ |
6e58e79db introduce copy_pa... |
1402 1403 |
static ssize_t do_generic_file_read(struct file *filp, loff_t *ppos, struct iov_iter *iter, ssize_t written) |
1da177e4c Linux-2.6.12-rc2 |
1404 |
{ |
36e789144 kill do_generic_m... |
1405 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
1406 |
struct inode *inode = mapping->host; |
36e789144 kill do_generic_m... |
1407 |
struct file_ra_state *ra = &filp->f_ra; |
57f6b96c0 filemap: convert ... |
1408 1409 1410 1411 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372 readahead: improv... |
1412 |
unsigned int prev_offset; |
6e58e79db introduce copy_pa... |
1413 |
int error = 0; |
1da177e4c Linux-2.6.12-rc2 |
1414 |
|
1da177e4c Linux-2.6.12-rc2 |
1415 |
index = *ppos >> PAGE_CACHE_SHIFT; |
7ff81078d readahead: remove... |
1416 1417 |
prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); |
6e58e79db introduce copy_pa... |
1418 |
last_index = (*ppos + iter->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
1419 |
offset = *ppos & ~PAGE_CACHE_MASK; |
1da177e4c Linux-2.6.12-rc2 |
1420 1421 |
for (;;) { struct page *page; |
57f6b96c0 filemap: convert ... |
1422 |
pgoff_t end_index; |
a32ea1e1f Fix read/truncate... |
1423 |
loff_t isize; |
1da177e4c Linux-2.6.12-rc2 |
1424 |
unsigned long nr, ret; |
1da177e4c Linux-2.6.12-rc2 |
1425 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
1426 1427 |
find_page: page = find_get_page(mapping, index); |
3ea89ee86 readahead: conver... |
1428 |
if (!page) { |
cf914a7d6 readahead: split ... |
1429 |
page_cache_sync_readahead(mapping, |
7ff81078d readahead: remove... |
1430 |
ra, filp, |
3ea89ee86 readahead: conver... |
1431 1432 1433 1434 1435 1436 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6 readahead: split ... |
1437 |
page_cache_async_readahead(mapping, |
7ff81078d readahead: remove... |
1438 |
ra, filp, page, |
3ea89ee86 readahead: conver... |
1439 |
index, last_index - index); |
1da177e4c Linux-2.6.12-rc2 |
1440 |
} |
8ab22b9ab vfs: pagecache us... |
1441 1442 1443 1444 |
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... |
1445 |
if (!trylock_page(page)) |
8ab22b9ab vfs: pagecache us... |
1446 |
goto page_not_up_to_date; |
8d056cb96 mm/vfs: revalidat... |
1447 1448 1449 |
/* Did it get truncated before we got the lock? */ if (!page->mapping) goto page_not_up_to_date_locked; |
8ab22b9ab vfs: pagecache us... |
1450 |
if (!mapping->a_ops->is_partially_uptodate(page, |
6e58e79db introduce copy_pa... |
1451 |
offset, iter->count)) |
8ab22b9ab vfs: pagecache us... |
1452 1453 1454 |
goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1455 |
page_ok: |
a32ea1e1f Fix read/truncate... |
1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 |
/* * 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 |
1482 1483 1484 1485 1486 1487 1488 1489 1490 |
/* 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... |
1491 1492 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c Linux-2.6.12-rc2 |
1493 |
*/ |
ec0f16372 readahead: improv... |
1494 |
if (prev_index != index || offset != prev_offset) |
1da177e4c Linux-2.6.12-rc2 |
1495 1496 1497 1498 1499 1500 |
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... |
1da177e4c Linux-2.6.12-rc2 |
1501 |
*/ |
6e58e79db introduce copy_pa... |
1502 1503 |
ret = copy_page_to_iter(page, offset, nr, iter); |
1da177e4c Linux-2.6.12-rc2 |
1504 1505 1506 |
offset += ret; index += offset >> PAGE_CACHE_SHIFT; offset &= ~PAGE_CACHE_MASK; |
6ce745ed3 readahead: code c... |
1507 |
prev_offset = offset; |
1da177e4c Linux-2.6.12-rc2 |
1508 1509 |
page_cache_release(page); |
6e58e79db introduce copy_pa... |
1510 1511 1512 1513 1514 1515 1516 1517 |
written += ret; if (!iov_iter_count(iter)) goto out; if (ret < nr) { error = -EFAULT; goto out; } continue; |
1da177e4c Linux-2.6.12-rc2 |
1518 1519 1520 |
page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235 do_generic_file_r... |
1521 1522 1523 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1524 |
|
8ab22b9ab vfs: pagecache us... |
1525 |
page_not_up_to_date_locked: |
da6052f7b [PATCH] update so... |
1526 |
/* Did it get truncated before we got the lock? */ |
1da177e4c Linux-2.6.12-rc2 |
1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 |
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... |
1540 1541 1542 1543 1544 1545 |
/* * 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 |
1546 1547 |
/* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7 [PATCH] add AOP_T... |
1548 1549 1550 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { page_cache_release(page); |
6e58e79db introduce copy_pa... |
1551 |
error = 0; |
994fc28c7 [PATCH] add AOP_T... |
1552 1553 |
goto find_page; } |
1da177e4c Linux-2.6.12-rc2 |
1554 |
goto readpage_error; |
994fc28c7 [PATCH] add AOP_T... |
1555 |
} |
1da177e4c Linux-2.6.12-rc2 |
1556 1557 |
if (!PageUptodate(page)) { |
854623235 do_generic_file_r... |
1558 1559 1560 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1561 1562 1563 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* |
2ecdc82ef kill unused inval... |
1564 |
* invalidate_mapping_pages got it |
1da177e4c Linux-2.6.12-rc2 |
1565 1566 1567 1568 1569 1570 |
*/ unlock_page(page); page_cache_release(page); goto find_page; } unlock_page(page); |
7ff81078d readahead: remove... |
1571 |
shrink_readahead_size_eio(filp, ra); |
854623235 do_generic_file_r... |
1572 1573 |
error = -EIO; goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1574 1575 1576 |
} unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1577 1578 1579 1580 |
goto page_ok; readpage_error: /* UHHUH! A synchronous read error occurred. Report it */ |
1da177e4c Linux-2.6.12-rc2 |
1581 1582 1583 1584 1585 1586 1587 1588 |
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... |
1589 1590 |
page = page_cache_alloc_cold(mapping); if (!page) { |
6e58e79db introduce copy_pa... |
1591 |
error = -ENOMEM; |
eb2be1893 mm: buffered writ... |
1592 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
1593 |
} |
eb2be1893 mm: buffered writ... |
1594 |
error = add_to_page_cache_lru(page, mapping, |
1da177e4c Linux-2.6.12-rc2 |
1595 1596 |
index, GFP_KERNEL); if (error) { |
eb2be1893 mm: buffered writ... |
1597 |
page_cache_release(page); |
6e58e79db introduce copy_pa... |
1598 1599 |
if (error == -EEXIST) { error = 0; |
1da177e4c Linux-2.6.12-rc2 |
1600 |
goto find_page; |
6e58e79db introduce copy_pa... |
1601 |
} |
1da177e4c Linux-2.6.12-rc2 |
1602 1603 |
goto out; } |
1da177e4c Linux-2.6.12-rc2 |
1604 1605 1606 1607 |
goto readpage; } out: |
7ff81078d readahead: remove... |
1608 1609 1610 |
ra->prev_pos = prev_index; ra->prev_pos <<= PAGE_CACHE_SHIFT; ra->prev_pos |= prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
1611 |
|
f4e6b498d readahead: combin... |
1612 |
*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; |
0c6aa2639 mm: do_generic_fi... |
1613 |
file_accessed(filp); |
6e58e79db introduce copy_pa... |
1614 |
return written ? written : error; |
1da177e4c Linux-2.6.12-rc2 |
1615 |
} |
0ceb33143 mm: move common s... |
1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 |
/* * 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... |
1654 |
/** |
b2abacf3a [PATCH] mm: fix i... |
1655 |
* generic_file_aio_read - generic filesystem read routine |
485bb99b4 [PATCH] kernel-do... |
1656 1657 1658 |
* @iocb: kernel I/O control block * @iov: io vector request * @nr_segs: number of segments in the iovec |
b2abacf3a [PATCH] mm: fix i... |
1659 |
* @pos: current file position |
485bb99b4 [PATCH] kernel-do... |
1660 |
* |
1da177e4c Linux-2.6.12-rc2 |
1661 1662 1663 1664 |
* This is the "read()" routine for all filesystems * that can use the page cache directly. */ ssize_t |
543ade1fc [PATCH] Streamlin... |
1665 1666 |
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c Linux-2.6.12-rc2 |
1667 1668 1669 |
{ struct file *filp = iocb->ki_filp; ssize_t retval; |
1da177e4c Linux-2.6.12-rc2 |
1670 |
size_t count; |
543ade1fc [PATCH] Streamlin... |
1671 |
loff_t *ppos = &iocb->ki_pos; |
6e58e79db introduce copy_pa... |
1672 |
struct iov_iter i; |
1da177e4c Linux-2.6.12-rc2 |
1673 1674 |
count = 0; |
0ceb33143 mm: move common s... |
1675 1676 1677 |
retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; |
6e58e79db introduce copy_pa... |
1678 |
iov_iter_init(&i, iov, nr_segs, count, 0); |
1da177e4c Linux-2.6.12-rc2 |
1679 1680 1681 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { |
543ade1fc [PATCH] Streamlin... |
1682 |
loff_t size; |
1da177e4c Linux-2.6.12-rc2 |
1683 1684 1685 1686 1687 |
struct address_space *mapping; struct inode *inode; mapping = filp->f_mapping; inode = mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
1688 1689 1690 |
if (!count) goto out; /* skip atime */ size = i_size_read(inode); |
9fe55eea7 Fix race when che... |
1691 |
retval = filemap_write_and_wait_range(mapping, pos, |
48b47c561 mm: direct IO sta... |
1692 |
pos + iov_length(iov, nr_segs) - 1); |
9fe55eea7 Fix race when che... |
1693 1694 1695 1696 1697 1698 1699 |
if (!retval) { retval = mapping->a_ops->direct_IO(READ, iocb, iov, pos, nr_segs); } if (retval > 0) { *ppos = pos + retval; count -= retval; |
6e58e79db introduce copy_pa... |
1700 1701 1702 1703 1704 |
/* * If we did a short DIO read we need to skip the * section of the iov that we've already read data into. */ iov_iter_advance(&i, retval); |
9fe55eea7 Fix race when che... |
1705 |
} |
66f998f61 fs: allow short d... |
1706 |
|
9fe55eea7 Fix race when che... |
1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 |
/* * 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) { file_accessed(filp); goto out; |
0e0bcae3b [GFS2] Fix direct... |
1718 |
} |
1da177e4c Linux-2.6.12-rc2 |
1719 |
} |
6e58e79db introduce copy_pa... |
1720 |
retval = do_generic_file_read(filp, ppos, &i, retval); |
1da177e4c Linux-2.6.12-rc2 |
1721 1722 1723 |
out: return retval; } |
1da177e4c Linux-2.6.12-rc2 |
1724 |
EXPORT_SYMBOL(generic_file_aio_read); |
1da177e4c Linux-2.6.12-rc2 |
1725 |
#ifdef CONFIG_MMU |
485bb99b4 [PATCH] kernel-do... |
1726 1727 1728 1729 1730 |
/** * 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 |
1731 1732 1733 |
* 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... |
1734 |
static int page_cache_read(struct file *file, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1735 1736 1737 |
{ struct address_space *mapping = file->f_mapping; struct page *page; |
994fc28c7 [PATCH] add AOP_T... |
1738 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1739 |
|
994fc28c7 [PATCH] add AOP_T... |
1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 |
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 |
1750 |
|
1da177e4c Linux-2.6.12-rc2 |
1751 |
page_cache_release(page); |
1da177e4c Linux-2.6.12-rc2 |
1752 |
|
994fc28c7 [PATCH] add AOP_T... |
1753 1754 1755 |
} while (ret == AOP_TRUNCATED_PAGE); return ret; |
1da177e4c Linux-2.6.12-rc2 |
1756 1757 1758 |
} #define MMAP_LOTSAMISS (100) |
ef00e08e2 readahead: clean ... |
1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 |
/* * 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... |
1772 |
if (vma->vm_flags & VM_RAND_READ) |
ef00e08e2 readahead: clean ... |
1773 |
return; |
275b12bf5 readahead: return... |
1774 1775 |
if (!ra->ra_pages) return; |
ef00e08e2 readahead: clean ... |
1776 |
|
64363aad5 mm: remove unused... |
1777 |
if (vma->vm_flags & VM_SEQ_READ) { |
7ffc59b4d readahead: enforc... |
1778 1779 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
1780 1781 |
return; } |
207d04baa readahead: reduce... |
1782 1783 |
/* Avoid banging the cache line if not needed */ if (ra->mmap_miss < MMAP_LOTSAMISS * 10) |
ef00e08e2 readahead: clean ... |
1784 1785 1786 1787 1788 1789 1790 1791 |
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... |
1792 1793 1794 |
/* * mmap read-around */ |
ef00e08e2 readahead: clean ... |
1795 |
ra_pages = max_sane_readahead(ra->ra_pages); |
275b12bf5 readahead: return... |
1796 1797 |
ra->start = max_t(long, 0, offset - ra_pages / 2); ra->size = ra_pages; |
2cbea1d3a readahead: trigge... |
1798 |
ra->async_size = ra_pages / 4; |
275b12bf5 readahead: return... |
1799 |
ra_submit(ra, mapping, file); |
ef00e08e2 readahead: clean ... |
1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 |
} /* * 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... |
1815 |
if (vma->vm_flags & VM_RAND_READ) |
ef00e08e2 readahead: clean ... |
1816 1817 1818 1819 |
return; if (ra->mmap_miss > 0) ra->mmap_miss--; if (PageReadahead(page)) |
2fad6f5de readahead: enforc... |
1820 1821 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
1822 |
} |
485bb99b4 [PATCH] kernel-do... |
1823 |
/** |
54cb8821d mm: merge populat... |
1824 |
* filemap_fault - read in file data for page fault handling |
d0217ac04 mm: fault feedbac... |
1825 1826 |
* @vma: vma in which the fault was taken * @vmf: struct vm_fault containing details of the fault |
485bb99b4 [PATCH] kernel-do... |
1827 |
* |
54cb8821d mm: merge populat... |
1828 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c Linux-2.6.12-rc2 |
1829 1830 1831 1832 1833 1834 |
* 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... |
1835 |
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
1da177e4c Linux-2.6.12-rc2 |
1836 1837 |
{ int error; |
54cb8821d mm: merge populat... |
1838 |
struct file *file = vma->vm_file; |
1da177e4c Linux-2.6.12-rc2 |
1839 1840 1841 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2 readahead: clean ... |
1842 |
pgoff_t offset = vmf->pgoff; |
1da177e4c Linux-2.6.12-rc2 |
1843 |
struct page *page; |
99e3e53f4 mm: cleanup size ... |
1844 |
loff_t size; |
83c54070e mm: fault feedbac... |
1845 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1846 |
|
99e3e53f4 mm: cleanup size ... |
1847 1848 |
size = round_up(i_size_read(inode), PAGE_CACHE_SIZE); if (offset >= size >> PAGE_CACHE_SHIFT) |
5307cc1aa Remove broken ptr... |
1849 |
return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
1850 |
|
1da177e4c Linux-2.6.12-rc2 |
1851 |
/* |
494264208 mm: memcg: handle... |
1852 |
* Do we have something in the page cache already? |
1da177e4c Linux-2.6.12-rc2 |
1853 |
*/ |
ef00e08e2 readahead: clean ... |
1854 |
page = find_get_page(mapping, offset); |
45cac65b0 readahead: fault ... |
1855 |
if (likely(page) && !(vmf->flags & FAULT_FLAG_TRIED)) { |
1da177e4c Linux-2.6.12-rc2 |
1856 |
/* |
ef00e08e2 readahead: clean ... |
1857 1858 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c Linux-2.6.12-rc2 |
1859 |
*/ |
ef00e08e2 readahead: clean ... |
1860 |
do_async_mmap_readahead(vma, ra, file, page, offset); |
45cac65b0 readahead: fault ... |
1861 |
} else if (!page) { |
ef00e08e2 readahead: clean ... |
1862 1863 1864 |
/* 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... |
1865 |
mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT); |
ef00e08e2 readahead: clean ... |
1866 1867 |
ret = VM_FAULT_MAJOR; retry_find: |
b522c94da mm: filemap_fault... |
1868 |
page = find_get_page(mapping, offset); |
1da177e4c Linux-2.6.12-rc2 |
1869 1870 1871 |
if (!page) goto no_cached_page; } |
d88c0922f Release page refe... |
1872 1873 |
if (!lock_page_or_retry(page, vma->vm_mm, vmf->flags)) { page_cache_release(page); |
d065bd810 mm: retry page fa... |
1874 |
return ret | VM_FAULT_RETRY; |
d88c0922f Release page refe... |
1875 |
} |
b522c94da mm: filemap_fault... |
1876 1877 1878 1879 1880 1881 1882 |
/* Did it get truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); put_page(page); goto retry_find; } |
309381fea mm: dump page whe... |
1883 |
VM_BUG_ON_PAGE(page->index != offset, page); |
b522c94da mm: filemap_fault... |
1884 |
|
1da177e4c Linux-2.6.12-rc2 |
1885 |
/* |
d00806b18 mm: fix fault vs ... |
1886 1887 |
* 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 |
1888 |
*/ |
d00806b18 mm: fix fault vs ... |
1889 |
if (unlikely(!PageUptodate(page))) |
1da177e4c Linux-2.6.12-rc2 |
1890 |
goto page_not_uptodate; |
ef00e08e2 readahead: clean ... |
1891 1892 1893 1894 |
/* * Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
99e3e53f4 mm: cleanup size ... |
1895 1896 |
size = round_up(i_size_read(inode), PAGE_CACHE_SIZE); if (unlikely(offset >= size >> PAGE_CACHE_SHIFT)) { |
d00806b18 mm: fix fault vs ... |
1897 |
unlock_page(page); |
745ad48e8 fix page release ... |
1898 |
page_cache_release(page); |
5307cc1aa Remove broken ptr... |
1899 |
return VM_FAULT_SIGBUS; |
d00806b18 mm: fix fault vs ... |
1900 |
} |
d0217ac04 mm: fault feedbac... |
1901 |
vmf->page = page; |
83c54070e mm: fault feedbac... |
1902 |
return ret | VM_FAULT_LOCKED; |
1da177e4c Linux-2.6.12-rc2 |
1903 |
|
1da177e4c Linux-2.6.12-rc2 |
1904 1905 1906 1907 1908 |
no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ |
ef00e08e2 readahead: clean ... |
1909 |
error = page_cache_read(file, offset); |
1da177e4c Linux-2.6.12-rc2 |
1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 |
/* * 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... |
1925 1926 |
return VM_FAULT_OOM; return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
1927 1928 |
page_not_uptodate: |
1da177e4c Linux-2.6.12-rc2 |
1929 1930 1931 1932 1933 1934 |
/* * 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 |
1935 |
ClearPageError(page); |
994fc28c7 [PATCH] add AOP_T... |
1936 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e mm: fix infinite ... |
1937 1938 1939 1940 1941 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
d00806b18 mm: fix fault vs ... |
1942 1943 1944 |
page_cache_release(page); if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7 [PATCH] add AOP_T... |
1945 |
goto retry_find; |
1da177e4c Linux-2.6.12-rc2 |
1946 |
|
d00806b18 mm: fix fault vs ... |
1947 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96 [PATCH] readahead... |
1948 |
shrink_readahead_size_eio(file, ra); |
d0217ac04 mm: fault feedbac... |
1949 |
return VM_FAULT_SIGBUS; |
54cb8821d mm: merge populat... |
1950 1951 |
} EXPORT_SYMBOL(filemap_fault); |
f1820361f mm: implement ->m... |
1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 |
void filemap_map_pages(struct vm_area_struct *vma, struct vm_fault *vmf) { struct radix_tree_iter iter; void **slot; struct file *file = vma->vm_file; struct address_space *mapping = file->f_mapping; loff_t size; struct page *page; unsigned long address = (unsigned long) vmf->virtual_address; unsigned long addr; pte_t *pte; rcu_read_lock(); radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, vmf->pgoff) { if (iter.index > vmf->max_pgoff) break; repeat: page = radix_tree_deref_slot(slot); if (unlikely(!page)) goto next; if (radix_tree_exception(page)) { if (radix_tree_deref_retry(page)) break; else goto next; } if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *slot)) { page_cache_release(page); goto repeat; } if (!PageUptodate(page) || PageReadahead(page) || PageHWPoison(page)) goto skip; if (!trylock_page(page)) goto skip; if (page->mapping != mapping || !PageUptodate(page)) goto unlock; |
99e3e53f4 mm: cleanup size ... |
1997 1998 |
size = round_up(i_size_read(mapping->host), PAGE_CACHE_SIZE); if (page->index >= size >> PAGE_CACHE_SHIFT) |
f1820361f mm: implement ->m... |
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 |
goto unlock; pte = vmf->pte + page->index - vmf->pgoff; if (!pte_none(*pte)) goto unlock; if (file->f_ra.mmap_miss > 0) file->f_ra.mmap_miss--; addr = address + (page->index - vmf->pgoff) * PAGE_SIZE; do_set_pte(vma, addr, page, pte, false, false); unlock_page(page); goto next; unlock: unlock_page(page); skip: page_cache_release(page); next: if (iter.index == vmf->max_pgoff) break; } rcu_read_unlock(); } EXPORT_SYMBOL(filemap_map_pages); |
4fcf1c620 mm: Make default ... |
2022 2023 2024 |
int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) { struct page *page = vmf->page; |
496ad9aa8 new helper: file_... |
2025 |
struct inode *inode = file_inode(vma->vm_file); |
4fcf1c620 mm: Make default ... |
2026 |
int ret = VM_FAULT_LOCKED; |
14da92001 fs: Protect write... |
2027 |
sb_start_pagefault(inode->i_sb); |
4fcf1c620 mm: Make default ... |
2028 2029 2030 2031 2032 2033 2034 |
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... |
2035 2036 2037 2038 2039 2040 |
/* * 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 ... |
2041 |
wait_for_stable_page(page); |
4fcf1c620 mm: Make default ... |
2042 |
out: |
14da92001 fs: Protect write... |
2043 |
sb_end_pagefault(inode->i_sb); |
4fcf1c620 mm: Make default ... |
2044 2045 2046 |
return ret; } EXPORT_SYMBOL(filemap_page_mkwrite); |
f0f37e2f7 const: mark struc... |
2047 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d mm: merge populat... |
2048 |
.fault = filemap_fault, |
f1820361f mm: implement ->m... |
2049 |
.map_pages = filemap_map_pages, |
4fcf1c620 mm: Make default ... |
2050 |
.page_mkwrite = filemap_page_mkwrite, |
0b173bc4d mm: kill vma flag... |
2051 |
.remap_pages = generic_file_remap_pages, |
1da177e4c Linux-2.6.12-rc2 |
2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 |
}; /* 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 |
2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 |
/* * 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); |
67f9fd91f mm: remove read_c... |
2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 |
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... |
2100 |
static struct page *__read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
2101 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2102 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2103 2104 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
2105 |
{ |
eb2be1893 mm: buffered writ... |
2106 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
2107 2108 2109 2110 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
0531b2aac mm: add new 'read... |
2111 |
page = __page_cache_alloc(gfp | __GFP_COLD); |
eb2be1893 mm: buffered writ... |
2112 2113 |
if (!page) return ERR_PTR(-ENOMEM); |
e6f67b8c0 vfs: __read_cache... |
2114 |
err = add_to_page_cache_lru(page, mapping, index, gfp); |
eb2be1893 mm: buffered writ... |
2115 2116 2117 2118 |
if (unlikely(err)) { page_cache_release(page); if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
2119 |
/* Presumably ENOMEM for radix tree node */ |
1da177e4c Linux-2.6.12-rc2 |
2120 2121 |
return ERR_PTR(err); } |
1da177e4c Linux-2.6.12-rc2 |
2122 2123 2124 2125 |
err = filler(data, page); if (err < 0) { page_cache_release(page); page = ERR_PTR(err); |
67f9fd91f mm: remove read_c... |
2126 2127 |
} else { page = wait_on_page_read(page); |
1da177e4c Linux-2.6.12-rc2 |
2128 2129 |
} } |
1da177e4c Linux-2.6.12-rc2 |
2130 2131 |
return page; } |
0531b2aac mm: add new 'read... |
2132 |
static struct page *do_read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
2133 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2134 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2135 2136 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
2137 2138 2139 2140 2141 |
{ struct page *page; int err; retry: |
0531b2aac mm: add new 'read... |
2142 |
page = __read_cache_page(mapping, index, filler, data, gfp); |
1da177e4c Linux-2.6.12-rc2 |
2143 |
if (IS_ERR(page)) |
c855ff371 Fix a bad error c... |
2144 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 |
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... |
2161 |
return ERR_PTR(err); |
67f9fd91f mm: remove read_c... |
2162 2163 2164 2165 |
} else { page = wait_on_page_read(page); if (IS_ERR(page)) return page; |
1da177e4c Linux-2.6.12-rc2 |
2166 |
} |
c855ff371 Fix a bad error c... |
2167 |
out: |
6fe6900e1 mm: make read_cac... |
2168 2169 2170 |
mark_page_accessed(page); return page; } |
0531b2aac mm: add new 'read... |
2171 2172 |
/** |
67f9fd91f mm: remove read_c... |
2173 |
* read_cache_page - read into page cache, fill it if needed |
0531b2aac mm: add new 'read... |
2174 2175 2176 |
* @mapping: the page's address_space * @index: the page index * @filler: function to perform the read |
5e5358e7c mm: cleanup descr... |
2177 |
* @data: first arg to filler(data, page) function, often left as NULL |
0531b2aac mm: add new 'read... |
2178 |
* |
0531b2aac mm: add new 'read... |
2179 |
* Read into the page cache. If a page already exists, and PageUptodate() is |
67f9fd91f mm: remove read_c... |
2180 |
* not set, try to fill the page and wait for it to become unlocked. |
0531b2aac mm: add new 'read... |
2181 2182 2183 |
* * If the page does not get brought uptodate, return -EIO. */ |
67f9fd91f mm: remove read_c... |
2184 |
struct page *read_cache_page(struct address_space *mapping, |
0531b2aac mm: add new 'read... |
2185 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2186 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2187 2188 2189 2190 |
void *data) { return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); } |
67f9fd91f mm: remove read_c... |
2191 |
EXPORT_SYMBOL(read_cache_page); |
0531b2aac mm: add new 'read... |
2192 2193 2194 2195 2196 2197 2198 2199 |
/** * 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... |
2200 |
* any new page allocations done using the specified allocation flags. |
0531b2aac mm: add new 'read... |
2201 2202 2203 2204 2205 2206 2207 2208 |
* * 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; |
67f9fd91f mm: remove read_c... |
2209 |
return do_read_cache_page(mapping, index, filler, NULL, gfp); |
0531b2aac mm: add new 'read... |
2210 2211 |
} EXPORT_SYMBOL(read_cache_page_gfp); |
2f718ffc1 mm: buffered writ... |
2212 |
/* |
1da177e4c Linux-2.6.12-rc2 |
2213 2214 |
* Performs necessary checks before doing a write * |
485bb99b4 [PATCH] kernel-do... |
2215 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c Linux-2.6.12-rc2 |
2216 2217 2218 2219 2220 2221 |
* 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... |
2222 |
unsigned long limit = rlimit(RLIMIT_FSIZE); |
1da177e4c Linux-2.6.12-rc2 |
2223 2224 2225 |
if (unlikely(*pos < 0)) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 |
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 |
2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 |
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 |
2265 2266 2267 2268 2269 2270 2271 2272 |
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... |
2273 |
#ifdef CONFIG_BLOCK |
1da177e4c Linux-2.6.12-rc2 |
2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 |
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... |
2285 2286 2287 |
#else return -EPERM; #endif |
1da177e4c Linux-2.6.12-rc2 |
2288 2289 2290 2291 |
} return 0; } EXPORT_SYMBOL(generic_write_checks); |
afddba49d fs: introduce wri... |
2292 2293 2294 2295 2296 |
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... |
2297 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d fs: introduce wri... |
2298 |
pagep, fsdata); |
afddba49d fs: introduce wri... |
2299 2300 2301 2302 2303 2304 2305 2306 |
} 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... |
2307 |
|
4e02ed4b4 fs: remove prepar... |
2308 2309 |
mark_page_accessed(page); return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d fs: introduce wri... |
2310 2311 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c Linux-2.6.12-rc2 |
2312 2313 |
ssize_t generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, |
5cb6c6c7e generic_file_dire... |
2314 |
unsigned long *nr_segs, loff_t pos, |
1da177e4c Linux-2.6.12-rc2 |
2315 2316 2317 2318 2319 2320 |
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... |
2321 2322 |
size_t write_len; pgoff_t end; |
1da177e4c Linux-2.6.12-rc2 |
2323 2324 2325 |
if (count != ocount) *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); |
a969e903a kill generic_file... |
2326 2327 |
write_len = iov_length(iov, *nr_segs); end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; |
a969e903a kill generic_file... |
2328 |
|
48b47c561 mm: direct IO sta... |
2329 |
written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); |
a969e903a kill generic_file... |
2330 2331 2332 2333 2334 2335 2336 |
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... |
2337 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a kill generic_file... |
2338 2339 2340 2341 |
*/ if (mapping->nrpages) { written = invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); |
6ccfa806a VFS: fix dio writ... |
2342 2343 2344 2345 2346 2347 2348 |
/* * 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... |
2349 |
goto out; |
6ccfa806a VFS: fix dio writ... |
2350 |
} |
a969e903a kill generic_file... |
2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 |
} 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 |
2367 |
if (written > 0) { |
0116651c8 mm: remove tempor... |
2368 2369 2370 |
pos += written; if (pos > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, pos); |
1da177e4c Linux-2.6.12-rc2 |
2371 2372 |
mark_inode_dirty(inode); } |
5cb6c6c7e generic_file_dire... |
2373 |
iocb->ki_pos = pos; |
1da177e4c Linux-2.6.12-rc2 |
2374 |
} |
a969e903a kill generic_file... |
2375 |
out: |
1da177e4c Linux-2.6.12-rc2 |
2376 2377 2378 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893 mm: buffered writ... |
2379 2380 2381 2382 |
/* * 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... |
2383 2384 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893 mm: buffered writ... |
2385 2386 |
{ int status; |
0faa70cb0 mm: filemap: pass... |
2387 |
gfp_t gfp_mask; |
eb2be1893 mm: buffered writ... |
2388 |
struct page *page; |
54566b2c1 fs: symlink write... |
2389 |
gfp_t gfp_notmask = 0; |
0faa70cb0 mm: filemap: pass... |
2390 |
|
1010bb1b8 mm: don't set __G... |
2391 2392 2393 |
gfp_mask = mapping_gfp_mask(mapping); if (mapping_cap_account_dirty(mapping)) gfp_mask |= __GFP_WRITE; |
54566b2c1 fs: symlink write... |
2394 2395 |
if (flags & AOP_FLAG_NOFS) gfp_notmask = __GFP_FS; |
eb2be1893 mm: buffered writ... |
2396 2397 |
repeat: page = find_lock_page(mapping, index); |
c585a2678 mm: remove likely... |
2398 |
if (page) |
3d08bcc88 mm: Wait for writ... |
2399 |
goto found; |
eb2be1893 mm: buffered writ... |
2400 |
|
0faa70cb0 mm: filemap: pass... |
2401 |
page = __page_cache_alloc(gfp_mask & ~gfp_notmask); |
eb2be1893 mm: buffered writ... |
2402 2403 |
if (!page) return NULL; |
54566b2c1 fs: symlink write... |
2404 2405 |
status = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL & ~gfp_notmask); |
eb2be1893 mm: buffered writ... |
2406 2407 2408 2409 2410 2411 |
if (unlikely(status)) { page_cache_release(page); if (status == -EEXIST) goto repeat; return NULL; } |
3d08bcc88 mm: Wait for writ... |
2412 |
found: |
1d1d1a767 mm: only enforce ... |
2413 |
wait_for_stable_page(page); |
eb2be1893 mm: buffered writ... |
2414 2415 |
return page; } |
54566b2c1 fs: symlink write... |
2416 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893 mm: buffered writ... |
2417 |
|
3b93f911d export generic_pe... |
2418 |
ssize_t generic_perform_write(struct file *file, |
afddba49d fs: introduce wri... |
2419 2420 2421 2422 2423 2424 |
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... |
2425 2426 2427 2428 2429 2430 2431 |
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... |
2432 2433 2434 |
do { struct page *page; |
afddba49d fs: introduce wri... |
2435 2436 2437 2438 2439 2440 |
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... |
2441 2442 2443 2444 |
bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_count(i)); again: |
afddba49d fs: introduce wri... |
2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 |
/* * 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... |
2459 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d fs: introduce wri... |
2460 2461 2462 |
&page, &fsdata); if (unlikely(status)) break; |
931e80e4b mm: flush dcache ... |
2463 2464 |
if (mapping_writably_mapped(mapping)) flush_dcache_page(page); |
afddba49d fs: introduce wri... |
2465 |
copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); |
afddba49d fs: introduce wri... |
2466 |
flush_dcache_page(page); |
c8236db9c mm: mark page acc... |
2467 |
mark_page_accessed(page); |
afddba49d fs: introduce wri... |
2468 2469 2470 2471 2472 2473 2474 |
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... |
2475 |
iov_iter_advance(i, copied); |
afddba49d fs: introduce wri... |
2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 |
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... |
2489 2490 2491 2492 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); |
a50527b19 fs: Make write(2)... |
2493 2494 2495 2496 |
if (fatal_signal_pending(current)) { status = -EINTR; break; } |
afddba49d fs: introduce wri... |
2497 2498 2499 2500 |
} while (iov_iter_count(i)); return written ? written : status; } |
3b93f911d export generic_pe... |
2501 |
EXPORT_SYMBOL(generic_perform_write); |
1da177e4c Linux-2.6.12-rc2 |
2502 |
|
e4dd9de3c vfs: Export __gen... |
2503 2504 2505 2506 2507 |
/** * __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 |
e4dd9de3c vfs: Export __gen... |
2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 |
* * 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, |
41fc56d57 kill the 4th argu... |
2522 |
unsigned long nr_segs) |
1da177e4c Linux-2.6.12-rc2 |
2523 2524 |
{ struct file *file = iocb->ki_filp; |
fb5527e68 [PATCH] direct-io... |
2525 |
struct address_space * mapping = file->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
2526 2527 2528 |
size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; |
41fc56d57 kill the 4th argu... |
2529 |
loff_t pos = iocb->ki_pos; |
3b93f911d export generic_pe... |
2530 |
ssize_t written = 0; |
1da177e4c Linux-2.6.12-rc2 |
2531 |
ssize_t err; |
3b93f911d export generic_pe... |
2532 2533 |
ssize_t status; struct iov_iter from; |
1da177e4c Linux-2.6.12-rc2 |
2534 2535 |
ocount = 0; |
0ceb33143 mm: move common s... |
2536 2537 2538 |
err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (err) return err; |
1da177e4c Linux-2.6.12-rc2 |
2539 2540 |
count = ocount; |
1da177e4c Linux-2.6.12-rc2 |
2541 |
|
1da177e4c Linux-2.6.12-rc2 |
2542 2543 |
/* We can write back this queue in page reclaim */ current->backing_dev_info = mapping->backing_dev_info; |
1da177e4c Linux-2.6.12-rc2 |
2544 2545 2546 2547 2548 2549 |
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: ... |
2550 |
err = file_remove_suid(file); |
1da177e4c Linux-2.6.12-rc2 |
2551 2552 |
if (err) goto out; |
c3b2da314 fs: introduce ino... |
2553 2554 2555 |
err = file_update_time(file); if (err) goto out; |
1da177e4c Linux-2.6.12-rc2 |
2556 |
|
3b93f911d export generic_pe... |
2557 |
iov_iter_init(&from, iov, nr_segs, count, 0); |
1da177e4c Linux-2.6.12-rc2 |
2558 2559 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { |
fb5527e68 [PATCH] direct-io... |
2560 |
loff_t endbyte; |
fb5527e68 [PATCH] direct-io... |
2561 |
|
3b93f911d export generic_pe... |
2562 |
written = generic_file_direct_write(iocb, iov, &from.nr_segs, pos, |
5cb6c6c7e generic_file_dire... |
2563 |
count, ocount); |
1da177e4c Linux-2.6.12-rc2 |
2564 2565 |
if (written < 0 || written == count) goto out; |
3b93f911d export generic_pe... |
2566 |
iov_iter_advance(&from, written); |
1da177e4c Linux-2.6.12-rc2 |
2567 2568 2569 2570 2571 2572 |
/* * direct-io write to a hole: fall through to buffered I/O * for completing the rest of the request. */ pos += written; count -= written; |
3b93f911d export generic_pe... |
2573 2574 |
status = generic_perform_write(file, &from, pos); |
fb5527e68 [PATCH] direct-io... |
2575 |
/* |
3b93f911d export generic_pe... |
2576 |
* If generic_perform_write() returned a synchronous error |
fb5527e68 [PATCH] direct-io... |
2577 2578 2579 2580 2581 |
* 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. */ |
3b93f911d export generic_pe... |
2582 2583 |
if (unlikely(status < 0) && !written) { err = status; |
fb5527e68 [PATCH] direct-io... |
2584 2585 |
goto out; } |
3b93f911d export generic_pe... |
2586 |
iocb->ki_pos = pos + status; |
fb5527e68 [PATCH] direct-io... |
2587 2588 2589 2590 2591 |
/* * We need to ensure that the page cache pages are written to * disk and invalidated to preserve the expected O_DIRECT * semantics. */ |
3b93f911d export generic_pe... |
2592 |
endbyte = pos + status - 1; |
c05c4edd8 direct I/O fallba... |
2593 |
err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); |
fb5527e68 [PATCH] direct-io... |
2594 |
if (err == 0) { |
3b93f911d export generic_pe... |
2595 |
written += status; |
fb5527e68 [PATCH] direct-io... |
2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 |
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 { |
3b93f911d export generic_pe... |
2606 2607 2608 |
written = generic_perform_write(file, &from, pos); if (likely(written >= 0)) iocb->ki_pos = pos + written; |
fb5527e68 [PATCH] direct-io... |
2609 |
} |
1da177e4c Linux-2.6.12-rc2 |
2610 2611 2612 2613 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
e4dd9de3c vfs: Export __gen... |
2614 |
EXPORT_SYMBOL(__generic_file_aio_write); |
e4dd9de3c vfs: Export __gen... |
2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 |
/** * 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... |
2626 2627 |
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 |
2628 2629 |
{ struct file *file = iocb->ki_filp; |
148f948ba vfs: Introduce ne... |
2630 |
struct inode *inode = file->f_mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
2631 |
ssize_t ret; |
1da177e4c Linux-2.6.12-rc2 |
2632 2633 |
BUG_ON(iocb->ki_pos != pos); |
1b1dcc1b5 [PATCH] mutex sub... |
2634 |
mutex_lock(&inode->i_mutex); |
41fc56d57 kill the 4th argu... |
2635 |
ret = __generic_file_aio_write(iocb, iov, nr_segs); |
1b1dcc1b5 [PATCH] mutex sub... |
2636 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2637 |
|
02afc27fa direct-io: Handle... |
2638 |
if (ret > 0) { |
1da177e4c Linux-2.6.12-rc2 |
2639 |
ssize_t err; |
d311d79de fix O_SYNC|O_APPE... |
2640 2641 |
err = generic_write_sync(file, iocb->ki_pos - ret, ret); if (err < 0) |
1da177e4c Linux-2.6.12-rc2 |
2642 2643 2644 2645 2646 |
ret = err; } return ret; } EXPORT_SYMBOL(generic_file_aio_write); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 |
/** * 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... |
2657 2658 2659 |
* 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... |
2660 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
3f31fddfa jbd: fix race bet... |
2661 |
* this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS). |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2662 |
* |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 |
*/ 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); |