Blame view
mm/filemap.c
95.8 KB
457c89965 treewide: Add SPD... |
1 |
// SPDX-License-Identifier: GPL-2.0-only |
1da177e4c Linux-2.6.12-rc2 |
2 3 4 5 6 7 8 9 10 11 12 |
/* * 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... |
13 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
14 |
#include <linux/compiler.h> |
f9fe48bec dax: support dirt... |
15 |
#include <linux/dax.h> |
1da177e4c Linux-2.6.12-rc2 |
16 |
#include <linux/fs.h> |
3f07c0144 sched/headers: Pr... |
17 |
#include <linux/sched/signal.h> |
c22ce143d [PATCH] x86: cach... |
18 |
#include <linux/uaccess.h> |
c59ede7b7 [PATCH] move capa... |
19 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
20 |
#include <linux/kernel_stat.h> |
5a0e3ad6a include cleanup: ... |
21 |
#include <linux/gfp.h> |
1da177e4c Linux-2.6.12-rc2 |
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> |
cfcbfb138 mm/filemap.c: ena... |
28 |
#include <linux/error-injection.h> |
1da177e4c Linux-2.6.12-rc2 |
29 30 |
#include <linux/hash.h> #include <linux/writeback.h> |
53253383f Include <linux/ba... |
31 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
32 33 34 |
#include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/security.h> |
44110fe38 [PATCH] cpuset me... |
35 |
#include <linux/cpuset.h> |
00501b531 mm: memcontrol: r... |
36 |
#include <linux/hugetlb.h> |
8a9f3ccd2 Memory controller... |
37 |
#include <linux/memcontrol.h> |
c515e1fd3 mm/fs: add hooks ... |
38 |
#include <linux/cleancache.h> |
c7df8ad29 mm, truncate: do ... |
39 |
#include <linux/shmem_fs.h> |
f1820361f mm: implement ->m... |
40 |
#include <linux/rmap.h> |
b1d29ba82 delayacct: track ... |
41 |
#include <linux/delayacct.h> |
eb414681d psi: pressure sta... |
42 |
#include <linux/psi.h> |
d0e6a5821 mm/filemap.c: inc... |
43 |
#include <linux/ramfs.h> |
0f8053a50 [PATCH] mm: make ... |
44 |
#include "internal.h" |
fe0bfaaff mm: trace filemap... |
45 46 |
#define CREATE_TRACE_POINTS #include <trace/events/filemap.h> |
1da177e4c Linux-2.6.12-rc2 |
47 |
/* |
1da177e4c Linux-2.6.12-rc2 |
48 49 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba vfs: Introduce ne... |
50 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c Linux-2.6.12-rc2 |
51 |
|
1da177e4c Linux-2.6.12-rc2 |
52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
#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: * |
c8c06efa8 mm: convert i_mma... |
69 |
* ->i_mmap_rwsem (truncate_pagecache) |
1da177e4c Linux-2.6.12-rc2 |
70 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919 [PATCH] swap: swa... |
71 |
* ->swap_lock (exclusive_swap_page, others) |
b93b01631 page cache: use x... |
72 |
* ->i_pages lock |
1da177e4c Linux-2.6.12-rc2 |
73 |
* |
1b1dcc1b5 [PATCH] mutex sub... |
74 |
* ->i_mutex |
c8c06efa8 mm: convert i_mma... |
75 |
* ->i_mmap_rwsem (truncate->unmap_mapping_range) |
1da177e4c Linux-2.6.12-rc2 |
76 77 |
* * ->mmap_sem |
c8c06efa8 mm: convert i_mma... |
78 |
* ->i_mmap_rwsem |
b8072f099 [PATCH] mm: updat... |
79 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
b93b01631 page cache: use x... |
80 |
* ->i_pages lock (arch-dependent flush_dcache_mmap_lock) |
1da177e4c Linux-2.6.12-rc2 |
81 82 83 84 |
* * ->mmap_sem * ->lock_page (access_process_vm) * |
ccad23656 kill generic_file... |
85 |
* ->i_mutex (generic_perform_write) |
82591e6ea [PATCH] mm: more ... |
86 |
* ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c Linux-2.6.12-rc2 |
87 |
* |
f758eeabe writeback: split ... |
88 |
* bdi->wb.list_lock |
a66979aba fs: move i_wb_lis... |
89 |
* sb_lock (fs/fs-writeback.c) |
b93b01631 page cache: use x... |
90 |
* ->i_pages lock (__sync_single_inode) |
1da177e4c Linux-2.6.12-rc2 |
91 |
* |
c8c06efa8 mm: convert i_mma... |
92 |
* ->i_mmap_rwsem |
1da177e4c Linux-2.6.12-rc2 |
93 94 95 |
* ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099 [PATCH] mm: updat... |
96 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c Linux-2.6.12-rc2 |
97 |
* |
b8072f099 [PATCH] mm: updat... |
98 |
* ->page_table_lock or pte_lock |
5d337b919 [PATCH] swap: swa... |
99 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c Linux-2.6.12-rc2 |
100 |
* ->private_lock (try_to_unmap_one) |
b93b01631 page cache: use x... |
101 |
* ->i_pages lock (try_to_unmap_one) |
f4b7e272b mm: remove zone_l... |
102 103 |
* ->pgdat->lru_lock (follow_page->mark_page_accessed) * ->pgdat->lru_lock (check_pte_range->isolate_lru_page) |
1da177e4c Linux-2.6.12-rc2 |
104 |
* ->private_lock (page_remove_rmap->set_page_dirty) |
b93b01631 page cache: use x... |
105 |
* ->i_pages lock (page_remove_rmap->set_page_dirty) |
f758eeabe writeback: split ... |
106 |
* bdi.wb->list_lock (page_remove_rmap->set_page_dirty) |
250df6ed2 fs: protect inode... |
107 |
* ->inode->i_lock (page_remove_rmap->set_page_dirty) |
81f8c3a46 mm: memcontrol: g... |
108 |
* ->memcg->move_lock (page_remove_rmap->lock_page_memcg) |
f758eeabe writeback: split ... |
109 |
* bdi.wb->list_lock (zap_pte_range->set_page_dirty) |
250df6ed2 fs: protect inode... |
110 |
* ->inode->i_lock (zap_pte_range->set_page_dirty) |
1da177e4c Linux-2.6.12-rc2 |
111 112 |
* ->private_lock (zap_pte_range->__set_page_dirty_buffers) * |
c8c06efa8 mm: convert i_mma... |
113 |
* ->i_mmap_rwsem |
9a3c531df mm: update stale ... |
114 |
* ->tasklist_lock (memory_failure, collect_procs_ao) |
1da177e4c Linux-2.6.12-rc2 |
115 |
*/ |
5c024e6a4 page cache: Conve... |
116 |
static void page_cache_delete(struct address_space *mapping, |
91b0abe36 mm + fs: store sh... |
117 118 |
struct page *page, void *shadow) { |
5c024e6a4 page cache: Conve... |
119 120 |
XA_STATE(xas, &mapping->i_pages, page->index); unsigned int nr = 1; |
c70b647d3 mm/filemap.c: add... |
121 |
|
5c024e6a4 page cache: Conve... |
122 |
mapping_set_update(&xas, mapping); |
c70b647d3 mm/filemap.c: add... |
123 |
|
5c024e6a4 page cache: Conve... |
124 125 126 |
/* hugetlb pages are represented by a single entry in the xarray */ if (!PageHuge(page)) { xas_set_order(&xas, page->index, compound_order(page)); |
d8c6546b1 mm: introduce com... |
127 |
nr = compound_nr(page); |
5c024e6a4 page cache: Conve... |
128 |
} |
91b0abe36 mm + fs: store sh... |
129 |
|
83929372f filemap: prepare ... |
130 131 132 |
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageTail(page), page); VM_BUG_ON_PAGE(nr != 1 && shadow, page); |
449dd6984 mm: keep page cac... |
133 |
|
5c024e6a4 page cache: Conve... |
134 135 |
xas_store(&xas, shadow); xas_init_marks(&xas); |
d3798ae8c mm: filemap: don'... |
136 |
|
2300638b1 mm: move clearing... |
137 138 |
page->mapping = NULL; /* Leave page->index set: truncation lookup relies upon it */ |
d3798ae8c mm: filemap: don'... |
139 140 141 142 143 144 145 146 147 148 149 |
if (shadow) { mapping->nrexceptional += nr; /* * Make sure the nrexceptional 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(); } mapping->nrpages -= nr; |
91b0abe36 mm + fs: store sh... |
150 |
} |
5ecc4d852 mm: factor out ch... |
151 152 |
static void unaccount_page_cache_page(struct address_space *mapping, struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
153 |
{ |
5ecc4d852 mm: factor out ch... |
154 |
int nr; |
1da177e4c Linux-2.6.12-rc2 |
155 |
|
c515e1fd3 mm/fs: add hooks ... |
156 157 158 159 160 161 162 163 |
/* * 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... |
164 |
cleancache_invalidate_page(mapping, page); |
c515e1fd3 mm/fs: add hooks ... |
165 |
|
83929372f filemap: prepare ... |
166 |
VM_BUG_ON_PAGE(PageTail(page), page); |
06b241f32 mm: __delete_from... |
167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 |
VM_BUG_ON_PAGE(page_mapped(page), page); if (!IS_ENABLED(CONFIG_DEBUG_VM) && unlikely(page_mapped(page))) { int mapcount; pr_alert("BUG: Bad page cache in process %s pfn:%05lx ", current->comm, page_to_pfn(page)); dump_page(page, "still mapped when deleted"); dump_stack(); add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE); mapcount = page_mapcount(page); if (mapping_exiting(mapping) && page_count(page) >= mapcount + 2) { /* * All vmas have already been torn down, so it's * a good bet that actually the page is unmapped, * and we'd prefer not to leak it: if we're wrong, * some other bad page check should catch it later. */ page_mapcount_reset(page); |
6d061f9f6 mm/page_ref: use ... |
188 |
page_ref_sub(page, mapcount); |
06b241f32 mm: __delete_from... |
189 190 |
} } |
4165b9b46 hugetlb: do not a... |
191 |
/* hugetlb pages do not participate in page cache accounting. */ |
5ecc4d852 mm: factor out ch... |
192 193 |
if (PageHuge(page)) return; |
09612fa65 mm: hugetlb: retu... |
194 |
|
5ecc4d852 mm: factor out ch... |
195 196 197 198 199 200 201 |
nr = hpage_nr_pages(page); __mod_node_page_state(page_pgdat(page), NR_FILE_PAGES, -nr); if (PageSwapBacked(page)) { __mod_node_page_state(page_pgdat(page), NR_SHMEM, -nr); if (PageTransHuge(page)) __dec_node_page_state(page, NR_SHMEM_THPS); |
99cb0dbd4 mm,thp: add read-... |
202 203 |
} else if (PageTransHuge(page)) { __dec_node_page_state(page, NR_FILE_THPS); |
09d91cda0 mm,thp: avoid wri... |
204 |
filemap_nr_thps_dec(mapping); |
800d8c63b shmem: add huge p... |
205 |
} |
5ecc4d852 mm: factor out ch... |
206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 |
/* * At this point page must be either written or cleaned by * truncate. Dirty page here signals a bug and loss of * unwritten data. * * This fixes dirty accounting after removing the page entirely * but leaves PageDirty set: it has no effect for truncated * page and anyway will be cleared before returning page into * buddy allocator. */ if (WARN_ON_ONCE(PageDirty(page))) account_page_cleaned(page, mapping, inode_to_wb(mapping->host)); } /* * Delete a page from the page cache and free it. Caller has to make * sure the page is locked and that nobody else uses it - or that usage |
b93b01631 page cache: use x... |
224 |
* is safe. The caller must hold the i_pages lock. |
5ecc4d852 mm: factor out ch... |
225 226 227 228 229 230 231 232 |
*/ void __delete_from_page_cache(struct page *page, void *shadow) { struct address_space *mapping = page->mapping; trace_mm_filemap_delete_from_page_cache(page); unaccount_page_cache_page(mapping, page); |
5c024e6a4 page cache: Conve... |
233 |
page_cache_delete(mapping, page, shadow); |
1da177e4c Linux-2.6.12-rc2 |
234 |
} |
59c66c5f8 mm: factor out pa... |
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 |
static void page_cache_free_page(struct address_space *mapping, struct page *page) { void (*freepage)(struct page *); freepage = mapping->a_ops->freepage; if (freepage) freepage(page); if (PageTransHuge(page) && !PageHuge(page)) { page_ref_sub(page, HPAGE_PMD_NR); VM_BUG_ON_PAGE(page_count(page) <= 0, page); } else { put_page(page); } } |
702cfbf93 mm: goodbye remov... |
251 252 253 254 255 256 257 258 259 |
/** * 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 |
260 |
{ |
83929372f filemap: prepare ... |
261 |
struct address_space *mapping = page_mapping(page); |
c4843a759 memcg: add per cg... |
262 |
unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
263 |
|
cd7619d6b [PATCH] Extermina... |
264 |
BUG_ON(!PageLocked(page)); |
b93b01631 page cache: use x... |
265 |
xa_lock_irqsave(&mapping->i_pages, flags); |
62cccb8c8 mm: simplify lock... |
266 |
__delete_from_page_cache(page, NULL); |
b93b01631 page cache: use x... |
267 |
xa_unlock_irqrestore(&mapping->i_pages, flags); |
6072d13c4 Call the filesyst... |
268 |
|
59c66c5f8 mm: factor out pa... |
269 |
page_cache_free_page(mapping, page); |
97cecb5a2 mm: introduce del... |
270 271 |
} EXPORT_SYMBOL(delete_from_page_cache); |
aa65c29ce mm: batch radix t... |
272 |
/* |
ef8e5717d page cache: Conve... |
273 |
* page_cache_delete_batch - delete several pages from page cache |
aa65c29ce mm: batch radix t... |
274 275 276 |
* @mapping: the mapping to which pages belong * @pvec: pagevec with pages to delete * |
b93b01631 page cache: use x... |
277 |
* The function walks over mapping->i_pages and removes pages passed in @pvec |
4101196b1 mm: page cache: s... |
278 279 |
* from the mapping. The function expects @pvec to be sorted by page index * and is optimised for it to be dense. |
b93b01631 page cache: use x... |
280 |
* It tolerates holes in @pvec (mapping entries at those indices are not |
aa65c29ce mm: batch radix t... |
281 |
* modified). The function expects only THP head pages to be present in the |
4101196b1 mm: page cache: s... |
282 |
* @pvec. |
aa65c29ce mm: batch radix t... |
283 |
* |
b93b01631 page cache: use x... |
284 |
* The function expects the i_pages lock to be held. |
aa65c29ce mm: batch radix t... |
285 |
*/ |
ef8e5717d page cache: Conve... |
286 |
static void page_cache_delete_batch(struct address_space *mapping, |
aa65c29ce mm: batch radix t... |
287 288 |
struct pagevec *pvec) { |
ef8e5717d page cache: Conve... |
289 |
XA_STATE(xas, &mapping->i_pages, pvec->pages[0]->index); |
aa65c29ce mm: batch radix t... |
290 |
int total_pages = 0; |
4101196b1 mm: page cache: s... |
291 |
int i = 0; |
aa65c29ce mm: batch radix t... |
292 |
struct page *page; |
aa65c29ce mm: batch radix t... |
293 |
|
ef8e5717d page cache: Conve... |
294 295 |
mapping_set_update(&xas, mapping); xas_for_each(&xas, page, ULONG_MAX) { |
4101196b1 mm: page cache: s... |
296 |
if (i >= pagevec_count(pvec)) |
aa65c29ce mm: batch radix t... |
297 |
break; |
4101196b1 mm: page cache: s... |
298 299 |
/* A swap/dax/shadow entry got inserted? Skip it. */ |
3159f943a xarray: Replace e... |
300 |
if (xa_is_value(page)) |
aa65c29ce mm: batch radix t... |
301 |
continue; |
4101196b1 mm: page cache: s... |
302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 |
/* * A page got inserted in our range? Skip it. We have our * pages locked so they are protected from being removed. * If we see a page whose index is higher than ours, it * means our page has been removed, which shouldn't be * possible because we're holding the PageLock. */ if (page != pvec->pages[i]) { VM_BUG_ON_PAGE(page->index > pvec->pages[i]->index, page); continue; } WARN_ON_ONCE(!PageLocked(page)); if (page->index == xas.xa_index) |
aa65c29ce mm: batch radix t... |
318 |
page->mapping = NULL; |
4101196b1 mm: page cache: s... |
319 320 321 322 323 324 325 326 |
/* Leave page->index set: truncation lookup relies on it */ /* * Move to the next page in the vector if this is a regular * page or the index is of the last sub-page of this compound * page. */ if (page->index + compound_nr(page) - 1 == xas.xa_index) |
aa65c29ce mm: batch radix t... |
327 |
i++; |
ef8e5717d page cache: Conve... |
328 |
xas_store(&xas, NULL); |
aa65c29ce mm: batch radix t... |
329 330 331 332 333 334 335 336 337 338 339 340 341 |
total_pages++; } mapping->nrpages -= total_pages; } void delete_from_page_cache_batch(struct address_space *mapping, struct pagevec *pvec) { int i; unsigned long flags; if (!pagevec_count(pvec)) return; |
b93b01631 page cache: use x... |
342 |
xa_lock_irqsave(&mapping->i_pages, flags); |
aa65c29ce mm: batch radix t... |
343 344 345 346 347 |
for (i = 0; i < pagevec_count(pvec); i++) { trace_mm_filemap_delete_from_page_cache(pvec->pages[i]); unaccount_page_cache_page(mapping, pvec->pages[i]); } |
ef8e5717d page cache: Conve... |
348 |
page_cache_delete_batch(mapping, pvec); |
b93b01631 page cache: use x... |
349 |
xa_unlock_irqrestore(&mapping->i_pages, flags); |
aa65c29ce mm: batch radix t... |
350 351 352 353 |
for (i = 0; i < pagevec_count(pvec); i++) page_cache_free_page(mapping, pvec->pages[i]); } |
d72d9e2a5 mm: export filema... |
354 |
int filemap_check_errors(struct address_space *mapping) |
865ffef37 fs: fix fsync() e... |
355 356 357 |
{ int ret = 0; /* Check for outstanding write errors */ |
7fcbbaf18 mm/filemap.c: avo... |
358 359 |
if (test_bit(AS_ENOSPC, &mapping->flags) && test_and_clear_bit(AS_ENOSPC, &mapping->flags)) |
865ffef37 fs: fix fsync() e... |
360 |
ret = -ENOSPC; |
7fcbbaf18 mm/filemap.c: avo... |
361 362 |
if (test_bit(AS_EIO, &mapping->flags) && test_and_clear_bit(AS_EIO, &mapping->flags)) |
865ffef37 fs: fix fsync() e... |
363 364 365 |
ret = -EIO; return ret; } |
d72d9e2a5 mm: export filema... |
366 |
EXPORT_SYMBOL(filemap_check_errors); |
865ffef37 fs: fix fsync() e... |
367 |
|
76341cabb jbd2: don't clear... |
368 369 370 371 372 373 374 375 376 |
static int filemap_check_and_keep_errors(struct address_space *mapping) { /* Check for outstanding write errors */ if (test_bit(AS_EIO, &mapping->flags)) return -EIO; if (test_bit(AS_ENOSPC, &mapping->flags)) return -ENOSPC; return 0; } |
1da177e4c Linux-2.6.12-rc2 |
377 |
/** |
485bb99b4 [PATCH] kernel-do... |
378 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b [PATCH] DocBook: ... |
379 380 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03 [PATCH] filemap_f... |
381 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b [PATCH] DocBook: ... |
382 |
* @sync_mode: enable synchronous operation |
1da177e4c Linux-2.6.12-rc2 |
383 |
* |
485bb99b4 [PATCH] kernel-do... |
384 385 386 |
* 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 |
387 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4 [PATCH] kernel-do... |
388 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c Linux-2.6.12-rc2 |
389 390 |
* these two operations is that if a dirty page/buffer is encountered, it must * be waited upon, and not just skipped over. |
a862f68a8 docs/core-api/mm:... |
391 392 |
* * Return: %0 on success, negative error code otherwise. |
1da177e4c Linux-2.6.12-rc2 |
393 |
*/ |
ebcf28e1c [PATCH] fadvise()... |
394 395 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c Linux-2.6.12-rc2 |
396 397 398 399 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd mm: write_cache_p... |
400 |
.nr_to_write = LONG_MAX, |
111ebb6e6 [PATCH] writeback... |
401 402 |
.range_start = start, .range_end = end, |
1da177e4c Linux-2.6.12-rc2 |
403 |
}; |
c3aab9a0b mm/filemap.c: don... |
404 405 |
if (!mapping_cap_writeback_dirty(mapping) || !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) |
1da177e4c Linux-2.6.12-rc2 |
406 |
return 0; |
b16b1deb5 writeback: make w... |
407 |
wbc_attach_fdatawrite_inode(&wbc, mapping->host); |
1da177e4c Linux-2.6.12-rc2 |
408 |
ret = do_writepages(mapping, &wbc); |
b16b1deb5 writeback: make w... |
409 |
wbc_detach_inode(&wbc); |
1da177e4c Linux-2.6.12-rc2 |
410 411 412 413 414 415 |
return ret; } static inline int __filemap_fdatawrite(struct address_space *mapping, int sync_mode) { |
111ebb6e6 [PATCH] writeback... |
416 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c Linux-2.6.12-rc2 |
417 418 419 420 421 422 423 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf vfs: export filem... |
424 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c [PATCH] fadvise()... |
425 |
loff_t end) |
1da177e4c Linux-2.6.12-rc2 |
426 427 428 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf vfs: export filem... |
429 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c Linux-2.6.12-rc2 |
430 |
|
485bb99b4 [PATCH] kernel-do... |
431 432 433 434 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c Linux-2.6.12-rc2 |
435 436 |
* This is a mostly non-blocking flush. Not suitable for data-integrity * purposes - I/O may not be started against all dirty pages. |
a862f68a8 docs/core-api/mm:... |
437 438 |
* * Return: %0 on success, negative error code otherwise. |
1da177e4c Linux-2.6.12-rc2 |
439 440 441 442 443 444 |
*/ int filemap_flush(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_NONE); } EXPORT_SYMBOL(filemap_flush); |
7fc9e4722 fs: Introduce fil... |
445 446 447 448 449 450 451 452 |
/** * filemap_range_has_page - check if a page exists in range. * @mapping: address space within which to check * @start_byte: offset in bytes where the range starts * @end_byte: offset in bytes where the range ends (inclusive) * * Find at least one page in the range supplied, usually used to check if * direct writing in this range will trigger a writeback. |
a862f68a8 docs/core-api/mm:... |
453 454 455 |
* * Return: %true if at least one page exists in the specified range, * %false otherwise. |
7fc9e4722 fs: Introduce fil... |
456 457 458 459 |
*/ bool filemap_range_has_page(struct address_space *mapping, loff_t start_byte, loff_t end_byte) { |
f7b680468 mm: use find_get_... |
460 |
struct page *page; |
8fa8e538e page cache: Conve... |
461 462 |
XA_STATE(xas, &mapping->i_pages, start_byte >> PAGE_SHIFT); pgoff_t max = end_byte >> PAGE_SHIFT; |
7fc9e4722 fs: Introduce fil... |
463 464 465 |
if (end_byte < start_byte) return false; |
8fa8e538e page cache: Conve... |
466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 |
rcu_read_lock(); for (;;) { page = xas_find(&xas, max); if (xas_retry(&xas, page)) continue; /* Shadow entries don't count */ if (xa_is_value(page)) continue; /* * We don't need to try to pin this page; we're about to * release the RCU lock anyway. It is enough to know that * there was a page here recently. */ break; } rcu_read_unlock(); |
7fc9e4722 fs: Introduce fil... |
482 |
|
8fa8e538e page cache: Conve... |
483 |
return page != NULL; |
7fc9e4722 fs: Introduce fil... |
484 485 |
} EXPORT_SYMBOL(filemap_range_has_page); |
5e8fcc1a0 mm: don't TestCle... |
486 |
static void __filemap_fdatawait_range(struct address_space *mapping, |
aa750fd71 mm/filemap.c: mak... |
487 |
loff_t start_byte, loff_t end_byte) |
1da177e4c Linux-2.6.12-rc2 |
488 |
{ |
09cbfeaf1 mm, fs: get rid o... |
489 490 |
pgoff_t index = start_byte >> PAGE_SHIFT; pgoff_t end = end_byte >> PAGE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
491 492 |
struct pagevec pvec; int nr_pages; |
1da177e4c Linux-2.6.12-rc2 |
493 |
|
94004ed72 kill wait_on_page... |
494 |
if (end_byte < start_byte) |
5e8fcc1a0 mm: don't TestCle... |
495 |
return; |
1da177e4c Linux-2.6.12-rc2 |
496 |
|
866798201 mm, pagevec: remo... |
497 |
pagevec_init(&pvec); |
312e9d2f7 mm: use pagevec_l... |
498 |
while (index <= end) { |
1da177e4c Linux-2.6.12-rc2 |
499 |
unsigned i; |
312e9d2f7 mm: use pagevec_l... |
500 |
nr_pages = pagevec_lookup_range_tag(&pvec, mapping, &index, |
67fd707f4 mm: remove nr_pag... |
501 |
end, PAGECACHE_TAG_WRITEBACK); |
312e9d2f7 mm: use pagevec_l... |
502 503 |
if (!nr_pages) break; |
1da177e4c Linux-2.6.12-rc2 |
504 505 |
for (i = 0; i < nr_pages; i++) { struct page *page = pvec.pages[i]; |
1da177e4c Linux-2.6.12-rc2 |
506 |
wait_on_page_writeback(page); |
5e8fcc1a0 mm: don't TestCle... |
507 |
ClearPageError(page); |
1da177e4c Linux-2.6.12-rc2 |
508 509 510 511 |
} pagevec_release(&pvec); cond_resched(); } |
aa750fd71 mm/filemap.c: mak... |
512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 |
} /** * 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) * * Walk the list of under-writeback pages of the given address space * in the given range and wait for all of them. Check error status of * the address space and return it. * * Since the error status of the address space is cleared by this function, * callers are responsible for checking the return value and handling and/or * reporting the error. |
a862f68a8 docs/core-api/mm:... |
527 528 |
* * Return: error status of the address space. |
aa750fd71 mm/filemap.c: mak... |
529 530 531 532 |
*/ int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) { |
5e8fcc1a0 mm: don't TestCle... |
533 534 |
__filemap_fdatawait_range(mapping, start_byte, end_byte); return filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
535 |
} |
d3bccb6f4 vfs: Introduce fi... |
536 537 538 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
aa0bfcd93 mm: add filemap_f... |
539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 |
* filemap_fdatawait_range_keep_errors - 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) * * Walk the list of under-writeback pages of the given address space in the * given range and wait for all of them. Unlike filemap_fdatawait_range(), * this function does not clear error status of the address space. * * Use this function if callers don't handle errors themselves. Expected * call sites are system-wide / filesystem-wide data flushers: e.g. sync(2), * fsfreeze(8) */ int filemap_fdatawait_range_keep_errors(struct address_space *mapping, loff_t start_byte, loff_t end_byte) { __filemap_fdatawait_range(mapping, start_byte, end_byte); return filemap_check_and_keep_errors(mapping); } EXPORT_SYMBOL(filemap_fdatawait_range_keep_errors); /** |
a823e4589 mm: add file_fdat... |
561 562 563 564 565 566 567 568 569 570 571 572 |
* file_fdatawait_range - wait for writeback to complete * @file: file pointing to 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) * * Walk the list of under-writeback pages of the address space that file * refers to, in the given range and wait for all of them. Check error * status of the address space vs. the file->f_wb_err cursor and return it. * * Since the error status of the file is advanced by this function, * callers are responsible for checking the return value and handling and/or * reporting the error. |
a862f68a8 docs/core-api/mm:... |
573 574 |
* * Return: error status of the address space vs. the file->f_wb_err cursor. |
a823e4589 mm: add file_fdat... |
575 576 577 578 579 580 581 582 583 |
*/ int file_fdatawait_range(struct file *file, loff_t start_byte, loff_t end_byte) { struct address_space *mapping = file->f_mapping; __filemap_fdatawait_range(mapping, start_byte, end_byte); return file_check_and_advance_wb_err(file); } EXPORT_SYMBOL(file_fdatawait_range); |
d3bccb6f4 vfs: Introduce fi... |
584 585 |
/** |
aa750fd71 mm/filemap.c: mak... |
586 587 588 589 590 591 592 593 594 595 |
* filemap_fdatawait_keep_errors - wait for writeback without clearing errors * @mapping: address space structure to wait for * * Walk the list of under-writeback pages of the given address space * and wait for all of them. Unlike filemap_fdatawait(), this function * does not clear error status of the address space. * * Use this function if callers don't handle errors themselves. Expected * call sites are system-wide / filesystem-wide data flushers: e.g. sync(2), * fsfreeze(8) |
a862f68a8 docs/core-api/mm:... |
596 597 |
* * Return: error status of the address space. |
aa750fd71 mm/filemap.c: mak... |
598 |
*/ |
76341cabb jbd2: don't clear... |
599 |
int filemap_fdatawait_keep_errors(struct address_space *mapping) |
aa750fd71 mm/filemap.c: mak... |
600 |
{ |
ffb959bbd mm: remove optimi... |
601 |
__filemap_fdatawait_range(mapping, 0, LLONG_MAX); |
76341cabb jbd2: don't clear... |
602 |
return filemap_check_and_keep_errors(mapping); |
aa750fd71 mm/filemap.c: mak... |
603 |
} |
76341cabb jbd2: don't clear... |
604 |
EXPORT_SYMBOL(filemap_fdatawait_keep_errors); |
aa750fd71 mm/filemap.c: mak... |
605 |
|
875d91b11 mm/filemap.c: rew... |
606 |
/* Returns true if writeback might be needed or already in progress. */ |
9326c9b20 mm: consolidate d... |
607 |
static bool mapping_needs_writeback(struct address_space *mapping) |
1da177e4c Linux-2.6.12-rc2 |
608 |
{ |
875d91b11 mm/filemap.c: rew... |
609 610 611 612 |
if (dax_mapping(mapping)) return mapping->nrexceptional; return mapping->nrpages; |
1da177e4c Linux-2.6.12-rc2 |
613 |
} |
1da177e4c Linux-2.6.12-rc2 |
614 615 616 |
int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982 [PATCH] Fix and a... |
617 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
618 |
|
9326c9b20 mm: consolidate d... |
619 |
if (mapping_needs_writeback(mapping)) { |
28fd12982 [PATCH] Fix and a... |
620 621 622 623 624 625 626 627 628 629 630 |
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; |
cbeaf9510 mm: clear AS_EIO/... |
631 632 633 |
} else { /* Clear any previously stored errors */ filemap_check_errors(mapping); |
28fd12982 [PATCH] Fix and a... |
634 |
} |
865ffef37 fs: fix fsync() e... |
635 636 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
637 |
} |
28fd12982 [PATCH] Fix and a... |
638 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
639 |
} |
28fd12982 [PATCH] Fix and a... |
640 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c Linux-2.6.12-rc2 |
641 |
|
485bb99b4 [PATCH] kernel-do... |
642 643 644 645 646 647 |
/** * 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... |
648 649 |
* Write out and wait upon file offsets lstart->lend, inclusive. * |
0e056eb55 kernel-api.rst: f... |
650 |
* Note that @lend is inclusive (describes the last byte to be written) so |
469eb4d03 [PATCH] filemap_f... |
651 |
* that this function can be used to write to the very end-of-file (end = -1). |
a862f68a8 docs/core-api/mm:... |
652 653 |
* * Return: error status of the address space. |
469eb4d03 [PATCH] filemap_f... |
654 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
655 656 657 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982 [PATCH] Fix and a... |
658 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
659 |
|
9326c9b20 mm: consolidate d... |
660 |
if (mapping_needs_writeback(mapping)) { |
28fd12982 [PATCH] Fix and a... |
661 662 663 664 |
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... |
665 666 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982 [PATCH] Fix and a... |
667 668 |
if (!err) err = err2; |
cbeaf9510 mm: clear AS_EIO/... |
669 670 671 |
} else { /* Clear any previously stored errors */ filemap_check_errors(mapping); |
28fd12982 [PATCH] Fix and a... |
672 |
} |
865ffef37 fs: fix fsync() e... |
673 674 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
675 |
} |
28fd12982 [PATCH] Fix and a... |
676 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
677 |
} |
f69955855 Export filemap_wr... |
678 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c Linux-2.6.12-rc2 |
679 |
|
5660e13d2 fs: new infrastru... |
680 681 |
void __filemap_set_wb_err(struct address_space *mapping, int err) { |
3acdfd280 errseq: rename __... |
682 |
errseq_t eseq = errseq_set(&mapping->wb_err, err); |
5660e13d2 fs: new infrastru... |
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 |
trace_filemap_set_wb_err(mapping, eseq); } EXPORT_SYMBOL(__filemap_set_wb_err); /** * file_check_and_advance_wb_err - report wb error (if any) that was previously * and advance wb_err to current one * @file: struct file on which the error is being reported * * When userland calls fsync (or something like nfsd does the equivalent), we * want to report any writeback errors that occurred since the last fsync (or * since the file was opened if there haven't been any). * * Grab the wb_err from the mapping. If it matches what we have in the file, * then just quickly return 0. The file is all caught up. * * If it doesn't match, then take the mapping value, set the "seen" flag in * it and try to swap it into place. If it works, or another task beat us * to it with the new value, then update the f_wb_err and return the error * portion. The error at this point must be reported via proper channels * (a'la fsync, or NFS COMMIT operation, etc.). * * While we handle mapping->wb_err with atomic operations, the f_wb_err * value is protected by the f_lock since we must ensure that it reflects * the latest value swapped in for this file descriptor. |
a862f68a8 docs/core-api/mm:... |
709 710 |
* * Return: %0 on success, negative error code otherwise. |
5660e13d2 fs: new infrastru... |
711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 |
*/ int file_check_and_advance_wb_err(struct file *file) { int err = 0; errseq_t old = READ_ONCE(file->f_wb_err); struct address_space *mapping = file->f_mapping; /* Locklessly handle the common case where nothing has changed */ if (errseq_check(&mapping->wb_err, old)) { /* Something changed, must use slow path */ spin_lock(&file->f_lock); old = file->f_wb_err; err = errseq_check_and_advance(&mapping->wb_err, &file->f_wb_err); trace_file_check_and_advance_wb_err(file, old); spin_unlock(&file->f_lock); } |
f4e222c56 mm: have filemap_... |
728 729 730 731 732 733 734 735 |
/* * We're mostly using this function as a drop in replacement for * filemap_check_errors. Clear AS_EIO/AS_ENOSPC to emulate the effect * that the legacy code would have had on these flags. */ clear_bit(AS_EIO, &mapping->flags); clear_bit(AS_ENOSPC, &mapping->flags); |
5660e13d2 fs: new infrastru... |
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 |
return err; } EXPORT_SYMBOL(file_check_and_advance_wb_err); /** * file_write_and_wait_range - write out & wait on a file range * @file: file pointing to address_space with pages * @lstart: offset in bytes where the range starts * @lend: offset in bytes where the range ends (inclusive) * * 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). * * After writing out and waiting on the data, we check and advance the * f_wb_err cursor to the latest value, and return any errors detected there. |
a862f68a8 docs/core-api/mm:... |
753 754 |
* * Return: %0 on success, negative error code otherwise. |
5660e13d2 fs: new infrastru... |
755 756 757 758 759 |
*/ int file_write_and_wait_range(struct file *file, loff_t lstart, loff_t lend) { int err = 0, err2; struct address_space *mapping = file->f_mapping; |
9326c9b20 mm: consolidate d... |
760 |
if (mapping_needs_writeback(mapping)) { |
5660e13d2 fs: new infrastru... |
761 762 763 764 765 766 767 768 769 770 771 772 |
err = __filemap_fdatawrite_range(mapping, lstart, lend, WB_SYNC_ALL); /* See comment of filemap_write_and_wait() */ if (err != -EIO) __filemap_fdatawait_range(mapping, lstart, lend); } err2 = file_check_and_advance_wb_err(file); if (!err) err = err2; return err; } EXPORT_SYMBOL(file_write_and_wait_range); |
485bb99b4 [PATCH] kernel-do... |
773 |
/** |
ef6a3c631 mm: add replace_p... |
774 775 776 777 778 779 780 781 782 783 784 |
* 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. * |
74d609585 page cache: Add a... |
785 |
* The remove + add is atomic. This function cannot fail. |
a862f68a8 docs/core-api/mm:... |
786 787 |
* * Return: %0 |
ef6a3c631 mm: add replace_p... |
788 789 790 |
*/ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) { |
74d609585 page cache: Add a... |
791 792 793 794 795 |
struct address_space *mapping = old->mapping; void (*freepage)(struct page *) = mapping->a_ops->freepage; pgoff_t offset = old->index; XA_STATE(xas, &mapping->i_pages, offset); unsigned long flags; |
ef6a3c631 mm: add replace_p... |
796 |
|
309381fea mm: dump page whe... |
797 798 799 |
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... |
800 |
|
74d609585 page cache: Add a... |
801 802 803 |
get_page(new); new->mapping = mapping; new->index = offset; |
ef6a3c631 mm: add replace_p... |
804 |
|
74d609585 page cache: Add a... |
805 806 |
xas_lock_irqsave(&xas, flags); xas_store(&xas, new); |
4165b9b46 hugetlb: do not a... |
807 |
|
74d609585 page cache: Add a... |
808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 |
old->mapping = NULL; /* hugetlb pages do not participate in page cache accounting. */ if (!PageHuge(old)) __dec_node_page_state(new, NR_FILE_PAGES); if (!PageHuge(new)) __inc_node_page_state(new, NR_FILE_PAGES); if (PageSwapBacked(old)) __dec_node_page_state(new, NR_SHMEM); if (PageSwapBacked(new)) __inc_node_page_state(new, NR_SHMEM); xas_unlock_irqrestore(&xas, flags); mem_cgroup_migrate(old, new); if (freepage) freepage(old); put_page(old); |
ef6a3c631 mm: add replace_p... |
823 |
|
74d609585 page cache: Add a... |
824 |
return 0; |
ef6a3c631 mm: add replace_p... |
825 826 |
} EXPORT_SYMBOL_GPL(replace_page_cache_page); |
a528910e1 mm: thrash detect... |
827 828 829 830 |
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 |
831 |
{ |
74d609585 page cache: Add a... |
832 |
XA_STATE(xas, &mapping->i_pages, offset); |
00501b531 mm: memcontrol: r... |
833 834 |
int huge = PageHuge(page); struct mem_cgroup *memcg; |
e286781d5 mm: speculative p... |
835 |
int error; |
74d609585 page cache: Add a... |
836 |
void *old; |
e286781d5 mm: speculative p... |
837 |
|
309381fea mm: dump page whe... |
838 839 |
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageSwapBacked(page), page); |
74d609585 page cache: Add a... |
840 |
mapping_set_update(&xas, mapping); |
e286781d5 mm: speculative p... |
841 |
|
00501b531 mm: memcontrol: r... |
842 843 |
if (!huge) { error = mem_cgroup_try_charge(page, current->mm, |
f627c2f53 memcg: adjust to ... |
844 |
gfp_mask, &memcg, false); |
00501b531 mm: memcontrol: r... |
845 846 847 |
if (error) return error; } |
1da177e4c Linux-2.6.12-rc2 |
848 |
|
09cbfeaf1 mm, fs: get rid o... |
849 |
get_page(page); |
66a0c8ee3 mm: cleanup add_t... |
850 851 |
page->mapping = mapping; page->index = offset; |
74d609585 page cache: Add a... |
852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 |
do { xas_lock_irq(&xas); old = xas_load(&xas); if (old && !xa_is_value(old)) xas_set_err(&xas, -EEXIST); xas_store(&xas, page); if (xas_error(&xas)) goto unlock; if (xa_is_value(old)) { mapping->nrexceptional--; if (shadowp) *shadowp = old; } mapping->nrpages++; /* hugetlb pages do not participate in page cache accounting */ if (!huge) __inc_node_page_state(page, NR_FILE_PAGES); unlock: xas_unlock_irq(&xas); } while (xas_nomem(&xas, gfp_mask & GFP_RECLAIM_MASK)); if (xas_error(&xas)) goto error; |
4165b9b46 hugetlb: do not a... |
877 |
|
00501b531 mm: memcontrol: r... |
878 |
if (!huge) |
f627c2f53 memcg: adjust to ... |
879 |
mem_cgroup_commit_charge(page, memcg, false, false); |
66a0c8ee3 mm: cleanup add_t... |
880 881 |
trace_mm_filemap_add_to_page_cache(page); return 0; |
74d609585 page cache: Add a... |
882 |
error: |
66a0c8ee3 mm: cleanup add_t... |
883 884 |
page->mapping = NULL; /* Leave page->index set: truncation relies upon it */ |
00501b531 mm: memcontrol: r... |
885 |
if (!huge) |
f627c2f53 memcg: adjust to ... |
886 |
mem_cgroup_cancel_charge(page, memcg, false); |
09cbfeaf1 mm, fs: get rid o... |
887 |
put_page(page); |
74d609585 page cache: Add a... |
888 |
return xas_error(&xas); |
1da177e4c Linux-2.6.12-rc2 |
889 |
} |
cfcbfb138 mm/filemap.c: ena... |
890 |
ALLOW_ERROR_INJECTION(__add_to_page_cache_locked, ERRNO); |
a528910e1 mm: thrash detect... |
891 892 893 894 895 896 897 898 899 900 |
/** * 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. |
a862f68a8 docs/core-api/mm:... |
901 902 |
* * Return: %0 on success, negative error code otherwise. |
a528910e1 mm: thrash detect... |
903 904 905 906 907 908 909 |
*/ 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... |
910 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c Linux-2.6.12-rc2 |
911 912 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
913 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
914 |
{ |
a528910e1 mm: thrash detect... |
915 |
void *shadow = NULL; |
4f98a2fee vmscan: split LRU... |
916 |
int ret; |
48c935ad8 page-flags: defin... |
917 |
__SetPageLocked(page); |
a528910e1 mm: thrash detect... |
918 919 920 |
ret = __add_to_page_cache_locked(page, mapping, offset, gfp_mask, &shadow); if (unlikely(ret)) |
48c935ad8 page-flags: defin... |
921 |
__ClearPageLocked(page); |
a528910e1 mm: thrash detect... |
922 923 924 925 926 |
else { /* * The page might have been evicted from cache only * recently, in which case it should be activated like * any other repeatedly accessed page. |
f0281a00f mm: workingset: o... |
927 928 929 |
* The exception is pages getting rewritten; evicting other * data from the working set, only to cache data that will * get overwritten with something else, is a waste of memory. |
a528910e1 mm: thrash detect... |
930 |
*/ |
1899ad18c mm: workingset: t... |
931 932 933 |
WARN_ON_ONCE(PageActive(page)); if (!(gfp_mask & __GFP_WRITE) && shadow) workingset_refault(page, shadow); |
a528910e1 mm: thrash detect... |
934 935 |
lru_cache_add(page); } |
1da177e4c Linux-2.6.12-rc2 |
936 937 |
return ret; } |
18bc0bbd1 Staging: pohmelfs... |
938 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c Linux-2.6.12-rc2 |
939 |
|
44110fe38 [PATCH] cpuset me... |
940 |
#ifdef CONFIG_NUMA |
2ae88149a [PATCH] mm: clean... |
941 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38 [PATCH] cpuset me... |
942 |
{ |
c0ff7453b cpuset,mm: fix no... |
943 944 |
int n; struct page *page; |
44110fe38 [PATCH] cpuset me... |
945 |
if (cpuset_do_page_mem_spread()) { |
cc9a6c877 cpuset: mm: reduc... |
946 947 |
unsigned int cpuset_mems_cookie; do { |
d26914d11 mm: optimize put_... |
948 |
cpuset_mems_cookie = read_mems_allowed_begin(); |
cc9a6c877 cpuset: mm: reduc... |
949 |
n = cpuset_mem_spread_node(); |
96db800f5 mm: rename alloc_... |
950 |
page = __alloc_pages_node(n, gfp, 0); |
d26914d11 mm: optimize put_... |
951 |
} while (!page && read_mems_allowed_retry(cpuset_mems_cookie)); |
cc9a6c877 cpuset: mm: reduc... |
952 |
|
c0ff7453b cpuset,mm: fix no... |
953 |
return page; |
44110fe38 [PATCH] cpuset me... |
954 |
} |
2ae88149a [PATCH] mm: clean... |
955 |
return alloc_pages(gfp, 0); |
44110fe38 [PATCH] cpuset me... |
956 |
} |
2ae88149a [PATCH] mm: clean... |
957 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38 [PATCH] cpuset me... |
958 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
959 960 961 962 963 964 965 966 967 968 |
/* * 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. */ |
629060270 mm: add PageWaite... |
969 970 971 972 973 |
#define PAGE_WAIT_TABLE_BITS 8 #define PAGE_WAIT_TABLE_SIZE (1 << PAGE_WAIT_TABLE_BITS) static wait_queue_head_t page_wait_table[PAGE_WAIT_TABLE_SIZE] __cacheline_aligned; static wait_queue_head_t *page_waitqueue(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
974 |
{ |
629060270 mm: add PageWaite... |
975 |
return &page_wait_table[hash_ptr(page, PAGE_WAIT_TABLE_BITS)]; |
1da177e4c Linux-2.6.12-rc2 |
976 |
} |
1da177e4c Linux-2.6.12-rc2 |
977 |
|
629060270 mm: add PageWaite... |
978 |
void __init pagecache_init(void) |
1da177e4c Linux-2.6.12-rc2 |
979 |
{ |
629060270 mm: add PageWaite... |
980 |
int i; |
1da177e4c Linux-2.6.12-rc2 |
981 |
|
629060270 mm: add PageWaite... |
982 983 984 985 |
for (i = 0; i < PAGE_WAIT_TABLE_SIZE; i++) init_waitqueue_head(&page_wait_table[i]); page_writeback_init(); |
1da177e4c Linux-2.6.12-rc2 |
986 |
} |
1da177e4c Linux-2.6.12-rc2 |
987 |
|
3510ca20e Minor page waitqu... |
988 |
/* This has the same layout as wait_bit_key - see fs/cachefiles/rdwr.c */ |
629060270 mm: add PageWaite... |
989 990 991 992 993 994 995 996 997 |
struct wait_page_key { struct page *page; int bit_nr; int page_match; }; struct wait_page_queue { struct page *page; int bit_nr; |
ac6424b98 sched/wait: Renam... |
998 |
wait_queue_entry_t wait; |
629060270 mm: add PageWaite... |
999 |
}; |
ac6424b98 sched/wait: Renam... |
1000 |
static int wake_page_function(wait_queue_entry_t *wait, unsigned mode, int sync, void *arg) |
f62e00cc3 mm: introduce wai... |
1001 |
{ |
629060270 mm: add PageWaite... |
1002 1003 1004 1005 1006 1007 1008 |
struct wait_page_key *key = arg; struct wait_page_queue *wait_page = container_of(wait, struct wait_page_queue, wait); if (wait_page->page != key->page) return 0; key->page_match = 1; |
f62e00cc3 mm: introduce wai... |
1009 |
|
629060270 mm: add PageWaite... |
1010 1011 |
if (wait_page->bit_nr != key->bit_nr) return 0; |
3510ca20e Minor page waitqu... |
1012 |
|
9a1ea439b mm: put_and_wait_... |
1013 1014 1015 1016 1017 1018 1019 1020 |
/* * Stop walking if it's locked. * Is this safe if put_and_wait_on_page_locked() is in use? * Yes: the waker must hold a reference to this page, and if PG_locked * has now already been set by another task, that task must also hold * a reference to the *same usage* of this page; so there is no need * to walk on to wake even the put_and_wait_on_page_locked() callers. */ |
629060270 mm: add PageWaite... |
1021 |
if (test_bit(key->bit_nr, &key->page->flags)) |
3510ca20e Minor page waitqu... |
1022 |
return -1; |
f62e00cc3 mm: introduce wai... |
1023 |
|
629060270 mm: add PageWaite... |
1024 |
return autoremove_wake_function(wait, mode, sync, key); |
f62e00cc3 mm: introduce wai... |
1025 |
} |
74d81bfae mm: un-export wak... |
1026 |
static void wake_up_page_bit(struct page *page, int bit_nr) |
cbbce8220 SCHED: add some "... |
1027 |
{ |
629060270 mm: add PageWaite... |
1028 1029 1030 |
wait_queue_head_t *q = page_waitqueue(page); struct wait_page_key key; unsigned long flags; |
11a19c7b0 sched/wait: Intro... |
1031 |
wait_queue_entry_t bookmark; |
cbbce8220 SCHED: add some "... |
1032 |
|
629060270 mm: add PageWaite... |
1033 1034 1035 |
key.page = page; key.bit_nr = bit_nr; key.page_match = 0; |
11a19c7b0 sched/wait: Intro... |
1036 1037 1038 1039 |
bookmark.flags = 0; bookmark.private = NULL; bookmark.func = NULL; INIT_LIST_HEAD(&bookmark.entry); |
629060270 mm: add PageWaite... |
1040 |
spin_lock_irqsave(&q->lock, flags); |
11a19c7b0 sched/wait: Intro... |
1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 |
__wake_up_locked_key_bookmark(q, TASK_NORMAL, &key, &bookmark); while (bookmark.flags & WQ_FLAG_BOOKMARK) { /* * Take a breather from holding the lock, * allow pages that finish wake up asynchronously * to acquire the lock and remove themselves * from wait queue */ spin_unlock_irqrestore(&q->lock, flags); cpu_relax(); spin_lock_irqsave(&q->lock, flags); __wake_up_locked_key_bookmark(q, TASK_NORMAL, &key, &bookmark); } |
629060270 mm: add PageWaite... |
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 |
/* * It is possible for other pages to have collided on the waitqueue * hash, so in that case check for a page match. That prevents a long- * term waiter * * It is still possible to miss a case here, when we woke page waiters * and removed them from the waitqueue, but there are still other * page waiters. */ if (!waitqueue_active(q) || !key.page_match) { ClearPageWaiters(page); /* * It's possible to miss clearing Waiters here, when we woke * our page waiters, but the hashed waitqueue has waiters for * other pages on it. * * That's okay, it's a rare case. The next waker will clear it. */ } spin_unlock_irqrestore(&q->lock, flags); } |
74d81bfae mm: un-export wak... |
1076 1077 1078 1079 1080 1081 1082 |
static void wake_up_page(struct page *page, int bit) { if (!PageWaiters(page)) return; wake_up_page_bit(page, bit); } |
629060270 mm: add PageWaite... |
1083 |
|
9a1ea439b mm: put_and_wait_... |
1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 |
/* * A choice of three behaviors for wait_on_page_bit_common(): */ enum behavior { EXCLUSIVE, /* Hold ref to page and take the bit when woken, like * __lock_page() waiting on then setting PG_locked. */ SHARED, /* Hold ref to page and check the bit when woken, like * wait_on_page_writeback() waiting on PG_writeback. */ DROP, /* Drop ref to page before wait, no check when woken, * like put_and_wait_on_page_locked() on PG_locked. */ }; |
629060270 mm: add PageWaite... |
1098 |
static inline int wait_on_page_bit_common(wait_queue_head_t *q, |
9a1ea439b mm: put_and_wait_... |
1099 |
struct page *page, int bit_nr, int state, enum behavior behavior) |
629060270 mm: add PageWaite... |
1100 1101 |
{ struct wait_page_queue wait_page; |
ac6424b98 sched/wait: Renam... |
1102 |
wait_queue_entry_t *wait = &wait_page.wait; |
9a1ea439b mm: put_and_wait_... |
1103 |
bool bit_is_set; |
b1d29ba82 delayacct: track ... |
1104 |
bool thrashing = false; |
9a1ea439b mm: put_and_wait_... |
1105 |
bool delayacct = false; |
eb414681d psi: pressure sta... |
1106 |
unsigned long pflags; |
629060270 mm: add PageWaite... |
1107 |
int ret = 0; |
eb414681d psi: pressure sta... |
1108 |
if (bit_nr == PG_locked && |
b1d29ba82 delayacct: track ... |
1109 |
!PageUptodate(page) && PageWorkingset(page)) { |
9a1ea439b mm: put_and_wait_... |
1110 |
if (!PageSwapBacked(page)) { |
eb414681d psi: pressure sta... |
1111 |
delayacct_thrashing_start(); |
9a1ea439b mm: put_and_wait_... |
1112 1113 |
delayacct = true; } |
eb414681d psi: pressure sta... |
1114 |
psi_memstall_enter(&pflags); |
b1d29ba82 delayacct: track ... |
1115 1116 |
thrashing = true; } |
629060270 mm: add PageWaite... |
1117 |
init_wait(wait); |
9a1ea439b mm: put_and_wait_... |
1118 |
wait->flags = behavior == EXCLUSIVE ? WQ_FLAG_EXCLUSIVE : 0; |
629060270 mm: add PageWaite... |
1119 1120 1121 1122 1123 1124 |
wait->func = wake_page_function; wait_page.page = page; wait_page.bit_nr = bit_nr; for (;;) { spin_lock_irq(&q->lock); |
2055da973 sched/wait: Disam... |
1125 |
if (likely(list_empty(&wait->entry))) { |
3510ca20e Minor page waitqu... |
1126 |
__add_wait_queue_entry_tail(q, wait); |
629060270 mm: add PageWaite... |
1127 1128 1129 1130 1131 1132 |
SetPageWaiters(page); } set_current_state(state); spin_unlock_irq(&q->lock); |
9a1ea439b mm: put_and_wait_... |
1133 1134 1135 1136 1137 |
bit_is_set = test_bit(bit_nr, &page->flags); if (behavior == DROP) put_page(page); if (likely(bit_is_set)) |
629060270 mm: add PageWaite... |
1138 |
io_schedule(); |
629060270 mm: add PageWaite... |
1139 |
|
9a1ea439b mm: put_and_wait_... |
1140 |
if (behavior == EXCLUSIVE) { |
629060270 mm: add PageWaite... |
1141 1142 |
if (!test_and_set_bit_lock(bit_nr, &page->flags)) break; |
9a1ea439b mm: put_and_wait_... |
1143 |
} else if (behavior == SHARED) { |
629060270 mm: add PageWaite... |
1144 1145 1146 |
if (!test_bit(bit_nr, &page->flags)) break; } |
a8b169afb Avoid page waitqu... |
1147 |
|
fa45f1162 mm/: remove calle... |
1148 |
if (signal_pending_state(state, current)) { |
a8b169afb Avoid page waitqu... |
1149 1150 1151 |
ret = -EINTR; break; } |
9a1ea439b mm: put_and_wait_... |
1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 |
if (behavior == DROP) { /* * We can no longer safely access page->flags: * even if CONFIG_MEMORY_HOTREMOVE is not enabled, * there is a risk of waiting forever on a page reused * for something that keeps it locked indefinitely. * But best check for -EINTR above before breaking. */ break; } |
629060270 mm: add PageWaite... |
1163 1164 1165 |
} finish_wait(q, wait); |
eb414681d psi: pressure sta... |
1166 |
if (thrashing) { |
9a1ea439b mm: put_and_wait_... |
1167 |
if (delayacct) |
eb414681d psi: pressure sta... |
1168 1169 1170 |
delayacct_thrashing_end(); psi_memstall_leave(&pflags); } |
b1d29ba82 delayacct: track ... |
1171 |
|
629060270 mm: add PageWaite... |
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 |
/* * A signal could leave PageWaiters set. Clearing it here if * !waitqueue_active would be possible (by open-coding finish_wait), * but still fail to catch it in the case of wait hash collision. We * already can fail to clear wait hash collision cases, so don't * bother with signals either. */ return ret; } void wait_on_page_bit(struct page *page, int bit_nr) { wait_queue_head_t *q = page_waitqueue(page); |
9a1ea439b mm: put_and_wait_... |
1186 |
wait_on_page_bit_common(q, page, bit_nr, TASK_UNINTERRUPTIBLE, SHARED); |
629060270 mm: add PageWaite... |
1187 1188 1189 1190 1191 1192 |
} EXPORT_SYMBOL(wait_on_page_bit); int wait_on_page_bit_killable(struct page *page, int bit_nr) { wait_queue_head_t *q = page_waitqueue(page); |
9a1ea439b mm: put_and_wait_... |
1193 |
return wait_on_page_bit_common(q, page, bit_nr, TASK_KILLABLE, SHARED); |
cbbce8220 SCHED: add some "... |
1194 |
} |
4343d0087 afs: Get rid of t... |
1195 |
EXPORT_SYMBOL(wait_on_page_bit_killable); |
cbbce8220 SCHED: add some "... |
1196 |
|
1da177e4c Linux-2.6.12-rc2 |
1197 |
/** |
9a1ea439b mm: put_and_wait_... |
1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 |
* put_and_wait_on_page_locked - Drop a reference and wait for it to be unlocked * @page: The page to wait for. * * The caller should hold a reference on @page. They expect the page to * become unlocked relatively soon, but do not wish to hold up migration * (for example) by holding the reference while waiting for the page to * come unlocked. After this function returns, the caller should not * dereference @page. */ void put_and_wait_on_page_locked(struct page *page) { wait_queue_head_t *q; page = compound_head(page); q = page_waitqueue(page); wait_on_page_bit_common(q, page, PG_locked, TASK_UNINTERRUPTIBLE, DROP); } /** |
385e1ca5f CacheFiles: Permi... |
1217 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc filemap: fix kern... |
1218 1219 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f CacheFiles: Permi... |
1220 1221 1222 |
* * Add an arbitrary @waiter to the wait queue for the nominated @page. */ |
ac6424b98 sched/wait: Renam... |
1223 |
void add_page_wait_queue(struct page *page, wait_queue_entry_t *waiter) |
385e1ca5f CacheFiles: Permi... |
1224 1225 1226 1227 1228 |
{ wait_queue_head_t *q = page_waitqueue(page); unsigned long flags; spin_lock_irqsave(&q->lock, flags); |
9c3a815f4 page waitqueue: a... |
1229 |
__add_wait_queue_entry_tail(q, waiter); |
629060270 mm: add PageWaite... |
1230 |
SetPageWaiters(page); |
385e1ca5f CacheFiles: Permi... |
1231 1232 1233 |
spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL_GPL(add_page_wait_queue); |
b91e1302a mm: optimize Page... |
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 |
#ifndef clear_bit_unlock_is_negative_byte /* * PG_waiters is the high bit in the same byte as PG_lock. * * On x86 (and on many other architectures), we can clear PG_lock and * test the sign bit at the same time. But if the architecture does * not support that special operation, we just do this all by hand * instead. * * The read of PG_waiters has to be after (or concurrently with) PG_locked * being cleared, but a memory barrier should be unneccssary since it is * in the same byte as PG_locked. */ static inline bool clear_bit_unlock_is_negative_byte(long nr, volatile void *mem) { clear_bit_unlock(nr, mem); /* smp_mb__after_atomic(); */ |
98473f9f3 mm/filemap: fix p... |
1252 |
return test_bit(PG_waiters, mem); |
b91e1302a mm: optimize Page... |
1253 1254 1255 |
} #endif |
385e1ca5f CacheFiles: Permi... |
1256 |
/** |
485bb99b4 [PATCH] kernel-do... |
1257 |
* unlock_page - unlock a locked page |
1da177e4c Linux-2.6.12-rc2 |
1258 1259 1260 1261 |
* @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 |
da3dae54e Documentation: Do... |
1262 |
* mechanism between PageLocked pages and PageWriteback pages is shared. |
1da177e4c Linux-2.6.12-rc2 |
1263 1264 |
* But that's OK - sleepers in wait_on_page_writeback() just go back to sleep. * |
b91e1302a mm: optimize Page... |
1265 1266 1267 1268 1269 |
* Note that this depends on PG_waiters being the sign bit in the byte * that contains PG_locked - thus the BUILD_BUG_ON(). That allows us to * clear the PG_locked bit and test PG_waiters at the same time fairly * portably (architectures that do LL/SC can test any bit, while x86 can * test the sign bit). |
1da177e4c Linux-2.6.12-rc2 |
1270 |
*/ |
920c7a5d0 mm: remove fastca... |
1271 |
void unlock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
1272 |
{ |
b91e1302a mm: optimize Page... |
1273 |
BUILD_BUG_ON(PG_waiters != 7); |
48c935ad8 page-flags: defin... |
1274 |
page = compound_head(page); |
309381fea mm: dump page whe... |
1275 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
b91e1302a mm: optimize Page... |
1276 1277 |
if (clear_bit_unlock_is_negative_byte(PG_locked, &page->flags)) wake_up_page_bit(page, PG_locked); |
1da177e4c Linux-2.6.12-rc2 |
1278 1279 |
} EXPORT_SYMBOL(unlock_page); |
485bb99b4 [PATCH] kernel-do... |
1280 1281 1282 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c Linux-2.6.12-rc2 |
1283 1284 1285 |
*/ void end_page_writeback(struct page *page) { |
888cf2db4 mm: avoid unneces... |
1286 1287 1288 1289 1290 1291 1292 1293 1294 |
/* * TestClearPageReclaim could be used here but it is an atomic * operation and overkill in this particular case. Failing to * shuffle a page marked for immediate reclaim is too mild to * justify taking an atomic operation penalty at the end of * ever page writeback. */ if (PageReclaim(page)) { ClearPageReclaim(page); |
ac6aadb24 mm: rotate_reclai... |
1295 |
rotate_reclaimable_page(page); |
888cf2db4 mm: avoid unneces... |
1296 |
} |
ac6aadb24 mm: rotate_reclai... |
1297 1298 1299 |
if (!test_clear_page_writeback(page)) BUG(); |
4e857c58e arch: Mass conver... |
1300 |
smp_mb__after_atomic(); |
1da177e4c Linux-2.6.12-rc2 |
1301 1302 1303 |
wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
57d998456 fs/mpage.c: facto... |
1304 1305 1306 1307 |
/* * After completing I/O on a page, call this routine to update the page * flags appropriately */ |
c11f0c0b5 block/mm: make bd... |
1308 |
void page_endio(struct page *page, bool is_write, int err) |
57d998456 fs/mpage.c: facto... |
1309 |
{ |
c11f0c0b5 block/mm: make bd... |
1310 |
if (!is_write) { |
57d998456 fs/mpage.c: facto... |
1311 1312 1313 1314 1315 1316 1317 |
if (!err) { SetPageUptodate(page); } else { ClearPageUptodate(page); SetPageError(page); } unlock_page(page); |
abf545484 mm/block: convert... |
1318 |
} else { |
57d998456 fs/mpage.c: facto... |
1319 |
if (err) { |
dd8416c47 mm: do not access... |
1320 |
struct address_space *mapping; |
57d998456 fs/mpage.c: facto... |
1321 |
SetPageError(page); |
dd8416c47 mm: do not access... |
1322 1323 1324 |
mapping = page_mapping(page); if (mapping) mapping_set_error(mapping, err); |
57d998456 fs/mpage.c: facto... |
1325 1326 1327 1328 1329 |
} end_page_writeback(page); } } EXPORT_SYMBOL_GPL(page_endio); |
485bb99b4 [PATCH] kernel-do... |
1330 1331 |
/** * __lock_page - get a lock on the page, assuming we need to sleep to get it |
870667553 mm: fix filemap.c... |
1332 |
* @__page: the page to lock |
1da177e4c Linux-2.6.12-rc2 |
1333 |
*/ |
629060270 mm: add PageWaite... |
1334 |
void __lock_page(struct page *__page) |
1da177e4c Linux-2.6.12-rc2 |
1335 |
{ |
629060270 mm: add PageWaite... |
1336 1337 |
struct page *page = compound_head(__page); wait_queue_head_t *q = page_waitqueue(page); |
9a1ea439b mm: put_and_wait_... |
1338 1339 |
wait_on_page_bit_common(q, page, PG_locked, TASK_UNINTERRUPTIBLE, EXCLUSIVE); |
1da177e4c Linux-2.6.12-rc2 |
1340 1341 |
} EXPORT_SYMBOL(__lock_page); |
629060270 mm: add PageWaite... |
1342 |
int __lock_page_killable(struct page *__page) |
2687a3569 Add lock_page_kil... |
1343 |
{ |
629060270 mm: add PageWaite... |
1344 1345 |
struct page *page = compound_head(__page); wait_queue_head_t *q = page_waitqueue(page); |
9a1ea439b mm: put_and_wait_... |
1346 1347 |
return wait_on_page_bit_common(q, page, PG_locked, TASK_KILLABLE, EXCLUSIVE); |
2687a3569 Add lock_page_kil... |
1348 |
} |
18bc0bbd1 Staging: pohmelfs... |
1349 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569 Add lock_page_kil... |
1350 |
|
9a95f3cf7 mm: describe mmap... |
1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 |
/* * Return values: * 1 - page is locked; mmap_sem is still held. * 0 - page is not locked. * mmap_sem has been released (up_read()), unless flags had both * FAULT_FLAG_ALLOW_RETRY and FAULT_FLAG_RETRY_NOWAIT set, in * which case mmap_sem is still held. * * If neither ALLOW_RETRY nor KILLABLE are set, will always return 1 * with the page locked and the mmap_sem unperturbed. */ |
d065bd810 mm: retry page fa... |
1362 1363 1364 |
int __lock_page_or_retry(struct page *page, struct mm_struct *mm, unsigned int flags) { |
37b23e052 x86,mm: make page... |
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 |
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 ... |
1377 |
wait_on_page_locked(page); |
d065bd810 mm: retry page fa... |
1378 |
return 0; |
37b23e052 x86,mm: make page... |
1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 |
} 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... |
1391 1392 |
} } |
485bb99b4 [PATCH] kernel-do... |
1393 |
/** |
0d3f92966 page cache: Conve... |
1394 1395 1396 1397 |
* page_cache_next_miss() - Find the next gap in the page cache. * @mapping: Mapping. * @index: Index. * @max_scan: Maximum range to search. |
e7b563bb2 mm: filemap: move... |
1398 |
* |
0d3f92966 page cache: Conve... |
1399 1400 |
* Search the range [index, min(index + max_scan - 1, ULONG_MAX)] for the * gap with the lowest index. |
e7b563bb2 mm: filemap: move... |
1401 |
* |
0d3f92966 page cache: Conve... |
1402 1403 1404 1405 1406 |
* This function may be called under the rcu_read_lock. However, this will * not atomically search a snapshot of the cache at a single point in time. * For example, if a gap is created at index 5, then subsequently a gap is * created at index 10, page_cache_next_miss covering both indices may * return 10 if called under the rcu_read_lock. |
e7b563bb2 mm: filemap: move... |
1407 |
* |
0d3f92966 page cache: Conve... |
1408 1409 1410 |
* Return: The index of the gap if found, otherwise an index outside the * range specified (in which case 'return - index >= max_scan' will be true). * In the rare case of index wrap-around, 0 will be returned. |
e7b563bb2 mm: filemap: move... |
1411 |
*/ |
0d3f92966 page cache: Conve... |
1412 |
pgoff_t page_cache_next_miss(struct address_space *mapping, |
e7b563bb2 mm: filemap: move... |
1413 1414 |
pgoff_t index, unsigned long max_scan) { |
0d3f92966 page cache: Conve... |
1415 |
XA_STATE(xas, &mapping->i_pages, index); |
e7b563bb2 mm: filemap: move... |
1416 |
|
0d3f92966 page cache: Conve... |
1417 1418 1419 |
while (max_scan--) { void *entry = xas_next(&xas); if (!entry || xa_is_value(entry)) |
e7b563bb2 mm: filemap: move... |
1420 |
break; |
0d3f92966 page cache: Conve... |
1421 |
if (xas.xa_index == 0) |
e7b563bb2 mm: filemap: move... |
1422 1423 |
break; } |
0d3f92966 page cache: Conve... |
1424 |
return xas.xa_index; |
e7b563bb2 mm: filemap: move... |
1425 |
} |
0d3f92966 page cache: Conve... |
1426 |
EXPORT_SYMBOL(page_cache_next_miss); |
e7b563bb2 mm: filemap: move... |
1427 1428 |
/** |
2346a5605 mm/filemap.c: fix... |
1429 |
* page_cache_prev_miss() - Find the previous gap in the page cache. |
0d3f92966 page cache: Conve... |
1430 1431 1432 |
* @mapping: Mapping. * @index: Index. * @max_scan: Maximum range to search. |
e7b563bb2 mm: filemap: move... |
1433 |
* |
0d3f92966 page cache: Conve... |
1434 1435 |
* Search the range [max(index - max_scan + 1, 0), index] for the * gap with the highest index. |
e7b563bb2 mm: filemap: move... |
1436 |
* |
0d3f92966 page cache: Conve... |
1437 1438 1439 1440 1441 |
* This function may be called under the rcu_read_lock. However, this will * not atomically search a snapshot of the cache at a single point in time. * For example, if a gap is created at index 10, then subsequently a gap is * created at index 5, page_cache_prev_miss() covering both indices may * return 5 if called under the rcu_read_lock. |
e7b563bb2 mm: filemap: move... |
1442 |
* |
0d3f92966 page cache: Conve... |
1443 1444 1445 |
* Return: The index of the gap if found, otherwise an index outside the * range specified (in which case 'index - return >= max_scan' will be true). * In the rare case of wrap-around, ULONG_MAX will be returned. |
e7b563bb2 mm: filemap: move... |
1446 |
*/ |
0d3f92966 page cache: Conve... |
1447 |
pgoff_t page_cache_prev_miss(struct address_space *mapping, |
e7b563bb2 mm: filemap: move... |
1448 1449 |
pgoff_t index, unsigned long max_scan) { |
0d3f92966 page cache: Conve... |
1450 |
XA_STATE(xas, &mapping->i_pages, index); |
e7b563bb2 mm: filemap: move... |
1451 |
|
0d3f92966 page cache: Conve... |
1452 1453 1454 |
while (max_scan--) { void *entry = xas_prev(&xas); if (!entry || xa_is_value(entry)) |
e7b563bb2 mm: filemap: move... |
1455 |
break; |
0d3f92966 page cache: Conve... |
1456 |
if (xas.xa_index == ULONG_MAX) |
e7b563bb2 mm: filemap: move... |
1457 1458 |
break; } |
0d3f92966 page cache: Conve... |
1459 |
return xas.xa_index; |
e7b563bb2 mm: filemap: move... |
1460 |
} |
0d3f92966 page cache: Conve... |
1461 |
EXPORT_SYMBOL(page_cache_prev_miss); |
e7b563bb2 mm: filemap: move... |
1462 1463 |
/** |
0cd6144aa mm + fs: prepare ... |
1464 |
* find_get_entry - find and get a page cache entry |
485bb99b4 [PATCH] kernel-do... |
1465 |
* @mapping: the address_space to search |
0cd6144aa mm + fs: prepare ... |
1466 1467 1468 1469 |
* @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... |
1470 |
* |
139b6a6fb mm: filemap: upda... |
1471 1472 |
* If the slot holds a shadow entry of a previously evicted page, or a * swap entry from shmem/tmpfs, it is returned. |
0cd6144aa mm + fs: prepare ... |
1473 |
* |
a862f68a8 docs/core-api/mm:... |
1474 |
* Return: the found page or shadow entry, %NULL if nothing is found. |
1da177e4c Linux-2.6.12-rc2 |
1475 |
*/ |
0cd6144aa mm + fs: prepare ... |
1476 |
struct page *find_get_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1477 |
{ |
4c7472c0d page cache: Conve... |
1478 |
XA_STATE(xas, &mapping->i_pages, offset); |
4101196b1 mm: page cache: s... |
1479 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1480 |
|
a60637c85 mm: lockless page... |
1481 1482 |
rcu_read_lock(); repeat: |
4c7472c0d page cache: Conve... |
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 |
xas_reset(&xas); page = xas_load(&xas); if (xas_retry(&xas, page)) goto repeat; /* * A shadow entry of a recently evicted page, or a swap entry from * shmem/tmpfs. Return it without attempting to raise page count. */ if (!page || xa_is_value(page)) goto out; |
83929372f filemap: prepare ... |
1493 |
|
4101196b1 mm: page cache: s... |
1494 |
if (!page_cache_get_speculative(page)) |
4c7472c0d page cache: Conve... |
1495 |
goto repeat; |
83929372f filemap: prepare ... |
1496 |
|
4c7472c0d page cache: Conve... |
1497 |
/* |
4101196b1 mm: page cache: s... |
1498 |
* Has the page moved or been split? |
4c7472c0d page cache: Conve... |
1499 1500 1501 1502 |
* This is part of the lockless pagecache protocol. See * include/linux/pagemap.h for details. */ if (unlikely(page != xas_reload(&xas))) { |
4101196b1 mm: page cache: s... |
1503 |
put_page(page); |
4c7472c0d page cache: Conve... |
1504 |
goto repeat; |
a60637c85 mm: lockless page... |
1505 |
} |
4101196b1 mm: page cache: s... |
1506 |
page = find_subpage(page, offset); |
27d20fddc radix-tree: fix R... |
1507 |
out: |
a60637c85 mm: lockless page... |
1508 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1509 1510 |
return page; } |
0cd6144aa mm + fs: prepare ... |
1511 |
EXPORT_SYMBOL(find_get_entry); |
1da177e4c Linux-2.6.12-rc2 |
1512 |
|
485bb99b4 [PATCH] kernel-do... |
1513 |
/** |
0cd6144aa mm + fs: prepare ... |
1514 1515 1516 1517 1518 1519 1520 1521 |
* 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. * |
139b6a6fb mm: filemap: upda... |
1522 1523 |
* If the slot holds a shadow entry of a previously evicted page, or a * swap entry from shmem/tmpfs, it is returned. |
0cd6144aa mm + fs: prepare ... |
1524 |
* |
0cd6144aa mm + fs: prepare ... |
1525 |
* find_lock_entry() may sleep. |
a862f68a8 docs/core-api/mm:... |
1526 1527 |
* * Return: the found page or shadow entry, %NULL if nothing is found. |
0cd6144aa mm + fs: prepare ... |
1528 1529 |
*/ struct page *find_lock_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1530 1531 |
{ struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1532 |
repeat: |
0cd6144aa mm + fs: prepare ... |
1533 |
page = find_get_entry(mapping, offset); |
4c7472c0d page cache: Conve... |
1534 |
if (page && !xa_is_value(page)) { |
a60637c85 mm: lockless page... |
1535 1536 |
lock_page(page); /* Has the page been truncated? */ |
83929372f filemap: prepare ... |
1537 |
if (unlikely(page_mapping(page) != mapping)) { |
a60637c85 mm: lockless page... |
1538 |
unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
1539 |
put_page(page); |
a60637c85 mm: lockless page... |
1540 |
goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1541 |
} |
83929372f filemap: prepare ... |
1542 |
VM_BUG_ON_PAGE(page_to_pgoff(page) != offset, page); |
1da177e4c Linux-2.6.12-rc2 |
1543 |
} |
1da177e4c Linux-2.6.12-rc2 |
1544 1545 |
return page; } |
0cd6144aa mm + fs: prepare ... |
1546 1547 1548 |
EXPORT_SYMBOL(find_lock_entry); /** |
2457aec63 mm: non-atomicall... |
1549 |
* pagecache_get_page - find and get a page reference |
0cd6144aa mm + fs: prepare ... |
1550 1551 |
* @mapping: the address_space to search * @offset: the page index |
2457aec63 mm: non-atomicall... |
1552 |
* @fgp_flags: PCG flags |
45f87de57 mm: get rid of ra... |
1553 |
* @gfp_mask: gfp mask to use for the page cache data page allocation |
0cd6144aa mm + fs: prepare ... |
1554 |
* |
2457aec63 mm: non-atomicall... |
1555 |
* Looks up the page cache slot at @mapping & @offset. |
1da177e4c Linux-2.6.12-rc2 |
1556 |
* |
75325189c mm: fix filemap.c... |
1557 |
* PCG flags modify how the page is returned. |
0cd6144aa mm + fs: prepare ... |
1558 |
* |
0e056eb55 kernel-api.rst: f... |
1559 1560 1561 1562 1563 1564 1565 |
* @fgp_flags can be: * * - FGP_ACCESSED: the page will be marked accessed * - FGP_LOCK: Page is return locked * - FGP_CREAT: If page is not present then a new page is allocated using * @gfp_mask and added to the page cache and the VM's LRU * list. The page is returned locked and with an increased |
a862f68a8 docs/core-api/mm:... |
1566 |
* refcount. |
a75d4c333 filemap: kill pag... |
1567 1568 1569 |
* - FGP_FOR_MMAP: Similar to FGP_CREAT, only we want to allow the caller to do * its own locking dance if the page is already in cache, or unlock the page * before returning if we had to add the page to pagecache. |
1da177e4c Linux-2.6.12-rc2 |
1570 |
* |
2457aec63 mm: non-atomicall... |
1571 1572 |
* If FGP_LOCK or FGP_CREAT are specified then the function may sleep even * if the GFP flags specified for FGP_CREAT are atomic. |
1da177e4c Linux-2.6.12-rc2 |
1573 |
* |
2457aec63 mm: non-atomicall... |
1574 |
* If there is a page cache page, it is returned with an increased refcount. |
a862f68a8 docs/core-api/mm:... |
1575 1576 |
* * Return: the found page or %NULL otherwise. |
1da177e4c Linux-2.6.12-rc2 |
1577 |
*/ |
2457aec63 mm: non-atomicall... |
1578 |
struct page *pagecache_get_page(struct address_space *mapping, pgoff_t offset, |
45f87de57 mm: get rid of ra... |
1579 |
int fgp_flags, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
1580 |
{ |
eb2be1893 mm: buffered writ... |
1581 |
struct page *page; |
2457aec63 mm: non-atomicall... |
1582 |
|
1da177e4c Linux-2.6.12-rc2 |
1583 |
repeat: |
2457aec63 mm: non-atomicall... |
1584 |
page = find_get_entry(mapping, offset); |
3159f943a xarray: Replace e... |
1585 |
if (xa_is_value(page)) |
2457aec63 mm: non-atomicall... |
1586 1587 1588 1589 1590 1591 1592 |
page = NULL; if (!page) goto no_page; if (fgp_flags & FGP_LOCK) { if (fgp_flags & FGP_NOWAIT) { if (!trylock_page(page)) { |
09cbfeaf1 mm, fs: get rid o... |
1593 |
put_page(page); |
2457aec63 mm: non-atomicall... |
1594 1595 1596 1597 1598 1599 1600 |
return NULL; } } else { lock_page(page); } /* Has the page been truncated? */ |
31895438e filemap: check co... |
1601 |
if (unlikely(compound_head(page)->mapping != mapping)) { |
2457aec63 mm: non-atomicall... |
1602 |
unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
1603 |
put_page(page); |
2457aec63 mm: non-atomicall... |
1604 1605 1606 1607 |
goto repeat; } VM_BUG_ON_PAGE(page->index != offset, page); } |
c16eb000c mm/filemap.c: rem... |
1608 |
if (fgp_flags & FGP_ACCESSED) |
2457aec63 mm: non-atomicall... |
1609 1610 1611 1612 1613 1614 |
mark_page_accessed(page); no_page: if (!page && (fgp_flags & FGP_CREAT)) { int err; if ((fgp_flags & FGP_WRITE) && mapping_cap_account_dirty(mapping)) |
45f87de57 mm: get rid of ra... |
1615 1616 1617 |
gfp_mask |= __GFP_WRITE; if (fgp_flags & FGP_NOFS) gfp_mask &= ~__GFP_FS; |
2457aec63 mm: non-atomicall... |
1618 |
|
45f87de57 mm: get rid of ra... |
1619 |
page = __page_cache_alloc(gfp_mask); |
eb2be1893 mm: buffered writ... |
1620 1621 |
if (!page) return NULL; |
2457aec63 mm: non-atomicall... |
1622 |
|
a75d4c333 filemap: kill pag... |
1623 |
if (WARN_ON_ONCE(!(fgp_flags & (FGP_LOCK | FGP_FOR_MMAP)))) |
2457aec63 mm: non-atomicall... |
1624 |
fgp_flags |= FGP_LOCK; |
eb39d618f mm: replace init_... |
1625 |
/* Init accessed so avoid atomic mark_page_accessed later */ |
2457aec63 mm: non-atomicall... |
1626 |
if (fgp_flags & FGP_ACCESSED) |
eb39d618f mm: replace init_... |
1627 |
__SetPageReferenced(page); |
2457aec63 mm: non-atomicall... |
1628 |
|
abc1be13f mm/filemap.c: fix... |
1629 |
err = add_to_page_cache_lru(page, mapping, offset, gfp_mask); |
eb2be1893 mm: buffered writ... |
1630 |
if (unlikely(err)) { |
09cbfeaf1 mm, fs: get rid o... |
1631 |
put_page(page); |
eb2be1893 mm: buffered writ... |
1632 1633 1634 |
page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1635 |
} |
a75d4c333 filemap: kill pag... |
1636 1637 1638 1639 1640 1641 1642 |
/* * add_to_page_cache_lru locks the page, and for mmap we expect * an unlocked page. */ if (page && (fgp_flags & FGP_FOR_MMAP)) unlock_page(page); |
1da177e4c Linux-2.6.12-rc2 |
1643 |
} |
2457aec63 mm: non-atomicall... |
1644 |
|
1da177e4c Linux-2.6.12-rc2 |
1645 1646 |
return page; } |
2457aec63 mm: non-atomicall... |
1647 |
EXPORT_SYMBOL(pagecache_get_page); |
1da177e4c Linux-2.6.12-rc2 |
1648 1649 |
/** |
0cd6144aa mm + fs: prepare ... |
1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 |
* 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. * |
139b6a6fb mm: filemap: upda... |
1666 1667 |
* Any shadow entries of evicted pages, or swap entries from * shmem/tmpfs, are included in the returned array. |
0cd6144aa mm + fs: prepare ... |
1668 |
* |
a862f68a8 docs/core-api/mm:... |
1669 |
* Return: the number of pages and shadow entries which were found. |
0cd6144aa mm + fs: prepare ... |
1670 1671 1672 1673 1674 |
*/ unsigned find_get_entries(struct address_space *mapping, pgoff_t start, unsigned int nr_entries, struct page **entries, pgoff_t *indices) { |
f280bf092 page cache: Conve... |
1675 1676 |
XA_STATE(xas, &mapping->i_pages, start); struct page *page; |
0cd6144aa mm + fs: prepare ... |
1677 |
unsigned int ret = 0; |
0cd6144aa mm + fs: prepare ... |
1678 1679 1680 1681 1682 |
if (!nr_entries) return 0; rcu_read_lock(); |
f280bf092 page cache: Conve... |
1683 |
xas_for_each(&xas, page, ULONG_MAX) { |
f280bf092 page cache: Conve... |
1684 |
if (xas_retry(&xas, page)) |
0cd6144aa mm + fs: prepare ... |
1685 |
continue; |
f280bf092 page cache: Conve... |
1686 1687 1688 1689 1690 1691 |
/* * A shadow entry of a recently evicted page, a swap * entry from shmem/tmpfs or a DAX entry. Return it * without attempting to raise page count. */ if (xa_is_value(page)) |
0cd6144aa mm + fs: prepare ... |
1692 |
goto export; |
83929372f filemap: prepare ... |
1693 |
|
4101196b1 mm: page cache: s... |
1694 |
if (!page_cache_get_speculative(page)) |
f280bf092 page cache: Conve... |
1695 |
goto retry; |
83929372f filemap: prepare ... |
1696 |
|
4101196b1 mm: page cache: s... |
1697 |
/* Has the page moved or been split? */ |
f280bf092 page cache: Conve... |
1698 1699 |
if (unlikely(page != xas_reload(&xas))) goto put_page; |
4101196b1 mm: page cache: s... |
1700 |
page = find_subpage(page, xas.xa_index); |
f280bf092 page cache: Conve... |
1701 |
|
0cd6144aa mm + fs: prepare ... |
1702 |
export: |
f280bf092 page cache: Conve... |
1703 |
indices[ret] = xas.xa_index; |
0cd6144aa mm + fs: prepare ... |
1704 1705 1706 |
entries[ret] = page; if (++ret == nr_entries) break; |
f280bf092 page cache: Conve... |
1707 1708 |
continue; put_page: |
4101196b1 mm: page cache: s... |
1709 |
put_page(page); |
f280bf092 page cache: Conve... |
1710 1711 |
retry: xas_reset(&xas); |
0cd6144aa mm + fs: prepare ... |
1712 1713 1714 1715 1716 1717 |
} rcu_read_unlock(); return ret; } /** |
b947cee4b mm: implement fin... |
1718 |
* find_get_pages_range - gang pagecache lookup |
1da177e4c Linux-2.6.12-rc2 |
1719 1720 |
* @mapping: The address_space to search * @start: The starting page index |
b947cee4b mm: implement fin... |
1721 |
* @end: The final page index (inclusive) |
1da177e4c Linux-2.6.12-rc2 |
1722 1723 1724 |
* @nr_pages: The maximum number of pages * @pages: Where the resulting pages are placed * |
b947cee4b mm: implement fin... |
1725 1726 1727 1728 |
* find_get_pages_range() will search for and return a group of up to @nr_pages * pages in the mapping starting at index @start and up to index @end * (inclusive). The pages are placed at @pages. find_get_pages_range() takes * a reference against the returned pages. |
1da177e4c Linux-2.6.12-rc2 |
1729 1730 1731 |
* * The search returns a group of mapping-contiguous pages with ascending * indexes. There may be holes in the indices due to not-present pages. |
d72dc8a25 mm: make pagevec_... |
1732 |
* We also update @start to index the next page for the traversal. |
1da177e4c Linux-2.6.12-rc2 |
1733 |
* |
a862f68a8 docs/core-api/mm:... |
1734 1735 |
* Return: the number of pages which were found. If this number is * smaller than @nr_pages, the end of specified range has been |
b947cee4b mm: implement fin... |
1736 |
* reached. |
1da177e4c Linux-2.6.12-rc2 |
1737 |
*/ |
b947cee4b mm: implement fin... |
1738 1739 1740 |
unsigned find_get_pages_range(struct address_space *mapping, pgoff_t *start, pgoff_t end, unsigned int nr_pages, struct page **pages) |
1da177e4c Linux-2.6.12-rc2 |
1741 |
{ |
fd1b3cee2 page cache: Conve... |
1742 1743 |
XA_STATE(xas, &mapping->i_pages, *start); struct page *page; |
0fc9d1040 radix-tree: use i... |
1744 1745 1746 1747 |
unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1748 1749 |
rcu_read_lock(); |
fd1b3cee2 page cache: Conve... |
1750 |
xas_for_each(&xas, page, end) { |
fd1b3cee2 page cache: Conve... |
1751 |
if (xas_retry(&xas, page)) |
a60637c85 mm: lockless page... |
1752 |
continue; |
fd1b3cee2 page cache: Conve... |
1753 1754 |
/* Skip over shadow, swap and DAX entries */ if (xa_is_value(page)) |
8079b1c85 mm: clarify the r... |
1755 |
continue; |
a60637c85 mm: lockless page... |
1756 |
|
4101196b1 mm: page cache: s... |
1757 |
if (!page_cache_get_speculative(page)) |
fd1b3cee2 page cache: Conve... |
1758 |
goto retry; |
83929372f filemap: prepare ... |
1759 |
|
4101196b1 mm: page cache: s... |
1760 |
/* Has the page moved or been split? */ |
fd1b3cee2 page cache: Conve... |
1761 1762 |
if (unlikely(page != xas_reload(&xas))) goto put_page; |
1da177e4c Linux-2.6.12-rc2 |
1763 |
|
4101196b1 mm: page cache: s... |
1764 |
pages[ret] = find_subpage(page, xas.xa_index); |
b947cee4b mm: implement fin... |
1765 |
if (++ret == nr_pages) { |
5d3ee42f8 mm/shmem: make fi... |
1766 |
*start = xas.xa_index + 1; |
b947cee4b mm: implement fin... |
1767 1768 |
goto out; } |
fd1b3cee2 page cache: Conve... |
1769 1770 |
continue; put_page: |
4101196b1 mm: page cache: s... |
1771 |
put_page(page); |
fd1b3cee2 page cache: Conve... |
1772 1773 |
retry: xas_reset(&xas); |
a60637c85 mm: lockless page... |
1774 |
} |
5b280c0cc mm: don't return ... |
1775 |
|
b947cee4b mm: implement fin... |
1776 1777 1778 |
/* * We come here when there is no page beyond @end. We take care to not * overflow the index @start as it confuses some of the callers. This |
fd1b3cee2 page cache: Conve... |
1779 |
* breaks the iteration when there is a page at index -1 but that is |
b947cee4b mm: implement fin... |
1780 1781 1782 1783 1784 1785 1786 |
* already broken anyway. */ if (end == (pgoff_t)-1) *start = (pgoff_t)-1; else *start = end + 1; out: |
a60637c85 mm: lockless page... |
1787 |
rcu_read_unlock(); |
d72dc8a25 mm: make pagevec_... |
1788 |
|
1da177e4c Linux-2.6.12-rc2 |
1789 1790 |
return ret; } |
ebf43500e [PATCH] Add find_... |
1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 |
/** * 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. * |
a862f68a8 docs/core-api/mm:... |
1801 |
* Return: the number of pages which were found. |
ebf43500e [PATCH] Add find_... |
1802 1803 1804 1805 |
*/ unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index, unsigned int nr_pages, struct page **pages) { |
3ece58a27 page cache: Conve... |
1806 1807 |
XA_STATE(xas, &mapping->i_pages, index); struct page *page; |
0fc9d1040 radix-tree: use i... |
1808 1809 1810 1811 |
unsigned int ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1812 1813 |
rcu_read_lock(); |
3ece58a27 page cache: Conve... |
1814 |
for (page = xas_load(&xas); page; page = xas_next(&xas)) { |
3ece58a27 page cache: Conve... |
1815 1816 1817 1818 1819 1820 1821 |
if (xas_retry(&xas, page)) continue; /* * If the entry has been swapped out, we can stop looking. * No current caller is looking for DAX entries. */ if (xa_is_value(page)) |
8079b1c85 mm: clarify the r... |
1822 |
break; |
ebf43500e [PATCH] Add find_... |
1823 |
|
4101196b1 mm: page cache: s... |
1824 |
if (!page_cache_get_speculative(page)) |
3ece58a27 page cache: Conve... |
1825 |
goto retry; |
83929372f filemap: prepare ... |
1826 |
|
4101196b1 mm: page cache: s... |
1827 |
/* Has the page moved or been split? */ |
3ece58a27 page cache: Conve... |
1828 1829 |
if (unlikely(page != xas_reload(&xas))) goto put_page; |
a60637c85 mm: lockless page... |
1830 |
|
4101196b1 mm: page cache: s... |
1831 |
pages[ret] = find_subpage(page, xas.xa_index); |
0fc9d1040 radix-tree: use i... |
1832 1833 |
if (++ret == nr_pages) break; |
3ece58a27 page cache: Conve... |
1834 1835 |
continue; put_page: |
4101196b1 mm: page cache: s... |
1836 |
put_page(page); |
3ece58a27 page cache: Conve... |
1837 1838 |
retry: xas_reset(&xas); |
ebf43500e [PATCH] Add find_... |
1839 |
} |
a60637c85 mm: lockless page... |
1840 1841 |
rcu_read_unlock(); return ret; |
ebf43500e [PATCH] Add find_... |
1842 |
} |
ef71c15c4 AFS: export a cou... |
1843 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e [PATCH] Add find_... |
1844 |
|
485bb99b4 [PATCH] kernel-do... |
1845 |
/** |
72b045aec mm: implement fin... |
1846 |
* find_get_pages_range_tag - find and return pages in given range matching @tag |
485bb99b4 [PATCH] kernel-do... |
1847 1848 |
* @mapping: the address_space to search * @index: the starting page index |
72b045aec mm: implement fin... |
1849 |
* @end: The final page index (inclusive) |
485bb99b4 [PATCH] kernel-do... |
1850 1851 1852 1853 |
* @tag: the tag index * @nr_pages: the maximum number of pages * @pages: where the resulting pages are placed * |
1da177e4c Linux-2.6.12-rc2 |
1854 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4 [PATCH] kernel-do... |
1855 |
* @tag. We update @index to index the next page for the traversal. |
a862f68a8 docs/core-api/mm:... |
1856 1857 |
* * Return: the number of pages which were found. |
1da177e4c Linux-2.6.12-rc2 |
1858 |
*/ |
72b045aec mm: implement fin... |
1859 |
unsigned find_get_pages_range_tag(struct address_space *mapping, pgoff_t *index, |
a6906972f page cache; Conve... |
1860 |
pgoff_t end, xa_mark_t tag, unsigned int nr_pages, |
72b045aec mm: implement fin... |
1861 |
struct page **pages) |
1da177e4c Linux-2.6.12-rc2 |
1862 |
{ |
a6906972f page cache; Conve... |
1863 1864 |
XA_STATE(xas, &mapping->i_pages, *index); struct page *page; |
0fc9d1040 radix-tree: use i... |
1865 1866 1867 1868 |
unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1869 1870 |
rcu_read_lock(); |
a6906972f page cache; Conve... |
1871 |
xas_for_each_marked(&xas, page, end, tag) { |
a6906972f page cache; Conve... |
1872 |
if (xas_retry(&xas, page)) |
a60637c85 mm: lockless page... |
1873 |
continue; |
a6906972f page cache; Conve... |
1874 1875 1876 1877 1878 1879 |
/* * Shadow entries should never be tagged, but this iteration * is lockless so there is a window for page reclaim to evict * a page we saw tagged. Skip over it. */ if (xa_is_value(page)) |
139b6a6fb mm: filemap: upda... |
1880 |
continue; |
a60637c85 mm: lockless page... |
1881 |
|
4101196b1 mm: page cache: s... |
1882 |
if (!page_cache_get_speculative(page)) |
a6906972f page cache; Conve... |
1883 |
goto retry; |
a60637c85 mm: lockless page... |
1884 |
|
4101196b1 mm: page cache: s... |
1885 |
/* Has the page moved or been split? */ |
a6906972f page cache; Conve... |
1886 1887 |
if (unlikely(page != xas_reload(&xas))) goto put_page; |
a60637c85 mm: lockless page... |
1888 |
|
4101196b1 mm: page cache: s... |
1889 |
pages[ret] = find_subpage(page, xas.xa_index); |
72b045aec mm: implement fin... |
1890 |
if (++ret == nr_pages) { |
5d3ee42f8 mm/shmem: make fi... |
1891 |
*index = xas.xa_index + 1; |
72b045aec mm: implement fin... |
1892 1893 |
goto out; } |
a6906972f page cache; Conve... |
1894 1895 |
continue; put_page: |
4101196b1 mm: page cache: s... |
1896 |
put_page(page); |
a6906972f page cache; Conve... |
1897 1898 |
retry: xas_reset(&xas); |
a60637c85 mm: lockless page... |
1899 |
} |
5b280c0cc mm: don't return ... |
1900 |
|
72b045aec mm: implement fin... |
1901 |
/* |
a6906972f page cache; Conve... |
1902 |
* We come here when we got to @end. We take care to not overflow the |
72b045aec mm: implement fin... |
1903 |
* index @index as it confuses some of the callers. This breaks the |
a6906972f page cache; Conve... |
1904 1905 |
* iteration when there is a page at index -1 but that is already * broken anyway. |
72b045aec mm: implement fin... |
1906 1907 1908 1909 1910 1911 |
*/ if (end == (pgoff_t)-1) *index = (pgoff_t)-1; else *index = end + 1; out: |
a60637c85 mm: lockless page... |
1912 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1913 |
|
1da177e4c Linux-2.6.12-rc2 |
1914 1915 |
return ret; } |
72b045aec mm: implement fin... |
1916 |
EXPORT_SYMBOL(find_get_pages_range_tag); |
1da177e4c Linux-2.6.12-rc2 |
1917 |
|
76d42bd96 [PATCH] readahead... |
1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 |
/* * 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... |
1936 |
ra->ra_pages /= 4; |
76d42bd96 [PATCH] readahead... |
1937 |
} |
485bb99b4 [PATCH] kernel-do... |
1938 |
/** |
47c27bc46 fs: pass iocb to ... |
1939 1940 |
* generic_file_buffered_read - generic file read routine * @iocb: the iocb to read |
6e58e79db introduce copy_pa... |
1941 1942 |
* @iter: data destination * @written: already copied |
485bb99b4 [PATCH] kernel-do... |
1943 |
* |
1da177e4c Linux-2.6.12-rc2 |
1944 |
* This is a generic file read routine, and uses the |
485bb99b4 [PATCH] kernel-do... |
1945 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c Linux-2.6.12-rc2 |
1946 1947 1948 |
* * This is really ugly. But the goto's actually try to clarify some * of the logic when it comes to error handling etc. |
a862f68a8 docs/core-api/mm:... |
1949 1950 1951 1952 |
* * Return: * * total number of bytes copied, including those the were already @written * * negative error code if nothing was copied |
1da177e4c Linux-2.6.12-rc2 |
1953 |
*/ |
47c27bc46 fs: pass iocb to ... |
1954 |
static ssize_t generic_file_buffered_read(struct kiocb *iocb, |
6e58e79db introduce copy_pa... |
1955 |
struct iov_iter *iter, ssize_t written) |
1da177e4c Linux-2.6.12-rc2 |
1956 |
{ |
47c27bc46 fs: pass iocb to ... |
1957 |
struct file *filp = iocb->ki_filp; |
36e789144 kill do_generic_m... |
1958 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
1959 |
struct inode *inode = mapping->host; |
36e789144 kill do_generic_m... |
1960 |
struct file_ra_state *ra = &filp->f_ra; |
47c27bc46 fs: pass iocb to ... |
1961 |
loff_t *ppos = &iocb->ki_pos; |
57f6b96c0 filemap: convert ... |
1962 1963 1964 1965 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372 readahead: improv... |
1966 |
unsigned int prev_offset; |
6e58e79db introduce copy_pa... |
1967 |
int error = 0; |
1da177e4c Linux-2.6.12-rc2 |
1968 |
|
c2a9737f4 vfs,mm: fix a dea... |
1969 |
if (unlikely(*ppos >= inode->i_sb->s_maxbytes)) |
d05c5f7ba vfs,mm: fix retur... |
1970 |
return 0; |
c2a9737f4 vfs,mm: fix a dea... |
1971 |
iov_iter_truncate(iter, inode->i_sb->s_maxbytes); |
09cbfeaf1 mm, fs: get rid o... |
1972 1973 1974 1975 1976 |
index = *ppos >> PAGE_SHIFT; prev_index = ra->prev_pos >> PAGE_SHIFT; prev_offset = ra->prev_pos & (PAGE_SIZE-1); last_index = (*ppos + iter->count + PAGE_SIZE-1) >> PAGE_SHIFT; offset = *ppos & ~PAGE_MASK; |
1da177e4c Linux-2.6.12-rc2 |
1977 |
|
1da177e4c Linux-2.6.12-rc2 |
1978 1979 |
for (;;) { struct page *page; |
57f6b96c0 filemap: convert ... |
1980 |
pgoff_t end_index; |
a32ea1e1f Fix read/truncate... |
1981 |
loff_t isize; |
1da177e4c Linux-2.6.12-rc2 |
1982 |
unsigned long nr, ret; |
1da177e4c Linux-2.6.12-rc2 |
1983 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
1984 |
find_page: |
5abf186a3 mm, fs: check for... |
1985 1986 1987 1988 |
if (fatal_signal_pending(current)) { error = -EINTR; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
1989 |
page = find_get_page(mapping, index); |
3ea89ee86 readahead: conver... |
1990 |
if (!page) { |
3239d8348 fs: support IOCB_... |
1991 1992 |
if (iocb->ki_flags & IOCB_NOWAIT) goto would_block; |
cf914a7d6 readahead: split ... |
1993 |
page_cache_sync_readahead(mapping, |
7ff81078d readahead: remove... |
1994 |
ra, filp, |
3ea89ee86 readahead: conver... |
1995 1996 1997 1998 1999 2000 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6 readahead: split ... |
2001 |
page_cache_async_readahead(mapping, |
7ff81078d readahead: remove... |
2002 |
ra, filp, page, |
3ea89ee86 readahead: conver... |
2003 |
index, last_index - index); |
1da177e4c Linux-2.6.12-rc2 |
2004 |
} |
8ab22b9ab vfs: pagecache us... |
2005 |
if (!PageUptodate(page)) { |
3239d8348 fs: support IOCB_... |
2006 2007 2008 2009 |
if (iocb->ki_flags & IOCB_NOWAIT) { put_page(page); goto would_block; } |
ebded0278 mm: filemap: avoi... |
2010 2011 2012 2013 2014 |
/* * See comment in do_read_cache_page on why * wait_on_page_locked is used to avoid unnecessarily * serialisations and why it's safe. */ |
c4b209a42 do_generic_file_r... |
2015 2016 2017 |
error = wait_on_page_locked_killable(page); if (unlikely(error)) goto readpage_error; |
ebded0278 mm: filemap: avoi... |
2018 2019 |
if (PageUptodate(page)) goto page_ok; |
09cbfeaf1 mm, fs: get rid o... |
2020 |
if (inode->i_blkbits == PAGE_SHIFT || |
8ab22b9ab vfs: pagecache us... |
2021 2022 |
!mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; |
6d6d36bc6 mm/filemap: don't... |
2023 |
/* pipes can't handle partially uptodate pages */ |
00e237074 iov_iter: Use acc... |
2024 |
if (unlikely(iov_iter_is_pipe(iter))) |
6d6d36bc6 mm/filemap: don't... |
2025 |
goto page_not_up_to_date; |
529ae9aaa mm: rename page t... |
2026 |
if (!trylock_page(page)) |
8ab22b9ab vfs: pagecache us... |
2027 |
goto page_not_up_to_date; |
8d056cb96 mm/vfs: revalidat... |
2028 2029 2030 |
/* Did it get truncated before we got the lock? */ if (!page->mapping) goto page_not_up_to_date_locked; |
8ab22b9ab vfs: pagecache us... |
2031 |
if (!mapping->a_ops->is_partially_uptodate(page, |
6e58e79db introduce copy_pa... |
2032 |
offset, iter->count)) |
8ab22b9ab vfs: pagecache us... |
2033 2034 2035 |
goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
2036 |
page_ok: |
a32ea1e1f Fix read/truncate... |
2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 |
/* * 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); |
09cbfeaf1 mm, fs: get rid o... |
2047 |
end_index = (isize - 1) >> PAGE_SHIFT; |
a32ea1e1f Fix read/truncate... |
2048 |
if (unlikely(!isize || index > end_index)) { |
09cbfeaf1 mm, fs: get rid o... |
2049 |
put_page(page); |
a32ea1e1f Fix read/truncate... |
2050 2051 2052 2053 |
goto out; } /* nr is the maximum number of bytes to copy from this page */ |
09cbfeaf1 mm, fs: get rid o... |
2054 |
nr = PAGE_SIZE; |
a32ea1e1f Fix read/truncate... |
2055 |
if (index == end_index) { |
09cbfeaf1 mm, fs: get rid o... |
2056 |
nr = ((isize - 1) & ~PAGE_MASK) + 1; |
a32ea1e1f Fix read/truncate... |
2057 |
if (nr <= offset) { |
09cbfeaf1 mm, fs: get rid o... |
2058 |
put_page(page); |
a32ea1e1f Fix read/truncate... |
2059 2060 2061 2062 |
goto out; } } nr = nr - offset; |
1da177e4c Linux-2.6.12-rc2 |
2063 2064 2065 2066 2067 2068 2069 2070 2071 |
/* 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... |
2072 2073 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c Linux-2.6.12-rc2 |
2074 |
*/ |
ec0f16372 readahead: improv... |
2075 |
if (prev_index != index || offset != prev_offset) |
1da177e4c Linux-2.6.12-rc2 |
2076 2077 2078 2079 2080 2081 |
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 |
2082 |
*/ |
6e58e79db introduce copy_pa... |
2083 2084 |
ret = copy_page_to_iter(page, offset, nr, iter); |
1da177e4c Linux-2.6.12-rc2 |
2085 |
offset += ret; |
09cbfeaf1 mm, fs: get rid o... |
2086 2087 |
index += offset >> PAGE_SHIFT; offset &= ~PAGE_MASK; |
6ce745ed3 readahead: code c... |
2088 |
prev_offset = offset; |
1da177e4c Linux-2.6.12-rc2 |
2089 |
|
09cbfeaf1 mm, fs: get rid o... |
2090 |
put_page(page); |
6e58e79db introduce copy_pa... |
2091 2092 2093 2094 2095 2096 2097 2098 |
written += ret; if (!iov_iter_count(iter)) goto out; if (ret < nr) { error = -EFAULT; goto out; } continue; |
1da177e4c Linux-2.6.12-rc2 |
2099 2100 2101 |
page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235 do_generic_file_r... |
2102 2103 2104 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
2105 |
|
8ab22b9ab vfs: pagecache us... |
2106 |
page_not_up_to_date_locked: |
da6052f7b [PATCH] update so... |
2107 |
/* Did it get truncated before we got the lock? */ |
1da177e4c Linux-2.6.12-rc2 |
2108 2109 |
if (!page->mapping) { unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2110 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 |
continue; } /* Did somebody else fill it already? */ if (PageUptodate(page)) { unlock_page(page); goto page_ok; } readpage: |
91803b499 do_generic_file_r... |
2121 2122 2123 2124 2125 2126 |
/* * 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 |
2127 2128 |
/* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7 [PATCH] add AOP_T... |
2129 2130 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { |
09cbfeaf1 mm, fs: get rid o... |
2131 |
put_page(page); |
6e58e79db introduce copy_pa... |
2132 |
error = 0; |
994fc28c7 [PATCH] add AOP_T... |
2133 2134 |
goto find_page; } |
1da177e4c Linux-2.6.12-rc2 |
2135 |
goto readpage_error; |
994fc28c7 [PATCH] add AOP_T... |
2136 |
} |
1da177e4c Linux-2.6.12-rc2 |
2137 2138 |
if (!PageUptodate(page)) { |
854623235 do_generic_file_r... |
2139 2140 2141 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
2142 2143 2144 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* |
2ecdc82ef kill unused inval... |
2145 |
* invalidate_mapping_pages got it |
1da177e4c Linux-2.6.12-rc2 |
2146 2147 |
*/ unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2148 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2149 2150 2151 |
goto find_page; } unlock_page(page); |
7ff81078d readahead: remove... |
2152 |
shrink_readahead_size_eio(filp, ra); |
854623235 do_generic_file_r... |
2153 2154 |
error = -EIO; goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
2155 2156 2157 |
} unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
2158 2159 2160 2161 |
goto page_ok; readpage_error: /* UHHUH! A synchronous read error occurred. Report it */ |
09cbfeaf1 mm, fs: get rid o... |
2162 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2163 2164 2165 2166 2167 2168 2169 |
goto out; no_cached_page: /* * Ok, it wasn't cached, so we need to create a new * page.. */ |
453f85d43 mm: remove __GFP_... |
2170 |
page = page_cache_alloc(mapping); |
eb2be1893 mm: buffered writ... |
2171 |
if (!page) { |
6e58e79db introduce copy_pa... |
2172 |
error = -ENOMEM; |
eb2be1893 mm: buffered writ... |
2173 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
2174 |
} |
6afdb859b mm: do not ignore... |
2175 |
error = add_to_page_cache_lru(page, mapping, index, |
c62d25556 mm, fs: introduce... |
2176 |
mapping_gfp_constraint(mapping, GFP_KERNEL)); |
1da177e4c Linux-2.6.12-rc2 |
2177 |
if (error) { |
09cbfeaf1 mm, fs: get rid o... |
2178 |
put_page(page); |
6e58e79db introduce copy_pa... |
2179 2180 |
if (error == -EEXIST) { error = 0; |
1da177e4c Linux-2.6.12-rc2 |
2181 |
goto find_page; |
6e58e79db introduce copy_pa... |
2182 |
} |
1da177e4c Linux-2.6.12-rc2 |
2183 2184 |
goto out; } |
1da177e4c Linux-2.6.12-rc2 |
2185 2186 |
goto readpage; } |
3239d8348 fs: support IOCB_... |
2187 2188 |
would_block: error = -EAGAIN; |
1da177e4c Linux-2.6.12-rc2 |
2189 |
out: |
7ff81078d readahead: remove... |
2190 |
ra->prev_pos = prev_index; |
09cbfeaf1 mm, fs: get rid o... |
2191 |
ra->prev_pos <<= PAGE_SHIFT; |
7ff81078d readahead: remove... |
2192 |
ra->prev_pos |= prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
2193 |
|
09cbfeaf1 mm, fs: get rid o... |
2194 |
*ppos = ((loff_t)index << PAGE_SHIFT) + offset; |
0c6aa2639 mm: do_generic_fi... |
2195 |
file_accessed(filp); |
6e58e79db introduce copy_pa... |
2196 |
return written ? written : error; |
1da177e4c Linux-2.6.12-rc2 |
2197 |
} |
485bb99b4 [PATCH] kernel-do... |
2198 |
/** |
6abd23227 bury generic_file... |
2199 |
* generic_file_read_iter - generic filesystem read routine |
485bb99b4 [PATCH] kernel-do... |
2200 |
* @iocb: kernel I/O control block |
6abd23227 bury generic_file... |
2201 |
* @iter: destination for the data read |
485bb99b4 [PATCH] kernel-do... |
2202 |
* |
6abd23227 bury generic_file... |
2203 |
* This is the "read_iter()" routine for all filesystems |
1da177e4c Linux-2.6.12-rc2 |
2204 |
* that can use the page cache directly. |
a862f68a8 docs/core-api/mm:... |
2205 2206 2207 |
* Return: * * number of bytes copied, even for partial reads * * negative error code if nothing was read |
1da177e4c Linux-2.6.12-rc2 |
2208 2209 |
*/ ssize_t |
ed978a811 new helper: gener... |
2210 |
generic_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) |
1da177e4c Linux-2.6.12-rc2 |
2211 |
{ |
e7080a439 mm/filemap: gener... |
2212 |
size_t count = iov_iter_count(iter); |
47c27bc46 fs: pass iocb to ... |
2213 |
ssize_t retval = 0; |
e7080a439 mm/filemap: gener... |
2214 2215 2216 |
if (!count) goto out; /* skip atime */ |
1da177e4c Linux-2.6.12-rc2 |
2217 |
|
2ba48ce51 mirror O_APPEND a... |
2218 |
if (iocb->ki_flags & IOCB_DIRECT) { |
47c27bc46 fs: pass iocb to ... |
2219 |
struct file *file = iocb->ki_filp; |
ed978a811 new helper: gener... |
2220 2221 |
struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; |
543ade1fc [PATCH] Streamlin... |
2222 |
loff_t size; |
1da177e4c Linux-2.6.12-rc2 |
2223 |
|
1da177e4c Linux-2.6.12-rc2 |
2224 |
size = i_size_read(inode); |
6be96d3ad fs: return if dir... |
2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 |
if (iocb->ki_flags & IOCB_NOWAIT) { if (filemap_range_has_page(mapping, iocb->ki_pos, iocb->ki_pos + count - 1)) return -EAGAIN; } else { retval = filemap_write_and_wait_range(mapping, iocb->ki_pos, iocb->ki_pos + count - 1); if (retval < 0) goto out; } |
d8d3d94b8 pass iov_iter to ... |
2236 |
|
0d5b0cf24 fs: update atime ... |
2237 |
file_accessed(file); |
5ecda1371 generic_file_read... |
2238 |
retval = mapping->a_ops->direct_IO(iocb, iter); |
c3a690240 fix ITER_PIPE int... |
2239 |
if (retval >= 0) { |
c64fb5c74 filemap: remove p... |
2240 |
iocb->ki_pos += retval; |
5ecda1371 generic_file_read... |
2241 |
count -= retval; |
9fe55eea7 Fix race when che... |
2242 |
} |
5b47d59af fix braino in gen... |
2243 |
iov_iter_revert(iter, count - iov_iter_count(iter)); |
66f998f61 fs: allow short d... |
2244 |
|
9fe55eea7 Fix race when che... |
2245 2246 2247 2248 2249 2250 |
/* * 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 |
fbbbad4bc vfs,ext2: introdu... |
2251 2252 |
* the rest of the read. Buffered reads will not work for * DAX files, so don't bother trying. |
9fe55eea7 Fix race when che... |
2253 |
*/ |
5ecda1371 generic_file_read... |
2254 |
if (retval < 0 || !count || iocb->ki_pos >= size || |
0d5b0cf24 fs: update atime ... |
2255 |
IS_DAX(inode)) |
9fe55eea7 Fix race when che... |
2256 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
2257 |
} |
47c27bc46 fs: pass iocb to ... |
2258 |
retval = generic_file_buffered_read(iocb, iter, retval); |
1da177e4c Linux-2.6.12-rc2 |
2259 2260 2261 |
out: return retval; } |
ed978a811 new helper: gener... |
2262 |
EXPORT_SYMBOL(generic_file_read_iter); |
1da177e4c Linux-2.6.12-rc2 |
2263 |
|
1da177e4c Linux-2.6.12-rc2 |
2264 |
#ifdef CONFIG_MMU |
1da177e4c Linux-2.6.12-rc2 |
2265 |
#define MMAP_LOTSAMISS (100) |
6b4c9f446 filemap: drop the... |
2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 |
/* * lock_page_maybe_drop_mmap - lock the page, possibly dropping the mmap_sem * @vmf - the vm_fault for this fault. * @page - the page to lock. * @fpin - the pointer to the file we may pin (or is already pinned). * * This works similar to lock_page_or_retry in that it can drop the mmap_sem. * It differs in that it actually returns the page locked if it returns 1 and 0 * if it couldn't lock the page. If we did have to drop the mmap_sem then fpin * will point to the pinned file and needs to be fput()'ed at a later point. */ static int lock_page_maybe_drop_mmap(struct vm_fault *vmf, struct page *page, struct file **fpin) { if (trylock_page(page)) return 1; |
8b0f9fa2e filemap: add a co... |
2282 2283 2284 2285 2286 |
/* * NOTE! This will make us return with VM_FAULT_RETRY, but with * the mmap_sem still held. That's how FAULT_FLAG_RETRY_NOWAIT * is supposed to work. We have way too many special cases.. */ |
6b4c9f446 filemap: drop the... |
2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 |
if (vmf->flags & FAULT_FLAG_RETRY_NOWAIT) return 0; *fpin = maybe_unlock_mmap_for_io(vmf, *fpin); if (vmf->flags & FAULT_FLAG_KILLABLE) { if (__lock_page_killable(page)) { /* * We didn't have the right flags to drop the mmap_sem, * but all fault_handlers only check for fatal signals * if we return VM_FAULT_RETRY, so we need to drop the * mmap_sem here and return 0 if we don't have a fpin. */ if (*fpin == NULL) up_read(&vmf->vma->vm_mm->mmap_sem); return 0; } } else __lock_page(page); return 1; } |
1da177e4c Linux-2.6.12-rc2 |
2307 |
|
ef00e08e2 readahead: clean ... |
2308 |
/* |
6b4c9f446 filemap: drop the... |
2309 2310 2311 2312 2313 |
* Synchronous readahead happens when we don't even find a page in the page * cache at all. We don't want to perform IO under the mmap sem, so if we have * to drop the mmap sem we return the file that was pinned in order for us to do * that. If we didn't pin a file then we return NULL. The file that is * returned needs to be fput()'ed when we're done with it. |
ef00e08e2 readahead: clean ... |
2314 |
*/ |
6b4c9f446 filemap: drop the... |
2315 |
static struct file *do_sync_mmap_readahead(struct vm_fault *vmf) |
ef00e08e2 readahead: clean ... |
2316 |
{ |
2a1180f1b filemap: pass vm_... |
2317 2318 |
struct file *file = vmf->vma->vm_file; struct file_ra_state *ra = &file->f_ra; |
ef00e08e2 readahead: clean ... |
2319 |
struct address_space *mapping = file->f_mapping; |
6b4c9f446 filemap: drop the... |
2320 |
struct file *fpin = NULL; |
2a1180f1b filemap: pass vm_... |
2321 |
pgoff_t offset = vmf->pgoff; |
ef00e08e2 readahead: clean ... |
2322 2323 |
/* If we don't want any read-ahead, don't bother */ |
2a1180f1b filemap: pass vm_... |
2324 |
if (vmf->vma->vm_flags & VM_RAND_READ) |
6b4c9f446 filemap: drop the... |
2325 |
return fpin; |
275b12bf5 readahead: return... |
2326 |
if (!ra->ra_pages) |
6b4c9f446 filemap: drop the... |
2327 |
return fpin; |
ef00e08e2 readahead: clean ... |
2328 |
|
2a1180f1b filemap: pass vm_... |
2329 |
if (vmf->vma->vm_flags & VM_SEQ_READ) { |
6b4c9f446 filemap: drop the... |
2330 |
fpin = maybe_unlock_mmap_for_io(vmf, fpin); |
7ffc59b4d readahead: enforc... |
2331 2332 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
6b4c9f446 filemap: drop the... |
2333 |
return fpin; |
ef00e08e2 readahead: clean ... |
2334 |
} |
207d04baa readahead: reduce... |
2335 2336 |
/* Avoid banging the cache line if not needed */ if (ra->mmap_miss < MMAP_LOTSAMISS * 10) |
ef00e08e2 readahead: clean ... |
2337 2338 2339 2340 2341 2342 2343 |
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) |
6b4c9f446 filemap: drop the... |
2344 |
return fpin; |
ef00e08e2 readahead: clean ... |
2345 |
|
d30a11004 readahead: record... |
2346 2347 2348 |
/* * mmap read-around */ |
6b4c9f446 filemap: drop the... |
2349 |
fpin = maybe_unlock_mmap_for_io(vmf, fpin); |
600e19afc mm: use only per-... |
2350 2351 2352 |
ra->start = max_t(long, 0, offset - ra->ra_pages / 2); ra->size = ra->ra_pages; ra->async_size = ra->ra_pages / 4; |
275b12bf5 readahead: return... |
2353 |
ra_submit(ra, mapping, file); |
6b4c9f446 filemap: drop the... |
2354 |
return fpin; |
ef00e08e2 readahead: clean ... |
2355 2356 2357 2358 |
} /* * Asynchronous readahead happens when we find the page and PG_readahead, |
6b4c9f446 filemap: drop the... |
2359 2360 |
* so we want to possibly extend the readahead further. We return the file that * was pinned if we have to drop the mmap_sem in order to do IO. |
ef00e08e2 readahead: clean ... |
2361 |
*/ |
6b4c9f446 filemap: drop the... |
2362 2363 |
static struct file *do_async_mmap_readahead(struct vm_fault *vmf, struct page *page) |
ef00e08e2 readahead: clean ... |
2364 |
{ |
2a1180f1b filemap: pass vm_... |
2365 2366 |
struct file *file = vmf->vma->vm_file; struct file_ra_state *ra = &file->f_ra; |
ef00e08e2 readahead: clean ... |
2367 |
struct address_space *mapping = file->f_mapping; |
6b4c9f446 filemap: drop the... |
2368 |
struct file *fpin = NULL; |
2a1180f1b filemap: pass vm_... |
2369 |
pgoff_t offset = vmf->pgoff; |
ef00e08e2 readahead: clean ... |
2370 2371 |
/* If we don't want any read-ahead, don't bother */ |
2a1180f1b filemap: pass vm_... |
2372 |
if (vmf->vma->vm_flags & VM_RAND_READ) |
6b4c9f446 filemap: drop the... |
2373 |
return fpin; |
ef00e08e2 readahead: clean ... |
2374 2375 |
if (ra->mmap_miss > 0) ra->mmap_miss--; |
6b4c9f446 filemap: drop the... |
2376 2377 |
if (PageReadahead(page)) { fpin = maybe_unlock_mmap_for_io(vmf, fpin); |
2fad6f5de readahead: enforc... |
2378 2379 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
6b4c9f446 filemap: drop the... |
2380 2381 |
} return fpin; |
ef00e08e2 readahead: clean ... |
2382 |
} |
485bb99b4 [PATCH] kernel-do... |
2383 |
/** |
54cb8821d mm: merge populat... |
2384 |
* filemap_fault - read in file data for page fault handling |
d0217ac04 mm: fault feedbac... |
2385 |
* @vmf: struct vm_fault containing details of the fault |
485bb99b4 [PATCH] kernel-do... |
2386 |
* |
54cb8821d mm: merge populat... |
2387 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c Linux-2.6.12-rc2 |
2388 2389 2390 2391 2392 |
* 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. |
9a95f3cf7 mm: describe mmap... |
2393 2394 2395 |
* * vma->vm_mm->mmap_sem must be held on entry. * |
a49858338 mm/filemap.c: cor... |
2396 2397 |
* If our return value has VM_FAULT_RETRY set, it's because the mmap_sem * may be dropped before doing I/O or by lock_page_maybe_drop_mmap(). |
9a95f3cf7 mm: describe mmap... |
2398 2399 2400 2401 2402 |
* * If our return value does not have VM_FAULT_RETRY set, the mmap_sem * has not been released. * * We never return with VM_FAULT_RETRY and a bit from VM_FAULT_ERROR set. |
a862f68a8 docs/core-api/mm:... |
2403 2404 |
* * Return: bitwise-OR of %VM_FAULT_ codes. |
1da177e4c Linux-2.6.12-rc2 |
2405 |
*/ |
2bcd6454b mm: use new retur... |
2406 |
vm_fault_t filemap_fault(struct vm_fault *vmf) |
1da177e4c Linux-2.6.12-rc2 |
2407 2408 |
{ int error; |
11bac8000 mm, fs: reduce fa... |
2409 |
struct file *file = vmf->vma->vm_file; |
6b4c9f446 filemap: drop the... |
2410 |
struct file *fpin = NULL; |
1da177e4c Linux-2.6.12-rc2 |
2411 2412 2413 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2 readahead: clean ... |
2414 |
pgoff_t offset = vmf->pgoff; |
9ab2594fe mm: tighten up th... |
2415 |
pgoff_t max_off; |
1da177e4c Linux-2.6.12-rc2 |
2416 |
struct page *page; |
2bcd6454b mm: use new retur... |
2417 |
vm_fault_t ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
2418 |
|
9ab2594fe mm: tighten up th... |
2419 2420 |
max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); if (unlikely(offset >= max_off)) |
5307cc1aa Remove broken ptr... |
2421 |
return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
2422 |
|
1da177e4c Linux-2.6.12-rc2 |
2423 |
/* |
494264208 mm: memcg: handle... |
2424 |
* Do we have something in the page cache already? |
1da177e4c Linux-2.6.12-rc2 |
2425 |
*/ |
ef00e08e2 readahead: clean ... |
2426 |
page = find_get_page(mapping, offset); |
45cac65b0 readahead: fault ... |
2427 |
if (likely(page) && !(vmf->flags & FAULT_FLAG_TRIED)) { |
1da177e4c Linux-2.6.12-rc2 |
2428 |
/* |
ef00e08e2 readahead: clean ... |
2429 2430 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c Linux-2.6.12-rc2 |
2431 |
*/ |
6b4c9f446 filemap: drop the... |
2432 |
fpin = do_async_mmap_readahead(vmf, page); |
45cac65b0 readahead: fault ... |
2433 |
} else if (!page) { |
ef00e08e2 readahead: clean ... |
2434 |
/* No page in the page cache at all */ |
ef00e08e2 readahead: clean ... |
2435 |
count_vm_event(PGMAJFAULT); |
2262185c5 mm: per-cgroup me... |
2436 |
count_memcg_event_mm(vmf->vma->vm_mm, PGMAJFAULT); |
ef00e08e2 readahead: clean ... |
2437 |
ret = VM_FAULT_MAJOR; |
6b4c9f446 filemap: drop the... |
2438 |
fpin = do_sync_mmap_readahead(vmf); |
ef00e08e2 readahead: clean ... |
2439 |
retry_find: |
a75d4c333 filemap: kill pag... |
2440 2441 2442 |
page = pagecache_get_page(mapping, offset, FGP_CREAT|FGP_FOR_MMAP, vmf->gfp_mask); |
6b4c9f446 filemap: drop the... |
2443 2444 2445 |
if (!page) { if (fpin) goto out_retry; |
a75d4c333 filemap: kill pag... |
2446 |
return vmf_error(-ENOMEM); |
6b4c9f446 filemap: drop the... |
2447 |
} |
1da177e4c Linux-2.6.12-rc2 |
2448 |
} |
6b4c9f446 filemap: drop the... |
2449 2450 |
if (!lock_page_maybe_drop_mmap(vmf, page, &fpin)) goto out_retry; |
b522c94da mm: filemap_fault... |
2451 2452 |
/* Did it get truncated? */ |
585e5a7ba filemap: check co... |
2453 |
if (unlikely(compound_head(page)->mapping != mapping)) { |
b522c94da mm: filemap_fault... |
2454 2455 2456 2457 |
unlock_page(page); put_page(page); goto retry_find; } |
520e5ba41 filemap: update o... |
2458 |
VM_BUG_ON_PAGE(page_to_pgoff(page) != offset, page); |
b522c94da mm: filemap_fault... |
2459 |
|
1da177e4c Linux-2.6.12-rc2 |
2460 |
/* |
d00806b18 mm: fix fault vs ... |
2461 2462 |
* 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 |
2463 |
*/ |
d00806b18 mm: fix fault vs ... |
2464 |
if (unlikely(!PageUptodate(page))) |
1da177e4c Linux-2.6.12-rc2 |
2465 |
goto page_not_uptodate; |
ef00e08e2 readahead: clean ... |
2466 |
/* |
6b4c9f446 filemap: drop the... |
2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 |
* We've made it this far and we had to drop our mmap_sem, now is the * time to return to the upper layer and have it re-find the vma and * redo the fault. */ if (fpin) { unlock_page(page); goto out_retry; } /* |
ef00e08e2 readahead: clean ... |
2477 2478 2479 |
* Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
9ab2594fe mm: tighten up th... |
2480 2481 |
max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); if (unlikely(offset >= max_off)) { |
d00806b18 mm: fix fault vs ... |
2482 |
unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2483 |
put_page(page); |
5307cc1aa Remove broken ptr... |
2484 |
return VM_FAULT_SIGBUS; |
d00806b18 mm: fix fault vs ... |
2485 |
} |
d0217ac04 mm: fault feedbac... |
2486 |
vmf->page = page; |
83c54070e mm: fault feedbac... |
2487 |
return ret | VM_FAULT_LOCKED; |
1da177e4c Linux-2.6.12-rc2 |
2488 |
|
1da177e4c Linux-2.6.12-rc2 |
2489 |
page_not_uptodate: |
1da177e4c Linux-2.6.12-rc2 |
2490 2491 2492 2493 2494 2495 |
/* * 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 |
2496 |
ClearPageError(page); |
6b4c9f446 filemap: drop the... |
2497 |
fpin = maybe_unlock_mmap_for_io(vmf, fpin); |
994fc28c7 [PATCH] add AOP_T... |
2498 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e mm: fix infinite ... |
2499 2500 2501 2502 2503 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
6b4c9f446 filemap: drop the... |
2504 2505 |
if (fpin) goto out_retry; |
09cbfeaf1 mm, fs: get rid o... |
2506 |
put_page(page); |
d00806b18 mm: fix fault vs ... |
2507 2508 |
if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7 [PATCH] add AOP_T... |
2509 |
goto retry_find; |
1da177e4c Linux-2.6.12-rc2 |
2510 |
|
d00806b18 mm: fix fault vs ... |
2511 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96 [PATCH] readahead... |
2512 |
shrink_readahead_size_eio(file, ra); |
d0217ac04 mm: fault feedbac... |
2513 |
return VM_FAULT_SIGBUS; |
6b4c9f446 filemap: drop the... |
2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 |
out_retry: /* * We dropped the mmap_sem, we need to return to the fault handler to * re-find the vma and come back and find our hopefully still populated * page. */ if (page) put_page(page); if (fpin) fput(fpin); return ret | VM_FAULT_RETRY; |
54cb8821d mm: merge populat... |
2526 2527 |
} EXPORT_SYMBOL(filemap_fault); |
82b0f8c39 mm: join struct f... |
2528 |
void filemap_map_pages(struct vm_fault *vmf, |
bae473a42 mm: introduce fau... |
2529 |
pgoff_t start_pgoff, pgoff_t end_pgoff) |
f1820361f mm: implement ->m... |
2530 |
{ |
82b0f8c39 mm: join struct f... |
2531 |
struct file *file = vmf->vma->vm_file; |
f1820361f mm: implement ->m... |
2532 |
struct address_space *mapping = file->f_mapping; |
bae473a42 mm: introduce fau... |
2533 |
pgoff_t last_pgoff = start_pgoff; |
9ab2594fe mm: tighten up th... |
2534 |
unsigned long max_idx; |
070e807c6 page cache: Conve... |
2535 |
XA_STATE(xas, &mapping->i_pages, start_pgoff); |
4101196b1 mm: page cache: s... |
2536 |
struct page *page; |
f1820361f mm: implement ->m... |
2537 2538 |
rcu_read_lock(); |
070e807c6 page cache: Conve... |
2539 2540 2541 2542 |
xas_for_each(&xas, page, end_pgoff) { if (xas_retry(&xas, page)) continue; if (xa_is_value(page)) |
2cf938aae mm: use radix_tre... |
2543 |
goto next; |
f1820361f mm: implement ->m... |
2544 |
|
e0975b2aa mm, fault_around:... |
2545 2546 2547 2548 |
/* * Check for a locked page first, as a speculative * reference may adversely influence page migration. */ |
4101196b1 mm: page cache: s... |
2549 |
if (PageLocked(page)) |
e0975b2aa mm, fault_around:... |
2550 |
goto next; |
4101196b1 mm: page cache: s... |
2551 |
if (!page_cache_get_speculative(page)) |
070e807c6 page cache: Conve... |
2552 |
goto next; |
f1820361f mm: implement ->m... |
2553 |
|
4101196b1 mm: page cache: s... |
2554 |
/* Has the page moved or been split? */ |
070e807c6 page cache: Conve... |
2555 2556 |
if (unlikely(page != xas_reload(&xas))) goto skip; |
4101196b1 mm: page cache: s... |
2557 |
page = find_subpage(page, xas.xa_index); |
f1820361f mm: implement ->m... |
2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 |
if (!PageUptodate(page) || PageReadahead(page) || PageHWPoison(page)) goto skip; if (!trylock_page(page)) goto skip; if (page->mapping != mapping || !PageUptodate(page)) goto unlock; |
9ab2594fe mm: tighten up th... |
2568 2569 |
max_idx = DIV_ROUND_UP(i_size_read(mapping->host), PAGE_SIZE); if (page->index >= max_idx) |
f1820361f mm: implement ->m... |
2570 |
goto unlock; |
f1820361f mm: implement ->m... |
2571 2572 |
if (file->f_ra.mmap_miss > 0) file->f_ra.mmap_miss--; |
7267ec008 mm: postpone page... |
2573 |
|
070e807c6 page cache: Conve... |
2574 |
vmf->address += (xas.xa_index - last_pgoff) << PAGE_SHIFT; |
82b0f8c39 mm: join struct f... |
2575 |
if (vmf->pte) |
070e807c6 page cache: Conve... |
2576 2577 |
vmf->pte += xas.xa_index - last_pgoff; last_pgoff = xas.xa_index; |
82b0f8c39 mm: join struct f... |
2578 |
if (alloc_set_pte(vmf, NULL, page)) |
7267ec008 mm: postpone page... |
2579 |
goto unlock; |
f1820361f mm: implement ->m... |
2580 2581 2582 2583 2584 |
unlock_page(page); goto next; unlock: unlock_page(page); skip: |
09cbfeaf1 mm, fs: get rid o... |
2585 |
put_page(page); |
f1820361f mm: implement ->m... |
2586 |
next: |
7267ec008 mm: postpone page... |
2587 |
/* Huge page is mapped? No need to proceed. */ |
82b0f8c39 mm: join struct f... |
2588 |
if (pmd_trans_huge(*vmf->pmd)) |
7267ec008 mm: postpone page... |
2589 |
break; |
f1820361f mm: implement ->m... |
2590 2591 2592 2593 |
} rcu_read_unlock(); } EXPORT_SYMBOL(filemap_map_pages); |
2bcd6454b mm: use new retur... |
2594 |
vm_fault_t filemap_page_mkwrite(struct vm_fault *vmf) |
4fcf1c620 mm: Make default ... |
2595 2596 |
{ struct page *page = vmf->page; |
11bac8000 mm, fs: reduce fa... |
2597 |
struct inode *inode = file_inode(vmf->vma->vm_file); |
2bcd6454b mm: use new retur... |
2598 |
vm_fault_t ret = VM_FAULT_LOCKED; |
4fcf1c620 mm: Make default ... |
2599 |
|
14da92001 fs: Protect write... |
2600 |
sb_start_pagefault(inode->i_sb); |
11bac8000 mm, fs: reduce fa... |
2601 |
file_update_time(vmf->vma->vm_file); |
4fcf1c620 mm: Make default ... |
2602 2603 2604 2605 2606 2607 |
lock_page(page); if (page->mapping != inode->i_mapping) { unlock_page(page); ret = VM_FAULT_NOPAGE; goto out; } |
14da92001 fs: Protect write... |
2608 2609 2610 2611 2612 2613 |
/* * 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 ... |
2614 |
wait_for_stable_page(page); |
4fcf1c620 mm: Make default ... |
2615 |
out: |
14da92001 fs: Protect write... |
2616 |
sb_end_pagefault(inode->i_sb); |
4fcf1c620 mm: Make default ... |
2617 2618 |
return ret; } |
4fcf1c620 mm: Make default ... |
2619 |
|
f0f37e2f7 const: mark struc... |
2620 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d mm: merge populat... |
2621 |
.fault = filemap_fault, |
f1820361f mm: implement ->m... |
2622 |
.map_pages = filemap_map_pages, |
4fcf1c620 mm: Make default ... |
2623 |
.page_mkwrite = filemap_page_mkwrite, |
1da177e4c Linux-2.6.12-rc2 |
2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 |
}; /* 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 |
2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 |
/* * 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 |
4b96a37d1 mm: convert to us... |
2649 |
vm_fault_t filemap_page_mkwrite(struct vm_fault *vmf) |
453972283 mm/filemap.c: pro... |
2650 |
{ |
4b96a37d1 mm: convert to us... |
2651 |
return VM_FAULT_SIGBUS; |
453972283 mm/filemap.c: pro... |
2652 |
} |
1da177e4c Linux-2.6.12-rc2 |
2653 2654 2655 2656 2657 2658 2659 2660 2661 |
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 */ |
453972283 mm/filemap.c: pro... |
2662 |
EXPORT_SYMBOL(filemap_page_mkwrite); |
1da177e4c Linux-2.6.12-rc2 |
2663 2664 |
EXPORT_SYMBOL(generic_file_mmap); EXPORT_SYMBOL(generic_file_readonly_mmap); |
67f9fd91f mm: remove read_c... |
2665 2666 2667 2668 2669 |
static struct page *wait_on_page_read(struct page *page) { if (!IS_ERR(page)) { wait_on_page_locked(page); if (!PageUptodate(page)) { |
09cbfeaf1 mm, fs: get rid o... |
2670 |
put_page(page); |
67f9fd91f mm: remove read_c... |
2671 2672 2673 2674 2675 |
page = ERR_PTR(-EIO); } } return page; } |
32b635298 mm: filemap: remo... |
2676 |
static struct page *do_read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
2677 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2678 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2679 2680 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
2681 |
{ |
eb2be1893 mm: buffered writ... |
2682 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
2683 2684 2685 2686 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
453f85d43 mm: remove __GFP_... |
2687 |
page = __page_cache_alloc(gfp); |
eb2be1893 mm: buffered writ... |
2688 2689 |
if (!page) return ERR_PTR(-ENOMEM); |
e6f67b8c0 vfs: __read_cache... |
2690 |
err = add_to_page_cache_lru(page, mapping, index, gfp); |
eb2be1893 mm: buffered writ... |
2691 |
if (unlikely(err)) { |
09cbfeaf1 mm, fs: get rid o... |
2692 |
put_page(page); |
eb2be1893 mm: buffered writ... |
2693 2694 |
if (err == -EEXIST) goto repeat; |
22ecdb4f8 page cache: Remov... |
2695 |
/* Presumably ENOMEM for xarray node */ |
1da177e4c Linux-2.6.12-rc2 |
2696 2697 |
return ERR_PTR(err); } |
32b635298 mm: filemap: remo... |
2698 2699 |
filler: |
6c45b4541 mm/filemap: don't... |
2700 2701 2702 2703 |
if (filler) err = filler(data, page); else err = mapping->a_ops->readpage(data, page); |
1da177e4c Linux-2.6.12-rc2 |
2704 |
if (err < 0) { |
09cbfeaf1 mm, fs: get rid o... |
2705 |
put_page(page); |
32b635298 mm: filemap: remo... |
2706 |
return ERR_PTR(err); |
1da177e4c Linux-2.6.12-rc2 |
2707 |
} |
1da177e4c Linux-2.6.12-rc2 |
2708 |
|
32b635298 mm: filemap: remo... |
2709 2710 2711 2712 2713 |
page = wait_on_page_read(page); if (IS_ERR(page)) return page; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
2714 2715 |
if (PageUptodate(page)) goto out; |
ebded0278 mm: filemap: avoi... |
2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 |
/* * Page is not up to date and may be locked due one of the following * case a: Page is being filled and the page lock is held * case b: Read/write error clearing the page uptodate status * case c: Truncation in progress (page locked) * case d: Reclaim in progress * * Case a, the page will be up to date when the page is unlocked. * There is no need to serialise on the page lock here as the page * is pinned so the lock gives no additional protection. Even if the * the page is truncated, the data is still valid if PageUptodate as * it's a race vs truncate race. * Case b, the page will not be up to date * Case c, the page may be truncated but in itself, the data may still * be valid after IO completes as it's a read vs truncate race. The * operation must restart if the page is not uptodate on unlock but * otherwise serialising on page lock to stabilise the mapping gives * no additional guarantees to the caller as the page lock is * released before return. * Case d, similar to truncation. If reclaim holds the page lock, it * will be a race with remove_mapping that determines if the mapping * is valid on unlock but otherwise the data is valid and there is * no need to serialise with page lock. * * As the page lock gives no additional guarantee, we optimistically * wait on the page to be unlocked and check if it's up to date and * use the page if it is. Otherwise, the page lock is required to * distinguish between the different cases. The motivation is that we * avoid spurious serialisations and wakeups when multiple processes * wait on the same page for IO to complete. */ wait_on_page_locked(page); if (PageUptodate(page)) goto out; /* Distinguish between all the cases under the safety of the lock */ |
1da177e4c Linux-2.6.12-rc2 |
2752 |
lock_page(page); |
ebded0278 mm: filemap: avoi... |
2753 2754 |
/* Case c or d, restart the operation */ |
1da177e4c Linux-2.6.12-rc2 |
2755 2756 |
if (!page->mapping) { unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2757 |
put_page(page); |
32b635298 mm: filemap: remo... |
2758 |
goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
2759 |
} |
ebded0278 mm: filemap: avoi... |
2760 2761 |
/* Someone else locked and filled the page in a very small window */ |
1da177e4c Linux-2.6.12-rc2 |
2762 2763 2764 2765 |
if (PageUptodate(page)) { unlock_page(page); goto out; } |
32b635298 mm: filemap: remo... |
2766 |
goto filler; |
c855ff371 Fix a bad error c... |
2767 |
out: |
6fe6900e1 mm: make read_cac... |
2768 2769 2770 |
mark_page_accessed(page); return page; } |
0531b2aac mm: add new 'read... |
2771 2772 |
/** |
67f9fd91f mm: remove read_c... |
2773 |
* read_cache_page - read into page cache, fill it if needed |
0531b2aac mm: add new 'read... |
2774 2775 2776 |
* @mapping: the page's address_space * @index: the page index * @filler: function to perform the read |
5e5358e7c mm: cleanup descr... |
2777 |
* @data: first arg to filler(data, page) function, often left as NULL |
0531b2aac mm: add new 'read... |
2778 |
* |
0531b2aac mm: add new 'read... |
2779 |
* Read into the page cache. If a page already exists, and PageUptodate() is |
67f9fd91f mm: remove read_c... |
2780 |
* not set, try to fill the page and wait for it to become unlocked. |
0531b2aac mm: add new 'read... |
2781 2782 |
* * If the page does not get brought uptodate, return -EIO. |
a862f68a8 docs/core-api/mm:... |
2783 2784 |
* * Return: up to date page on success, ERR_PTR() on failure. |
0531b2aac mm: add new 'read... |
2785 |
*/ |
67f9fd91f mm: remove read_c... |
2786 |
struct page *read_cache_page(struct address_space *mapping, |
0531b2aac mm: add new 'read... |
2787 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2788 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2789 2790 |
void *data) { |
d322a8e5e mm/filemap.c: fix... |
2791 2792 |
return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); |
0531b2aac mm: add new 'read... |
2793 |
} |
67f9fd91f mm: remove read_c... |
2794 |
EXPORT_SYMBOL(read_cache_page); |
0531b2aac mm: add new 'read... |
2795 2796 2797 2798 2799 2800 2801 2802 |
/** * 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... |
2803 |
* any new page allocations done using the specified allocation flags. |
0531b2aac mm: add new 'read... |
2804 2805 |
* * If the page does not get brought uptodate, return -EIO. |
a862f68a8 docs/core-api/mm:... |
2806 2807 |
* * Return: up to date page on success, ERR_PTR() on failure. |
0531b2aac mm: add new 'read... |
2808 2809 2810 2811 2812 |
*/ struct page *read_cache_page_gfp(struct address_space *mapping, pgoff_t index, gfp_t gfp) { |
6c45b4541 mm/filemap: don't... |
2813 |
return do_read_cache_page(mapping, index, NULL, NULL, gfp); |
0531b2aac mm: add new 'read... |
2814 2815 |
} EXPORT_SYMBOL(read_cache_page_gfp); |
2f718ffc1 mm: buffered writ... |
2816 |
/* |
9fd91a90c vfs: strengthen c... |
2817 2818 2819 2820 |
* Don't operate on ranges the page cache doesn't support, and don't exceed the * LFS limits. If pos is under the limit it becomes a short access. If it * exceeds the limit we return -EFBIG. */ |
9fd91a90c vfs: strengthen c... |
2821 2822 2823 |
static int generic_write_check_limits(struct file *file, loff_t pos, loff_t *count) { |
646955cd5 vfs: remove redun... |
2824 2825 |
struct inode *inode = file->f_mapping->host; loff_t max_size = inode->i_sb->s_maxbytes; |
9fd91a90c vfs: strengthen c... |
2826 2827 2828 2829 2830 2831 2832 2833 2834 |
loff_t limit = rlimit(RLIMIT_FSIZE); if (limit != RLIM_INFINITY) { if (pos >= limit) { send_sig(SIGXFSZ, current, 0); return -EFBIG; } *count = min(*count, limit - pos); } |
646955cd5 vfs: remove redun... |
2835 2836 2837 2838 2839 2840 2841 2842 2843 |
if (!(file->f_flags & O_LARGEFILE)) max_size = MAX_NON_LFS; if (unlikely(pos >= max_size)) return -EFBIG; *count = min(*count, max_size - pos); return 0; |
9fd91a90c vfs: strengthen c... |
2844 2845 2846 |
} /* |
1da177e4c Linux-2.6.12-rc2 |
2847 2848 |
* Performs necessary checks before doing a write * |
485bb99b4 [PATCH] kernel-do... |
2849 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c Linux-2.6.12-rc2 |
2850 2851 2852 |
* Returns appropriate error code that caller should return or * zero in case that write should be allowed. */ |
3309dd04c switch generic_wr... |
2853 |
inline ssize_t generic_write_checks(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
2854 |
{ |
3309dd04c switch generic_wr... |
2855 |
struct file *file = iocb->ki_filp; |
1da177e4c Linux-2.6.12-rc2 |
2856 |
struct inode *inode = file->f_mapping->host; |
9fd91a90c vfs: strengthen c... |
2857 2858 |
loff_t count; int ret; |
1da177e4c Linux-2.6.12-rc2 |
2859 |
|
dc617f29d vfs: don't allow ... |
2860 2861 |
if (IS_SWAPFILE(inode)) return -ETXTBSY; |
3309dd04c switch generic_wr... |
2862 2863 |
if (!iov_iter_count(from)) return 0; |
1da177e4c Linux-2.6.12-rc2 |
2864 |
|
0fa6b005a generic_write_che... |
2865 |
/* FIXME: this is for backwards compatibility with 2.4 */ |
2ba48ce51 mirror O_APPEND a... |
2866 |
if (iocb->ki_flags & IOCB_APPEND) |
3309dd04c switch generic_wr... |
2867 |
iocb->ki_pos = i_size_read(inode); |
1da177e4c Linux-2.6.12-rc2 |
2868 |
|
6be96d3ad fs: return if dir... |
2869 2870 |
if ((iocb->ki_flags & IOCB_NOWAIT) && !(iocb->ki_flags & IOCB_DIRECT)) return -EINVAL; |
9fd91a90c vfs: strengthen c... |
2871 2872 2873 2874 |
count = iov_iter_count(from); ret = generic_write_check_limits(file, iocb->ki_pos, &count); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
2875 |
|
9fd91a90c vfs: strengthen c... |
2876 |
iov_iter_truncate(from, count); |
3309dd04c switch generic_wr... |
2877 |
return iov_iter_count(from); |
1da177e4c Linux-2.6.12-rc2 |
2878 2879 |
} EXPORT_SYMBOL(generic_write_checks); |
1383a7ed6 vfs: check file r... |
2880 2881 2882 |
/* * Performs necessary checks before doing a clone. * |
646955cd5 vfs: remove redun... |
2883 |
* Can adjust amount of bytes to clone via @req_count argument. |
1383a7ed6 vfs: check file r... |
2884 2885 2886 2887 2888 |
* Returns appropriate error code that caller should return or * zero in case the clone should be allowed. */ int generic_remap_checks(struct file *file_in, loff_t pos_in, struct file *file_out, loff_t pos_out, |
42ec3d4c0 vfs: make remap_f... |
2889 |
loff_t *req_count, unsigned int remap_flags) |
1383a7ed6 vfs: check file r... |
2890 2891 2892 2893 2894 2895 2896 |
{ struct inode *inode_in = file_in->f_mapping->host; struct inode *inode_out = file_out->f_mapping->host; uint64_t count = *req_count; uint64_t bcount; loff_t size_in, size_out; loff_t bs = inode_out->i_sb->s_blocksize; |
9fd91a90c vfs: strengthen c... |
2897 |
int ret; |
1383a7ed6 vfs: check file r... |
2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 |
/* The start of both ranges must be aligned to an fs block. */ if (!IS_ALIGNED(pos_in, bs) || !IS_ALIGNED(pos_out, bs)) return -EINVAL; /* Ensure offsets don't wrap. */ if (pos_in + count < pos_in || pos_out + count < pos_out) return -EINVAL; size_in = i_size_read(inode_in); size_out = i_size_read(inode_out); /* Dedupe requires both ranges to be within EOF. */ |
3d28193e1 vfs: pass remap f... |
2911 |
if ((remap_flags & REMAP_FILE_DEDUP) && |
1383a7ed6 vfs: check file r... |
2912 2913 2914 2915 2916 2917 2918 2919 |
(pos_in >= size_in || pos_in + count > size_in || pos_out >= size_out || pos_out + count > size_out)) return -EINVAL; /* Ensure the infile range is within the infile. */ if (pos_in >= size_in) return -EINVAL; count = min(count, size_in - (uint64_t)pos_in); |
9fd91a90c vfs: strengthen c... |
2920 2921 2922 |
ret = generic_write_check_limits(file_out, pos_out, &count); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
2923 2924 |
/* |
1383a7ed6 vfs: check file r... |
2925 2926 2927 2928 2929 |
* If the user wanted us to link to the infile's EOF, round up to the * next block boundary for this check. * * Otherwise, make sure the count is also block-aligned, having * already confirmed the starting offsets' block alignment. |
1da177e4c Linux-2.6.12-rc2 |
2930 |
*/ |
1383a7ed6 vfs: check file r... |
2931 2932 2933 2934 |
if (pos_in + count == size_in) { bcount = ALIGN(size_in, bs) - pos_in; } else { if (!IS_ALIGNED(count, bs)) |
eca3654e3 vfs: enable remap... |
2935 |
count = ALIGN_DOWN(count, bs); |
1383a7ed6 vfs: check file r... |
2936 |
bcount = count; |
1da177e4c Linux-2.6.12-rc2 |
2937 |
} |
1383a7ed6 vfs: check file r... |
2938 2939 2940 2941 2942 |
/* Don't allow overlapped cloning within the same file. */ if (inode_in == inode_out && pos_out + bcount > pos_in && pos_out < pos_in + bcount) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
2943 |
/* |
eca3654e3 vfs: enable remap... |
2944 2945 |
* We shortened the request but the caller can't deal with that, so * bounce the request back to userspace. |
1da177e4c Linux-2.6.12-rc2 |
2946 |
*/ |
eca3654e3 vfs: enable remap... |
2947 |
if (*req_count != count && !(remap_flags & REMAP_FILE_CAN_SHORTEN)) |
1383a7ed6 vfs: check file r... |
2948 |
return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
2949 |
|
eca3654e3 vfs: enable remap... |
2950 |
*req_count = count; |
1383a7ed6 vfs: check file r... |
2951 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
2952 |
} |
1da177e4c Linux-2.6.12-rc2 |
2953 |
|
a31713517 vfs: introduce ge... |
2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 |
/* * Performs common checks before doing a file copy/clone * from @file_in to @file_out. */ int generic_file_rw_checks(struct file *file_in, struct file *file_out) { struct inode *inode_in = file_inode(file_in); struct inode *inode_out = file_inode(file_out); /* Don't copy dirs, pipes, sockets... */ if (S_ISDIR(inode_in->i_mode) || S_ISDIR(inode_out->i_mode)) return -EISDIR; if (!S_ISREG(inode_in->i_mode) || !S_ISREG(inode_out->i_mode)) return -EINVAL; if (!(file_in->f_mode & FMODE_READ) || !(file_out->f_mode & FMODE_WRITE) || (file_out->f_flags & O_APPEND)) return -EBADF; return 0; } |
96e6e8f4a vfs: add missing ... |
2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 |
/* * Performs necessary checks before doing a file copy * * Can adjust amount of bytes to copy via @req_count argument. * Returns appropriate error code that caller should return or * zero in case the copy should be allowed. */ int generic_copy_file_checks(struct file *file_in, loff_t pos_in, struct file *file_out, loff_t pos_out, size_t *req_count, unsigned int flags) { struct inode *inode_in = file_inode(file_in); struct inode *inode_out = file_inode(file_out); uint64_t count = *req_count; loff_t size_in; int ret; ret = generic_file_rw_checks(file_in, file_out); if (ret) return ret; /* Don't touch certain kinds of inodes */ if (IS_IMMUTABLE(inode_out)) return -EPERM; if (IS_SWAPFILE(inode_in) || IS_SWAPFILE(inode_out)) return -ETXTBSY; /* Ensure offsets don't wrap. */ if (pos_in + count < pos_in || pos_out + count < pos_out) return -EOVERFLOW; /* Shorten the copy to EOF */ size_in = i_size_read(inode_in); if (pos_in >= size_in) count = 0; else count = min(count, size_in - (uint64_t)pos_in); ret = generic_write_check_limits(file_out, pos_out, &count); if (ret) return ret; /* Don't allow overlapped copying within the same file. */ if (inode_in == inode_out && pos_out + count > pos_in && pos_out < pos_in + count) return -EINVAL; *req_count = count; return 0; } |
afddba49d fs: introduce wri... |
3029 3030 3031 3032 3033 |
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... |
3034 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d fs: introduce wri... |
3035 |
pagep, fsdata); |
afddba49d fs: introduce wri... |
3036 3037 3038 3039 3040 3041 3042 3043 |
} 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... |
3044 |
|
4e02ed4b4 fs: remove prepar... |
3045 |
return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d fs: introduce wri... |
3046 3047 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c Linux-2.6.12-rc2 |
3048 |
ssize_t |
1af5bb491 filemap: remove t... |
3049 |
generic_file_direct_write(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
3050 3051 3052 3053 |
{ struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; |
1af5bb491 filemap: remove t... |
3054 |
loff_t pos = iocb->ki_pos; |
1da177e4c Linux-2.6.12-rc2 |
3055 |
ssize_t written; |
a969e903a kill generic_file... |
3056 3057 |
size_t write_len; pgoff_t end; |
1da177e4c Linux-2.6.12-rc2 |
3058 |
|
0c949334a iov_iter_truncate() |
3059 |
write_len = iov_iter_count(from); |
09cbfeaf1 mm, fs: get rid o... |
3060 |
end = (pos + write_len - 1) >> PAGE_SHIFT; |
a969e903a kill generic_file... |
3061 |
|
6be96d3ad fs: return if dir... |
3062 3063 3064 |
if (iocb->ki_flags & IOCB_NOWAIT) { /* If there are pages to writeback, return */ if (filemap_range_has_page(inode->i_mapping, pos, |
35f12f0f5 mm/filemap: pass ... |
3065 |
pos + write_len - 1)) |
6be96d3ad fs: return if dir... |
3066 3067 3068 3069 3070 3071 3072 |
return -EAGAIN; } else { written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); if (written) goto out; } |
a969e903a kill generic_file... |
3073 3074 3075 3076 3077 |
/* * 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... |
3078 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a kill generic_file... |
3079 |
*/ |
55635ba76 fs: fix data inva... |
3080 |
written = invalidate_inode_pages2_range(mapping, |
09cbfeaf1 mm, fs: get rid o... |
3081 |
pos >> PAGE_SHIFT, end); |
55635ba76 fs: fix data inva... |
3082 3083 3084 3085 3086 3087 3088 3089 |
/* * If a page can not be invalidated, return 0 to fall back * to buffered write. */ if (written) { if (written == -EBUSY) return 0; goto out; |
a969e903a kill generic_file... |
3090 |
} |
639a93a52 generic_file_dire... |
3091 |
written = mapping->a_ops->direct_IO(iocb, from); |
a969e903a kill generic_file... |
3092 3093 3094 3095 3096 3097 3098 3099 |
/* * 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... |
332391a99 fs: Fix page cach... |
3100 3101 3102 3103 3104 |
* * Most of the time we do not need this since dio_complete() will do * the invalidation for us. However there are some file systems that * do not end up with dio_complete() being called, so let's not break * them by removing it completely |
a969e903a kill generic_file... |
3105 |
*/ |
332391a99 fs: Fix page cach... |
3106 3107 3108 |
if (mapping->nrpages) invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT, end); |
a969e903a kill generic_file... |
3109 |
|
1da177e4c Linux-2.6.12-rc2 |
3110 |
if (written > 0) { |
0116651c8 mm: remove tempor... |
3111 |
pos += written; |
639a93a52 generic_file_dire... |
3112 |
write_len -= written; |
0116651c8 mm: remove tempor... |
3113 3114 |
if (pos > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, pos); |
1da177e4c Linux-2.6.12-rc2 |
3115 3116 |
mark_inode_dirty(inode); } |
5cb6c6c7e generic_file_dire... |
3117 |
iocb->ki_pos = pos; |
1da177e4c Linux-2.6.12-rc2 |
3118 |
} |
639a93a52 generic_file_dire... |
3119 |
iov_iter_revert(from, write_len - iov_iter_count(from)); |
a969e903a kill generic_file... |
3120 |
out: |
1da177e4c Linux-2.6.12-rc2 |
3121 3122 3123 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893 mm: buffered writ... |
3124 3125 3126 3127 |
/* * 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... |
3128 3129 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893 mm: buffered writ... |
3130 |
{ |
eb2be1893 mm: buffered writ... |
3131 |
struct page *page; |
bbddabe2e mm: filemap: only... |
3132 |
int fgp_flags = FGP_LOCK|FGP_WRITE|FGP_CREAT; |
0faa70cb0 mm: filemap: pass... |
3133 |
|
54566b2c1 fs: symlink write... |
3134 |
if (flags & AOP_FLAG_NOFS) |
2457aec63 mm: non-atomicall... |
3135 3136 3137 |
fgp_flags |= FGP_NOFS; page = pagecache_get_page(mapping, index, fgp_flags, |
45f87de57 mm: get rid of ra... |
3138 |
mapping_gfp_mask(mapping)); |
c585a2678 mm: remove likely... |
3139 |
if (page) |
2457aec63 mm: non-atomicall... |
3140 |
wait_for_stable_page(page); |
eb2be1893 mm: buffered writ... |
3141 |
|
eb2be1893 mm: buffered writ... |
3142 3143 |
return page; } |
54566b2c1 fs: symlink write... |
3144 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893 mm: buffered writ... |
3145 |
|
3b93f911d export generic_pe... |
3146 |
ssize_t generic_perform_write(struct file *file, |
afddba49d fs: introduce wri... |
3147 3148 3149 3150 3151 3152 |
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... |
3153 |
unsigned int flags = 0; |
afddba49d fs: introduce wri... |
3154 3155 |
do { struct page *page; |
afddba49d fs: introduce wri... |
3156 3157 3158 3159 |
unsigned long offset; /* Offset into pagecache page */ unsigned long bytes; /* Bytes to write to page */ size_t copied; /* Bytes copied from user */ void *fsdata; |
09cbfeaf1 mm, fs: get rid o... |
3160 3161 |
offset = (pos & (PAGE_SIZE - 1)); bytes = min_t(unsigned long, PAGE_SIZE - offset, |
afddba49d fs: introduce wri... |
3162 3163 3164 |
iov_iter_count(i)); again: |
00a3d660c Revert "fs: do no... |
3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 |
/* * 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; } |
296291cdd mm: make sendfile... |
3179 3180 3181 3182 |
if (fatal_signal_pending(current)) { status = -EINTR; break; } |
674b892ed mm: restore KERNE... |
3183 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d fs: introduce wri... |
3184 |
&page, &fsdata); |
2457aec63 mm: non-atomicall... |
3185 |
if (unlikely(status < 0)) |
afddba49d fs: introduce wri... |
3186 |
break; |
931e80e4b mm: flush dcache ... |
3187 3188 |
if (mapping_writably_mapped(mapping)) flush_dcache_page(page); |
00a3d660c Revert "fs: do no... |
3189 |
|
afddba49d fs: introduce wri... |
3190 |
copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); |
afddba49d fs: introduce wri... |
3191 3192 3193 3194 3195 3196 3197 3198 3199 |
flush_dcache_page(page); status = a_ops->write_end(file, mapping, pos, bytes, copied, page, fsdata); if (unlikely(status < 0)) break; copied = status; cond_resched(); |
124d3b704 fix writev regres... |
3200 |
iov_iter_advance(i, copied); |
afddba49d fs: introduce wri... |
3201 3202 3203 3204 3205 3206 3207 3208 3209 |
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. */ |
09cbfeaf1 mm, fs: get rid o... |
3210 |
bytes = min_t(unsigned long, PAGE_SIZE - offset, |
afddba49d fs: introduce wri... |
3211 3212 3213 |
iov_iter_single_seg_count(i)); goto again; } |
afddba49d fs: introduce wri... |
3214 3215 3216 3217 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); |
afddba49d fs: introduce wri... |
3218 3219 3220 3221 |
} while (iov_iter_count(i)); return written ? written : status; } |
3b93f911d export generic_pe... |
3222 |
EXPORT_SYMBOL(generic_perform_write); |
1da177e4c Linux-2.6.12-rc2 |
3223 |
|
e4dd9de3c vfs: Export __gen... |
3224 |
/** |
8174202b3 write_iter varian... |
3225 |
* __generic_file_write_iter - write data to a file |
e4dd9de3c vfs: Export __gen... |
3226 |
* @iocb: IO state structure (file, offset, etc.) |
8174202b3 write_iter varian... |
3227 |
* @from: iov_iter with data to write |
e4dd9de3c vfs: Export __gen... |
3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 |
* * 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. |
a862f68a8 docs/core-api/mm:... |
3240 3241 3242 3243 |
* * Return: * * number of bytes written, even for truncated writes * * negative error code if no data has been written at all |
e4dd9de3c vfs: Export __gen... |
3244 |
*/ |
8174202b3 write_iter varian... |
3245 |
ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
3246 3247 |
{ struct file *file = iocb->ki_filp; |
fb5527e68 [PATCH] direct-io... |
3248 |
struct address_space * mapping = file->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
3249 |
struct inode *inode = mapping->host; |
3b93f911d export generic_pe... |
3250 |
ssize_t written = 0; |
1da177e4c Linux-2.6.12-rc2 |
3251 |
ssize_t err; |
3b93f911d export generic_pe... |
3252 |
ssize_t status; |
1da177e4c Linux-2.6.12-rc2 |
3253 |
|
1da177e4c Linux-2.6.12-rc2 |
3254 |
/* We can write back this queue in page reclaim */ |
de1414a65 fs: export inode_... |
3255 |
current->backing_dev_info = inode_to_bdi(inode); |
5fa8e0a1c fs: Rename file_r... |
3256 |
err = file_remove_privs(file); |
1da177e4c Linux-2.6.12-rc2 |
3257 3258 |
if (err) goto out; |
c3b2da314 fs: introduce ino... |
3259 3260 3261 |
err = file_update_time(file); if (err) goto out; |
1da177e4c Linux-2.6.12-rc2 |
3262 |
|
2ba48ce51 mirror O_APPEND a... |
3263 |
if (iocb->ki_flags & IOCB_DIRECT) { |
0b8def9d6 __generic_file_wr... |
3264 |
loff_t pos, endbyte; |
fb5527e68 [PATCH] direct-io... |
3265 |
|
1af5bb491 filemap: remove t... |
3266 |
written = generic_file_direct_write(iocb, from); |
1da177e4c Linux-2.6.12-rc2 |
3267 |
/* |
fbbbad4bc vfs,ext2: introdu... |
3268 3269 3270 3271 3272 |
* If the write stopped short of completing, fall back to * buffered writes. Some filesystems do this for writes to * holes, for example. For DAX files, a buffered write will * not succeed (even if it did, DAX does not handle dirty * page-cache pages correctly). |
1da177e4c Linux-2.6.12-rc2 |
3273 |
*/ |
0b8def9d6 __generic_file_wr... |
3274 |
if (written < 0 || !iov_iter_count(from) || IS_DAX(inode)) |
fbbbad4bc vfs,ext2: introdu... |
3275 |
goto out; |
0b8def9d6 __generic_file_wr... |
3276 |
status = generic_perform_write(file, from, pos = iocb->ki_pos); |
fb5527e68 [PATCH] direct-io... |
3277 |
/* |
3b93f911d export generic_pe... |
3278 |
* If generic_perform_write() returned a synchronous error |
fb5527e68 [PATCH] direct-io... |
3279 3280 3281 3282 3283 |
* 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. */ |
60bb45297 __generic_file_wr... |
3284 |
if (unlikely(status < 0)) { |
3b93f911d export generic_pe... |
3285 |
err = status; |
fb5527e68 [PATCH] direct-io... |
3286 3287 |
goto out; } |
fb5527e68 [PATCH] direct-io... |
3288 3289 3290 3291 3292 |
/* * 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... |
3293 |
endbyte = pos + status - 1; |
0b8def9d6 __generic_file_wr... |
3294 |
err = filemap_write_and_wait_range(mapping, pos, endbyte); |
fb5527e68 [PATCH] direct-io... |
3295 |
if (err == 0) { |
0b8def9d6 __generic_file_wr... |
3296 |
iocb->ki_pos = endbyte + 1; |
3b93f911d export generic_pe... |
3297 |
written += status; |
fb5527e68 [PATCH] direct-io... |
3298 |
invalidate_mapping_pages(mapping, |
09cbfeaf1 mm, fs: get rid o... |
3299 3300 |
pos >> PAGE_SHIFT, endbyte >> PAGE_SHIFT); |
fb5527e68 [PATCH] direct-io... |
3301 3302 3303 3304 3305 3306 3307 |
} else { /* * We don't know how much we wrote, so just return * the number of bytes which were direct-written */ } } else { |
0b8def9d6 __generic_file_wr... |
3308 3309 3310 |
written = generic_perform_write(file, from, iocb->ki_pos); if (likely(written > 0)) iocb->ki_pos += written; |
fb5527e68 [PATCH] direct-io... |
3311 |
} |
1da177e4c Linux-2.6.12-rc2 |
3312 3313 3314 3315 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
8174202b3 write_iter varian... |
3316 |
EXPORT_SYMBOL(__generic_file_write_iter); |
e4dd9de3c vfs: Export __gen... |
3317 |
|
e4dd9de3c vfs: Export __gen... |
3318 |
/** |
8174202b3 write_iter varian... |
3319 |
* generic_file_write_iter - write data to a file |
e4dd9de3c vfs: Export __gen... |
3320 |
* @iocb: IO state structure |
8174202b3 write_iter varian... |
3321 |
* @from: iov_iter with data to write |
e4dd9de3c vfs: Export __gen... |
3322 |
* |
8174202b3 write_iter varian... |
3323 |
* This is a wrapper around __generic_file_write_iter() to be used by most |
e4dd9de3c vfs: Export __gen... |
3324 3325 |
* filesystems. It takes care of syncing the file in case of O_SYNC file * and acquires i_mutex as needed. |
a862f68a8 docs/core-api/mm:... |
3326 3327 3328 3329 |
* Return: * * negative error code if no data has been written at all of * vfs_fsync_range() failed for a synchronous write * * number of bytes written, even for truncated writes |
e4dd9de3c vfs: Export __gen... |
3330 |
*/ |
8174202b3 write_iter varian... |
3331 |
ssize_t generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
3332 3333 |
{ struct file *file = iocb->ki_filp; |
148f948ba vfs: Introduce ne... |
3334 |
struct inode *inode = file->f_mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
3335 |
ssize_t ret; |
1da177e4c Linux-2.6.12-rc2 |
3336 |
|
5955102c9 wrappers for ->i_... |
3337 |
inode_lock(inode); |
3309dd04c switch generic_wr... |
3338 3339 |
ret = generic_write_checks(iocb, from); if (ret > 0) |
5f380c7fa lift generic_writ... |
3340 |
ret = __generic_file_write_iter(iocb, from); |
5955102c9 wrappers for ->i_... |
3341 |
inode_unlock(inode); |
1da177e4c Linux-2.6.12-rc2 |
3342 |
|
e25922176 fs: simplify the ... |
3343 3344 |
if (ret > 0) ret = generic_write_sync(iocb, ret); |
1da177e4c Linux-2.6.12-rc2 |
3345 3346 |
return ret; } |
8174202b3 write_iter varian... |
3347 |
EXPORT_SYMBOL(generic_file_write_iter); |
1da177e4c Linux-2.6.12-rc2 |
3348 |
|
cf9a2ae8d [PATCH] BLOCK: Mo... |
3349 3350 3351 3352 3353 3354 3355 |
/** * 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 |
a862f68a8 docs/core-api/mm:... |
3356 |
* (presumably at page->private). |
cf9a2ae8d [PATCH] BLOCK: Mo... |
3357 |
* |
266cf658e FS-Cache: Recruit... |
3358 3359 3360 |
* 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... |
3361 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
71baba4b9 mm, page_alloc: r... |
3362 |
* this page (__GFP_IO), and whether the call may block (__GFP_RECLAIM & __GFP_FS). |
cf9a2ae8d [PATCH] BLOCK: Mo... |
3363 |
* |
a862f68a8 docs/core-api/mm:... |
3364 |
* Return: %1 if the release was successful, otherwise return zero. |
cf9a2ae8d [PATCH] BLOCK: Mo... |
3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 |
*/ 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); |