Blame view
mm/filemap.c
89.3 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 7 8 9 10 11 |
/* * linux/mm/filemap.c * * Copyright (C) 1994-1999 Linus Torvalds */ /* * This file handles the generic file mmap semantics used by * most "normal" filesystems (but you don't /have/ to use this: * the NFS filesystem used to do this differently, for example) */ |
b95f1b31b mm: Map most file... |
12 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
13 |
#include <linux/compiler.h> |
f9fe48bec dax: support dirt... |
14 |
#include <linux/dax.h> |
1da177e4c Linux-2.6.12-rc2 |
15 |
#include <linux/fs.h> |
3f07c0144 sched/headers: Pr... |
16 |
#include <linux/sched/signal.h> |
c22ce143d [PATCH] x86: cach... |
17 |
#include <linux/uaccess.h> |
c59ede7b7 [PATCH] move capa... |
18 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
19 |
#include <linux/kernel_stat.h> |
5a0e3ad6a include cleanup: ... |
20 |
#include <linux/gfp.h> |
1da177e4c Linux-2.6.12-rc2 |
21 22 23 24 25 26 27 28 |
#include <linux/mm.h> #include <linux/swap.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/file.h> #include <linux/uio.h> #include <linux/hash.h> #include <linux/writeback.h> |
53253383f Include <linux/ba... |
29 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
30 31 32 |
#include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/security.h> |
44110fe38 [PATCH] cpuset me... |
33 |
#include <linux/cpuset.h> |
00501b531 mm: memcontrol: r... |
34 |
#include <linux/hugetlb.h> |
8a9f3ccd2 Memory controller... |
35 |
#include <linux/memcontrol.h> |
c515e1fd3 mm/fs: add hooks ... |
36 |
#include <linux/cleancache.h> |
c7df8ad29 mm, truncate: do ... |
37 |
#include <linux/shmem_fs.h> |
f1820361f mm: implement ->m... |
38 |
#include <linux/rmap.h> |
0f8053a50 [PATCH] mm: make ... |
39 |
#include "internal.h" |
fe0bfaaff mm: trace filemap... |
40 41 |
#define CREATE_TRACE_POINTS #include <trace/events/filemap.h> |
1da177e4c Linux-2.6.12-rc2 |
42 |
/* |
1da177e4c Linux-2.6.12-rc2 |
43 44 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba vfs: Introduce ne... |
45 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c Linux-2.6.12-rc2 |
46 |
|
1da177e4c Linux-2.6.12-rc2 |
47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
#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... |
64 |
* ->i_mmap_rwsem (truncate_pagecache) |
1da177e4c Linux-2.6.12-rc2 |
65 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919 [PATCH] swap: swa... |
66 |
* ->swap_lock (exclusive_swap_page, others) |
b93b01631 page cache: use x... |
67 |
* ->i_pages lock |
1da177e4c Linux-2.6.12-rc2 |
68 |
* |
1b1dcc1b5 [PATCH] mutex sub... |
69 |
* ->i_mutex |
c8c06efa8 mm: convert i_mma... |
70 |
* ->i_mmap_rwsem (truncate->unmap_mapping_range) |
1da177e4c Linux-2.6.12-rc2 |
71 72 |
* * ->mmap_sem |
c8c06efa8 mm: convert i_mma... |
73 |
* ->i_mmap_rwsem |
b8072f099 [PATCH] mm: updat... |
74 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
b93b01631 page cache: use x... |
75 |
* ->i_pages lock (arch-dependent flush_dcache_mmap_lock) |
1da177e4c Linux-2.6.12-rc2 |
76 77 78 79 |
* * ->mmap_sem * ->lock_page (access_process_vm) * |
ccad23656 kill generic_file... |
80 |
* ->i_mutex (generic_perform_write) |
82591e6ea [PATCH] mm: more ... |
81 |
* ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c Linux-2.6.12-rc2 |
82 |
* |
f758eeabe writeback: split ... |
83 |
* bdi->wb.list_lock |
a66979aba fs: move i_wb_lis... |
84 |
* sb_lock (fs/fs-writeback.c) |
b93b01631 page cache: use x... |
85 |
* ->i_pages lock (__sync_single_inode) |
1da177e4c Linux-2.6.12-rc2 |
86 |
* |
c8c06efa8 mm: convert i_mma... |
87 |
* ->i_mmap_rwsem |
1da177e4c Linux-2.6.12-rc2 |
88 89 90 |
* ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099 [PATCH] mm: updat... |
91 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c Linux-2.6.12-rc2 |
92 |
* |
b8072f099 [PATCH] mm: updat... |
93 |
* ->page_table_lock or pte_lock |
5d337b919 [PATCH] swap: swa... |
94 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c Linux-2.6.12-rc2 |
95 |
* ->private_lock (try_to_unmap_one) |
b93b01631 page cache: use x... |
96 |
* ->i_pages lock (try_to_unmap_one) |
a52633d8e mm, vmscan: move ... |
97 98 |
* ->zone_lru_lock(zone) (follow_page->mark_page_accessed) * ->zone_lru_lock(zone) (check_pte_range->isolate_lru_page) |
1da177e4c Linux-2.6.12-rc2 |
99 |
* ->private_lock (page_remove_rmap->set_page_dirty) |
b93b01631 page cache: use x... |
100 |
* ->i_pages lock (page_remove_rmap->set_page_dirty) |
f758eeabe writeback: split ... |
101 |
* bdi.wb->list_lock (page_remove_rmap->set_page_dirty) |
250df6ed2 fs: protect inode... |
102 |
* ->inode->i_lock (page_remove_rmap->set_page_dirty) |
81f8c3a46 mm: memcontrol: g... |
103 |
* ->memcg->move_lock (page_remove_rmap->lock_page_memcg) |
f758eeabe writeback: split ... |
104 |
* bdi.wb->list_lock (zap_pte_range->set_page_dirty) |
250df6ed2 fs: protect inode... |
105 |
* ->inode->i_lock (zap_pte_range->set_page_dirty) |
1da177e4c Linux-2.6.12-rc2 |
106 107 |
* ->private_lock (zap_pte_range->__set_page_dirty_buffers) * |
c8c06efa8 mm: convert i_mma... |
108 |
* ->i_mmap_rwsem |
9a3c531df mm: update stale ... |
109 |
* ->tasklist_lock (memory_failure, collect_procs_ao) |
1da177e4c Linux-2.6.12-rc2 |
110 |
*/ |
22f2ac51b mm: workingset: f... |
111 112 113 114 115 116 |
static int page_cache_tree_insert(struct address_space *mapping, struct page *page, void **shadowp) { struct radix_tree_node *node; void **slot; int error; |
b93b01631 page cache: use x... |
117 |
error = __radix_tree_create(&mapping->i_pages, page->index, 0, |
22f2ac51b mm: workingset: f... |
118 119 120 121 122 |
&node, &slot); if (error) return error; if (*slot) { void *p; |
b93b01631 page cache: use x... |
123 124 |
p = radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); |
22f2ac51b mm: workingset: f... |
125 126 127 128 |
if (!radix_tree_exceptional_entry(p)) return -EEXIST; mapping->nrexceptional--; |
d01ad197a dax: remove DAX c... |
129 130 |
if (shadowp) *shadowp = p; |
22f2ac51b mm: workingset: f... |
131 |
} |
b93b01631 page cache: use x... |
132 |
__radix_tree_replace(&mapping->i_pages, node, slot, page, |
c7df8ad29 mm, truncate: do ... |
133 |
workingset_lookup_update(mapping)); |
22f2ac51b mm: workingset: f... |
134 |
mapping->nrpages++; |
22f2ac51b mm: workingset: f... |
135 136 |
return 0; } |
91b0abe36 mm + fs: store sh... |
137 138 139 |
static void page_cache_tree_delete(struct address_space *mapping, struct page *page, void *shadow) { |
c70b647d3 mm/filemap.c: add... |
140 141 142 143 |
int i, nr; /* hugetlb pages are represented by one entry in the radix tree */ nr = PageHuge(page) ? 1 : hpage_nr_pages(page); |
91b0abe36 mm + fs: store sh... |
144 |
|
83929372f filemap: prepare ... |
145 146 147 |
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... |
148 |
|
83929372f filemap: prepare ... |
149 |
for (i = 0; i < nr; i++) { |
d3798ae8c mm: filemap: don'... |
150 151 |
struct radix_tree_node *node; void **slot; |
b93b01631 page cache: use x... |
152 |
__radix_tree_lookup(&mapping->i_pages, page->index + i, |
d3798ae8c mm: filemap: don'... |
153 |
&node, &slot); |
dbc446b88 mm: workingset: r... |
154 |
VM_BUG_ON_PAGE(!node && nr != 1, page); |
449dd6984 mm: keep page cac... |
155 |
|
b93b01631 page cache: use x... |
156 157 |
radix_tree_clear_tags(&mapping->i_pages, node, slot); __radix_tree_replace(&mapping->i_pages, node, slot, shadow, |
c7df8ad29 mm, truncate: do ... |
158 |
workingset_lookup_update(mapping)); |
449dd6984 mm: keep page cac... |
159 |
} |
d3798ae8c mm: filemap: don'... |
160 |
|
2300638b1 mm: move clearing... |
161 162 |
page->mapping = NULL; /* Leave page->index set: truncation lookup relies upon it */ |
d3798ae8c mm: filemap: don'... |
163 164 165 166 167 168 169 170 171 172 173 |
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... |
174 |
} |
5ecc4d852 mm: factor out ch... |
175 176 |
static void unaccount_page_cache_page(struct address_space *mapping, struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
177 |
{ |
5ecc4d852 mm: factor out ch... |
178 |
int nr; |
1da177e4c Linux-2.6.12-rc2 |
179 |
|
c515e1fd3 mm/fs: add hooks ... |
180 181 182 183 184 185 186 187 |
/* * 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... |
188 |
cleancache_invalidate_page(mapping, page); |
c515e1fd3 mm/fs: add hooks ... |
189 |
|
83929372f filemap: prepare ... |
190 |
VM_BUG_ON_PAGE(PageTail(page), page); |
06b241f32 mm: __delete_from... |
191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 |
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 ... |
212 |
page_ref_sub(page, mapcount); |
06b241f32 mm: __delete_from... |
213 214 |
} } |
4165b9b46 hugetlb: do not a... |
215 |
/* hugetlb pages do not participate in page cache accounting. */ |
5ecc4d852 mm: factor out ch... |
216 217 |
if (PageHuge(page)) return; |
09612fa65 mm: hugetlb: retu... |
218 |
|
5ecc4d852 mm: factor out ch... |
219 220 221 222 223 224 225 226 227 |
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); } else { VM_BUG_ON_PAGE(PageTransHuge(page), page); |
800d8c63b shmem: add huge p... |
228 |
} |
5ecc4d852 mm: factor out ch... |
229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
/* * 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... |
247 |
* is safe. The caller must hold the i_pages lock. |
5ecc4d852 mm: factor out ch... |
248 249 250 251 252 253 254 255 |
*/ 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); |
76253fbc8 mm: move accounti... |
256 |
page_cache_tree_delete(mapping, page, shadow); |
1da177e4c Linux-2.6.12-rc2 |
257 |
} |
59c66c5f8 mm: factor out pa... |
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 |
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... |
274 275 276 277 278 279 280 281 282 |
/** * 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 |
283 |
{ |
83929372f filemap: prepare ... |
284 |
struct address_space *mapping = page_mapping(page); |
c4843a759 memcg: add per cg... |
285 |
unsigned long flags; |
1da177e4c Linux-2.6.12-rc2 |
286 |
|
cd7619d6b [PATCH] Extermina... |
287 |
BUG_ON(!PageLocked(page)); |
b93b01631 page cache: use x... |
288 |
xa_lock_irqsave(&mapping->i_pages, flags); |
62cccb8c8 mm: simplify lock... |
289 |
__delete_from_page_cache(page, NULL); |
b93b01631 page cache: use x... |
290 |
xa_unlock_irqrestore(&mapping->i_pages, flags); |
6072d13c4 Call the filesyst... |
291 |
|
59c66c5f8 mm: factor out pa... |
292 |
page_cache_free_page(mapping, page); |
97cecb5a2 mm: introduce del... |
293 294 |
} EXPORT_SYMBOL(delete_from_page_cache); |
aa65c29ce mm: batch radix t... |
295 296 297 298 299 |
/* * page_cache_tree_delete_batch - delete several pages from page cache * @mapping: the mapping to which pages belong * @pvec: pagevec with pages to delete * |
b93b01631 page cache: use x... |
300 301 302 |
* The function walks over mapping->i_pages and removes pages passed in @pvec * from the mapping. The function expects @pvec to be sorted by page index. * It tolerates holes in @pvec (mapping entries at those indices are not |
aa65c29ce mm: batch radix t... |
303 |
* modified). The function expects only THP head pages to be present in the |
b93b01631 page cache: use x... |
304 305 |
* @pvec and takes care to delete all corresponding tail pages from the * mapping as well. |
aa65c29ce mm: batch radix t... |
306 |
* |
b93b01631 page cache: use x... |
307 |
* The function expects the i_pages lock to be held. |
aa65c29ce mm: batch radix t... |
308 309 310 311 312 313 314 315 316 317 318 319 320 |
*/ static void page_cache_tree_delete_batch(struct address_space *mapping, struct pagevec *pvec) { struct radix_tree_iter iter; void **slot; int total_pages = 0; int i = 0, tail_pages = 0; struct page *page; pgoff_t start; start = pvec->pages[0]->index; |
b93b01631 page cache: use x... |
321 |
radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { |
aa65c29ce mm: batch radix t... |
322 323 324 |
if (i >= pagevec_count(pvec) && !tail_pages) break; page = radix_tree_deref_slot_protected(slot, |
b93b01631 page cache: use x... |
325 |
&mapping->i_pages.xa_lock); |
aa65c29ce mm: batch radix t... |
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 |
if (radix_tree_exceptional_entry(page)) continue; if (!tail_pages) { /* * Some page got inserted in our range? Skip it. We * have our pages locked so they are protected from * being removed. */ if (page != pvec->pages[i]) continue; WARN_ON_ONCE(!PageLocked(page)); if (PageTransHuge(page) && !PageHuge(page)) tail_pages = HPAGE_PMD_NR - 1; page->mapping = NULL; /* * Leave page->index set: truncation lookup relies * upon it */ i++; } else { tail_pages--; } |
b93b01631 page cache: use x... |
348 349 |
radix_tree_clear_tags(&mapping->i_pages, iter.node, slot); __radix_tree_replace(&mapping->i_pages, iter.node, slot, NULL, |
c7df8ad29 mm, truncate: do ... |
350 |
workingset_lookup_update(mapping)); |
aa65c29ce mm: batch radix t... |
351 352 353 354 355 356 357 358 359 360 361 362 363 |
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... |
364 |
xa_lock_irqsave(&mapping->i_pages, flags); |
aa65c29ce mm: batch radix t... |
365 366 367 368 369 370 |
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]); } page_cache_tree_delete_batch(mapping, pvec); |
b93b01631 page cache: use x... |
371 |
xa_unlock_irqrestore(&mapping->i_pages, flags); |
aa65c29ce mm: batch radix t... |
372 373 374 375 |
for (i = 0; i < pagevec_count(pvec); i++) page_cache_free_page(mapping, pvec->pages[i]); } |
d72d9e2a5 mm: export filema... |
376 |
int filemap_check_errors(struct address_space *mapping) |
865ffef37 fs: fix fsync() e... |
377 378 379 |
{ int ret = 0; /* Check for outstanding write errors */ |
7fcbbaf18 mm/filemap.c: avo... |
380 381 |
if (test_bit(AS_ENOSPC, &mapping->flags) && test_and_clear_bit(AS_ENOSPC, &mapping->flags)) |
865ffef37 fs: fix fsync() e... |
382 |
ret = -ENOSPC; |
7fcbbaf18 mm/filemap.c: avo... |
383 384 |
if (test_bit(AS_EIO, &mapping->flags) && test_and_clear_bit(AS_EIO, &mapping->flags)) |
865ffef37 fs: fix fsync() e... |
385 386 387 |
ret = -EIO; return ret; } |
d72d9e2a5 mm: export filema... |
388 |
EXPORT_SYMBOL(filemap_check_errors); |
865ffef37 fs: fix fsync() e... |
389 |
|
76341cabb jbd2: don't clear... |
390 391 392 393 394 395 396 397 398 |
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 |
399 |
/** |
485bb99b4 [PATCH] kernel-do... |
400 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b [PATCH] DocBook: ... |
401 402 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03 [PATCH] filemap_f... |
403 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b [PATCH] DocBook: ... |
404 |
* @sync_mode: enable synchronous operation |
1da177e4c Linux-2.6.12-rc2 |
405 |
* |
485bb99b4 [PATCH] kernel-do... |
406 407 408 |
* 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 |
409 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4 [PATCH] kernel-do... |
410 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c Linux-2.6.12-rc2 |
411 412 413 |
* these two operations is that if a dirty page/buffer is encountered, it must * be waited upon, and not just skipped over. */ |
ebcf28e1c [PATCH] fadvise()... |
414 415 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c Linux-2.6.12-rc2 |
416 417 418 419 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd mm: write_cache_p... |
420 |
.nr_to_write = LONG_MAX, |
111ebb6e6 [PATCH] writeback... |
421 422 |
.range_start = start, .range_end = end, |
1da177e4c Linux-2.6.12-rc2 |
423 424 425 426 |
}; if (!mapping_cap_writeback_dirty(mapping)) return 0; |
b16b1deb5 writeback: make w... |
427 |
wbc_attach_fdatawrite_inode(&wbc, mapping->host); |
1da177e4c Linux-2.6.12-rc2 |
428 |
ret = do_writepages(mapping, &wbc); |
b16b1deb5 writeback: make w... |
429 |
wbc_detach_inode(&wbc); |
1da177e4c Linux-2.6.12-rc2 |
430 431 432 433 434 435 |
return ret; } static inline int __filemap_fdatawrite(struct address_space *mapping, int sync_mode) { |
111ebb6e6 [PATCH] writeback... |
436 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c Linux-2.6.12-rc2 |
437 438 439 440 441 442 443 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf vfs: export filem... |
444 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c [PATCH] fadvise()... |
445 |
loff_t end) |
1da177e4c Linux-2.6.12-rc2 |
446 447 448 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf vfs: export filem... |
449 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c Linux-2.6.12-rc2 |
450 |
|
485bb99b4 [PATCH] kernel-do... |
451 452 453 454 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c Linux-2.6.12-rc2 |
455 456 457 458 459 460 461 462 |
* This is a mostly non-blocking flush. Not suitable for data-integrity * purposes - I/O may not be started against all dirty pages. */ int filemap_flush(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_NONE); } EXPORT_SYMBOL(filemap_flush); |
7fc9e4722 fs: Introduce fil... |
463 464 465 466 467 468 469 470 471 472 473 474 475 476 |
/** * 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. */ bool filemap_range_has_page(struct address_space *mapping, loff_t start_byte, loff_t end_byte) { pgoff_t index = start_byte >> PAGE_SHIFT; pgoff_t end = end_byte >> PAGE_SHIFT; |
f7b680468 mm: use find_get_... |
477 |
struct page *page; |
7fc9e4722 fs: Introduce fil... |
478 479 480 481 482 483 |
if (end_byte < start_byte) return false; if (mapping->nrpages == 0) return false; |
f7b680468 mm: use find_get_... |
484 |
if (!find_get_pages_range(mapping, &index, end, 1, &page)) |
7fc9e4722 fs: Introduce fil... |
485 |
return false; |
f7b680468 mm: use find_get_... |
486 487 |
put_page(page); return true; |
7fc9e4722 fs: Introduce fil... |
488 489 |
} EXPORT_SYMBOL(filemap_range_has_page); |
5e8fcc1a0 mm: don't TestCle... |
490 |
static void __filemap_fdatawait_range(struct address_space *mapping, |
aa750fd71 mm/filemap.c: mak... |
491 |
loff_t start_byte, loff_t end_byte) |
1da177e4c Linux-2.6.12-rc2 |
492 |
{ |
09cbfeaf1 mm, fs: get rid o... |
493 494 |
pgoff_t index = start_byte >> PAGE_SHIFT; pgoff_t end = end_byte >> PAGE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
495 496 |
struct pagevec pvec; int nr_pages; |
1da177e4c Linux-2.6.12-rc2 |
497 |
|
94004ed72 kill wait_on_page... |
498 |
if (end_byte < start_byte) |
5e8fcc1a0 mm: don't TestCle... |
499 |
return; |
1da177e4c Linux-2.6.12-rc2 |
500 |
|
866798201 mm, pagevec: remo... |
501 |
pagevec_init(&pvec); |
312e9d2f7 mm: use pagevec_l... |
502 |
while (index <= end) { |
1da177e4c Linux-2.6.12-rc2 |
503 |
unsigned i; |
312e9d2f7 mm: use pagevec_l... |
504 |
nr_pages = pagevec_lookup_range_tag(&pvec, mapping, &index, |
67fd707f4 mm: remove nr_pag... |
505 |
end, PAGECACHE_TAG_WRITEBACK); |
312e9d2f7 mm: use pagevec_l... |
506 507 |
if (!nr_pages) break; |
1da177e4c Linux-2.6.12-rc2 |
508 509 |
for (i = 0; i < nr_pages; i++) { struct page *page = pvec.pages[i]; |
1da177e4c Linux-2.6.12-rc2 |
510 |
wait_on_page_writeback(page); |
5e8fcc1a0 mm: don't TestCle... |
511 |
ClearPageError(page); |
1da177e4c Linux-2.6.12-rc2 |
512 513 514 515 |
} pagevec_release(&pvec); cond_resched(); } |
aa750fd71 mm/filemap.c: mak... |
516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 |
} /** * 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. */ int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) { |
5e8fcc1a0 mm: don't TestCle... |
535 536 |
__filemap_fdatawait_range(mapping, start_byte, end_byte); return filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
537 |
} |
d3bccb6f4 vfs: Introduce fi... |
538 539 540 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
a823e4589 mm: add file_fdat... |
541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 |
* 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. */ 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... |
562 563 |
/** |
aa750fd71 mm/filemap.c: mak... |
564 565 566 567 568 569 570 571 572 573 574 |
* 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) */ |
76341cabb jbd2: don't clear... |
575 |
int filemap_fdatawait_keep_errors(struct address_space *mapping) |
aa750fd71 mm/filemap.c: mak... |
576 |
{ |
ffb959bbd mm: remove optimi... |
577 |
__filemap_fdatawait_range(mapping, 0, LLONG_MAX); |
76341cabb jbd2: don't clear... |
578 |
return filemap_check_and_keep_errors(mapping); |
aa750fd71 mm/filemap.c: mak... |
579 |
} |
76341cabb jbd2: don't clear... |
580 |
EXPORT_SYMBOL(filemap_fdatawait_keep_errors); |
aa750fd71 mm/filemap.c: mak... |
581 |
|
9326c9b20 mm: consolidate d... |
582 |
static bool mapping_needs_writeback(struct address_space *mapping) |
1da177e4c Linux-2.6.12-rc2 |
583 |
{ |
9326c9b20 mm: consolidate d... |
584 585 |
return (!dax_mapping(mapping) && mapping->nrpages) || (dax_mapping(mapping) && mapping->nrexceptional); |
1da177e4c Linux-2.6.12-rc2 |
586 |
} |
1da177e4c Linux-2.6.12-rc2 |
587 588 589 |
int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982 [PATCH] Fix and a... |
590 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
591 |
|
9326c9b20 mm: consolidate d... |
592 |
if (mapping_needs_writeback(mapping)) { |
28fd12982 [PATCH] Fix and a... |
593 594 595 596 597 598 599 600 601 602 603 |
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/... |
604 605 606 |
} else { /* Clear any previously stored errors */ filemap_check_errors(mapping); |
28fd12982 [PATCH] Fix and a... |
607 |
} |
865ffef37 fs: fix fsync() e... |
608 609 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
610 |
} |
28fd12982 [PATCH] Fix and a... |
611 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
612 |
} |
28fd12982 [PATCH] Fix and a... |
613 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c Linux-2.6.12-rc2 |
614 |
|
485bb99b4 [PATCH] kernel-do... |
615 616 617 618 619 620 |
/** * 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... |
621 622 |
* Write out and wait upon file offsets lstart->lend, inclusive. * |
0e056eb55 kernel-api.rst: f... |
623 |
* Note that @lend is inclusive (describes the last byte to be written) so |
469eb4d03 [PATCH] filemap_f... |
624 625 |
* that this function can be used to write to the very end-of-file (end = -1). */ |
1da177e4c Linux-2.6.12-rc2 |
626 627 628 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982 [PATCH] Fix and a... |
629 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
630 |
|
9326c9b20 mm: consolidate d... |
631 |
if (mapping_needs_writeback(mapping)) { |
28fd12982 [PATCH] Fix and a... |
632 633 634 635 |
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... |
636 637 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982 [PATCH] Fix and a... |
638 639 |
if (!err) err = err2; |
cbeaf9510 mm: clear AS_EIO/... |
640 641 642 |
} else { /* Clear any previously stored errors */ filemap_check_errors(mapping); |
28fd12982 [PATCH] Fix and a... |
643 |
} |
865ffef37 fs: fix fsync() e... |
644 645 |
} else { err = filemap_check_errors(mapping); |
1da177e4c Linux-2.6.12-rc2 |
646 |
} |
28fd12982 [PATCH] Fix and a... |
647 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
648 |
} |
f69955855 Export filemap_wr... |
649 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c Linux-2.6.12-rc2 |
650 |
|
5660e13d2 fs: new infrastru... |
651 652 |
void __filemap_set_wb_err(struct address_space *mapping, int err) { |
3acdfd280 errseq: rename __... |
653 |
errseq_t eseq = errseq_set(&mapping->wb_err, err); |
5660e13d2 fs: new infrastru... |
654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 |
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. */ 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_... |
697 698 699 700 701 702 703 704 |
/* * 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... |
705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 |
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. */ 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... |
727 |
if (mapping_needs_writeback(mapping)) { |
5660e13d2 fs: new infrastru... |
728 729 730 731 732 733 734 735 736 737 738 739 |
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... |
740 |
/** |
ef6a3c631 mm: add replace_p... |
741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 |
* replace_page_cache_page - replace a pagecache page with a new one * @old: page to be replaced * @new: page to replace with * @gfp_mask: allocation mode * * This function replaces a page in the pagecache with a new one. On * success it acquires the pagecache reference for the new page and * drops it for the old page. Both the old and new pages must be * locked. This function does not add the new page to the LRU, the * caller must do that. * * The remove + add is atomic. The only way this function can fail is * memory allocation failure. */ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) { int error; |
ef6a3c631 mm: add replace_p... |
758 |
|
309381fea mm: dump page whe... |
759 760 761 |
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... |
762 |
|
abc1be13f mm/filemap.c: fix... |
763 |
error = radix_tree_preload(gfp_mask & GFP_RECLAIM_MASK); |
ef6a3c631 mm: add replace_p... |
764 765 766 |
if (!error) { struct address_space *mapping = old->mapping; void (*freepage)(struct page *); |
c4843a759 memcg: add per cg... |
767 |
unsigned long flags; |
ef6a3c631 mm: add replace_p... |
768 769 770 |
pgoff_t offset = old->index; freepage = mapping->a_ops->freepage; |
09cbfeaf1 mm, fs: get rid o... |
771 |
get_page(new); |
ef6a3c631 mm: add replace_p... |
772 773 |
new->mapping = mapping; new->index = offset; |
b93b01631 page cache: use x... |
774 |
xa_lock_irqsave(&mapping->i_pages, flags); |
62cccb8c8 mm: simplify lock... |
775 |
__delete_from_page_cache(old, NULL); |
22f2ac51b mm: workingset: f... |
776 |
error = page_cache_tree_insert(mapping, new, NULL); |
ef6a3c631 mm: add replace_p... |
777 |
BUG_ON(error); |
4165b9b46 hugetlb: do not a... |
778 779 780 781 782 |
/* * hugetlb pages do not participate in page cache accounting. */ if (!PageHuge(new)) |
11fb99898 mm: move most fil... |
783 |
__inc_node_page_state(new, NR_FILE_PAGES); |
ef6a3c631 mm: add replace_p... |
784 |
if (PageSwapBacked(new)) |
11fb99898 mm: move most fil... |
785 |
__inc_node_page_state(new, NR_SHMEM); |
b93b01631 page cache: use x... |
786 |
xa_unlock_irqrestore(&mapping->i_pages, flags); |
6a93ca8fd mm: migrate: do n... |
787 |
mem_cgroup_migrate(old, new); |
ef6a3c631 mm: add replace_p... |
788 789 790 |
radix_tree_preload_end(); if (freepage) freepage(old); |
09cbfeaf1 mm, fs: get rid o... |
791 |
put_page(old); |
ef6a3c631 mm: add replace_p... |
792 793 794 795 796 |
} return error; } EXPORT_SYMBOL_GPL(replace_page_cache_page); |
a528910e1 mm: thrash detect... |
797 798 799 800 |
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 |
801 |
{ |
00501b531 mm: memcontrol: r... |
802 803 |
int huge = PageHuge(page); struct mem_cgroup *memcg; |
e286781d5 mm: speculative p... |
804 |
int error; |
309381fea mm: dump page whe... |
805 806 |
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageSwapBacked(page), page); |
e286781d5 mm: speculative p... |
807 |
|
00501b531 mm: memcontrol: r... |
808 809 |
if (!huge) { error = mem_cgroup_try_charge(page, current->mm, |
f627c2f53 memcg: adjust to ... |
810 |
gfp_mask, &memcg, false); |
00501b531 mm: memcontrol: r... |
811 812 813 |
if (error) return error; } |
1da177e4c Linux-2.6.12-rc2 |
814 |
|
abc1be13f mm/filemap.c: fix... |
815 |
error = radix_tree_maybe_preload(gfp_mask & GFP_RECLAIM_MASK); |
66a0c8ee3 mm: cleanup add_t... |
816 |
if (error) { |
00501b531 mm: memcontrol: r... |
817 |
if (!huge) |
f627c2f53 memcg: adjust to ... |
818 |
mem_cgroup_cancel_charge(page, memcg, false); |
66a0c8ee3 mm: cleanup add_t... |
819 820 |
return error; } |
09cbfeaf1 mm, fs: get rid o... |
821 |
get_page(page); |
66a0c8ee3 mm: cleanup add_t... |
822 823 |
page->mapping = mapping; page->index = offset; |
b93b01631 page cache: use x... |
824 |
xa_lock_irq(&mapping->i_pages); |
a528910e1 mm: thrash detect... |
825 |
error = page_cache_tree_insert(mapping, page, shadowp); |
66a0c8ee3 mm: cleanup add_t... |
826 827 828 |
radix_tree_preload_end(); if (unlikely(error)) goto err_insert; |
4165b9b46 hugetlb: do not a... |
829 830 831 |
/* hugetlb pages do not participate in page cache accounting. */ if (!huge) |
11fb99898 mm: move most fil... |
832 |
__inc_node_page_state(page, NR_FILE_PAGES); |
b93b01631 page cache: use x... |
833 |
xa_unlock_irq(&mapping->i_pages); |
00501b531 mm: memcontrol: r... |
834 |
if (!huge) |
f627c2f53 memcg: adjust to ... |
835 |
mem_cgroup_commit_charge(page, memcg, false, false); |
66a0c8ee3 mm: cleanup add_t... |
836 837 838 839 840 |
trace_mm_filemap_add_to_page_cache(page); return 0; err_insert: page->mapping = NULL; /* Leave page->index set: truncation relies upon it */ |
b93b01631 page cache: use x... |
841 |
xa_unlock_irq(&mapping->i_pages); |
00501b531 mm: memcontrol: r... |
842 |
if (!huge) |
f627c2f53 memcg: adjust to ... |
843 |
mem_cgroup_cancel_charge(page, memcg, false); |
09cbfeaf1 mm, fs: get rid o... |
844 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
845 846 |
return error; } |
a528910e1 mm: thrash detect... |
847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 |
/** * add_to_page_cache_locked - add a locked page to the pagecache * @page: page to add * @mapping: the page's address_space * @offset: page index * @gfp_mask: page allocation mode * * This function is used to add a page to the pagecache. It must be locked. * This function does not add the page to the LRU. The caller must do that. */ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, pgoff_t offset, gfp_t gfp_mask) { return __add_to_page_cache_locked(page, mapping, offset, gfp_mask, NULL); } |
e286781d5 mm: speculative p... |
864 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c Linux-2.6.12-rc2 |
865 866 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
867 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
868 |
{ |
a528910e1 mm: thrash detect... |
869 |
void *shadow = NULL; |
4f98a2fee vmscan: split LRU... |
870 |
int ret; |
48c935ad8 page-flags: defin... |
871 |
__SetPageLocked(page); |
a528910e1 mm: thrash detect... |
872 873 874 |
ret = __add_to_page_cache_locked(page, mapping, offset, gfp_mask, &shadow); if (unlikely(ret)) |
48c935ad8 page-flags: defin... |
875 |
__ClearPageLocked(page); |
a528910e1 mm: thrash detect... |
876 877 878 879 880 |
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... |
881 882 883 |
* 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... |
884 |
*/ |
f0281a00f mm: workingset: o... |
885 886 |
if (!(gfp_mask & __GFP_WRITE) && shadow && workingset_refault(shadow)) { |
a528910e1 mm: thrash detect... |
887 888 889 890 891 892 |
SetPageActive(page); workingset_activation(page); } else ClearPageActive(page); lru_cache_add(page); } |
1da177e4c Linux-2.6.12-rc2 |
893 894 |
return ret; } |
18bc0bbd1 Staging: pohmelfs... |
895 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c Linux-2.6.12-rc2 |
896 |
|
44110fe38 [PATCH] cpuset me... |
897 |
#ifdef CONFIG_NUMA |
2ae88149a [PATCH] mm: clean... |
898 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38 [PATCH] cpuset me... |
899 |
{ |
c0ff7453b cpuset,mm: fix no... |
900 901 |
int n; struct page *page; |
44110fe38 [PATCH] cpuset me... |
902 |
if (cpuset_do_page_mem_spread()) { |
cc9a6c877 cpuset: mm: reduc... |
903 904 |
unsigned int cpuset_mems_cookie; do { |
d26914d11 mm: optimize put_... |
905 |
cpuset_mems_cookie = read_mems_allowed_begin(); |
cc9a6c877 cpuset: mm: reduc... |
906 |
n = cpuset_mem_spread_node(); |
96db800f5 mm: rename alloc_... |
907 |
page = __alloc_pages_node(n, gfp, 0); |
d26914d11 mm: optimize put_... |
908 |
} while (!page && read_mems_allowed_retry(cpuset_mems_cookie)); |
cc9a6c877 cpuset: mm: reduc... |
909 |
|
c0ff7453b cpuset,mm: fix no... |
910 |
return page; |
44110fe38 [PATCH] cpuset me... |
911 |
} |
2ae88149a [PATCH] mm: clean... |
912 |
return alloc_pages(gfp, 0); |
44110fe38 [PATCH] cpuset me... |
913 |
} |
2ae88149a [PATCH] mm: clean... |
914 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38 [PATCH] cpuset me... |
915 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
916 917 918 919 920 921 922 923 924 925 |
/* * 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... |
926 927 928 929 930 |
#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 |
931 |
{ |
629060270 mm: add PageWaite... |
932 |
return &page_wait_table[hash_ptr(page, PAGE_WAIT_TABLE_BITS)]; |
1da177e4c Linux-2.6.12-rc2 |
933 |
} |
1da177e4c Linux-2.6.12-rc2 |
934 |
|
629060270 mm: add PageWaite... |
935 |
void __init pagecache_init(void) |
1da177e4c Linux-2.6.12-rc2 |
936 |
{ |
629060270 mm: add PageWaite... |
937 |
int i; |
1da177e4c Linux-2.6.12-rc2 |
938 |
|
629060270 mm: add PageWaite... |
939 940 941 942 |
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 |
943 |
} |
1da177e4c Linux-2.6.12-rc2 |
944 |
|
3510ca20e Minor page waitqu... |
945 |
/* This has the same layout as wait_bit_key - see fs/cachefiles/rdwr.c */ |
629060270 mm: add PageWaite... |
946 947 948 949 950 951 952 953 954 |
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... |
955 |
wait_queue_entry_t wait; |
629060270 mm: add PageWaite... |
956 |
}; |
ac6424b98 sched/wait: Renam... |
957 |
static int wake_page_function(wait_queue_entry_t *wait, unsigned mode, int sync, void *arg) |
f62e00cc3 mm: introduce wai... |
958 |
{ |
629060270 mm: add PageWaite... |
959 960 961 962 963 964 965 |
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... |
966 |
|
629060270 mm: add PageWaite... |
967 968 |
if (wait_page->bit_nr != key->bit_nr) return 0; |
3510ca20e Minor page waitqu... |
969 970 |
/* Stop walking if it's locked */ |
629060270 mm: add PageWaite... |
971 |
if (test_bit(key->bit_nr, &key->page->flags)) |
3510ca20e Minor page waitqu... |
972 |
return -1; |
f62e00cc3 mm: introduce wai... |
973 |
|
629060270 mm: add PageWaite... |
974 |
return autoremove_wake_function(wait, mode, sync, key); |
f62e00cc3 mm: introduce wai... |
975 |
} |
74d81bfae mm: un-export wak... |
976 |
static void wake_up_page_bit(struct page *page, int bit_nr) |
cbbce8220 SCHED: add some "... |
977 |
{ |
629060270 mm: add PageWaite... |
978 979 980 |
wait_queue_head_t *q = page_waitqueue(page); struct wait_page_key key; unsigned long flags; |
11a19c7b0 sched/wait: Intro... |
981 |
wait_queue_entry_t bookmark; |
cbbce8220 SCHED: add some "... |
982 |
|
629060270 mm: add PageWaite... |
983 984 985 |
key.page = page; key.bit_nr = bit_nr; key.page_match = 0; |
11a19c7b0 sched/wait: Intro... |
986 987 988 989 |
bookmark.flags = 0; bookmark.private = NULL; bookmark.func = NULL; INIT_LIST_HEAD(&bookmark.entry); |
629060270 mm: add PageWaite... |
990 |
spin_lock_irqsave(&q->lock, flags); |
11a19c7b0 sched/wait: Intro... |
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 |
__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... |
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 |
/* * 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... |
1026 1027 1028 1029 1030 1031 1032 |
static void wake_up_page(struct page *page, int bit) { if (!PageWaiters(page)) return; wake_up_page_bit(page, bit); } |
629060270 mm: add PageWaite... |
1033 1034 1035 1036 1037 |
static inline int wait_on_page_bit_common(wait_queue_head_t *q, struct page *page, int bit_nr, int state, bool lock) { struct wait_page_queue wait_page; |
ac6424b98 sched/wait: Renam... |
1038 |
wait_queue_entry_t *wait = &wait_page.wait; |
629060270 mm: add PageWaite... |
1039 1040 1041 |
int ret = 0; init_wait(wait); |
3510ca20e Minor page waitqu... |
1042 |
wait->flags = lock ? WQ_FLAG_EXCLUSIVE : 0; |
629060270 mm: add PageWaite... |
1043 1044 1045 1046 1047 1048 |
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... |
1049 |
if (likely(list_empty(&wait->entry))) { |
3510ca20e Minor page waitqu... |
1050 |
__add_wait_queue_entry_tail(q, wait); |
629060270 mm: add PageWaite... |
1051 1052 1053 1054 1055 1056 1057 1058 1059 |
SetPageWaiters(page); } set_current_state(state); spin_unlock_irq(&q->lock); if (likely(test_bit(bit_nr, &page->flags))) { io_schedule(); |
629060270 mm: add PageWaite... |
1060 1061 1062 1063 1064 1065 1066 1067 1068 |
} if (lock) { if (!test_and_set_bit_lock(bit_nr, &page->flags)) break; } else { if (!test_bit(bit_nr, &page->flags)) break; } |
a8b169afb Avoid page waitqu... |
1069 1070 1071 1072 1073 |
if (unlikely(signal_pending_state(state, current))) { ret = -EINTR; break; } |
629060270 mm: add PageWaite... |
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 |
} finish_wait(q, wait); /* * 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); wait_on_page_bit_common(q, page, bit_nr, TASK_UNINTERRUPTIBLE, false); } 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); return wait_on_page_bit_common(q, page, bit_nr, TASK_KILLABLE, false); |
cbbce8220 SCHED: add some "... |
1100 |
} |
4343d0087 afs: Get rid of t... |
1101 |
EXPORT_SYMBOL(wait_on_page_bit_killable); |
cbbce8220 SCHED: add some "... |
1102 |
|
1da177e4c Linux-2.6.12-rc2 |
1103 |
/** |
385e1ca5f CacheFiles: Permi... |
1104 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc filemap: fix kern... |
1105 1106 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f CacheFiles: Permi... |
1107 1108 1109 |
* * Add an arbitrary @waiter to the wait queue for the nominated @page. */ |
ac6424b98 sched/wait: Renam... |
1110 |
void add_page_wait_queue(struct page *page, wait_queue_entry_t *waiter) |
385e1ca5f CacheFiles: Permi... |
1111 1112 1113 1114 1115 |
{ wait_queue_head_t *q = page_waitqueue(page); unsigned long flags; spin_lock_irqsave(&q->lock, flags); |
9c3a815f4 page waitqueue: a... |
1116 |
__add_wait_queue_entry_tail(q, waiter); |
629060270 mm: add PageWaite... |
1117 |
SetPageWaiters(page); |
385e1ca5f CacheFiles: Permi... |
1118 1119 1120 |
spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL_GPL(add_page_wait_queue); |
b91e1302a mm: optimize Page... |
1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 |
#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... |
1139 |
return test_bit(PG_waiters, mem); |
b91e1302a mm: optimize Page... |
1140 1141 1142 |
} #endif |
385e1ca5f CacheFiles: Permi... |
1143 |
/** |
485bb99b4 [PATCH] kernel-do... |
1144 |
* unlock_page - unlock a locked page |
1da177e4c Linux-2.6.12-rc2 |
1145 1146 1147 1148 |
* @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... |
1149 |
* mechanism between PageLocked pages and PageWriteback pages is shared. |
1da177e4c Linux-2.6.12-rc2 |
1150 1151 |
* But that's OK - sleepers in wait_on_page_writeback() just go back to sleep. * |
b91e1302a mm: optimize Page... |
1152 1153 1154 1155 1156 |
* 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 |
1157 |
*/ |
920c7a5d0 mm: remove fastca... |
1158 |
void unlock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
1159 |
{ |
b91e1302a mm: optimize Page... |
1160 |
BUILD_BUG_ON(PG_waiters != 7); |
48c935ad8 page-flags: defin... |
1161 |
page = compound_head(page); |
309381fea mm: dump page whe... |
1162 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
b91e1302a mm: optimize Page... |
1163 1164 |
if (clear_bit_unlock_is_negative_byte(PG_locked, &page->flags)) wake_up_page_bit(page, PG_locked); |
1da177e4c Linux-2.6.12-rc2 |
1165 1166 |
} EXPORT_SYMBOL(unlock_page); |
485bb99b4 [PATCH] kernel-do... |
1167 1168 1169 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c Linux-2.6.12-rc2 |
1170 1171 1172 |
*/ void end_page_writeback(struct page *page) { |
888cf2db4 mm: avoid unneces... |
1173 1174 1175 1176 1177 1178 1179 1180 1181 |
/* * 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... |
1182 |
rotate_reclaimable_page(page); |
888cf2db4 mm: avoid unneces... |
1183 |
} |
ac6aadb24 mm: rotate_reclai... |
1184 1185 1186 |
if (!test_clear_page_writeback(page)) BUG(); |
4e857c58e arch: Mass conver... |
1187 |
smp_mb__after_atomic(); |
1da177e4c Linux-2.6.12-rc2 |
1188 1189 1190 |
wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
57d998456 fs/mpage.c: facto... |
1191 1192 1193 1194 |
/* * After completing I/O on a page, call this routine to update the page * flags appropriately */ |
c11f0c0b5 block/mm: make bd... |
1195 |
void page_endio(struct page *page, bool is_write, int err) |
57d998456 fs/mpage.c: facto... |
1196 |
{ |
c11f0c0b5 block/mm: make bd... |
1197 |
if (!is_write) { |
57d998456 fs/mpage.c: facto... |
1198 1199 1200 1201 1202 1203 1204 |
if (!err) { SetPageUptodate(page); } else { ClearPageUptodate(page); SetPageError(page); } unlock_page(page); |
abf545484 mm/block: convert... |
1205 |
} else { |
57d998456 fs/mpage.c: facto... |
1206 |
if (err) { |
dd8416c47 mm: do not access... |
1207 |
struct address_space *mapping; |
57d998456 fs/mpage.c: facto... |
1208 |
SetPageError(page); |
dd8416c47 mm: do not access... |
1209 1210 1211 |
mapping = page_mapping(page); if (mapping) mapping_set_error(mapping, err); |
57d998456 fs/mpage.c: facto... |
1212 1213 1214 1215 1216 |
} end_page_writeback(page); } } EXPORT_SYMBOL_GPL(page_endio); |
485bb99b4 [PATCH] kernel-do... |
1217 1218 |
/** * __lock_page - get a lock on the page, assuming we need to sleep to get it |
870667553 mm: fix filemap.c... |
1219 |
* @__page: the page to lock |
1da177e4c Linux-2.6.12-rc2 |
1220 |
*/ |
629060270 mm: add PageWaite... |
1221 |
void __lock_page(struct page *__page) |
1da177e4c Linux-2.6.12-rc2 |
1222 |
{ |
629060270 mm: add PageWaite... |
1223 1224 1225 |
struct page *page = compound_head(__page); wait_queue_head_t *q = page_waitqueue(page); wait_on_page_bit_common(q, page, PG_locked, TASK_UNINTERRUPTIBLE, true); |
1da177e4c Linux-2.6.12-rc2 |
1226 1227 |
} EXPORT_SYMBOL(__lock_page); |
629060270 mm: add PageWaite... |
1228 |
int __lock_page_killable(struct page *__page) |
2687a3569 Add lock_page_kil... |
1229 |
{ |
629060270 mm: add PageWaite... |
1230 1231 1232 |
struct page *page = compound_head(__page); wait_queue_head_t *q = page_waitqueue(page); return wait_on_page_bit_common(q, page, PG_locked, TASK_KILLABLE, true); |
2687a3569 Add lock_page_kil... |
1233 |
} |
18bc0bbd1 Staging: pohmelfs... |
1234 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569 Add lock_page_kil... |
1235 |
|
9a95f3cf7 mm: describe mmap... |
1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 |
/* * 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... |
1247 1248 1249 |
int __lock_page_or_retry(struct page *page, struct mm_struct *mm, unsigned int flags) { |
37b23e052 x86,mm: make page... |
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 |
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 ... |
1262 |
wait_on_page_locked(page); |
d065bd810 mm: retry page fa... |
1263 |
return 0; |
37b23e052 x86,mm: make page... |
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 |
} 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... |
1276 1277 |
} } |
485bb99b4 [PATCH] kernel-do... |
1278 |
/** |
e7b563bb2 mm: filemap: move... |
1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 |
* page_cache_next_hole - find the next hole (not-present entry) * @mapping: mapping * @index: index * @max_scan: maximum range to search * * Search the set [index, min(index+max_scan-1, MAX_INDEX)] for the * lowest indexed hole. * * Returns: the index of the hole if found, otherwise returns an index * outside of the set specified (in which case 'return - index >= * max_scan' will be true). In rare cases of index wrap-around, 0 will * be returned. * * page_cache_next_hole may be called under rcu_read_lock. However, * like radix_tree_gang_lookup, this will not atomically search a * snapshot of the tree at a single point in time. For example, if a * hole is created at index 5, then subsequently a hole is created at * index 10, page_cache_next_hole covering both indexes may return 10 * if called under rcu_read_lock. */ pgoff_t page_cache_next_hole(struct address_space *mapping, pgoff_t index, unsigned long max_scan) { unsigned long i; for (i = 0; i < max_scan; i++) { |
0cd6144aa mm + fs: prepare ... |
1305 |
struct page *page; |
b93b01631 page cache: use x... |
1306 |
page = radix_tree_lookup(&mapping->i_pages, index); |
0cd6144aa mm + fs: prepare ... |
1307 |
if (!page || radix_tree_exceptional_entry(page)) |
e7b563bb2 mm: filemap: move... |
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 |
break; index++; if (index == 0) break; } return index; } EXPORT_SYMBOL(page_cache_next_hole); /** * page_cache_prev_hole - find the prev hole (not-present entry) * @mapping: mapping * @index: index * @max_scan: maximum range to search * * Search backwards in the range [max(index-max_scan+1, 0), index] for * the first hole. * * Returns: the index of the hole if found, otherwise returns an index * outside of the set specified (in which case 'index - return >= * max_scan' will be true). In rare cases of wrap-around, ULONG_MAX * will be returned. * * page_cache_prev_hole may be called under rcu_read_lock. However, * like radix_tree_gang_lookup, this will not atomically search a * snapshot of the tree at a single point in time. For example, if a * hole is created at index 10, then subsequently a hole is created at * index 5, page_cache_prev_hole covering both indexes may return 5 if * called under rcu_read_lock. */ pgoff_t page_cache_prev_hole(struct address_space *mapping, pgoff_t index, unsigned long max_scan) { unsigned long i; for (i = 0; i < max_scan; i++) { |
0cd6144aa mm + fs: prepare ... |
1345 |
struct page *page; |
b93b01631 page cache: use x... |
1346 |
page = radix_tree_lookup(&mapping->i_pages, index); |
0cd6144aa mm + fs: prepare ... |
1347 |
if (!page || radix_tree_exceptional_entry(page)) |
e7b563bb2 mm: filemap: move... |
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 |
break; index--; if (index == ULONG_MAX) break; } return index; } EXPORT_SYMBOL(page_cache_prev_hole); /** |
0cd6144aa mm + fs: prepare ... |
1359 |
* find_get_entry - find and get a page cache entry |
485bb99b4 [PATCH] kernel-do... |
1360 |
* @mapping: the address_space to search |
0cd6144aa mm + fs: prepare ... |
1361 1362 1363 1364 |
* @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... |
1365 |
* |
139b6a6fb mm: filemap: upda... |
1366 1367 |
* 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 ... |
1368 1369 |
* * Otherwise, %NULL is returned. |
1da177e4c Linux-2.6.12-rc2 |
1370 |
*/ |
0cd6144aa mm + fs: prepare ... |
1371 |
struct page *find_get_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1372 |
{ |
a60637c85 mm: lockless page... |
1373 |
void **pagep; |
83929372f filemap: prepare ... |
1374 |
struct page *head, *page; |
1da177e4c Linux-2.6.12-rc2 |
1375 |
|
a60637c85 mm: lockless page... |
1376 1377 1378 |
rcu_read_lock(); repeat: page = NULL; |
b93b01631 page cache: use x... |
1379 |
pagep = radix_tree_lookup_slot(&mapping->i_pages, offset); |
a60637c85 mm: lockless page... |
1380 1381 |
if (pagep) { page = radix_tree_deref_slot(pagep); |
27d20fddc radix-tree: fix R... |
1382 1383 |
if (unlikely(!page)) goto out; |
a2c16d6cb mm: let swap use ... |
1384 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1385 1386 1387 |
if (radix_tree_deref_retry(page)) goto repeat; /* |
139b6a6fb mm: filemap: upda... |
1388 1389 1390 |
* A shadow entry of a recently evicted page, * or a swap entry from shmem/tmpfs. Return * it without attempting to raise page count. |
8079b1c85 mm: clarify the r... |
1391 1392 |
*/ goto out; |
a2c16d6cb mm: let swap use ... |
1393 |
} |
83929372f filemap: prepare ... |
1394 1395 1396 1397 1398 1399 1400 1401 |
head = compound_head(page); if (!page_cache_get_speculative(head)) goto repeat; /* The page was split under us? */ if (compound_head(page) != head) { put_page(head); |
a60637c85 mm: lockless page... |
1402 |
goto repeat; |
83929372f filemap: prepare ... |
1403 |
} |
a60637c85 mm: lockless page... |
1404 1405 1406 1407 1408 1409 1410 |
/* * Has the page moved? * This is part of the lockless pagecache protocol. See * include/linux/pagemap.h for details. */ if (unlikely(page != *pagep)) { |
83929372f filemap: prepare ... |
1411 |
put_page(head); |
a60637c85 mm: lockless page... |
1412 1413 1414 |
goto repeat; } } |
27d20fddc radix-tree: fix R... |
1415 |
out: |
a60637c85 mm: lockless page... |
1416 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1417 1418 |
return page; } |
0cd6144aa mm + fs: prepare ... |
1419 |
EXPORT_SYMBOL(find_get_entry); |
1da177e4c Linux-2.6.12-rc2 |
1420 |
|
485bb99b4 [PATCH] kernel-do... |
1421 |
/** |
0cd6144aa mm + fs: prepare ... |
1422 1423 1424 1425 1426 1427 1428 1429 |
* 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... |
1430 1431 |
* 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 ... |
1432 1433 1434 1435 1436 1437 |
* * Otherwise, %NULL is returned. * * find_lock_entry() may sleep. */ struct page *find_lock_entry(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1438 1439 |
{ struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1440 |
repeat: |
0cd6144aa mm + fs: prepare ... |
1441 |
page = find_get_entry(mapping, offset); |
a2c16d6cb mm: let swap use ... |
1442 |
if (page && !radix_tree_exception(page)) { |
a60637c85 mm: lockless page... |
1443 1444 |
lock_page(page); /* Has the page been truncated? */ |
83929372f filemap: prepare ... |
1445 |
if (unlikely(page_mapping(page) != mapping)) { |
a60637c85 mm: lockless page... |
1446 |
unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
1447 |
put_page(page); |
a60637c85 mm: lockless page... |
1448 |
goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1449 |
} |
83929372f filemap: prepare ... |
1450 |
VM_BUG_ON_PAGE(page_to_pgoff(page) != offset, page); |
1da177e4c Linux-2.6.12-rc2 |
1451 |
} |
1da177e4c Linux-2.6.12-rc2 |
1452 1453 |
return page; } |
0cd6144aa mm + fs: prepare ... |
1454 1455 1456 |
EXPORT_SYMBOL(find_lock_entry); /** |
2457aec63 mm: non-atomicall... |
1457 |
* pagecache_get_page - find and get a page reference |
0cd6144aa mm + fs: prepare ... |
1458 1459 |
* @mapping: the address_space to search * @offset: the page index |
2457aec63 mm: non-atomicall... |
1460 |
* @fgp_flags: PCG flags |
45f87de57 mm: get rid of ra... |
1461 |
* @gfp_mask: gfp mask to use for the page cache data page allocation |
0cd6144aa mm + fs: prepare ... |
1462 |
* |
2457aec63 mm: non-atomicall... |
1463 |
* Looks up the page cache slot at @mapping & @offset. |
1da177e4c Linux-2.6.12-rc2 |
1464 |
* |
75325189c mm: fix filemap.c... |
1465 |
* PCG flags modify how the page is returned. |
0cd6144aa mm + fs: prepare ... |
1466 |
* |
0e056eb55 kernel-api.rst: f... |
1467 1468 1469 1470 1471 1472 1473 1474 |
* @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 * refcount. Otherwise, NULL is returned. |
1da177e4c Linux-2.6.12-rc2 |
1475 |
* |
2457aec63 mm: non-atomicall... |
1476 1477 |
* 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 |
1478 |
* |
2457aec63 mm: non-atomicall... |
1479 |
* If there is a page cache page, it is returned with an increased refcount. |
1da177e4c Linux-2.6.12-rc2 |
1480 |
*/ |
2457aec63 mm: non-atomicall... |
1481 |
struct page *pagecache_get_page(struct address_space *mapping, pgoff_t offset, |
45f87de57 mm: get rid of ra... |
1482 |
int fgp_flags, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
1483 |
{ |
eb2be1893 mm: buffered writ... |
1484 |
struct page *page; |
2457aec63 mm: non-atomicall... |
1485 |
|
1da177e4c Linux-2.6.12-rc2 |
1486 |
repeat: |
2457aec63 mm: non-atomicall... |
1487 1488 1489 1490 1491 1492 1493 1494 1495 |
page = find_get_entry(mapping, offset); if (radix_tree_exceptional_entry(page)) page = NULL; if (!page) goto no_page; if (fgp_flags & FGP_LOCK) { if (fgp_flags & FGP_NOWAIT) { if (!trylock_page(page)) { |
09cbfeaf1 mm, fs: get rid o... |
1496 |
put_page(page); |
2457aec63 mm: non-atomicall... |
1497 1498 1499 1500 1501 1502 1503 1504 1505 |
return NULL; } } else { lock_page(page); } /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
1506 |
put_page(page); |
2457aec63 mm: non-atomicall... |
1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 |
goto repeat; } VM_BUG_ON_PAGE(page->index != offset, page); } if (page && (fgp_flags & FGP_ACCESSED)) mark_page_accessed(page); no_page: if (!page && (fgp_flags & FGP_CREAT)) { int err; if ((fgp_flags & FGP_WRITE) && mapping_cap_account_dirty(mapping)) |
45f87de57 mm: get rid of ra... |
1519 1520 1521 |
gfp_mask |= __GFP_WRITE; if (fgp_flags & FGP_NOFS) gfp_mask &= ~__GFP_FS; |
2457aec63 mm: non-atomicall... |
1522 |
|
45f87de57 mm: get rid of ra... |
1523 |
page = __page_cache_alloc(gfp_mask); |
eb2be1893 mm: buffered writ... |
1524 1525 |
if (!page) return NULL; |
2457aec63 mm: non-atomicall... |
1526 1527 1528 |
if (WARN_ON_ONCE(!(fgp_flags & FGP_LOCK))) fgp_flags |= FGP_LOCK; |
eb39d618f mm: replace init_... |
1529 |
/* Init accessed so avoid atomic mark_page_accessed later */ |
2457aec63 mm: non-atomicall... |
1530 |
if (fgp_flags & FGP_ACCESSED) |
eb39d618f mm: replace init_... |
1531 |
__SetPageReferenced(page); |
2457aec63 mm: non-atomicall... |
1532 |
|
abc1be13f mm/filemap.c: fix... |
1533 |
err = add_to_page_cache_lru(page, mapping, offset, gfp_mask); |
eb2be1893 mm: buffered writ... |
1534 |
if (unlikely(err)) { |
09cbfeaf1 mm, fs: get rid o... |
1535 |
put_page(page); |
eb2be1893 mm: buffered writ... |
1536 1537 1538 |
page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1539 |
} |
1da177e4c Linux-2.6.12-rc2 |
1540 |
} |
2457aec63 mm: non-atomicall... |
1541 |
|
1da177e4c Linux-2.6.12-rc2 |
1542 1543 |
return page; } |
2457aec63 mm: non-atomicall... |
1544 |
EXPORT_SYMBOL(pagecache_get_page); |
1da177e4c Linux-2.6.12-rc2 |
1545 1546 |
/** |
0cd6144aa mm + fs: prepare ... |
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 |
* 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... |
1563 1564 |
* Any shadow entries of evicted pages, or swap entries from * shmem/tmpfs, are included in the returned array. |
0cd6144aa mm + fs: prepare ... |
1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 |
* * find_get_entries() returns the number of pages and shadow entries * which were found. */ unsigned find_get_entries(struct address_space *mapping, pgoff_t start, unsigned int nr_entries, struct page **entries, pgoff_t *indices) { void **slot; unsigned int ret = 0; struct radix_tree_iter iter; if (!nr_entries) return 0; rcu_read_lock(); |
b93b01631 page cache: use x... |
1581 |
radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { |
83929372f filemap: prepare ... |
1582 |
struct page *head, *page; |
0cd6144aa mm + fs: prepare ... |
1583 1584 1585 1586 1587 |
repeat: page = radix_tree_deref_slot(slot); if (unlikely(!page)) continue; if (radix_tree_exception(page)) { |
2cf938aae mm: use radix_tre... |
1588 1589 1590 1591 |
if (radix_tree_deref_retry(page)) { slot = radix_tree_iter_retry(&iter); continue; } |
0cd6144aa mm + fs: prepare ... |
1592 |
/* |
f9fe48bec dax: support dirt... |
1593 1594 1595 |
* 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. |
0cd6144aa mm + fs: prepare ... |
1596 1597 1598 |
*/ goto export; } |
83929372f filemap: prepare ... |
1599 1600 1601 1602 1603 1604 1605 1606 |
head = compound_head(page); if (!page_cache_get_speculative(head)) goto repeat; /* The page was split under us? */ if (compound_head(page) != head) { put_page(head); |
0cd6144aa mm + fs: prepare ... |
1607 |
goto repeat; |
83929372f filemap: prepare ... |
1608 |
} |
0cd6144aa mm + fs: prepare ... |
1609 1610 1611 |
/* Has the page moved? */ if (unlikely(page != *slot)) { |
83929372f filemap: prepare ... |
1612 |
put_page(head); |
0cd6144aa mm + fs: prepare ... |
1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 |
goto repeat; } export: indices[ret] = iter.index; entries[ret] = page; if (++ret == nr_entries) break; } rcu_read_unlock(); return ret; } /** |
b947cee4b mm: implement fin... |
1626 |
* find_get_pages_range - gang pagecache lookup |
1da177e4c Linux-2.6.12-rc2 |
1627 1628 |
* @mapping: The address_space to search * @start: The starting page index |
b947cee4b mm: implement fin... |
1629 |
* @end: The final page index (inclusive) |
1da177e4c Linux-2.6.12-rc2 |
1630 1631 1632 |
* @nr_pages: The maximum number of pages * @pages: Where the resulting pages are placed * |
b947cee4b mm: implement fin... |
1633 1634 1635 1636 |
* 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 |
1637 1638 1639 |
* * 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_... |
1640 |
* We also update @start to index the next page for the traversal. |
1da177e4c Linux-2.6.12-rc2 |
1641 |
* |
b947cee4b mm: implement fin... |
1642 1643 1644 |
* find_get_pages_range() returns the number of pages which were found. If this * number is smaller than @nr_pages, the end of specified range has been * reached. |
1da177e4c Linux-2.6.12-rc2 |
1645 |
*/ |
b947cee4b mm: implement fin... |
1646 1647 1648 |
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 |
1649 |
{ |
0fc9d1040 radix-tree: use i... |
1650 1651 1652 1653 1654 1655 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1656 1657 |
rcu_read_lock(); |
b93b01631 page cache: use x... |
1658 |
radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, *start) { |
83929372f filemap: prepare ... |
1659 |
struct page *head, *page; |
b947cee4b mm: implement fin... |
1660 1661 1662 |
if (iter.index > end) break; |
a60637c85 mm: lockless page... |
1663 |
repeat: |
0fc9d1040 radix-tree: use i... |
1664 |
page = radix_tree_deref_slot(slot); |
a60637c85 mm: lockless page... |
1665 1666 |
if (unlikely(!page)) continue; |
9d8aa4ea8 mm: remove worryi... |
1667 |
|
a2c16d6cb mm: let swap use ... |
1668 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1669 |
if (radix_tree_deref_retry(page)) { |
2cf938aae mm: use radix_tre... |
1670 1671 |
slot = radix_tree_iter_retry(&iter); continue; |
8079b1c85 mm: clarify the r... |
1672 |
} |
a2c16d6cb mm: let swap use ... |
1673 |
/* |
139b6a6fb mm: filemap: upda... |
1674 1675 1676 |
* A shadow entry of a recently evicted page, * or a swap entry from shmem/tmpfs. Skip * over it. |
a2c16d6cb mm: let swap use ... |
1677 |
*/ |
8079b1c85 mm: clarify the r... |
1678 |
continue; |
27d20fddc radix-tree: fix R... |
1679 |
} |
a60637c85 mm: lockless page... |
1680 |
|
83929372f filemap: prepare ... |
1681 1682 1683 1684 1685 1686 1687 |
head = compound_head(page); if (!page_cache_get_speculative(head)) goto repeat; /* The page was split under us? */ if (compound_head(page) != head) { put_page(head); |
a60637c85 mm: lockless page... |
1688 |
goto repeat; |
83929372f filemap: prepare ... |
1689 |
} |
a60637c85 mm: lockless page... |
1690 1691 |
/* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1692 |
if (unlikely(page != *slot)) { |
83929372f filemap: prepare ... |
1693 |
put_page(head); |
a60637c85 mm: lockless page... |
1694 1695 |
goto repeat; } |
1da177e4c Linux-2.6.12-rc2 |
1696 |
|
a60637c85 mm: lockless page... |
1697 |
pages[ret] = page; |
b947cee4b mm: implement fin... |
1698 1699 1700 1701 |
if (++ret == nr_pages) { *start = pages[ret - 1]->index + 1; goto out; } |
a60637c85 mm: lockless page... |
1702 |
} |
5b280c0cc mm: don't return ... |
1703 |
|
b947cee4b mm: implement fin... |
1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 |
/* * 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 * breaks the iteration when there is page at index -1 but that is * already broken anyway. */ if (end == (pgoff_t)-1) *start = (pgoff_t)-1; else *start = end + 1; out: |
a60637c85 mm: lockless page... |
1715 |
rcu_read_unlock(); |
d72dc8a25 mm: make pagevec_... |
1716 |
|
1da177e4c Linux-2.6.12-rc2 |
1717 1718 |
return ret; } |
ebf43500e [PATCH] Add find_... |
1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 |
/** * find_get_pages_contig - gang contiguous pagecache lookup * @mapping: The address_space to search * @index: The starting page index * @nr_pages: The maximum number of pages * @pages: Where the resulting pages are placed * * find_get_pages_contig() works exactly like find_get_pages(), except * that the returned number of pages are guaranteed to be contiguous. * * find_get_pages_contig() returns the number of pages which were found. */ unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index, unsigned int nr_pages, struct page **pages) { |
0fc9d1040 radix-tree: use i... |
1734 1735 1736 1737 1738 1739 |
struct radix_tree_iter iter; void **slot; unsigned int ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1740 1741 |
rcu_read_lock(); |
b93b01631 page cache: use x... |
1742 |
radix_tree_for_each_contig(slot, &mapping->i_pages, &iter, index) { |
83929372f filemap: prepare ... |
1743 |
struct page *head, *page; |
a60637c85 mm: lockless page... |
1744 |
repeat: |
0fc9d1040 radix-tree: use i... |
1745 1746 |
page = radix_tree_deref_slot(slot); /* The hole, there no reason to continue */ |
a60637c85 mm: lockless page... |
1747 |
if (unlikely(!page)) |
0fc9d1040 radix-tree: use i... |
1748 |
break; |
9d8aa4ea8 mm: remove worryi... |
1749 |
|
a2c16d6cb mm: let swap use ... |
1750 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1751 |
if (radix_tree_deref_retry(page)) { |
2cf938aae mm: use radix_tre... |
1752 1753 |
slot = radix_tree_iter_retry(&iter); continue; |
8079b1c85 mm: clarify the r... |
1754 |
} |
a2c16d6cb mm: let swap use ... |
1755 |
/* |
139b6a6fb mm: filemap: upda... |
1756 1757 1758 |
* A shadow entry of a recently evicted page, * or a swap entry from shmem/tmpfs. Stop * looking for contiguous pages. |
a2c16d6cb mm: let swap use ... |
1759 |
*/ |
8079b1c85 mm: clarify the r... |
1760 |
break; |
a2c16d6cb mm: let swap use ... |
1761 |
} |
ebf43500e [PATCH] Add find_... |
1762 |
|
83929372f filemap: prepare ... |
1763 1764 1765 1766 1767 1768 1769 |
head = compound_head(page); if (!page_cache_get_speculative(head)) goto repeat; /* The page was split under us? */ if (compound_head(page) != head) { put_page(head); |
a60637c85 mm: lockless page... |
1770 |
goto repeat; |
83929372f filemap: prepare ... |
1771 |
} |
a60637c85 mm: lockless page... |
1772 1773 |
/* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1774 |
if (unlikely(page != *slot)) { |
83929372f filemap: prepare ... |
1775 |
put_page(head); |
a60637c85 mm: lockless page... |
1776 1777 |
goto repeat; } |
9cbb4cb21 mm: find_get_page... |
1778 1779 1780 1781 1782 |
/* * must check mapping and index after taking the ref. * otherwise we can get both false positives and false * negatives, which is just confusing to the caller. */ |
83929372f filemap: prepare ... |
1783 |
if (page->mapping == NULL || page_to_pgoff(page) != iter.index) { |
09cbfeaf1 mm, fs: get rid o... |
1784 |
put_page(page); |
9cbb4cb21 mm: find_get_page... |
1785 1786 |
break; } |
a60637c85 mm: lockless page... |
1787 |
pages[ret] = page; |
0fc9d1040 radix-tree: use i... |
1788 1789 |
if (++ret == nr_pages) break; |
ebf43500e [PATCH] Add find_... |
1790 |
} |
a60637c85 mm: lockless page... |
1791 1792 |
rcu_read_unlock(); return ret; |
ebf43500e [PATCH] Add find_... |
1793 |
} |
ef71c15c4 AFS: export a cou... |
1794 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e [PATCH] Add find_... |
1795 |
|
485bb99b4 [PATCH] kernel-do... |
1796 |
/** |
72b045aec mm: implement fin... |
1797 |
* find_get_pages_range_tag - find and return pages in given range matching @tag |
485bb99b4 [PATCH] kernel-do... |
1798 1799 |
* @mapping: the address_space to search * @index: the starting page index |
72b045aec mm: implement fin... |
1800 |
* @end: The final page index (inclusive) |
485bb99b4 [PATCH] kernel-do... |
1801 1802 1803 1804 |
* @tag: the tag index * @nr_pages: the maximum number of pages * @pages: where the resulting pages are placed * |
1da177e4c Linux-2.6.12-rc2 |
1805 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4 [PATCH] kernel-do... |
1806 |
* @tag. We update @index to index the next page for the traversal. |
1da177e4c Linux-2.6.12-rc2 |
1807 |
*/ |
72b045aec mm: implement fin... |
1808 1809 1810 |
unsigned find_get_pages_range_tag(struct address_space *mapping, pgoff_t *index, pgoff_t end, int tag, unsigned int nr_pages, struct page **pages) |
1da177e4c Linux-2.6.12-rc2 |
1811 |
{ |
0fc9d1040 radix-tree: use i... |
1812 1813 1814 1815 1816 1817 |
struct radix_tree_iter iter; void **slot; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; |
a60637c85 mm: lockless page... |
1818 1819 |
rcu_read_lock(); |
b93b01631 page cache: use x... |
1820 |
radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, *index, tag) { |
83929372f filemap: prepare ... |
1821 |
struct page *head, *page; |
72b045aec mm: implement fin... |
1822 1823 1824 |
if (iter.index > end) break; |
a60637c85 mm: lockless page... |
1825 |
repeat: |
0fc9d1040 radix-tree: use i... |
1826 |
page = radix_tree_deref_slot(slot); |
a60637c85 mm: lockless page... |
1827 1828 |
if (unlikely(!page)) continue; |
9d8aa4ea8 mm: remove worryi... |
1829 |
|
a2c16d6cb mm: let swap use ... |
1830 |
if (radix_tree_exception(page)) { |
8079b1c85 mm: clarify the r... |
1831 |
if (radix_tree_deref_retry(page)) { |
2cf938aae mm: use radix_tre... |
1832 1833 |
slot = radix_tree_iter_retry(&iter); continue; |
8079b1c85 mm: clarify the r... |
1834 |
} |
a2c16d6cb mm: let swap use ... |
1835 |
/* |
139b6a6fb mm: filemap: upda... |
1836 1837 1838 1839 1840 1841 1842 1843 1844 |
* A shadow entry of a recently evicted page. * * Those entries should never be tagged, but * this tree walk is lockless and the tags are * looked up in bulk, one radix tree node at a * time, so there is a sizable window for page * reclaim to evict a page we saw tagged. * * Skip over it. |
a2c16d6cb mm: let swap use ... |
1845 |
*/ |
139b6a6fb mm: filemap: upda... |
1846 |
continue; |
a2c16d6cb mm: let swap use ... |
1847 |
} |
a60637c85 mm: lockless page... |
1848 |
|
83929372f filemap: prepare ... |
1849 1850 |
head = compound_head(page); if (!page_cache_get_speculative(head)) |
a60637c85 mm: lockless page... |
1851 |
goto repeat; |
83929372f filemap: prepare ... |
1852 1853 1854 1855 1856 |
/* The page was split under us? */ if (compound_head(page) != head) { put_page(head); goto repeat; } |
a60637c85 mm: lockless page... |
1857 |
/* Has the page moved? */ |
0fc9d1040 radix-tree: use i... |
1858 |
if (unlikely(page != *slot)) { |
83929372f filemap: prepare ... |
1859 |
put_page(head); |
a60637c85 mm: lockless page... |
1860 1861 1862 1863 |
goto repeat; } pages[ret] = page; |
72b045aec mm: implement fin... |
1864 1865 1866 1867 |
if (++ret == nr_pages) { *index = pages[ret - 1]->index + 1; goto out; } |
a60637c85 mm: lockless page... |
1868 |
} |
5b280c0cc mm: don't return ... |
1869 |
|
72b045aec mm: implement fin... |
1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 |
/* * We come here when we got at @end. We take care to not overflow the * index @index as it confuses some of the callers. This breaks the * iteration when there is page at index -1 but that is already broken * anyway. */ if (end == (pgoff_t)-1) *index = (pgoff_t)-1; else *index = end + 1; out: |
a60637c85 mm: lockless page... |
1881 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
1882 |
|
1da177e4c Linux-2.6.12-rc2 |
1883 1884 |
return ret; } |
72b045aec mm: implement fin... |
1885 |
EXPORT_SYMBOL(find_get_pages_range_tag); |
1da177e4c Linux-2.6.12-rc2 |
1886 |
|
7e7f77498 mm: add find_get_... |
1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 |
/** * find_get_entries_tag - find and return entries that match @tag * @mapping: the address_space to search * @start: the starting page cache index * @tag: the tag 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 * * Like find_get_entries, except we only return entries which are tagged with * @tag. */ unsigned find_get_entries_tag(struct address_space *mapping, pgoff_t start, int tag, unsigned int nr_entries, struct page **entries, pgoff_t *indices) { void **slot; unsigned int ret = 0; struct radix_tree_iter iter; if (!nr_entries) return 0; rcu_read_lock(); |
b93b01631 page cache: use x... |
1911 |
radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, start, tag) { |
83929372f filemap: prepare ... |
1912 |
struct page *head, *page; |
7e7f77498 mm: add find_get_... |
1913 1914 1915 1916 1917 1918 |
repeat: page = radix_tree_deref_slot(slot); if (unlikely(!page)) continue; if (radix_tree_exception(page)) { if (radix_tree_deref_retry(page)) { |
2cf938aae mm: use radix_tre... |
1919 1920 |
slot = radix_tree_iter_retry(&iter); continue; |
7e7f77498 mm: add find_get_... |
1921 1922 1923 1924 1925 1926 1927 1928 1929 |
} /* * 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. */ goto export; } |
83929372f filemap: prepare ... |
1930 1931 1932 |
head = compound_head(page); if (!page_cache_get_speculative(head)) |
7e7f77498 mm: add find_get_... |
1933 |
goto repeat; |
83929372f filemap: prepare ... |
1934 1935 1936 1937 1938 |
/* The page was split under us? */ if (compound_head(page) != head) { put_page(head); goto repeat; } |
7e7f77498 mm: add find_get_... |
1939 1940 |
/* Has the page moved? */ if (unlikely(page != *slot)) { |
83929372f filemap: prepare ... |
1941 |
put_page(head); |
7e7f77498 mm: add find_get_... |
1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 |
goto repeat; } export: indices[ret] = iter.index; entries[ret] = page; if (++ret == nr_entries) break; } rcu_read_unlock(); return ret; } EXPORT_SYMBOL(find_get_entries_tag); |
76d42bd96 [PATCH] readahead... |
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 |
/* * 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... |
1972 |
ra->ra_pages /= 4; |
76d42bd96 [PATCH] readahead... |
1973 |
} |
485bb99b4 [PATCH] kernel-do... |
1974 |
/** |
47c27bc46 fs: pass iocb to ... |
1975 1976 |
* generic_file_buffered_read - generic file read routine * @iocb: the iocb to read |
6e58e79db introduce copy_pa... |
1977 1978 |
* @iter: data destination * @written: already copied |
485bb99b4 [PATCH] kernel-do... |
1979 |
* |
1da177e4c Linux-2.6.12-rc2 |
1980 |
* This is a generic file read routine, and uses the |
485bb99b4 [PATCH] kernel-do... |
1981 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c Linux-2.6.12-rc2 |
1982 1983 1984 |
* * This is really ugly. But the goto's actually try to clarify some * of the logic when it comes to error handling etc. |
1da177e4c Linux-2.6.12-rc2 |
1985 |
*/ |
47c27bc46 fs: pass iocb to ... |
1986 |
static ssize_t generic_file_buffered_read(struct kiocb *iocb, |
6e58e79db introduce copy_pa... |
1987 |
struct iov_iter *iter, ssize_t written) |
1da177e4c Linux-2.6.12-rc2 |
1988 |
{ |
47c27bc46 fs: pass iocb to ... |
1989 |
struct file *filp = iocb->ki_filp; |
36e789144 kill do_generic_m... |
1990 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
1991 |
struct inode *inode = mapping->host; |
36e789144 kill do_generic_m... |
1992 |
struct file_ra_state *ra = &filp->f_ra; |
47c27bc46 fs: pass iocb to ... |
1993 |
loff_t *ppos = &iocb->ki_pos; |
57f6b96c0 filemap: convert ... |
1994 1995 1996 1997 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372 readahead: improv... |
1998 |
unsigned int prev_offset; |
6e58e79db introduce copy_pa... |
1999 |
int error = 0; |
1da177e4c Linux-2.6.12-rc2 |
2000 |
|
c2a9737f4 vfs,mm: fix a dea... |
2001 |
if (unlikely(*ppos >= inode->i_sb->s_maxbytes)) |
d05c5f7ba vfs,mm: fix retur... |
2002 |
return 0; |
c2a9737f4 vfs,mm: fix a dea... |
2003 |
iov_iter_truncate(iter, inode->i_sb->s_maxbytes); |
09cbfeaf1 mm, fs: get rid o... |
2004 2005 2006 2007 2008 |
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 |
2009 |
|
1da177e4c Linux-2.6.12-rc2 |
2010 2011 |
for (;;) { struct page *page; |
57f6b96c0 filemap: convert ... |
2012 |
pgoff_t end_index; |
a32ea1e1f Fix read/truncate... |
2013 |
loff_t isize; |
1da177e4c Linux-2.6.12-rc2 |
2014 |
unsigned long nr, ret; |
1da177e4c Linux-2.6.12-rc2 |
2015 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
2016 |
find_page: |
5abf186a3 mm, fs: check for... |
2017 2018 2019 2020 |
if (fatal_signal_pending(current)) { error = -EINTR; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
2021 |
page = find_get_page(mapping, index); |
3ea89ee86 readahead: conver... |
2022 |
if (!page) { |
3239d8348 fs: support IOCB_... |
2023 2024 |
if (iocb->ki_flags & IOCB_NOWAIT) goto would_block; |
cf914a7d6 readahead: split ... |
2025 |
page_cache_sync_readahead(mapping, |
7ff81078d readahead: remove... |
2026 |
ra, filp, |
3ea89ee86 readahead: conver... |
2027 2028 2029 2030 2031 2032 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6 readahead: split ... |
2033 |
page_cache_async_readahead(mapping, |
7ff81078d readahead: remove... |
2034 |
ra, filp, page, |
3ea89ee86 readahead: conver... |
2035 |
index, last_index - index); |
1da177e4c Linux-2.6.12-rc2 |
2036 |
} |
8ab22b9ab vfs: pagecache us... |
2037 |
if (!PageUptodate(page)) { |
3239d8348 fs: support IOCB_... |
2038 2039 2040 2041 |
if (iocb->ki_flags & IOCB_NOWAIT) { put_page(page); goto would_block; } |
ebded0278 mm: filemap: avoi... |
2042 2043 2044 2045 2046 |
/* * 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... |
2047 2048 2049 |
error = wait_on_page_locked_killable(page); if (unlikely(error)) goto readpage_error; |
ebded0278 mm: filemap: avoi... |
2050 2051 |
if (PageUptodate(page)) goto page_ok; |
09cbfeaf1 mm, fs: get rid o... |
2052 |
if (inode->i_blkbits == PAGE_SHIFT || |
8ab22b9ab vfs: pagecache us... |
2053 2054 |
!mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; |
6d6d36bc6 mm/filemap: don't... |
2055 2056 2057 |
/* pipes can't handle partially uptodate pages */ if (unlikely(iter->type & ITER_PIPE)) goto page_not_up_to_date; |
529ae9aaa mm: rename page t... |
2058 |
if (!trylock_page(page)) |
8ab22b9ab vfs: pagecache us... |
2059 |
goto page_not_up_to_date; |
8d056cb96 mm/vfs: revalidat... |
2060 2061 2062 |
/* Did it get truncated before we got the lock? */ if (!page->mapping) goto page_not_up_to_date_locked; |
8ab22b9ab vfs: pagecache us... |
2063 |
if (!mapping->a_ops->is_partially_uptodate(page, |
6e58e79db introduce copy_pa... |
2064 |
offset, iter->count)) |
8ab22b9ab vfs: pagecache us... |
2065 2066 2067 |
goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
2068 |
page_ok: |
a32ea1e1f Fix read/truncate... |
2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 |
/* * 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... |
2079 |
end_index = (isize - 1) >> PAGE_SHIFT; |
a32ea1e1f Fix read/truncate... |
2080 |
if (unlikely(!isize || index > end_index)) { |
09cbfeaf1 mm, fs: get rid o... |
2081 |
put_page(page); |
a32ea1e1f Fix read/truncate... |
2082 2083 2084 2085 |
goto out; } /* nr is the maximum number of bytes to copy from this page */ |
09cbfeaf1 mm, fs: get rid o... |
2086 |
nr = PAGE_SIZE; |
a32ea1e1f Fix read/truncate... |
2087 |
if (index == end_index) { |
09cbfeaf1 mm, fs: get rid o... |
2088 |
nr = ((isize - 1) & ~PAGE_MASK) + 1; |
a32ea1e1f Fix read/truncate... |
2089 |
if (nr <= offset) { |
09cbfeaf1 mm, fs: get rid o... |
2090 |
put_page(page); |
a32ea1e1f Fix read/truncate... |
2091 2092 2093 2094 |
goto out; } } nr = nr - offset; |
1da177e4c Linux-2.6.12-rc2 |
2095 2096 2097 2098 2099 2100 2101 2102 2103 |
/* 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... |
2104 2105 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c Linux-2.6.12-rc2 |
2106 |
*/ |
ec0f16372 readahead: improv... |
2107 |
if (prev_index != index || offset != prev_offset) |
1da177e4c Linux-2.6.12-rc2 |
2108 2109 2110 2111 2112 2113 |
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 |
2114 |
*/ |
6e58e79db introduce copy_pa... |
2115 2116 |
ret = copy_page_to_iter(page, offset, nr, iter); |
1da177e4c Linux-2.6.12-rc2 |
2117 |
offset += ret; |
09cbfeaf1 mm, fs: get rid o... |
2118 2119 |
index += offset >> PAGE_SHIFT; offset &= ~PAGE_MASK; |
6ce745ed3 readahead: code c... |
2120 |
prev_offset = offset; |
1da177e4c Linux-2.6.12-rc2 |
2121 |
|
09cbfeaf1 mm, fs: get rid o... |
2122 |
put_page(page); |
6e58e79db introduce copy_pa... |
2123 2124 2125 2126 2127 2128 2129 2130 |
written += ret; if (!iov_iter_count(iter)) goto out; if (ret < nr) { error = -EFAULT; goto out; } continue; |
1da177e4c Linux-2.6.12-rc2 |
2131 2132 2133 |
page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235 do_generic_file_r... |
2134 2135 2136 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
2137 |
|
8ab22b9ab vfs: pagecache us... |
2138 |
page_not_up_to_date_locked: |
da6052f7b [PATCH] update so... |
2139 |
/* Did it get truncated before we got the lock? */ |
1da177e4c Linux-2.6.12-rc2 |
2140 2141 |
if (!page->mapping) { unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2142 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 |
continue; } /* Did somebody else fill it already? */ if (PageUptodate(page)) { unlock_page(page); goto page_ok; } readpage: |
91803b499 do_generic_file_r... |
2153 2154 2155 2156 2157 2158 |
/* * 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 |
2159 2160 |
/* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7 [PATCH] add AOP_T... |
2161 2162 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { |
09cbfeaf1 mm, fs: get rid o... |
2163 |
put_page(page); |
6e58e79db introduce copy_pa... |
2164 |
error = 0; |
994fc28c7 [PATCH] add AOP_T... |
2165 2166 |
goto find_page; } |
1da177e4c Linux-2.6.12-rc2 |
2167 |
goto readpage_error; |
994fc28c7 [PATCH] add AOP_T... |
2168 |
} |
1da177e4c Linux-2.6.12-rc2 |
2169 2170 |
if (!PageUptodate(page)) { |
854623235 do_generic_file_r... |
2171 2172 2173 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
2174 2175 2176 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* |
2ecdc82ef kill unused inval... |
2177 |
* invalidate_mapping_pages got it |
1da177e4c Linux-2.6.12-rc2 |
2178 2179 |
*/ unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2180 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2181 2182 2183 |
goto find_page; } unlock_page(page); |
7ff81078d readahead: remove... |
2184 |
shrink_readahead_size_eio(filp, ra); |
854623235 do_generic_file_r... |
2185 2186 |
error = -EIO; goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
2187 2188 2189 |
} unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
2190 2191 2192 2193 |
goto page_ok; readpage_error: /* UHHUH! A synchronous read error occurred. Report it */ |
09cbfeaf1 mm, fs: get rid o... |
2194 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2195 2196 2197 2198 2199 2200 2201 |
goto out; no_cached_page: /* * Ok, it wasn't cached, so we need to create a new * page.. */ |
453f85d43 mm: remove __GFP_... |
2202 |
page = page_cache_alloc(mapping); |
eb2be1893 mm: buffered writ... |
2203 |
if (!page) { |
6e58e79db introduce copy_pa... |
2204 |
error = -ENOMEM; |
eb2be1893 mm: buffered writ... |
2205 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
2206 |
} |
6afdb859b mm: do not ignore... |
2207 |
error = add_to_page_cache_lru(page, mapping, index, |
c62d25556 mm, fs: introduce... |
2208 |
mapping_gfp_constraint(mapping, GFP_KERNEL)); |
1da177e4c Linux-2.6.12-rc2 |
2209 |
if (error) { |
09cbfeaf1 mm, fs: get rid o... |
2210 |
put_page(page); |
6e58e79db introduce copy_pa... |
2211 2212 |
if (error == -EEXIST) { error = 0; |
1da177e4c Linux-2.6.12-rc2 |
2213 |
goto find_page; |
6e58e79db introduce copy_pa... |
2214 |
} |
1da177e4c Linux-2.6.12-rc2 |
2215 2216 |
goto out; } |
1da177e4c Linux-2.6.12-rc2 |
2217 2218 |
goto readpage; } |
3239d8348 fs: support IOCB_... |
2219 2220 |
would_block: error = -EAGAIN; |
1da177e4c Linux-2.6.12-rc2 |
2221 |
out: |
7ff81078d readahead: remove... |
2222 |
ra->prev_pos = prev_index; |
09cbfeaf1 mm, fs: get rid o... |
2223 |
ra->prev_pos <<= PAGE_SHIFT; |
7ff81078d readahead: remove... |
2224 |
ra->prev_pos |= prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
2225 |
|
09cbfeaf1 mm, fs: get rid o... |
2226 |
*ppos = ((loff_t)index << PAGE_SHIFT) + offset; |
0c6aa2639 mm: do_generic_fi... |
2227 |
file_accessed(filp); |
6e58e79db introduce copy_pa... |
2228 |
return written ? written : error; |
1da177e4c Linux-2.6.12-rc2 |
2229 |
} |
485bb99b4 [PATCH] kernel-do... |
2230 |
/** |
6abd23227 bury generic_file... |
2231 |
* generic_file_read_iter - generic filesystem read routine |
485bb99b4 [PATCH] kernel-do... |
2232 |
* @iocb: kernel I/O control block |
6abd23227 bury generic_file... |
2233 |
* @iter: destination for the data read |
485bb99b4 [PATCH] kernel-do... |
2234 |
* |
6abd23227 bury generic_file... |
2235 |
* This is the "read_iter()" routine for all filesystems |
1da177e4c Linux-2.6.12-rc2 |
2236 2237 2238 |
* that can use the page cache directly. */ ssize_t |
ed978a811 new helper: gener... |
2239 |
generic_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) |
1da177e4c Linux-2.6.12-rc2 |
2240 |
{ |
e7080a439 mm/filemap: gener... |
2241 |
size_t count = iov_iter_count(iter); |
47c27bc46 fs: pass iocb to ... |
2242 |
ssize_t retval = 0; |
e7080a439 mm/filemap: gener... |
2243 2244 2245 |
if (!count) goto out; /* skip atime */ |
1da177e4c Linux-2.6.12-rc2 |
2246 |
|
2ba48ce51 mirror O_APPEND a... |
2247 |
if (iocb->ki_flags & IOCB_DIRECT) { |
47c27bc46 fs: pass iocb to ... |
2248 |
struct file *file = iocb->ki_filp; |
ed978a811 new helper: gener... |
2249 2250 |
struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; |
543ade1fc [PATCH] Streamlin... |
2251 |
loff_t size; |
1da177e4c Linux-2.6.12-rc2 |
2252 |
|
1da177e4c Linux-2.6.12-rc2 |
2253 |
size = i_size_read(inode); |
6be96d3ad fs: return if dir... |
2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 |
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 ... |
2265 |
|
0d5b0cf24 fs: update atime ... |
2266 |
file_accessed(file); |
5ecda1371 generic_file_read... |
2267 |
retval = mapping->a_ops->direct_IO(iocb, iter); |
c3a690240 fix ITER_PIPE int... |
2268 |
if (retval >= 0) { |
c64fb5c74 filemap: remove p... |
2269 |
iocb->ki_pos += retval; |
5ecda1371 generic_file_read... |
2270 |
count -= retval; |
9fe55eea7 Fix race when che... |
2271 |
} |
5b47d59af fix braino in gen... |
2272 |
iov_iter_revert(iter, count - iov_iter_count(iter)); |
66f998f61 fs: allow short d... |
2273 |
|
9fe55eea7 Fix race when che... |
2274 2275 2276 2277 2278 2279 |
/* * 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... |
2280 2281 |
* the rest of the read. Buffered reads will not work for * DAX files, so don't bother trying. |
9fe55eea7 Fix race when che... |
2282 |
*/ |
5ecda1371 generic_file_read... |
2283 |
if (retval < 0 || !count || iocb->ki_pos >= size || |
0d5b0cf24 fs: update atime ... |
2284 |
IS_DAX(inode)) |
9fe55eea7 Fix race when che... |
2285 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
2286 |
} |
47c27bc46 fs: pass iocb to ... |
2287 |
retval = generic_file_buffered_read(iocb, iter, retval); |
1da177e4c Linux-2.6.12-rc2 |
2288 2289 2290 |
out: return retval; } |
ed978a811 new helper: gener... |
2291 |
EXPORT_SYMBOL(generic_file_read_iter); |
1da177e4c Linux-2.6.12-rc2 |
2292 |
|
1da177e4c Linux-2.6.12-rc2 |
2293 |
#ifdef CONFIG_MMU |
485bb99b4 [PATCH] kernel-do... |
2294 2295 2296 2297 |
/** * page_cache_read - adds requested page to the page cache if not already there * @file: file to read * @offset: page index |
62eb320ab mm: fix filemap.c... |
2298 |
* @gfp_mask: memory allocation flags |
485bb99b4 [PATCH] kernel-do... |
2299 |
* |
1da177e4c Linux-2.6.12-rc2 |
2300 2301 2302 |
* This adds the requested page to the page cache if it isn't already there, * and schedules an I/O to read in its contents from disk. */ |
c20cd45eb mm: allow GFP_{FS... |
2303 |
static int page_cache_read(struct file *file, pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
2304 2305 |
{ struct address_space *mapping = file->f_mapping; |
99dadfdde mm/filemap.c: rem... |
2306 |
struct page *page; |
994fc28c7 [PATCH] add AOP_T... |
2307 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
2308 |
|
994fc28c7 [PATCH] add AOP_T... |
2309 |
do { |
453f85d43 mm: remove __GFP_... |
2310 |
page = __page_cache_alloc(gfp_mask); |
994fc28c7 [PATCH] add AOP_T... |
2311 2312 |
if (!page) return -ENOMEM; |
abc1be13f mm/filemap.c: fix... |
2313 |
ret = add_to_page_cache_lru(page, mapping, offset, gfp_mask); |
994fc28c7 [PATCH] add AOP_T... |
2314 2315 2316 2317 |
if (ret == 0) ret = mapping->a_ops->readpage(file, page); else if (ret == -EEXIST) ret = 0; /* losing race to add is OK */ |
1da177e4c Linux-2.6.12-rc2 |
2318 |
|
09cbfeaf1 mm, fs: get rid o... |
2319 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2320 |
|
994fc28c7 [PATCH] add AOP_T... |
2321 |
} while (ret == AOP_TRUNCATED_PAGE); |
99dadfdde mm/filemap.c: rem... |
2322 |
|
994fc28c7 [PATCH] add AOP_T... |
2323 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
2324 2325 2326 |
} #define MMAP_LOTSAMISS (100) |
ef00e08e2 readahead: clean ... |
2327 2328 2329 2330 2331 2332 2333 2334 2335 |
/* * Synchronous readahead happens when we don't even find * a page in the page cache at all. */ static void do_sync_mmap_readahead(struct vm_area_struct *vma, struct file_ra_state *ra, struct file *file, pgoff_t offset) { |
ef00e08e2 readahead: clean ... |
2336 2337 2338 |
struct address_space *mapping = file->f_mapping; /* If we don't want any read-ahead, don't bother */ |
64363aad5 mm: remove unused... |
2339 |
if (vma->vm_flags & VM_RAND_READ) |
ef00e08e2 readahead: clean ... |
2340 |
return; |
275b12bf5 readahead: return... |
2341 2342 |
if (!ra->ra_pages) return; |
ef00e08e2 readahead: clean ... |
2343 |
|
64363aad5 mm: remove unused... |
2344 |
if (vma->vm_flags & VM_SEQ_READ) { |
7ffc59b4d readahead: enforc... |
2345 2346 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
2347 2348 |
return; } |
207d04baa readahead: reduce... |
2349 2350 |
/* Avoid banging the cache line if not needed */ if (ra->mmap_miss < MMAP_LOTSAMISS * 10) |
ef00e08e2 readahead: clean ... |
2351 2352 2353 2354 2355 2356 2357 2358 |
ra->mmap_miss++; /* * Do we miss much more than hit in this file? If so, * stop bothering with read-ahead. It will only hurt. */ if (ra->mmap_miss > MMAP_LOTSAMISS) return; |
d30a11004 readahead: record... |
2359 2360 2361 |
/* * mmap read-around */ |
600e19afc mm: use only per-... |
2362 2363 2364 |
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... |
2365 |
ra_submit(ra, mapping, file); |
ef00e08e2 readahead: clean ... |
2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 |
} /* * Asynchronous readahead happens when we find the page and PG_readahead, * so we want to possibly extend the readahead further.. */ static void do_async_mmap_readahead(struct vm_area_struct *vma, struct file_ra_state *ra, struct file *file, struct page *page, pgoff_t offset) { struct address_space *mapping = file->f_mapping; /* If we don't want any read-ahead, don't bother */ |
64363aad5 mm: remove unused... |
2381 |
if (vma->vm_flags & VM_RAND_READ) |
ef00e08e2 readahead: clean ... |
2382 2383 2384 2385 |
return; if (ra->mmap_miss > 0) ra->mmap_miss--; if (PageReadahead(page)) |
2fad6f5de readahead: enforc... |
2386 2387 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
2388 |
} |
485bb99b4 [PATCH] kernel-do... |
2389 |
/** |
54cb8821d mm: merge populat... |
2390 |
* filemap_fault - read in file data for page fault handling |
d0217ac04 mm: fault feedbac... |
2391 |
* @vmf: struct vm_fault containing details of the fault |
485bb99b4 [PATCH] kernel-do... |
2392 |
* |
54cb8821d mm: merge populat... |
2393 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c Linux-2.6.12-rc2 |
2394 2395 2396 2397 2398 |
* 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... |
2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 |
* * vma->vm_mm->mmap_sem must be held on entry. * * If our return value has VM_FAULT_RETRY set, it's because * lock_page_or_retry() returned 0. * The mmap_sem has usually been released in this case. * See __lock_page_or_retry() for the exception. * * 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. |
1da177e4c Linux-2.6.12-rc2 |
2411 |
*/ |
2bcd6454b mm: use new retur... |
2412 |
vm_fault_t filemap_fault(struct vm_fault *vmf) |
1da177e4c Linux-2.6.12-rc2 |
2413 2414 |
{ int error; |
11bac8000 mm, fs: reduce fa... |
2415 |
struct file *file = vmf->vma->vm_file; |
1da177e4c Linux-2.6.12-rc2 |
2416 2417 2418 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2 readahead: clean ... |
2419 |
pgoff_t offset = vmf->pgoff; |
9ab2594fe mm: tighten up th... |
2420 |
pgoff_t max_off; |
1da177e4c Linux-2.6.12-rc2 |
2421 |
struct page *page; |
2bcd6454b mm: use new retur... |
2422 |
vm_fault_t ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
2423 |
|
9ab2594fe mm: tighten up th... |
2424 2425 |
max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); if (unlikely(offset >= max_off)) |
5307cc1aa Remove broken ptr... |
2426 |
return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
2427 |
|
1da177e4c Linux-2.6.12-rc2 |
2428 |
/* |
494264208 mm: memcg: handle... |
2429 |
* Do we have something in the page cache already? |
1da177e4c Linux-2.6.12-rc2 |
2430 |
*/ |
ef00e08e2 readahead: clean ... |
2431 |
page = find_get_page(mapping, offset); |
45cac65b0 readahead: fault ... |
2432 |
if (likely(page) && !(vmf->flags & FAULT_FLAG_TRIED)) { |
1da177e4c Linux-2.6.12-rc2 |
2433 |
/* |
ef00e08e2 readahead: clean ... |
2434 2435 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c Linux-2.6.12-rc2 |
2436 |
*/ |
11bac8000 mm, fs: reduce fa... |
2437 |
do_async_mmap_readahead(vmf->vma, ra, file, page, offset); |
45cac65b0 readahead: fault ... |
2438 |
} else if (!page) { |
ef00e08e2 readahead: clean ... |
2439 |
/* No page in the page cache at all */ |
11bac8000 mm, fs: reduce fa... |
2440 |
do_sync_mmap_readahead(vmf->vma, ra, file, offset); |
ef00e08e2 readahead: clean ... |
2441 |
count_vm_event(PGMAJFAULT); |
2262185c5 mm: per-cgroup me... |
2442 |
count_memcg_event_mm(vmf->vma->vm_mm, PGMAJFAULT); |
ef00e08e2 readahead: clean ... |
2443 2444 |
ret = VM_FAULT_MAJOR; retry_find: |
b522c94da mm: filemap_fault... |
2445 |
page = find_get_page(mapping, offset); |
1da177e4c Linux-2.6.12-rc2 |
2446 2447 2448 |
if (!page) goto no_cached_page; } |
11bac8000 mm, fs: reduce fa... |
2449 |
if (!lock_page_or_retry(page, vmf->vma->vm_mm, vmf->flags)) { |
09cbfeaf1 mm, fs: get rid o... |
2450 |
put_page(page); |
d065bd810 mm: retry page fa... |
2451 |
return ret | VM_FAULT_RETRY; |
d88c0922f Release page refe... |
2452 |
} |
b522c94da mm: filemap_fault... |
2453 2454 2455 2456 2457 2458 2459 |
/* Did it get truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); put_page(page); goto retry_find; } |
309381fea mm: dump page whe... |
2460 |
VM_BUG_ON_PAGE(page->index != offset, page); |
b522c94da mm: filemap_fault... |
2461 |
|
1da177e4c Linux-2.6.12-rc2 |
2462 |
/* |
d00806b18 mm: fix fault vs ... |
2463 2464 |
* 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 |
2465 |
*/ |
d00806b18 mm: fix fault vs ... |
2466 |
if (unlikely(!PageUptodate(page))) |
1da177e4c Linux-2.6.12-rc2 |
2467 |
goto page_not_uptodate; |
ef00e08e2 readahead: clean ... |
2468 2469 2470 2471 |
/* * Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
9ab2594fe mm: tighten up th... |
2472 2473 |
max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); if (unlikely(offset >= max_off)) { |
d00806b18 mm: fix fault vs ... |
2474 |
unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2475 |
put_page(page); |
5307cc1aa Remove broken ptr... |
2476 |
return VM_FAULT_SIGBUS; |
d00806b18 mm: fix fault vs ... |
2477 |
} |
d0217ac04 mm: fault feedbac... |
2478 |
vmf->page = page; |
83c54070e mm: fault feedbac... |
2479 |
return ret | VM_FAULT_LOCKED; |
1da177e4c Linux-2.6.12-rc2 |
2480 |
|
1da177e4c Linux-2.6.12-rc2 |
2481 2482 2483 2484 2485 |
no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ |
c20cd45eb mm: allow GFP_{FS... |
2486 |
error = page_cache_read(file, offset, vmf->gfp_mask); |
1da177e4c Linux-2.6.12-rc2 |
2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 |
/* * The page we want has now been added to the page cache. * In the unlikely event that someone removed it in the * meantime, we'll just come back here and read it again. */ if (error >= 0) goto retry_find; /* * An error return from page_cache_read can result if the * system is low on memory, or a problem occurs while trying * to schedule I/O. */ if (error == -ENOMEM) |
d0217ac04 mm: fault feedbac... |
2502 2503 |
return VM_FAULT_OOM; return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
2504 2505 |
page_not_uptodate: |
1da177e4c Linux-2.6.12-rc2 |
2506 2507 2508 2509 2510 2511 |
/* * 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 |
2512 |
ClearPageError(page); |
994fc28c7 [PATCH] add AOP_T... |
2513 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e mm: fix infinite ... |
2514 2515 2516 2517 2518 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
09cbfeaf1 mm, fs: get rid o... |
2519 |
put_page(page); |
d00806b18 mm: fix fault vs ... |
2520 2521 |
if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7 [PATCH] add AOP_T... |
2522 |
goto retry_find; |
1da177e4c Linux-2.6.12-rc2 |
2523 |
|
d00806b18 mm: fix fault vs ... |
2524 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96 [PATCH] readahead... |
2525 |
shrink_readahead_size_eio(file, ra); |
d0217ac04 mm: fault feedbac... |
2526 |
return VM_FAULT_SIGBUS; |
54cb8821d mm: merge populat... |
2527 2528 |
} EXPORT_SYMBOL(filemap_fault); |
82b0f8c39 mm: join struct f... |
2529 |
void filemap_map_pages(struct vm_fault *vmf, |
bae473a42 mm: introduce fau... |
2530 |
pgoff_t start_pgoff, pgoff_t end_pgoff) |
f1820361f mm: implement ->m... |
2531 2532 2533 |
{ struct radix_tree_iter iter; void **slot; |
82b0f8c39 mm: join struct f... |
2534 |
struct file *file = vmf->vma->vm_file; |
f1820361f mm: implement ->m... |
2535 |
struct address_space *mapping = file->f_mapping; |
bae473a42 mm: introduce fau... |
2536 |
pgoff_t last_pgoff = start_pgoff; |
9ab2594fe mm: tighten up th... |
2537 |
unsigned long max_idx; |
83929372f filemap: prepare ... |
2538 |
struct page *head, *page; |
f1820361f mm: implement ->m... |
2539 2540 |
rcu_read_lock(); |
b93b01631 page cache: use x... |
2541 |
radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start_pgoff) { |
bae473a42 mm: introduce fau... |
2542 |
if (iter.index > end_pgoff) |
f1820361f mm: implement ->m... |
2543 2544 2545 2546 2547 2548 |
break; repeat: page = radix_tree_deref_slot(slot); if (unlikely(!page)) goto next; if (radix_tree_exception(page)) { |
2cf938aae mm: use radix_tre... |
2549 2550 2551 2552 2553 |
if (radix_tree_deref_retry(page)) { slot = radix_tree_iter_retry(&iter); continue; } goto next; |
f1820361f mm: implement ->m... |
2554 |
} |
83929372f filemap: prepare ... |
2555 2556 |
head = compound_head(page); if (!page_cache_get_speculative(head)) |
f1820361f mm: implement ->m... |
2557 |
goto repeat; |
83929372f filemap: prepare ... |
2558 2559 2560 2561 2562 |
/* The page was split under us? */ if (compound_head(page) != head) { put_page(head); goto repeat; } |
f1820361f mm: implement ->m... |
2563 2564 |
/* Has the page moved? */ if (unlikely(page != *slot)) { |
83929372f filemap: prepare ... |
2565 |
put_page(head); |
f1820361f mm: implement ->m... |
2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 |
goto repeat; } if (!PageUptodate(page) || PageReadahead(page) || PageHWPoison(page)) goto skip; if (!trylock_page(page)) goto skip; if (page->mapping != mapping || !PageUptodate(page)) goto unlock; |
9ab2594fe mm: tighten up th... |
2578 2579 |
max_idx = DIV_ROUND_UP(i_size_read(mapping->host), PAGE_SIZE); if (page->index >= max_idx) |
f1820361f mm: implement ->m... |
2580 |
goto unlock; |
f1820361f mm: implement ->m... |
2581 2582 |
if (file->f_ra.mmap_miss > 0) file->f_ra.mmap_miss--; |
7267ec008 mm: postpone page... |
2583 |
|
82b0f8c39 mm: join struct f... |
2584 2585 2586 |
vmf->address += (iter.index - last_pgoff) << PAGE_SHIFT; if (vmf->pte) vmf->pte += iter.index - last_pgoff; |
7267ec008 mm: postpone page... |
2587 |
last_pgoff = iter.index; |
82b0f8c39 mm: join struct f... |
2588 |
if (alloc_set_pte(vmf, NULL, page)) |
7267ec008 mm: postpone page... |
2589 |
goto unlock; |
f1820361f mm: implement ->m... |
2590 2591 2592 2593 2594 |
unlock_page(page); goto next; unlock: unlock_page(page); skip: |
09cbfeaf1 mm, fs: get rid o... |
2595 |
put_page(page); |
f1820361f mm: implement ->m... |
2596 |
next: |
7267ec008 mm: postpone page... |
2597 |
/* Huge page is mapped? No need to proceed. */ |
82b0f8c39 mm: join struct f... |
2598 |
if (pmd_trans_huge(*vmf->pmd)) |
7267ec008 mm: postpone page... |
2599 |
break; |
bae473a42 mm: introduce fau... |
2600 |
if (iter.index == end_pgoff) |
f1820361f mm: implement ->m... |
2601 2602 2603 2604 2605 |
break; } rcu_read_unlock(); } EXPORT_SYMBOL(filemap_map_pages); |
2bcd6454b mm: use new retur... |
2606 |
vm_fault_t filemap_page_mkwrite(struct vm_fault *vmf) |
4fcf1c620 mm: Make default ... |
2607 2608 |
{ struct page *page = vmf->page; |
11bac8000 mm, fs: reduce fa... |
2609 |
struct inode *inode = file_inode(vmf->vma->vm_file); |
2bcd6454b mm: use new retur... |
2610 |
vm_fault_t ret = VM_FAULT_LOCKED; |
4fcf1c620 mm: Make default ... |
2611 |
|
14da92001 fs: Protect write... |
2612 |
sb_start_pagefault(inode->i_sb); |
11bac8000 mm, fs: reduce fa... |
2613 |
file_update_time(vmf->vma->vm_file); |
4fcf1c620 mm: Make default ... |
2614 2615 2616 2617 2618 2619 |
lock_page(page); if (page->mapping != inode->i_mapping) { unlock_page(page); ret = VM_FAULT_NOPAGE; goto out; } |
14da92001 fs: Protect write... |
2620 2621 2622 2623 2624 2625 |
/* * 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 ... |
2626 |
wait_for_stable_page(page); |
4fcf1c620 mm: Make default ... |
2627 |
out: |
14da92001 fs: Protect write... |
2628 |
sb_end_pagefault(inode->i_sb); |
4fcf1c620 mm: Make default ... |
2629 2630 |
return ret; } |
4fcf1c620 mm: Make default ... |
2631 |
|
f0f37e2f7 const: mark struc... |
2632 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d mm: merge populat... |
2633 |
.fault = filemap_fault, |
f1820361f mm: implement ->m... |
2634 |
.map_pages = filemap_map_pages, |
4fcf1c620 mm: Make default ... |
2635 |
.page_mkwrite = filemap_page_mkwrite, |
1da177e4c Linux-2.6.12-rc2 |
2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 |
}; /* 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 |
2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 |
/* * 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 |
453972283 mm/filemap.c: pro... |
2661 2662 2663 2664 |
int filemap_page_mkwrite(struct vm_fault *vmf) { return -ENOSYS; } |
1da177e4c Linux-2.6.12-rc2 |
2665 2666 2667 2668 2669 2670 2671 2672 2673 |
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... |
2674 |
EXPORT_SYMBOL(filemap_page_mkwrite); |
1da177e4c Linux-2.6.12-rc2 |
2675 2676 |
EXPORT_SYMBOL(generic_file_mmap); EXPORT_SYMBOL(generic_file_readonly_mmap); |
67f9fd91f mm: remove read_c... |
2677 2678 2679 2680 2681 |
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... |
2682 |
put_page(page); |
67f9fd91f mm: remove read_c... |
2683 2684 2685 2686 2687 |
page = ERR_PTR(-EIO); } } return page; } |
32b635298 mm: filemap: remo... |
2688 |
static struct page *do_read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
2689 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2690 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2691 2692 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
2693 |
{ |
eb2be1893 mm: buffered writ... |
2694 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
2695 2696 2697 2698 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
453f85d43 mm: remove __GFP_... |
2699 |
page = __page_cache_alloc(gfp); |
eb2be1893 mm: buffered writ... |
2700 2701 |
if (!page) return ERR_PTR(-ENOMEM); |
e6f67b8c0 vfs: __read_cache... |
2702 |
err = add_to_page_cache_lru(page, mapping, index, gfp); |
eb2be1893 mm: buffered writ... |
2703 |
if (unlikely(err)) { |
09cbfeaf1 mm, fs: get rid o... |
2704 |
put_page(page); |
eb2be1893 mm: buffered writ... |
2705 2706 |
if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
2707 |
/* Presumably ENOMEM for radix tree node */ |
1da177e4c Linux-2.6.12-rc2 |
2708 2709 |
return ERR_PTR(err); } |
32b635298 mm: filemap: remo... |
2710 2711 |
filler: |
1da177e4c Linux-2.6.12-rc2 |
2712 2713 |
err = filler(data, page); if (err < 0) { |
09cbfeaf1 mm, fs: get rid o... |
2714 |
put_page(page); |
32b635298 mm: filemap: remo... |
2715 |
return ERR_PTR(err); |
1da177e4c Linux-2.6.12-rc2 |
2716 |
} |
1da177e4c Linux-2.6.12-rc2 |
2717 |
|
32b635298 mm: filemap: remo... |
2718 2719 2720 2721 2722 |
page = wait_on_page_read(page); if (IS_ERR(page)) return page; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
2723 2724 |
if (PageUptodate(page)) goto out; |
ebded0278 mm: filemap: avoi... |
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 2752 2753 2754 2755 2756 2757 2758 2759 2760 |
/* * 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 |
2761 |
lock_page(page); |
ebded0278 mm: filemap: avoi... |
2762 2763 |
/* Case c or d, restart the operation */ |
1da177e4c Linux-2.6.12-rc2 |
2764 2765 |
if (!page->mapping) { unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2766 |
put_page(page); |
32b635298 mm: filemap: remo... |
2767 |
goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
2768 |
} |
ebded0278 mm: filemap: avoi... |
2769 2770 |
/* Someone else locked and filled the page in a very small window */ |
1da177e4c Linux-2.6.12-rc2 |
2771 2772 2773 2774 |
if (PageUptodate(page)) { unlock_page(page); goto out; } |
32b635298 mm: filemap: remo... |
2775 |
goto filler; |
c855ff371 Fix a bad error c... |
2776 |
out: |
6fe6900e1 mm: make read_cac... |
2777 2778 2779 |
mark_page_accessed(page); return page; } |
0531b2aac mm: add new 'read... |
2780 2781 |
/** |
67f9fd91f mm: remove read_c... |
2782 |
* read_cache_page - read into page cache, fill it if needed |
0531b2aac mm: add new 'read... |
2783 2784 2785 |
* @mapping: the page's address_space * @index: the page index * @filler: function to perform the read |
5e5358e7c mm: cleanup descr... |
2786 |
* @data: first arg to filler(data, page) function, often left as NULL |
0531b2aac mm: add new 'read... |
2787 |
* |
0531b2aac mm: add new 'read... |
2788 |
* Read into the page cache. If a page already exists, and PageUptodate() is |
67f9fd91f mm: remove read_c... |
2789 |
* not set, try to fill the page and wait for it to become unlocked. |
0531b2aac mm: add new 'read... |
2790 2791 2792 |
* * If the page does not get brought uptodate, return -EIO. */ |
67f9fd91f mm: remove read_c... |
2793 |
struct page *read_cache_page(struct address_space *mapping, |
0531b2aac mm: add new 'read... |
2794 |
pgoff_t index, |
5e5358e7c mm: cleanup descr... |
2795 |
int (*filler)(void *, struct page *), |
0531b2aac mm: add new 'read... |
2796 2797 2798 2799 |
void *data) { return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); } |
67f9fd91f mm: remove read_c... |
2800 |
EXPORT_SYMBOL(read_cache_page); |
0531b2aac mm: add new 'read... |
2801 2802 2803 2804 2805 2806 2807 2808 |
/** * 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... |
2809 |
* any new page allocations done using the specified allocation flags. |
0531b2aac mm: add new 'read... |
2810 2811 2812 2813 2814 2815 2816 2817 |
* * If the page does not get brought uptodate, return -EIO. */ struct page *read_cache_page_gfp(struct address_space *mapping, pgoff_t index, gfp_t gfp) { filler_t *filler = (filler_t *)mapping->a_ops->readpage; |
67f9fd91f mm: remove read_c... |
2818 |
return do_read_cache_page(mapping, index, filler, NULL, gfp); |
0531b2aac mm: add new 'read... |
2819 2820 |
} EXPORT_SYMBOL(read_cache_page_gfp); |
2f718ffc1 mm: buffered writ... |
2821 |
/* |
1da177e4c Linux-2.6.12-rc2 |
2822 2823 |
* Performs necessary checks before doing a write * |
485bb99b4 [PATCH] kernel-do... |
2824 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c Linux-2.6.12-rc2 |
2825 2826 2827 |
* Returns appropriate error code that caller should return or * zero in case that write should be allowed. */ |
3309dd04c switch generic_wr... |
2828 |
inline ssize_t generic_write_checks(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
2829 |
{ |
3309dd04c switch generic_wr... |
2830 |
struct file *file = iocb->ki_filp; |
1da177e4c Linux-2.6.12-rc2 |
2831 |
struct inode *inode = file->f_mapping->host; |
59e99e5b9 mm: use rlimit he... |
2832 |
unsigned long limit = rlimit(RLIMIT_FSIZE); |
3309dd04c switch generic_wr... |
2833 |
loff_t pos; |
1da177e4c Linux-2.6.12-rc2 |
2834 |
|
3309dd04c switch generic_wr... |
2835 2836 |
if (!iov_iter_count(from)) return 0; |
1da177e4c Linux-2.6.12-rc2 |
2837 |
|
0fa6b005a generic_write_che... |
2838 |
/* FIXME: this is for backwards compatibility with 2.4 */ |
2ba48ce51 mirror O_APPEND a... |
2839 |
if (iocb->ki_flags & IOCB_APPEND) |
3309dd04c switch generic_wr... |
2840 |
iocb->ki_pos = i_size_read(inode); |
1da177e4c Linux-2.6.12-rc2 |
2841 |
|
3309dd04c switch generic_wr... |
2842 |
pos = iocb->ki_pos; |
1da177e4c Linux-2.6.12-rc2 |
2843 |
|
6be96d3ad fs: return if dir... |
2844 2845 |
if ((iocb->ki_flags & IOCB_NOWAIT) && !(iocb->ki_flags & IOCB_DIRECT)) return -EINVAL; |
0fa6b005a generic_write_che... |
2846 |
if (limit != RLIM_INFINITY) { |
3309dd04c switch generic_wr... |
2847 |
if (iocb->ki_pos >= limit) { |
0fa6b005a generic_write_che... |
2848 2849 |
send_sig(SIGXFSZ, current, 0); return -EFBIG; |
1da177e4c Linux-2.6.12-rc2 |
2850 |
} |
3309dd04c switch generic_wr... |
2851 |
iov_iter_truncate(from, limit - (unsigned long)pos); |
1da177e4c Linux-2.6.12-rc2 |
2852 2853 2854 2855 2856 |
} /* * LFS rule */ |
3309dd04c switch generic_wr... |
2857 |
if (unlikely(pos + iov_iter_count(from) > MAX_NON_LFS && |
1da177e4c Linux-2.6.12-rc2 |
2858 |
!(file->f_flags & O_LARGEFILE))) { |
3309dd04c switch generic_wr... |
2859 |
if (pos >= MAX_NON_LFS) |
1da177e4c Linux-2.6.12-rc2 |
2860 |
return -EFBIG; |
3309dd04c switch generic_wr... |
2861 |
iov_iter_truncate(from, MAX_NON_LFS - (unsigned long)pos); |
1da177e4c Linux-2.6.12-rc2 |
2862 2863 2864 2865 2866 2867 2868 2869 2870 |
} /* * Are we about to exceed the fs block limit ? * * If we have written data it becomes a short write. If we have * exceeded without writing data we send a signal and return EFBIG. * Linus frestrict idea will clean these up nicely.. */ |
3309dd04c switch generic_wr... |
2871 2872 |
if (unlikely(pos >= inode->i_sb->s_maxbytes)) return -EFBIG; |
1da177e4c Linux-2.6.12-rc2 |
2873 |
|
3309dd04c switch generic_wr... |
2874 2875 |
iov_iter_truncate(from, inode->i_sb->s_maxbytes - pos); return iov_iter_count(from); |
1da177e4c Linux-2.6.12-rc2 |
2876 2877 |
} EXPORT_SYMBOL(generic_write_checks); |
afddba49d fs: introduce wri... |
2878 2879 2880 2881 2882 |
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... |
2883 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d fs: introduce wri... |
2884 |
pagep, fsdata); |
afddba49d fs: introduce wri... |
2885 2886 2887 2888 2889 2890 2891 2892 |
} 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... |
2893 |
|
4e02ed4b4 fs: remove prepar... |
2894 |
return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d fs: introduce wri... |
2895 2896 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c Linux-2.6.12-rc2 |
2897 |
ssize_t |
1af5bb491 filemap: remove t... |
2898 |
generic_file_direct_write(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
2899 2900 2901 2902 |
{ struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; |
1af5bb491 filemap: remove t... |
2903 |
loff_t pos = iocb->ki_pos; |
1da177e4c Linux-2.6.12-rc2 |
2904 |
ssize_t written; |
a969e903a kill generic_file... |
2905 2906 |
size_t write_len; pgoff_t end; |
1da177e4c Linux-2.6.12-rc2 |
2907 |
|
0c949334a iov_iter_truncate() |
2908 |
write_len = iov_iter_count(from); |
09cbfeaf1 mm, fs: get rid o... |
2909 |
end = (pos + write_len - 1) >> PAGE_SHIFT; |
a969e903a kill generic_file... |
2910 |
|
6be96d3ad fs: return if dir... |
2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 |
if (iocb->ki_flags & IOCB_NOWAIT) { /* If there are pages to writeback, return */ if (filemap_range_has_page(inode->i_mapping, pos, pos + iov_iter_count(from))) return -EAGAIN; } else { written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); if (written) goto out; } |
a969e903a kill generic_file... |
2922 2923 2924 2925 2926 |
/* * 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... |
2927 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a kill generic_file... |
2928 |
*/ |
55635ba76 fs: fix data inva... |
2929 |
written = invalidate_inode_pages2_range(mapping, |
09cbfeaf1 mm, fs: get rid o... |
2930 |
pos >> PAGE_SHIFT, end); |
55635ba76 fs: fix data inva... |
2931 2932 2933 2934 2935 2936 2937 2938 |
/* * 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... |
2939 |
} |
639a93a52 generic_file_dire... |
2940 |
written = mapping->a_ops->direct_IO(iocb, from); |
a969e903a kill generic_file... |
2941 2942 2943 2944 2945 2946 2947 2948 |
/* * 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... |
2949 2950 2951 2952 2953 |
* * 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... |
2954 |
*/ |
332391a99 fs: Fix page cach... |
2955 2956 2957 |
if (mapping->nrpages) invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT, end); |
a969e903a kill generic_file... |
2958 |
|
1da177e4c Linux-2.6.12-rc2 |
2959 |
if (written > 0) { |
0116651c8 mm: remove tempor... |
2960 |
pos += written; |
639a93a52 generic_file_dire... |
2961 |
write_len -= written; |
0116651c8 mm: remove tempor... |
2962 2963 |
if (pos > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, pos); |
1da177e4c Linux-2.6.12-rc2 |
2964 2965 |
mark_inode_dirty(inode); } |
5cb6c6c7e generic_file_dire... |
2966 |
iocb->ki_pos = pos; |
1da177e4c Linux-2.6.12-rc2 |
2967 |
} |
639a93a52 generic_file_dire... |
2968 |
iov_iter_revert(from, write_len - iov_iter_count(from)); |
a969e903a kill generic_file... |
2969 |
out: |
1da177e4c Linux-2.6.12-rc2 |
2970 2971 2972 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893 mm: buffered writ... |
2973 2974 2975 2976 |
/* * 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... |
2977 2978 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893 mm: buffered writ... |
2979 |
{ |
eb2be1893 mm: buffered writ... |
2980 |
struct page *page; |
bbddabe2e mm: filemap: only... |
2981 |
int fgp_flags = FGP_LOCK|FGP_WRITE|FGP_CREAT; |
0faa70cb0 mm: filemap: pass... |
2982 |
|
54566b2c1 fs: symlink write... |
2983 |
if (flags & AOP_FLAG_NOFS) |
2457aec63 mm: non-atomicall... |
2984 2985 2986 |
fgp_flags |= FGP_NOFS; page = pagecache_get_page(mapping, index, fgp_flags, |
45f87de57 mm: get rid of ra... |
2987 |
mapping_gfp_mask(mapping)); |
c585a2678 mm: remove likely... |
2988 |
if (page) |
2457aec63 mm: non-atomicall... |
2989 |
wait_for_stable_page(page); |
eb2be1893 mm: buffered writ... |
2990 |
|
eb2be1893 mm: buffered writ... |
2991 2992 |
return page; } |
54566b2c1 fs: symlink write... |
2993 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893 mm: buffered writ... |
2994 |
|
3b93f911d export generic_pe... |
2995 |
ssize_t generic_perform_write(struct file *file, |
afddba49d fs: introduce wri... |
2996 2997 2998 2999 3000 3001 |
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... |
3002 |
unsigned int flags = 0; |
afddba49d fs: introduce wri... |
3003 3004 |
do { struct page *page; |
afddba49d fs: introduce wri... |
3005 3006 3007 3008 |
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... |
3009 3010 |
offset = (pos & (PAGE_SIZE - 1)); bytes = min_t(unsigned long, PAGE_SIZE - offset, |
afddba49d fs: introduce wri... |
3011 3012 3013 |
iov_iter_count(i)); again: |
00a3d660c Revert "fs: do no... |
3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 |
/* * 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... |
3028 3029 3030 3031 |
if (fatal_signal_pending(current)) { status = -EINTR; break; } |
674b892ed mm: restore KERNE... |
3032 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d fs: introduce wri... |
3033 |
&page, &fsdata); |
2457aec63 mm: non-atomicall... |
3034 |
if (unlikely(status < 0)) |
afddba49d fs: introduce wri... |
3035 |
break; |
931e80e4b mm: flush dcache ... |
3036 3037 |
if (mapping_writably_mapped(mapping)) flush_dcache_page(page); |
00a3d660c Revert "fs: do no... |
3038 |
|
afddba49d fs: introduce wri... |
3039 |
copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); |
afddba49d fs: introduce wri... |
3040 3041 3042 3043 3044 3045 3046 3047 3048 |
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... |
3049 |
iov_iter_advance(i, copied); |
afddba49d fs: introduce wri... |
3050 3051 3052 3053 3054 3055 3056 3057 3058 |
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... |
3059 |
bytes = min_t(unsigned long, PAGE_SIZE - offset, |
afddba49d fs: introduce wri... |
3060 3061 3062 |
iov_iter_single_seg_count(i)); goto again; } |
afddba49d fs: introduce wri... |
3063 3064 3065 3066 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); |
afddba49d fs: introduce wri... |
3067 3068 3069 3070 |
} while (iov_iter_count(i)); return written ? written : status; } |
3b93f911d export generic_pe... |
3071 |
EXPORT_SYMBOL(generic_perform_write); |
1da177e4c Linux-2.6.12-rc2 |
3072 |
|
e4dd9de3c vfs: Export __gen... |
3073 |
/** |
8174202b3 write_iter varian... |
3074 |
* __generic_file_write_iter - write data to a file |
e4dd9de3c vfs: Export __gen... |
3075 |
* @iocb: IO state structure (file, offset, etc.) |
8174202b3 write_iter varian... |
3076 |
* @from: iov_iter with data to write |
e4dd9de3c vfs: Export __gen... |
3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 |
* * 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. */ |
8174202b3 write_iter varian... |
3090 |
ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
3091 3092 |
{ struct file *file = iocb->ki_filp; |
fb5527e68 [PATCH] direct-io... |
3093 |
struct address_space * mapping = file->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
3094 |
struct inode *inode = mapping->host; |
3b93f911d export generic_pe... |
3095 |
ssize_t written = 0; |
1da177e4c Linux-2.6.12-rc2 |
3096 |
ssize_t err; |
3b93f911d export generic_pe... |
3097 |
ssize_t status; |
1da177e4c Linux-2.6.12-rc2 |
3098 |
|
1da177e4c Linux-2.6.12-rc2 |
3099 |
/* We can write back this queue in page reclaim */ |
de1414a65 fs: export inode_... |
3100 |
current->backing_dev_info = inode_to_bdi(inode); |
5fa8e0a1c fs: Rename file_r... |
3101 |
err = file_remove_privs(file); |
1da177e4c Linux-2.6.12-rc2 |
3102 3103 |
if (err) goto out; |
c3b2da314 fs: introduce ino... |
3104 3105 3106 |
err = file_update_time(file); if (err) goto out; |
1da177e4c Linux-2.6.12-rc2 |
3107 |
|
2ba48ce51 mirror O_APPEND a... |
3108 |
if (iocb->ki_flags & IOCB_DIRECT) { |
0b8def9d6 __generic_file_wr... |
3109 |
loff_t pos, endbyte; |
fb5527e68 [PATCH] direct-io... |
3110 |
|
1af5bb491 filemap: remove t... |
3111 |
written = generic_file_direct_write(iocb, from); |
1da177e4c Linux-2.6.12-rc2 |
3112 |
/* |
fbbbad4bc vfs,ext2: introdu... |
3113 3114 3115 3116 3117 |
* 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 |
3118 |
*/ |
0b8def9d6 __generic_file_wr... |
3119 |
if (written < 0 || !iov_iter_count(from) || IS_DAX(inode)) |
fbbbad4bc vfs,ext2: introdu... |
3120 |
goto out; |
0b8def9d6 __generic_file_wr... |
3121 |
status = generic_perform_write(file, from, pos = iocb->ki_pos); |
fb5527e68 [PATCH] direct-io... |
3122 |
/* |
3b93f911d export generic_pe... |
3123 |
* If generic_perform_write() returned a synchronous error |
fb5527e68 [PATCH] direct-io... |
3124 3125 3126 3127 3128 |
* 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... |
3129 |
if (unlikely(status < 0)) { |
3b93f911d export generic_pe... |
3130 |
err = status; |
fb5527e68 [PATCH] direct-io... |
3131 3132 |
goto out; } |
fb5527e68 [PATCH] direct-io... |
3133 3134 3135 3136 3137 |
/* * 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... |
3138 |
endbyte = pos + status - 1; |
0b8def9d6 __generic_file_wr... |
3139 |
err = filemap_write_and_wait_range(mapping, pos, endbyte); |
fb5527e68 [PATCH] direct-io... |
3140 |
if (err == 0) { |
0b8def9d6 __generic_file_wr... |
3141 |
iocb->ki_pos = endbyte + 1; |
3b93f911d export generic_pe... |
3142 |
written += status; |
fb5527e68 [PATCH] direct-io... |
3143 |
invalidate_mapping_pages(mapping, |
09cbfeaf1 mm, fs: get rid o... |
3144 3145 |
pos >> PAGE_SHIFT, endbyte >> PAGE_SHIFT); |
fb5527e68 [PATCH] direct-io... |
3146 3147 3148 3149 3150 3151 3152 |
} 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... |
3153 3154 3155 |
written = generic_perform_write(file, from, iocb->ki_pos); if (likely(written > 0)) iocb->ki_pos += written; |
fb5527e68 [PATCH] direct-io... |
3156 |
} |
1da177e4c Linux-2.6.12-rc2 |
3157 3158 3159 3160 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
8174202b3 write_iter varian... |
3161 |
EXPORT_SYMBOL(__generic_file_write_iter); |
e4dd9de3c vfs: Export __gen... |
3162 |
|
e4dd9de3c vfs: Export __gen... |
3163 |
/** |
8174202b3 write_iter varian... |
3164 |
* generic_file_write_iter - write data to a file |
e4dd9de3c vfs: Export __gen... |
3165 |
* @iocb: IO state structure |
8174202b3 write_iter varian... |
3166 |
* @from: iov_iter with data to write |
e4dd9de3c vfs: Export __gen... |
3167 |
* |
8174202b3 write_iter varian... |
3168 |
* This is a wrapper around __generic_file_write_iter() to be used by most |
e4dd9de3c vfs: Export __gen... |
3169 3170 3171 |
* filesystems. It takes care of syncing the file in case of O_SYNC file * and acquires i_mutex as needed. */ |
8174202b3 write_iter varian... |
3172 |
ssize_t generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) |
1da177e4c Linux-2.6.12-rc2 |
3173 3174 |
{ struct file *file = iocb->ki_filp; |
148f948ba vfs: Introduce ne... |
3175 |
struct inode *inode = file->f_mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
3176 |
ssize_t ret; |
1da177e4c Linux-2.6.12-rc2 |
3177 |
|
5955102c9 wrappers for ->i_... |
3178 |
inode_lock(inode); |
3309dd04c switch generic_wr... |
3179 3180 |
ret = generic_write_checks(iocb, from); if (ret > 0) |
5f380c7fa lift generic_writ... |
3181 |
ret = __generic_file_write_iter(iocb, from); |
5955102c9 wrappers for ->i_... |
3182 |
inode_unlock(inode); |
1da177e4c Linux-2.6.12-rc2 |
3183 |
|
e25922176 fs: simplify the ... |
3184 3185 |
if (ret > 0) ret = generic_write_sync(iocb, ret); |
1da177e4c Linux-2.6.12-rc2 |
3186 3187 |
return ret; } |
8174202b3 write_iter varian... |
3188 |
EXPORT_SYMBOL(generic_file_write_iter); |
1da177e4c Linux-2.6.12-rc2 |
3189 |
|
cf9a2ae8d [PATCH] BLOCK: Mo... |
3190 3191 3192 3193 3194 3195 3196 |
/** * 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 |
0e056eb55 kernel-api.rst: f... |
3197 |
* (presumably at page->private). If the release was successful, return '1'. |
cf9a2ae8d [PATCH] BLOCK: Mo... |
3198 3199 |
* Otherwise return zero. * |
266cf658e FS-Cache: Recruit... |
3200 3201 3202 |
* 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... |
3203 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
71baba4b9 mm, page_alloc: r... |
3204 |
* this page (__GFP_IO), and whether the call may block (__GFP_RECLAIM & __GFP_FS). |
cf9a2ae8d [PATCH] BLOCK: Mo... |
3205 |
* |
cf9a2ae8d [PATCH] BLOCK: Mo... |
3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 |
*/ 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); |