Blame view
mm/filemap.c
65.2 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) */ |
1da177e4c Linux-2.6.12-rc2 |
12 13 14 15 |
#include <linux/module.h> #include <linux/slab.h> #include <linux/compiler.h> #include <linux/fs.h> |
c22ce143d [PATCH] x86: cach... |
16 |
#include <linux/uaccess.h> |
1da177e4c Linux-2.6.12-rc2 |
17 |
#include <linux/aio.h> |
c59ede7b7 [PATCH] move capa... |
18 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
19 20 21 22 23 24 25 26 27 |
#include <linux/kernel_stat.h> #include <linux/mm.h> #include <linux/swap.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/file.h> #include <linux/uio.h> #include <linux/hash.h> #include <linux/writeback.h> |
53253383f Include <linux/ba... |
28 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
29 30 31 32 |
#include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/security.h> #include <linux/syscalls.h> |
44110fe38 [PATCH] cpuset me... |
33 |
#include <linux/cpuset.h> |
2f718ffc1 mm: buffered writ... |
34 |
#include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */ |
8a9f3ccd2 Memory controller... |
35 |
#include <linux/memcontrol.h> |
4f98a2fee vmscan: split LRU... |
36 |
#include <linux/mm_inline.h> /* for page_is_file_cache() */ |
0f8053a50 [PATCH] mm: make ... |
37 |
#include "internal.h" |
1da177e4c Linux-2.6.12-rc2 |
38 |
/* |
1da177e4c Linux-2.6.12-rc2 |
39 40 |
* FIXME: remove all knowledge of the buffer layer from the core VM */ |
148f948ba vfs: Introduce ne... |
41 |
#include <linux/buffer_head.h> /* for try_to_free_buffers */ |
1da177e4c Linux-2.6.12-rc2 |
42 |
|
1da177e4c Linux-2.6.12-rc2 |
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
#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: * |
25d9e2d15 truncate: new hel... |
60 |
* ->i_mmap_lock (truncate_pagecache) |
1da177e4c Linux-2.6.12-rc2 |
61 |
* ->private_lock (__free_pte->__set_page_dirty_buffers) |
5d337b919 [PATCH] swap: swa... |
62 63 |
* ->swap_lock (exclusive_swap_page, others) * ->mapping->tree_lock |
1da177e4c Linux-2.6.12-rc2 |
64 |
* |
1b1dcc1b5 [PATCH] mutex sub... |
65 |
* ->i_mutex |
1da177e4c Linux-2.6.12-rc2 |
66 67 68 69 |
* ->i_mmap_lock (truncate->unmap_mapping_range) * * ->mmap_sem * ->i_mmap_lock |
b8072f099 [PATCH] mm: updat... |
70 |
* ->page_table_lock or pte_lock (various, mainly in memory.c) |
1da177e4c Linux-2.6.12-rc2 |
71 72 73 74 75 |
* ->mapping->tree_lock (arch-dependent flush_dcache_mmap_lock) * * ->mmap_sem * ->lock_page (access_process_vm) * |
82591e6ea [PATCH] mm: more ... |
76 77 |
* ->i_mutex (generic_file_buffered_write) * ->mmap_sem (fault_in_pages_readable->do_page_fault) |
1da177e4c Linux-2.6.12-rc2 |
78 |
* |
1b1dcc1b5 [PATCH] mutex sub... |
79 |
* ->i_mutex |
1da177e4c Linux-2.6.12-rc2 |
80 81 82 83 84 85 86 87 88 89 |
* ->i_alloc_sem (various) * * ->inode_lock * ->sb_lock (fs/fs-writeback.c) * ->mapping->tree_lock (__sync_single_inode) * * ->i_mmap_lock * ->anon_vma.lock (vma_adjust) * * ->anon_vma.lock |
b8072f099 [PATCH] mm: updat... |
90 |
* ->page_table_lock or pte_lock (anon_vma_prepare and various) |
1da177e4c Linux-2.6.12-rc2 |
91 |
* |
b8072f099 [PATCH] mm: updat... |
92 |
* ->page_table_lock or pte_lock |
5d337b919 [PATCH] swap: swa... |
93 |
* ->swap_lock (try_to_unmap_one) |
1da177e4c Linux-2.6.12-rc2 |
94 95 96 |
* ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) |
053837fce [PATCH] mm: migra... |
97 |
* ->zone.lru_lock (check_pte_range->isolate_lru_page) |
1da177e4c Linux-2.6.12-rc2 |
98 99 100 101 102 103 104 105 |
* ->private_lock (page_remove_rmap->set_page_dirty) * ->tree_lock (page_remove_rmap->set_page_dirty) * ->inode_lock (page_remove_rmap->set_page_dirty) * ->inode_lock (zap_pte_range->set_page_dirty) * ->private_lock (zap_pte_range->__set_page_dirty_buffers) * * ->task->proc_lock * ->dcache_lock (proc_pid_lookup) |
6a46079cf HWPOISON: The hig... |
106 107 108 109 |
* * (code doesn't rely on that order, so you could switch it around) * ->tasklist_lock (memory_failure, collect_procs_ao) * ->i_mmap_lock |
1da177e4c Linux-2.6.12-rc2 |
110 111 112 113 114 |
*/ /* * Remove 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 |
19fd62312 mm: spinlock tree... |
115 |
* is safe. The caller must hold the mapping's tree_lock. |
1da177e4c Linux-2.6.12-rc2 |
116 117 118 119 120 121 122 123 |
*/ void __remove_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; radix_tree_delete(&mapping->page_tree, page->index); page->mapping = NULL; mapping->nrpages--; |
347ce434d [PATCH] zoned vm ... |
124 |
__dec_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac mm: oom analysis:... |
125 126 |
if (PageSwapBacked(page)) __dec_zone_page_state(page, NR_SHMEM); |
45426812d mm: debug check f... |
127 |
BUG_ON(page_mapped(page)); |
3a6927906 Do dirty page acc... |
128 129 130 131 132 133 134 135 136 137 138 139 |
/* * Some filesystems seem to re-dirty the page even after * the VM has canceled the dirty bit (eg ext3 journaling). * * Fix it up by doing a final dirty accounting check after * having removed the page entirely. */ if (PageDirty(page) && mapping_cap_account_dirty(mapping)) { dec_zone_page_state(page, NR_FILE_DIRTY); dec_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); } |
1da177e4c Linux-2.6.12-rc2 |
140 141 142 143 144 |
} void remove_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; |
cd7619d6b [PATCH] Extermina... |
145 |
BUG_ON(!PageLocked(page)); |
1da177e4c Linux-2.6.12-rc2 |
146 |
|
19fd62312 mm: spinlock tree... |
147 |
spin_lock_irq(&mapping->tree_lock); |
1da177e4c Linux-2.6.12-rc2 |
148 |
__remove_from_page_cache(page); |
19fd62312 mm: spinlock tree... |
149 |
spin_unlock_irq(&mapping->tree_lock); |
e767e0561 memcg: fix deadlo... |
150 |
mem_cgroup_uncharge_cache_page(page); |
1da177e4c Linux-2.6.12-rc2 |
151 152 153 154 155 156 |
} static int sync_page(void *word) { struct address_space *mapping; struct page *page; |
07808b74e [PATCH] x86_64: R... |
157 |
page = container_of((unsigned long *)word, struct page, flags); |
1da177e4c Linux-2.6.12-rc2 |
158 159 |
/* |
dd1d5afca [PATCH] sync_page... |
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 |
* page_mapping() is being called without PG_locked held. * Some knowledge of the state and use of the page is used to * reduce the requirements down to a memory barrier. * The danger here is of a stale page_mapping() return value * indicating a struct address_space different from the one it's * associated with when it is associated with one. * After smp_mb(), it's either the correct page_mapping() for * the page, or an old page_mapping() and the page's own * page_mapping() has gone NULL. * The ->sync_page() address_space operation must tolerate * page_mapping() going NULL. By an amazing coincidence, * this comes about because none of the users of the page * in the ->sync_page() methods make essential use of the * page_mapping(), merely passing the page down to the backing * device's unplug functions when it's non-NULL, which in turn |
4c21e2f24 [PATCH] mm: split... |
175 |
* ignore it for all cases but swap, where only page_private(page) is |
dd1d5afca [PATCH] sync_page... |
176 177 178 |
* of interest. When page_mapping() does go NULL, the entire * call stack gracefully ignores the page and returns. * -- wli |
1da177e4c Linux-2.6.12-rc2 |
179 180 181 182 183 184 185 186 |
*/ smp_mb(); mapping = page_mapping(page); if (mapping && mapping->a_ops && mapping->a_ops->sync_page) mapping->a_ops->sync_page(page); io_schedule(); return 0; } |
2687a3569 Add lock_page_kil... |
187 188 189 190 191 |
static int sync_page_killable(void *word) { sync_page(word); return fatal_signal_pending(current) ? -EINTR : 0; } |
1da177e4c Linux-2.6.12-rc2 |
192 |
/** |
485bb99b4 [PATCH] kernel-do... |
193 |
* __filemap_fdatawrite_range - start writeback on mapping dirty pages in range |
67be2dd1b [PATCH] DocBook: ... |
194 195 |
* @mapping: address space structure to write * @start: offset in bytes where the range starts |
469eb4d03 [PATCH] filemap_f... |
196 |
* @end: offset in bytes where the range ends (inclusive) |
67be2dd1b [PATCH] DocBook: ... |
197 |
* @sync_mode: enable synchronous operation |
1da177e4c Linux-2.6.12-rc2 |
198 |
* |
485bb99b4 [PATCH] kernel-do... |
199 200 201 |
* 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 |
202 |
* If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as |
485bb99b4 [PATCH] kernel-do... |
203 |
* opposed to a regular memory cleansing writeback. The difference between |
1da177e4c Linux-2.6.12-rc2 |
204 205 206 |
* 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()... |
207 208 |
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode) |
1da177e4c Linux-2.6.12-rc2 |
209 210 211 212 |
{ int ret; struct writeback_control wbc = { .sync_mode = sync_mode, |
05fe478dd mm: write_cache_p... |
213 |
.nr_to_write = LONG_MAX, |
111ebb6e6 [PATCH] writeback... |
214 215 |
.range_start = start, .range_end = end, |
1da177e4c Linux-2.6.12-rc2 |
216 217 218 219 220 221 222 223 224 225 226 227 |
}; if (!mapping_cap_writeback_dirty(mapping)) return 0; ret = do_writepages(mapping, &wbc); return ret; } static inline int __filemap_fdatawrite(struct address_space *mapping, int sync_mode) { |
111ebb6e6 [PATCH] writeback... |
228 |
return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode); |
1da177e4c Linux-2.6.12-rc2 |
229 230 231 232 233 234 235 |
} int filemap_fdatawrite(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_ALL); } EXPORT_SYMBOL(filemap_fdatawrite); |
f4c0a0fdf vfs: export filem... |
236 |
int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, |
ebcf28e1c [PATCH] fadvise()... |
237 |
loff_t end) |
1da177e4c Linux-2.6.12-rc2 |
238 239 240 |
{ return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } |
f4c0a0fdf vfs: export filem... |
241 |
EXPORT_SYMBOL(filemap_fdatawrite_range); |
1da177e4c Linux-2.6.12-rc2 |
242 |
|
485bb99b4 [PATCH] kernel-do... |
243 244 245 246 |
/** * filemap_flush - mostly a non-blocking flush * @mapping: target address_space * |
1da177e4c Linux-2.6.12-rc2 |
247 248 249 250 251 252 253 254 |
* This is a mostly non-blocking flush. Not suitable for data-integrity * purposes - I/O may not be started against all dirty pages. */ int filemap_flush(struct address_space *mapping) { return __filemap_fdatawrite(mapping, WB_SYNC_NONE); } EXPORT_SYMBOL(filemap_flush); |
485bb99b4 [PATCH] kernel-do... |
255 |
/** |
94004ed72 kill wait_on_page... |
256 257 258 259 |
* filemap_fdatawait_range - wait for writeback to complete * @mapping: address space structure to wait for * @start_byte: offset in bytes where the range starts * @end_byte: offset in bytes where the range ends (inclusive) |
485bb99b4 [PATCH] kernel-do... |
260 |
* |
94004ed72 kill wait_on_page... |
261 262 |
* Walk the list of under-writeback pages of the given address space * in the given range and wait for all of them. |
1da177e4c Linux-2.6.12-rc2 |
263 |
*/ |
94004ed72 kill wait_on_page... |
264 265 |
int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, loff_t end_byte) |
1da177e4c Linux-2.6.12-rc2 |
266 |
{ |
94004ed72 kill wait_on_page... |
267 268 |
pgoff_t index = start_byte >> PAGE_CACHE_SHIFT; pgoff_t end = end_byte >> PAGE_CACHE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
269 270 271 |
struct pagevec pvec; int nr_pages; int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
272 |
|
94004ed72 kill wait_on_page... |
273 |
if (end_byte < start_byte) |
1da177e4c Linux-2.6.12-rc2 |
274 275 276 |
return 0; pagevec_init(&pvec, 0); |
1da177e4c Linux-2.6.12-rc2 |
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 |
while ((index <= end) && (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, PAGECACHE_TAG_WRITEBACK, min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1)) != 0) { unsigned i; for (i = 0; i < nr_pages; i++) { struct page *page = pvec.pages[i]; /* until radix tree lookup accepts end_index */ if (page->index > end) continue; wait_on_page_writeback(page); if (PageError(page)) ret = -EIO; } pagevec_release(&pvec); cond_resched(); } /* Check for outstanding write errors */ if (test_and_clear_bit(AS_ENOSPC, &mapping->flags)) ret = -ENOSPC; if (test_and_clear_bit(AS_EIO, &mapping->flags)) ret = -EIO; return ret; } |
d3bccb6f4 vfs: Introduce fi... |
306 307 308 |
EXPORT_SYMBOL(filemap_fdatawait_range); /** |
485bb99b4 [PATCH] kernel-do... |
309 |
* filemap_fdatawait - wait for all under-writeback pages to complete |
1da177e4c Linux-2.6.12-rc2 |
310 |
* @mapping: address space structure to wait for |
485bb99b4 [PATCH] kernel-do... |
311 312 313 |
* * Walk the list of under-writeback pages of the given address space * and wait for all of them. |
1da177e4c Linux-2.6.12-rc2 |
314 315 316 317 318 319 320 |
*/ int filemap_fdatawait(struct address_space *mapping) { loff_t i_size = i_size_read(mapping->host); if (i_size == 0) return 0; |
94004ed72 kill wait_on_page... |
321 |
return filemap_fdatawait_range(mapping, 0, i_size - 1); |
1da177e4c Linux-2.6.12-rc2 |
322 323 324 325 326 |
} EXPORT_SYMBOL(filemap_fdatawait); int filemap_write_and_wait(struct address_space *mapping) { |
28fd12982 [PATCH] Fix and a... |
327 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
328 329 |
if (mapping->nrpages) { |
28fd12982 [PATCH] Fix and a... |
330 331 332 333 334 335 336 337 338 339 340 341 |
err = filemap_fdatawrite(mapping); /* * Even if the above returned error, the pages may be * written partially (e.g. -ENOSPC), so we wait for it. * But the -EIO is special case, it may indicate the worst * thing (e.g. bug) happened, so we avoid waiting for it. */ if (err != -EIO) { int err2 = filemap_fdatawait(mapping); if (!err) err = err2; } |
1da177e4c Linux-2.6.12-rc2 |
342 |
} |
28fd12982 [PATCH] Fix and a... |
343 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
344 |
} |
28fd12982 [PATCH] Fix and a... |
345 |
EXPORT_SYMBOL(filemap_write_and_wait); |
1da177e4c Linux-2.6.12-rc2 |
346 |
|
485bb99b4 [PATCH] kernel-do... |
347 348 349 350 351 352 |
/** * 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... |
353 354 355 356 357 |
* Write out and wait upon file offsets lstart->lend, inclusive. * * Note that `lend' is inclusive (describes the last byte to be written) so * that this function can be used to write to the very end-of-file (end = -1). */ |
1da177e4c Linux-2.6.12-rc2 |
358 359 360 |
int filemap_write_and_wait_range(struct address_space *mapping, loff_t lstart, loff_t lend) { |
28fd12982 [PATCH] Fix and a... |
361 |
int err = 0; |
1da177e4c Linux-2.6.12-rc2 |
362 363 |
if (mapping->nrpages) { |
28fd12982 [PATCH] Fix and a... |
364 365 366 367 |
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... |
368 369 |
int err2 = filemap_fdatawait_range(mapping, lstart, lend); |
28fd12982 [PATCH] Fix and a... |
370 371 372 |
if (!err) err = err2; } |
1da177e4c Linux-2.6.12-rc2 |
373 |
} |
28fd12982 [PATCH] Fix and a... |
374 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
375 |
} |
f69955855 Export filemap_wr... |
376 |
EXPORT_SYMBOL(filemap_write_and_wait_range); |
1da177e4c Linux-2.6.12-rc2 |
377 |
|
485bb99b4 [PATCH] kernel-do... |
378 |
/** |
e286781d5 mm: speculative p... |
379 |
* add_to_page_cache_locked - add a locked page to the pagecache |
485bb99b4 [PATCH] kernel-do... |
380 381 382 383 384 |
* @page: page to add * @mapping: the page's address_space * @offset: page index * @gfp_mask: page allocation mode * |
e286781d5 mm: speculative p... |
385 |
* This function is used to add a page to the pagecache. It must be locked. |
1da177e4c Linux-2.6.12-rc2 |
386 387 |
* This function does not add the page to the LRU. The caller must do that. */ |
e286781d5 mm: speculative p... |
388 |
int add_to_page_cache_locked(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
389 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
390 |
{ |
e286781d5 mm: speculative p... |
391 392 393 394 395 |
int error; VM_BUG_ON(!PageLocked(page)); error = mem_cgroup_cache_charge(page, current->mm, |
2c26fdd70 memcg: revert gfp... |
396 |
gfp_mask & GFP_RECLAIM_MASK); |
35c754d79 memory controller... |
397 398 |
if (error) goto out; |
1da177e4c Linux-2.6.12-rc2 |
399 |
|
35c754d79 memory controller... |
400 |
error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM); |
1da177e4c Linux-2.6.12-rc2 |
401 |
if (error == 0) { |
e286781d5 mm: speculative p... |
402 403 404 |
page_cache_get(page); page->mapping = mapping; page->index = offset; |
19fd62312 mm: spinlock tree... |
405 |
spin_lock_irq(&mapping->tree_lock); |
1da177e4c Linux-2.6.12-rc2 |
406 |
error = radix_tree_insert(&mapping->page_tree, offset, page); |
e286781d5 mm: speculative p... |
407 |
if (likely(!error)) { |
1da177e4c Linux-2.6.12-rc2 |
408 |
mapping->nrpages++; |
347ce434d [PATCH] zoned vm ... |
409 |
__inc_zone_page_state(page, NR_FILE_PAGES); |
4b02108ac mm: oom analysis:... |
410 411 |
if (PageSwapBacked(page)) __inc_zone_page_state(page, NR_SHMEM); |
e767e0561 memcg: fix deadlo... |
412 |
spin_unlock_irq(&mapping->tree_lock); |
e286781d5 mm: speculative p... |
413 414 |
} else { page->mapping = NULL; |
e767e0561 memcg: fix deadlo... |
415 |
spin_unlock_irq(&mapping->tree_lock); |
69029cd55 memcg: remove ref... |
416 |
mem_cgroup_uncharge_cache_page(page); |
e286781d5 mm: speculative p... |
417 418 |
page_cache_release(page); } |
1da177e4c Linux-2.6.12-rc2 |
419 |
radix_tree_preload_end(); |
35c754d79 memory controller... |
420 |
} else |
69029cd55 memcg: remove ref... |
421 |
mem_cgroup_uncharge_cache_page(page); |
8a9f3ccd2 Memory controller... |
422 |
out: |
1da177e4c Linux-2.6.12-rc2 |
423 424 |
return error; } |
e286781d5 mm: speculative p... |
425 |
EXPORT_SYMBOL(add_to_page_cache_locked); |
1da177e4c Linux-2.6.12-rc2 |
426 427 |
int add_to_page_cache_lru(struct page *page, struct address_space *mapping, |
6daa0e286 [PATCH] gfp_t: mm... |
428 |
pgoff_t offset, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
429 |
{ |
4f98a2fee vmscan: split LRU... |
430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 |
int ret; /* * Splice_read and readahead add shmem/tmpfs pages into the page cache * before shmem_readpage has a chance to mark them as SwapBacked: they * need to go on the active_anon lru below, and mem_cgroup_cache_charge * (called in add_to_page_cache) needs to know where they're going too. */ if (mapping_cap_swap_backed(mapping)) SetPageSwapBacked(page); ret = add_to_page_cache(page, mapping, offset, gfp_mask); if (ret == 0) { if (page_is_file_cache(page)) lru_cache_add_file(page); else lru_cache_add_active_anon(page); } |
1da177e4c Linux-2.6.12-rc2 |
448 449 |
return ret; } |
18bc0bbd1 Staging: pohmelfs... |
450 |
EXPORT_SYMBOL_GPL(add_to_page_cache_lru); |
1da177e4c Linux-2.6.12-rc2 |
451 |
|
44110fe38 [PATCH] cpuset me... |
452 |
#ifdef CONFIG_NUMA |
2ae88149a [PATCH] mm: clean... |
453 |
struct page *__page_cache_alloc(gfp_t gfp) |
44110fe38 [PATCH] cpuset me... |
454 455 456 |
{ if (cpuset_do_page_mem_spread()) { int n = cpuset_mem_spread_node(); |
6484eb3e2 page allocator: d... |
457 |
return alloc_pages_exact_node(n, gfp, 0); |
44110fe38 [PATCH] cpuset me... |
458 |
} |
2ae88149a [PATCH] mm: clean... |
459 |
return alloc_pages(gfp, 0); |
44110fe38 [PATCH] cpuset me... |
460 |
} |
2ae88149a [PATCH] mm: clean... |
461 |
EXPORT_SYMBOL(__page_cache_alloc); |
44110fe38 [PATCH] cpuset me... |
462 |
#endif |
db37648cd [PATCH] mm: non s... |
463 464 465 466 467 |
static int __sleep_on_page_lock(void *word) { io_schedule(); return 0; } |
1da177e4c Linux-2.6.12-rc2 |
468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 |
/* * In order to wait for pages to become available there must be * waitqueues associated with pages. By using a hash table of * waitqueues where the bucket discipline is to maintain all * waiters on the same queue and wake all when any of the pages * become available, and for the woken contexts to check to be * sure the appropriate page became available, this saves space * at a cost of "thundering herd" phenomena during rare hash * collisions. */ static wait_queue_head_t *page_waitqueue(struct page *page) { const struct zone *zone = page_zone(page); return &zone->wait_table[hash_ptr(page, zone->wait_table_bits)]; } static inline void wake_up_page(struct page *page, int bit) { __wake_up_bit(page_waitqueue(page), &page->flags, bit); } |
920c7a5d0 mm: remove fastca... |
489 |
void wait_on_page_bit(struct page *page, int bit_nr) |
1da177e4c Linux-2.6.12-rc2 |
490 491 492 493 494 495 496 497 498 499 |
{ DEFINE_WAIT_BIT(wait, &page->flags, bit_nr); if (test_bit(bit_nr, &page->flags)) __wait_on_bit(page_waitqueue(page), &wait, sync_page, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(wait_on_page_bit); /** |
385e1ca5f CacheFiles: Permi... |
500 |
* add_page_wait_queue - Add an arbitrary waiter to a page's wait queue |
697f619fc filemap: fix kern... |
501 502 |
* @page: Page defining the wait queue of interest * @waiter: Waiter to add to the queue |
385e1ca5f CacheFiles: Permi... |
503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 |
* * Add an arbitrary @waiter to the wait queue for the nominated @page. */ void add_page_wait_queue(struct page *page, wait_queue_t *waiter) { wait_queue_head_t *q = page_waitqueue(page); unsigned long flags; spin_lock_irqsave(&q->lock, flags); __add_wait_queue(q, waiter); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL_GPL(add_page_wait_queue); /** |
485bb99b4 [PATCH] kernel-do... |
518 |
* unlock_page - unlock a locked page |
1da177e4c Linux-2.6.12-rc2 |
519 520 521 522 523 524 525 |
* @page: the page * * Unlocks the page and wakes up sleepers in ___wait_on_page_locked(). * Also wakes sleepers in wait_on_page_writeback() because the wakeup * mechananism between PageLocked pages and PageWriteback pages is shared. * But that's OK - sleepers in wait_on_page_writeback() just go back to sleep. * |
8413ac9d8 mm: page lock use... |
526 527 |
* The mb is necessary to enforce ordering between the clear_bit and the read * of the waitqueue (to avoid SMP races with a parallel wait_on_page_locked()). |
1da177e4c Linux-2.6.12-rc2 |
528 |
*/ |
920c7a5d0 mm: remove fastca... |
529 |
void unlock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
530 |
{ |
8413ac9d8 mm: page lock use... |
531 532 533 |
VM_BUG_ON(!PageLocked(page)); clear_bit_unlock(PG_locked, &page->flags); smp_mb__after_clear_bit(); |
1da177e4c Linux-2.6.12-rc2 |
534 535 536 |
wake_up_page(page, PG_locked); } EXPORT_SYMBOL(unlock_page); |
485bb99b4 [PATCH] kernel-do... |
537 538 539 |
/** * end_page_writeback - end writeback against a page * @page: the page |
1da177e4c Linux-2.6.12-rc2 |
540 541 542 |
*/ void end_page_writeback(struct page *page) { |
ac6aadb24 mm: rotate_reclai... |
543 544 545 546 547 |
if (TestClearPageReclaim(page)) rotate_reclaimable_page(page); if (!test_clear_page_writeback(page)) BUG(); |
1da177e4c Linux-2.6.12-rc2 |
548 549 550 551 |
smp_mb__after_clear_bit(); wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); |
485bb99b4 [PATCH] kernel-do... |
552 553 554 |
/** * __lock_page - get a lock on the page, assuming we need to sleep to get it * @page: the page to lock |
1da177e4c Linux-2.6.12-rc2 |
555 |
* |
485bb99b4 [PATCH] kernel-do... |
556 |
* Ugly. Running sync_page() in state TASK_UNINTERRUPTIBLE is scary. If some |
1da177e4c Linux-2.6.12-rc2 |
557 558 559 560 |
* random driver's requestfn sets TASK_RUNNING, we could busywait. However * chances are that on the second loop, the block layer's plug list is empty, * so sync_page() will then return in state TASK_UNINTERRUPTIBLE. */ |
920c7a5d0 mm: remove fastca... |
561 |
void __lock_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
562 563 564 565 566 567 568 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); __wait_on_bit_lock(page_waitqueue(page), &wait, sync_page, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(__lock_page); |
b5606c2d4 remove final fast... |
569 |
int __lock_page_killable(struct page *page) |
2687a3569 Add lock_page_kil... |
570 571 572 573 574 575 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); return __wait_on_bit_lock(page_waitqueue(page), &wait, sync_page_killable, TASK_KILLABLE); } |
18bc0bbd1 Staging: pohmelfs... |
576 |
EXPORT_SYMBOL_GPL(__lock_page_killable); |
2687a3569 Add lock_page_kil... |
577 |
|
7682486b3 mm: fix various k... |
578 579 580 581 |
/** * __lock_page_nosync - get a lock on the page, without calling sync_page() * @page: the page to lock * |
db37648cd [PATCH] mm: non s... |
582 583 584 |
* Variant of lock_page that does not require the caller to hold a reference * on the page's mapping. */ |
920c7a5d0 mm: remove fastca... |
585 |
void __lock_page_nosync(struct page *page) |
db37648cd [PATCH] mm: non s... |
586 587 588 589 590 |
{ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked); __wait_on_bit_lock(page_waitqueue(page), &wait, __sleep_on_page_lock, TASK_UNINTERRUPTIBLE); } |
485bb99b4 [PATCH] kernel-do... |
591 592 593 594 595 |
/** * find_get_page - find and get a page reference * @mapping: the address_space to search * @offset: the page index * |
da6052f7b [PATCH] update so... |
596 597 |
* Is there a pagecache struct page at the given (mapping, offset) tuple? * If yes, increment its refcount and return it; if no, return NULL. |
1da177e4c Linux-2.6.12-rc2 |
598 |
*/ |
a60637c85 mm: lockless page... |
599 |
struct page *find_get_page(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
600 |
{ |
a60637c85 mm: lockless page... |
601 |
void **pagep; |
1da177e4c Linux-2.6.12-rc2 |
602 |
struct page *page; |
a60637c85 mm: lockless page... |
603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 |
rcu_read_lock(); repeat: page = NULL; pagep = radix_tree_lookup_slot(&mapping->page_tree, offset); if (pagep) { page = radix_tree_deref_slot(pagep); if (unlikely(!page || page == RADIX_TREE_RETRY)) goto repeat; if (!page_cache_get_speculative(page)) goto repeat; /* * Has the page moved? * This is part of the lockless pagecache protocol. See * include/linux/pagemap.h for details. */ if (unlikely(page != *pagep)) { page_cache_release(page); goto repeat; } } rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
626 627 |
return page; } |
1da177e4c Linux-2.6.12-rc2 |
628 |
EXPORT_SYMBOL(find_get_page); |
485bb99b4 [PATCH] kernel-do... |
629 |
/** |
1da177e4c Linux-2.6.12-rc2 |
630 |
* find_lock_page - locate, pin and lock a pagecache page |
67be2dd1b [PATCH] DocBook: ... |
631 632 |
* @mapping: the address_space to search * @offset: the page index |
1da177e4c Linux-2.6.12-rc2 |
633 634 635 636 637 638 |
* * Locates the desired pagecache page, locks it, increments its reference * count and returns its address. * * Returns zero if the page was not present. find_lock_page() may sleep. */ |
a60637c85 mm: lockless page... |
639 |
struct page *find_lock_page(struct address_space *mapping, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
640 641 |
{ struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
642 |
repeat: |
a60637c85 mm: lockless page... |
643 |
page = find_get_page(mapping, offset); |
1da177e4c Linux-2.6.12-rc2 |
644 |
if (page) { |
a60637c85 mm: lockless page... |
645 646 647 648 649 650 |
lock_page(page); /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); page_cache_release(page); goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
651 |
} |
a60637c85 mm: lockless page... |
652 |
VM_BUG_ON(page->index != offset); |
1da177e4c Linux-2.6.12-rc2 |
653 |
} |
1da177e4c Linux-2.6.12-rc2 |
654 655 |
return page; } |
1da177e4c Linux-2.6.12-rc2 |
656 657 658 659 |
EXPORT_SYMBOL(find_lock_page); /** * find_or_create_page - locate or add a pagecache page |
67be2dd1b [PATCH] DocBook: ... |
660 661 662 |
* @mapping: the page's address_space * @index: the page's index into the mapping * @gfp_mask: page allocation mode |
1da177e4c Linux-2.6.12-rc2 |
663 664 665 666 667 668 669 670 671 672 673 674 675 |
* * Locates a page in the pagecache. If the page is not present, a new page * is allocated using @gfp_mask and is added to the pagecache and to the VM's * LRU list. The returned page is locked and has its reference count * incremented. * * find_or_create_page() may sleep, even if @gfp_flags specifies an atomic * allocation! * * find_or_create_page() returns the desired page's address, or zero on * memory exhaustion. */ struct page *find_or_create_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
676 |
pgoff_t index, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
677 |
{ |
eb2be1893 mm: buffered writ... |
678 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
679 680 681 682 |
int err; repeat: page = find_lock_page(mapping, index); if (!page) { |
eb2be1893 mm: buffered writ... |
683 684 685 |
page = __page_cache_alloc(gfp_mask); if (!page) return NULL; |
67d58ac47 mm: pagecache gfp... |
686 687 688 689 690 691 692 693 |
/* * We want a regular kernel memory (not highmem or DMA etc) * allocation for the radix tree nodes, but we need to honour * the context-specific requirements the caller has asked for. * GFP_RECLAIM_MASK collects those requirements. */ err = add_to_page_cache_lru(page, mapping, index, (gfp_mask & GFP_RECLAIM_MASK)); |
eb2be1893 mm: buffered writ... |
694 695 696 697 698 |
if (unlikely(err)) { page_cache_release(page); page = NULL; if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
699 |
} |
1da177e4c Linux-2.6.12-rc2 |
700 |
} |
1da177e4c Linux-2.6.12-rc2 |
701 702 |
return page; } |
1da177e4c Linux-2.6.12-rc2 |
703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 |
EXPORT_SYMBOL(find_or_create_page); /** * find_get_pages - gang pagecache lookup * @mapping: The address_space to search * @start: The starting page index * @nr_pages: The maximum number of pages * @pages: Where the resulting pages are placed * * find_get_pages() will search for and return a group of up to * @nr_pages pages in the mapping. The pages are placed at @pages. * find_get_pages() takes a reference against the returned pages. * * The search returns a group of mapping-contiguous pages with ascending * indexes. There may be holes in the indices due to not-present pages. * * find_get_pages() returns the number of pages which were found. */ unsigned find_get_pages(struct address_space *mapping, pgoff_t start, unsigned int nr_pages, struct page **pages) { unsigned int i; unsigned int ret; |
a60637c85 mm: lockless page... |
726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 |
unsigned int nr_found; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, (void ***)pages, start, nr_pages); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: page = radix_tree_deref_slot((void **)pages[i]); if (unlikely(!page)) continue; /* * this can only trigger if nr_found == 1, making livelock * a non issue. */ if (unlikely(page == RADIX_TREE_RETRY)) goto restart; if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *((void **)pages[i]))) { page_cache_release(page); goto repeat; } |
1da177e4c Linux-2.6.12-rc2 |
754 |
|
a60637c85 mm: lockless page... |
755 756 757 758 |
pages[ret] = page; ret++; } rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
759 760 |
return ret; } |
ebf43500e [PATCH] Add find_... |
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 |
/** * 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) { unsigned int i; unsigned int ret; |
a60637c85 mm: lockless page... |
778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 |
unsigned int nr_found; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, (void ***)pages, index, nr_pages); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: page = radix_tree_deref_slot((void **)pages[i]); if (unlikely(!page)) continue; /* * this can only trigger if nr_found == 1, making livelock * a non issue. */ if (unlikely(page == RADIX_TREE_RETRY)) goto restart; |
ebf43500e [PATCH] Add find_... |
797 |
|
a60637c85 mm: lockless page... |
798 |
if (page->mapping == NULL || page->index != index) |
ebf43500e [PATCH] Add find_... |
799 |
break; |
a60637c85 mm: lockless page... |
800 801 802 803 804 805 806 807 808 809 810 |
if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *((void **)pages[i]))) { page_cache_release(page); goto repeat; } pages[ret] = page; ret++; |
ebf43500e [PATCH] Add find_... |
811 812 |
index++; } |
a60637c85 mm: lockless page... |
813 814 |
rcu_read_unlock(); return ret; |
ebf43500e [PATCH] Add find_... |
815 |
} |
ef71c15c4 AFS: export a cou... |
816 |
EXPORT_SYMBOL(find_get_pages_contig); |
ebf43500e [PATCH] Add find_... |
817 |
|
485bb99b4 [PATCH] kernel-do... |
818 819 820 821 822 823 824 825 |
/** * find_get_pages_tag - find and return pages that match @tag * @mapping: the address_space to search * @index: the starting page index * @tag: the tag index * @nr_pages: the maximum number of pages * @pages: where the resulting pages are placed * |
1da177e4c Linux-2.6.12-rc2 |
826 |
* Like find_get_pages, except we only return pages which are tagged with |
485bb99b4 [PATCH] kernel-do... |
827 |
* @tag. We update @index to index the next page for the traversal. |
1da177e4c Linux-2.6.12-rc2 |
828 829 830 831 832 833 |
*/ unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, int tag, unsigned int nr_pages, struct page **pages) { unsigned int i; unsigned int ret; |
a60637c85 mm: lockless page... |
834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 |
unsigned int nr_found; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_tag_slot(&mapping->page_tree, (void ***)pages, *index, nr_pages, tag); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: page = radix_tree_deref_slot((void **)pages[i]); if (unlikely(!page)) continue; /* * this can only trigger if nr_found == 1, making livelock * a non issue. */ if (unlikely(page == RADIX_TREE_RETRY)) goto restart; if (!page_cache_get_speculative(page)) goto repeat; /* Has the page moved? */ if (unlikely(page != *((void **)pages[i]))) { page_cache_release(page); goto repeat; } pages[ret] = page; ret++; } rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
867 |
|
1da177e4c Linux-2.6.12-rc2 |
868 869 |
if (ret) *index = pages[ret - 1]->index + 1; |
a60637c85 mm: lockless page... |
870 |
|
1da177e4c Linux-2.6.12-rc2 |
871 872 |
return ret; } |
ef71c15c4 AFS: export a cou... |
873 |
EXPORT_SYMBOL(find_get_pages_tag); |
1da177e4c Linux-2.6.12-rc2 |
874 |
|
485bb99b4 [PATCH] kernel-do... |
875 876 877 878 879 |
/** * grab_cache_page_nowait - returns locked page at given index in given cache * @mapping: target address_space * @index: the page index * |
72fd4a35a [PATCH] Numerous ... |
880 |
* Same as grab_cache_page(), but do not wait if the page is unavailable. |
1da177e4c Linux-2.6.12-rc2 |
881 882 883 884 885 886 887 888 |
* This is intended for speculative data generators, where the data can * be regenerated if the page couldn't be grabbed. This routine should * be safe to call while holding the lock for another page. * * Clear __GFP_FS when allocating the page to avoid recursion into the fs * and deadlock against the caller's locked page. */ struct page * |
57f6b96c0 filemap: convert ... |
889 |
grab_cache_page_nowait(struct address_space *mapping, pgoff_t index) |
1da177e4c Linux-2.6.12-rc2 |
890 891 |
{ struct page *page = find_get_page(mapping, index); |
1da177e4c Linux-2.6.12-rc2 |
892 893 |
if (page) { |
529ae9aaa mm: rename page t... |
894 |
if (trylock_page(page)) |
1da177e4c Linux-2.6.12-rc2 |
895 896 897 898 |
return page; page_cache_release(page); return NULL; } |
2ae88149a [PATCH] mm: clean... |
899 |
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); |
67d58ac47 mm: pagecache gfp... |
900 |
if (page && add_to_page_cache_lru(page, mapping, index, GFP_NOFS)) { |
1da177e4c Linux-2.6.12-rc2 |
901 902 903 904 905 |
page_cache_release(page); page = NULL; } return page; } |
1da177e4c Linux-2.6.12-rc2 |
906 |
EXPORT_SYMBOL(grab_cache_page_nowait); |
76d42bd96 [PATCH] readahead... |
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 |
/* * 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... |
925 |
ra->ra_pages /= 4; |
76d42bd96 [PATCH] readahead... |
926 |
} |
485bb99b4 [PATCH] kernel-do... |
927 |
/** |
36e789144 kill do_generic_m... |
928 |
* do_generic_file_read - generic file read routine |
485bb99b4 [PATCH] kernel-do... |
929 930 931 932 933 |
* @filp: the file to read * @ppos: current file position * @desc: read_descriptor * @actor: read method * |
1da177e4c Linux-2.6.12-rc2 |
934 |
* This is a generic file read routine, and uses the |
485bb99b4 [PATCH] kernel-do... |
935 |
* mapping->a_ops->readpage() function for the actual low-level stuff. |
1da177e4c Linux-2.6.12-rc2 |
936 937 938 |
* * 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 |
939 |
*/ |
36e789144 kill do_generic_m... |
940 941 |
static void do_generic_file_read(struct file *filp, loff_t *ppos, read_descriptor_t *desc, read_actor_t actor) |
1da177e4c Linux-2.6.12-rc2 |
942 |
{ |
36e789144 kill do_generic_m... |
943 |
struct address_space *mapping = filp->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
944 |
struct inode *inode = mapping->host; |
36e789144 kill do_generic_m... |
945 |
struct file_ra_state *ra = &filp->f_ra; |
57f6b96c0 filemap: convert ... |
946 947 948 949 |
pgoff_t index; pgoff_t last_index; pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ |
ec0f16372 readahead: improv... |
950 |
unsigned int prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
951 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
952 |
|
1da177e4c Linux-2.6.12-rc2 |
953 |
index = *ppos >> PAGE_CACHE_SHIFT; |
7ff81078d readahead: remove... |
954 955 |
prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); |
1da177e4c Linux-2.6.12-rc2 |
956 957 |
last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; offset = *ppos & ~PAGE_CACHE_MASK; |
1da177e4c Linux-2.6.12-rc2 |
958 959 |
for (;;) { struct page *page; |
57f6b96c0 filemap: convert ... |
960 |
pgoff_t end_index; |
a32ea1e1f Fix read/truncate... |
961 |
loff_t isize; |
1da177e4c Linux-2.6.12-rc2 |
962 |
unsigned long nr, ret; |
1da177e4c Linux-2.6.12-rc2 |
963 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
964 965 |
find_page: page = find_get_page(mapping, index); |
3ea89ee86 readahead: conver... |
966 |
if (!page) { |
cf914a7d6 readahead: split ... |
967 |
page_cache_sync_readahead(mapping, |
7ff81078d readahead: remove... |
968 |
ra, filp, |
3ea89ee86 readahead: conver... |
969 970 971 972 973 974 |
index, last_index - index); page = find_get_page(mapping, index); if (unlikely(page == NULL)) goto no_cached_page; } if (PageReadahead(page)) { |
cf914a7d6 readahead: split ... |
975 |
page_cache_async_readahead(mapping, |
7ff81078d readahead: remove... |
976 |
ra, filp, page, |
3ea89ee86 readahead: conver... |
977 |
index, last_index - index); |
1da177e4c Linux-2.6.12-rc2 |
978 |
} |
8ab22b9ab vfs: pagecache us... |
979 980 981 982 |
if (!PageUptodate(page)) { if (inode->i_blkbits == PAGE_CACHE_SHIFT || !mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; |
529ae9aaa mm: rename page t... |
983 |
if (!trylock_page(page)) |
8ab22b9ab vfs: pagecache us... |
984 985 986 987 988 989 |
goto page_not_up_to_date; if (!mapping->a_ops->is_partially_uptodate(page, desc, offset)) goto page_not_up_to_date_locked; unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
990 |
page_ok: |
a32ea1e1f Fix read/truncate... |
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 |
/* * i_size must be checked after we know the page is Uptodate. * * Checking i_size after the check allows us to calculate * the correct value for "nr", which means the zero-filled * part of the page is not copied back to userspace (unless * another truncate extends the file - this is desired though). */ isize = i_size_read(inode); end_index = (isize - 1) >> PAGE_CACHE_SHIFT; if (unlikely(!isize || index > end_index)) { page_cache_release(page); goto out; } /* nr is the maximum number of bytes to copy from this page */ nr = PAGE_CACHE_SIZE; if (index == end_index) { nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1; if (nr <= offset) { page_cache_release(page); goto out; } } nr = nr - offset; |
1da177e4c Linux-2.6.12-rc2 |
1017 1018 1019 1020 1021 1022 1023 1024 1025 |
/* 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... |
1026 1027 |
* When a sequential read accesses a page several times, * only mark it as accessed the first time. |
1da177e4c Linux-2.6.12-rc2 |
1028 |
*/ |
ec0f16372 readahead: improv... |
1029 |
if (prev_index != index || offset != prev_offset) |
1da177e4c Linux-2.6.12-rc2 |
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 |
mark_page_accessed(page); prev_index = index; /* * Ok, we have the page, and it's up-to-date, so * now we can copy it to user space... * * The actor routine returns how many bytes were actually used.. * NOTE! This may not be the same as how much of a user buffer * we filled up (we may be padding etc), so we can only update * "pos" here (the actor routine has to update the user buffer * pointers and the remaining count). */ ret = actor(desc, page, offset, nr); offset += ret; index += offset >> PAGE_CACHE_SHIFT; offset &= ~PAGE_CACHE_MASK; |
6ce745ed3 readahead: code c... |
1047 |
prev_offset = offset; |
1da177e4c Linux-2.6.12-rc2 |
1048 1049 1050 1051 1052 1053 1054 1055 |
page_cache_release(page); if (ret == nr && desc->count) continue; goto out; page_not_up_to_date: /* Get exclusive access to the page ... */ |
854623235 do_generic_file_r... |
1056 1057 1058 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1059 |
|
8ab22b9ab vfs: pagecache us... |
1060 |
page_not_up_to_date_locked: |
da6052f7b [PATCH] update so... |
1061 |
/* Did it get truncated before we got the lock? */ |
1da177e4c Linux-2.6.12-rc2 |
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 |
if (!page->mapping) { unlock_page(page); page_cache_release(page); continue; } /* Did somebody else fill it already? */ if (PageUptodate(page)) { unlock_page(page); goto page_ok; } readpage: /* Start the actual read. The read will unlock the page. */ error = mapping->a_ops->readpage(filp, page); |
994fc28c7 [PATCH] add AOP_T... |
1077 1078 1079 1080 1081 |
if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { page_cache_release(page); goto find_page; } |
1da177e4c Linux-2.6.12-rc2 |
1082 |
goto readpage_error; |
994fc28c7 [PATCH] add AOP_T... |
1083 |
} |
1da177e4c Linux-2.6.12-rc2 |
1084 1085 |
if (!PageUptodate(page)) { |
854623235 do_generic_file_r... |
1086 1087 1088 |
error = lock_page_killable(page); if (unlikely(error)) goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 |
if (!PageUptodate(page)) { if (page->mapping == NULL) { /* * invalidate_inode_pages got it */ unlock_page(page); page_cache_release(page); goto find_page; } unlock_page(page); |
7ff81078d readahead: remove... |
1099 |
shrink_readahead_size_eio(filp, ra); |
854623235 do_generic_file_r... |
1100 1101 |
error = -EIO; goto readpage_error; |
1da177e4c Linux-2.6.12-rc2 |
1102 1103 1104 |
} unlock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 |
goto page_ok; readpage_error: /* UHHUH! A synchronous read error occurred. Report it */ desc->error = error; page_cache_release(page); goto out; no_cached_page: /* * Ok, it wasn't cached, so we need to create a new * page.. */ |
eb2be1893 mm: buffered writ... |
1118 1119 1120 1121 |
page = page_cache_alloc_cold(mapping); if (!page) { desc->error = -ENOMEM; goto out; |
1da177e4c Linux-2.6.12-rc2 |
1122 |
} |
eb2be1893 mm: buffered writ... |
1123 |
error = add_to_page_cache_lru(page, mapping, |
1da177e4c Linux-2.6.12-rc2 |
1124 1125 |
index, GFP_KERNEL); if (error) { |
eb2be1893 mm: buffered writ... |
1126 |
page_cache_release(page); |
1da177e4c Linux-2.6.12-rc2 |
1127 1128 1129 1130 1131 |
if (error == -EEXIST) goto find_page; desc->error = error; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
1132 1133 1134 1135 |
goto readpage; } out: |
7ff81078d readahead: remove... |
1136 1137 1138 |
ra->prev_pos = prev_index; ra->prev_pos <<= PAGE_CACHE_SHIFT; ra->prev_pos |= prev_offset; |
1da177e4c Linux-2.6.12-rc2 |
1139 |
|
f4e6b498d readahead: combin... |
1140 |
*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; |
0c6aa2639 mm: do_generic_fi... |
1141 |
file_accessed(filp); |
1da177e4c Linux-2.6.12-rc2 |
1142 |
} |
1da177e4c Linux-2.6.12-rc2 |
1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 |
int file_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset, unsigned long size) { char *kaddr; unsigned long left, count = desc->count; if (size > count) size = count; /* * Faults on the destination of a read are common, so do it before * taking the kmap. */ if (!fault_in_pages_writeable(desc->arg.buf, size)) { kaddr = kmap_atomic(page, KM_USER0); left = __copy_to_user_inatomic(desc->arg.buf, kaddr + offset, size); kunmap_atomic(kaddr, KM_USER0); if (left == 0) goto success; } /* Do it the slow way */ kaddr = kmap(page); left = __copy_to_user(desc->arg.buf, kaddr + offset, size); kunmap(page); if (left) { size -= left; desc->error = -EFAULT; } success: desc->count = count - size; desc->written += size; desc->arg.buf += size; return size; } |
0ceb33143 mm: move common s... |
1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 |
/* * Performs necessary checks before doing a write * @iov: io vector request * @nr_segs: number of segments in the iovec * @count: number of bytes to write * @access_flags: type of access: %VERIFY_READ or %VERIFY_WRITE * * Adjust number of segments and amount of bytes to write (nr_segs should be * properly initialized first). Returns appropriate error code that caller * should return or zero in case that write should be allowed. */ int generic_segment_checks(const struct iovec *iov, unsigned long *nr_segs, size_t *count, int access_flags) { unsigned long seg; size_t cnt = 0; for (seg = 0; seg < *nr_segs; seg++) { const struct iovec *iv = &iov[seg]; /* * If any segment has a negative length, or the cumulative * length ever wraps negative then return -EINVAL. */ cnt += iv->iov_len; if (unlikely((ssize_t)(cnt|iv->iov_len) < 0)) return -EINVAL; if (access_ok(access_flags, iv->iov_base, iv->iov_len)) continue; if (seg == 0) return -EFAULT; *nr_segs = seg; cnt -= iv->iov_len; /* This segment is no good */ break; } *count = cnt; return 0; } EXPORT_SYMBOL(generic_segment_checks); |
485bb99b4 [PATCH] kernel-do... |
1219 |
/** |
b2abacf3a [PATCH] mm: fix i... |
1220 |
* generic_file_aio_read - generic filesystem read routine |
485bb99b4 [PATCH] kernel-do... |
1221 1222 1223 |
* @iocb: kernel I/O control block * @iov: io vector request * @nr_segs: number of segments in the iovec |
b2abacf3a [PATCH] mm: fix i... |
1224 |
* @pos: current file position |
485bb99b4 [PATCH] kernel-do... |
1225 |
* |
1da177e4c Linux-2.6.12-rc2 |
1226 1227 1228 1229 |
* This is the "read()" routine for all filesystems * that can use the page cache directly. */ ssize_t |
543ade1fc [PATCH] Streamlin... |
1230 1231 |
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c Linux-2.6.12-rc2 |
1232 1233 1234 1235 1236 |
{ struct file *filp = iocb->ki_filp; ssize_t retval; unsigned long seg; size_t count; |
543ade1fc [PATCH] Streamlin... |
1237 |
loff_t *ppos = &iocb->ki_pos; |
1da177e4c Linux-2.6.12-rc2 |
1238 1239 |
count = 0; |
0ceb33143 mm: move common s... |
1240 1241 1242 |
retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; |
1da177e4c Linux-2.6.12-rc2 |
1243 1244 1245 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { |
543ade1fc [PATCH] Streamlin... |
1246 |
loff_t size; |
1da177e4c Linux-2.6.12-rc2 |
1247 1248 1249 1250 1251 |
struct address_space *mapping; struct inode *inode; mapping = filp->f_mapping; inode = mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
1252 1253 1254 1255 |
if (!count) goto out; /* skip atime */ size = i_size_read(inode); if (pos < size) { |
48b47c561 mm: direct IO sta... |
1256 1257 |
retval = filemap_write_and_wait_range(mapping, pos, pos + iov_length(iov, nr_segs) - 1); |
a969e903a kill generic_file... |
1258 1259 1260 1261 |
if (!retval) { retval = mapping->a_ops->direct_IO(READ, iocb, iov, pos, nr_segs); } |
1da177e4c Linux-2.6.12-rc2 |
1262 1263 |
if (retval > 0) *ppos = pos + retval; |
11fa977ec generic_file_aio_... |
1264 1265 1266 1267 |
if (retval) { file_accessed(filp); goto out; } |
0e0bcae3b [GFS2] Fix direct... |
1268 |
} |
1da177e4c Linux-2.6.12-rc2 |
1269 |
} |
11fa977ec generic_file_aio_... |
1270 1271 |
for (seg = 0; seg < nr_segs; seg++) { read_descriptor_t desc; |
1da177e4c Linux-2.6.12-rc2 |
1272 |
|
11fa977ec generic_file_aio_... |
1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 |
desc.written = 0; desc.arg.buf = iov[seg].iov_base; desc.count = iov[seg].iov_len; if (desc.count == 0) continue; desc.error = 0; do_generic_file_read(filp, ppos, &desc, file_read_actor); retval += desc.written; if (desc.error) { retval = retval ?: desc.error; break; |
1da177e4c Linux-2.6.12-rc2 |
1284 |
} |
11fa977ec generic_file_aio_... |
1285 1286 |
if (desc.count > 0) break; |
1da177e4c Linux-2.6.12-rc2 |
1287 1288 1289 1290 |
} out: return retval; } |
1da177e4c Linux-2.6.12-rc2 |
1291 |
EXPORT_SYMBOL(generic_file_aio_read); |
1da177e4c Linux-2.6.12-rc2 |
1292 1293 |
static ssize_t do_readahead(struct address_space *mapping, struct file *filp, |
57f6b96c0 filemap: convert ... |
1294 |
pgoff_t index, unsigned long nr) |
1da177e4c Linux-2.6.12-rc2 |
1295 1296 1297 |
{ if (!mapping || !mapping->a_ops || !mapping->a_ops->readpage) return -EINVAL; |
f7e839dd3 readahead: move m... |
1298 |
force_page_cache_readahead(mapping, filp, index, nr); |
1da177e4c Linux-2.6.12-rc2 |
1299 1300 |
return 0; } |
6673e0c3f [CVE-2009-0029] S... |
1301 |
SYSCALL_DEFINE(readahead)(int fd, loff_t offset, size_t count) |
1da177e4c Linux-2.6.12-rc2 |
1302 1303 1304 1305 1306 1307 1308 1309 1310 |
{ ssize_t ret; struct file *file; ret = -EBADF; file = fget(fd); if (file) { if (file->f_mode & FMODE_READ) { struct address_space *mapping = file->f_mapping; |
57f6b96c0 filemap: convert ... |
1311 1312 |
pgoff_t start = offset >> PAGE_CACHE_SHIFT; pgoff_t end = (offset + count - 1) >> PAGE_CACHE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
1313 1314 1315 1316 1317 1318 1319 |
unsigned long len = end - start + 1; ret = do_readahead(mapping, file, start, len); } fput(file); } return ret; } |
6673e0c3f [CVE-2009-0029] S... |
1320 1321 1322 1323 1324 1325 1326 |
#ifdef CONFIG_HAVE_SYSCALL_WRAPPERS asmlinkage long SyS_readahead(long fd, loff_t offset, long count) { return SYSC_readahead((int) fd, offset, (size_t) count); } SYSCALL_ALIAS(sys_readahead, SyS_readahead); #endif |
1da177e4c Linux-2.6.12-rc2 |
1327 1328 |
#ifdef CONFIG_MMU |
485bb99b4 [PATCH] kernel-do... |
1329 1330 1331 1332 1333 |
/** * page_cache_read - adds requested page to the page cache if not already there * @file: file to read * @offset: page index * |
1da177e4c Linux-2.6.12-rc2 |
1334 1335 1336 |
* This adds the requested page to the page cache if it isn't already there, * and schedules an I/O to read in its contents from disk. */ |
920c7a5d0 mm: remove fastca... |
1337 |
static int page_cache_read(struct file *file, pgoff_t offset) |
1da177e4c Linux-2.6.12-rc2 |
1338 1339 1340 |
{ struct address_space *mapping = file->f_mapping; struct page *page; |
994fc28c7 [PATCH] add AOP_T... |
1341 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1342 |
|
994fc28c7 [PATCH] add AOP_T... |
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 |
do { page = page_cache_alloc_cold(mapping); if (!page) return -ENOMEM; ret = add_to_page_cache_lru(page, mapping, offset, GFP_KERNEL); if (ret == 0) ret = mapping->a_ops->readpage(file, page); else if (ret == -EEXIST) ret = 0; /* losing race to add is OK */ |
1da177e4c Linux-2.6.12-rc2 |
1353 |
|
1da177e4c Linux-2.6.12-rc2 |
1354 |
page_cache_release(page); |
1da177e4c Linux-2.6.12-rc2 |
1355 |
|
994fc28c7 [PATCH] add AOP_T... |
1356 1357 1358 |
} while (ret == AOP_TRUNCATED_PAGE); return ret; |
1da177e4c Linux-2.6.12-rc2 |
1359 1360 1361 |
} #define MMAP_LOTSAMISS (100) |
ef00e08e2 readahead: clean ... |
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 |
/* * Synchronous readahead happens when we don't even find * a page in the page cache at all. */ static void do_sync_mmap_readahead(struct vm_area_struct *vma, struct file_ra_state *ra, struct file *file, pgoff_t offset) { unsigned long ra_pages; struct address_space *mapping = file->f_mapping; /* If we don't want any read-ahead, don't bother */ if (VM_RandomReadHint(vma)) return; |
70ac23cfa readahead: sequen... |
1377 1378 |
if (VM_SequentialReadHint(vma) || offset - 1 == (ra->prev_pos >> PAGE_CACHE_SHIFT)) { |
7ffc59b4d readahead: enforc... |
1379 1380 |
page_cache_sync_readahead(mapping, ra, file, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 |
return; } if (ra->mmap_miss < INT_MAX) 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... |
1393 1394 1395 |
/* * mmap read-around */ |
ef00e08e2 readahead: clean ... |
1396 1397 |
ra_pages = max_sane_readahead(ra->ra_pages); if (ra_pages) { |
d30a11004 readahead: record... |
1398 1399 1400 1401 |
ra->start = max_t(long, 0, offset - ra_pages/2); ra->size = ra_pages; ra->async_size = 0; ra_submit(ra, mapping, file); |
ef00e08e2 readahead: clean ... |
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 |
} } /* * Asynchronous readahead happens when we find the page and PG_readahead, * so we want to possibly extend the readahead further.. */ static void do_async_mmap_readahead(struct vm_area_struct *vma, struct file_ra_state *ra, struct file *file, struct page *page, pgoff_t offset) { struct address_space *mapping = file->f_mapping; /* If we don't want any read-ahead, don't bother */ if (VM_RandomReadHint(vma)) return; if (ra->mmap_miss > 0) ra->mmap_miss--; if (PageReadahead(page)) |
2fad6f5de readahead: enforc... |
1423 1424 |
page_cache_async_readahead(mapping, ra, file, page, offset, ra->ra_pages); |
ef00e08e2 readahead: clean ... |
1425 |
} |
485bb99b4 [PATCH] kernel-do... |
1426 |
/** |
54cb8821d mm: merge populat... |
1427 |
* filemap_fault - read in file data for page fault handling |
d0217ac04 mm: fault feedbac... |
1428 1429 |
* @vma: vma in which the fault was taken * @vmf: struct vm_fault containing details of the fault |
485bb99b4 [PATCH] kernel-do... |
1430 |
* |
54cb8821d mm: merge populat... |
1431 |
* filemap_fault() is invoked via the vma operations vector for a |
1da177e4c Linux-2.6.12-rc2 |
1432 1433 1434 1435 1436 1437 |
* mapped memory region to read in file data during a page fault. * * The goto's are kind of ugly, but this streamlines the normal case of having * it in the page cache, and handles the special cases reasonably without * having a lot of duplicated code. */ |
d0217ac04 mm: fault feedbac... |
1438 |
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
1da177e4c Linux-2.6.12-rc2 |
1439 1440 |
{ int error; |
54cb8821d mm: merge populat... |
1441 |
struct file *file = vma->vm_file; |
1da177e4c Linux-2.6.12-rc2 |
1442 1443 1444 |
struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; |
ef00e08e2 readahead: clean ... |
1445 |
pgoff_t offset = vmf->pgoff; |
1da177e4c Linux-2.6.12-rc2 |
1446 |
struct page *page; |
2004dc8ee Use pgoff_t inste... |
1447 |
pgoff_t size; |
83c54070e mm: fault feedbac... |
1448 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1449 |
|
1da177e4c Linux-2.6.12-rc2 |
1450 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2 readahead: clean ... |
1451 |
if (offset >= size) |
5307cc1aa Remove broken ptr... |
1452 |
return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
1453 |
|
1da177e4c Linux-2.6.12-rc2 |
1454 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1455 1456 |
* Do we have something in the page cache already? */ |
ef00e08e2 readahead: clean ... |
1457 1458 |
page = find_get_page(mapping, offset); if (likely(page)) { |
1da177e4c Linux-2.6.12-rc2 |
1459 |
/* |
ef00e08e2 readahead: clean ... |
1460 1461 |
* We found the page, so try async readahead before * waiting for the lock. |
1da177e4c Linux-2.6.12-rc2 |
1462 |
*/ |
ef00e08e2 readahead: clean ... |
1463 1464 |
do_async_mmap_readahead(vma, ra, file, page, offset); lock_page(page); |
1da177e4c Linux-2.6.12-rc2 |
1465 |
|
ef00e08e2 readahead: clean ... |
1466 1467 1468 1469 1470 |
/* Did it get truncated? */ if (unlikely(page->mapping != mapping)) { unlock_page(page); put_page(page); goto no_cached_page; |
1da177e4c Linux-2.6.12-rc2 |
1471 |
} |
ef00e08e2 readahead: clean ... |
1472 1473 1474 1475 1476 1477 1478 |
} else { /* No page in the page cache at all */ do_sync_mmap_readahead(vma, ra, file, offset); count_vm_event(PGMAJFAULT); ret = VM_FAULT_MAJOR; retry_find: page = find_lock_page(mapping, offset); |
1da177e4c Linux-2.6.12-rc2 |
1479 1480 1481 |
if (!page) goto no_cached_page; } |
1da177e4c Linux-2.6.12-rc2 |
1482 |
/* |
d00806b18 mm: fix fault vs ... |
1483 1484 |
* 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 |
1485 |
*/ |
d00806b18 mm: fix fault vs ... |
1486 |
if (unlikely(!PageUptodate(page))) |
1da177e4c Linux-2.6.12-rc2 |
1487 |
goto page_not_uptodate; |
ef00e08e2 readahead: clean ... |
1488 1489 1490 1491 |
/* * Found the page and have a reference on it. * We must recheck i_size under page lock. */ |
d00806b18 mm: fix fault vs ... |
1492 |
size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
ef00e08e2 readahead: clean ... |
1493 |
if (unlikely(offset >= size)) { |
d00806b18 mm: fix fault vs ... |
1494 |
unlock_page(page); |
745ad48e8 fix page release ... |
1495 |
page_cache_release(page); |
5307cc1aa Remove broken ptr... |
1496 |
return VM_FAULT_SIGBUS; |
d00806b18 mm: fix fault vs ... |
1497 |
} |
ef00e08e2 readahead: clean ... |
1498 |
ra->prev_pos = (loff_t)offset << PAGE_CACHE_SHIFT; |
d0217ac04 mm: fault feedbac... |
1499 |
vmf->page = page; |
83c54070e mm: fault feedbac... |
1500 |
return ret | VM_FAULT_LOCKED; |
1da177e4c Linux-2.6.12-rc2 |
1501 |
|
1da177e4c Linux-2.6.12-rc2 |
1502 1503 1504 1505 1506 |
no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ |
ef00e08e2 readahead: clean ... |
1507 |
error = page_cache_read(file, offset); |
1da177e4c Linux-2.6.12-rc2 |
1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 |
/* * 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... |
1523 1524 |
return VM_FAULT_OOM; return VM_FAULT_SIGBUS; |
1da177e4c Linux-2.6.12-rc2 |
1525 1526 |
page_not_uptodate: |
1da177e4c Linux-2.6.12-rc2 |
1527 1528 1529 1530 1531 1532 |
/* * 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 |
1533 |
ClearPageError(page); |
994fc28c7 [PATCH] add AOP_T... |
1534 |
error = mapping->a_ops->readpage(file, page); |
3ef0f720e mm: fix infinite ... |
1535 1536 1537 1538 1539 |
if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } |
d00806b18 mm: fix fault vs ... |
1540 1541 1542 |
page_cache_release(page); if (!error || error == AOP_TRUNCATED_PAGE) |
994fc28c7 [PATCH] add AOP_T... |
1543 |
goto retry_find; |
1da177e4c Linux-2.6.12-rc2 |
1544 |
|
d00806b18 mm: fix fault vs ... |
1545 |
/* Things didn't work out. Return zero to tell the mm layer so. */ |
76d42bd96 [PATCH] readahead... |
1546 |
shrink_readahead_size_eio(file, ra); |
d0217ac04 mm: fault feedbac... |
1547 |
return VM_FAULT_SIGBUS; |
54cb8821d mm: merge populat... |
1548 1549 |
} EXPORT_SYMBOL(filemap_fault); |
f0f37e2f7 const: mark struc... |
1550 |
const struct vm_operations_struct generic_file_vm_ops = { |
54cb8821d mm: merge populat... |
1551 |
.fault = filemap_fault, |
1da177e4c Linux-2.6.12-rc2 |
1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 |
}; /* This is used for a general mmap of a disk file */ int generic_file_mmap(struct file * file, struct vm_area_struct * vma) { struct address_space *mapping = file->f_mapping; if (!mapping->a_ops->readpage) return -ENOEXEC; file_accessed(file); vma->vm_ops = &generic_file_vm_ops; |
d0217ac04 mm: fault feedbac... |
1564 |
vma->vm_flags |= VM_CAN_NONLINEAR; |
1da177e4c Linux-2.6.12-rc2 |
1565 1566 |
return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 |
/* * This is for filesystems which do not implement ->writepage. */ int generic_file_readonly_mmap(struct file *file, struct vm_area_struct *vma) { if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE)) return -EINVAL; return generic_file_mmap(file, vma); } #else int generic_file_mmap(struct file * file, struct vm_area_struct * vma) { return -ENOSYS; } int generic_file_readonly_mmap(struct file * file, struct vm_area_struct * vma) { return -ENOSYS; } #endif /* CONFIG_MMU */ EXPORT_SYMBOL(generic_file_mmap); EXPORT_SYMBOL(generic_file_readonly_mmap); |
6fe6900e1 mm: make read_cac... |
1590 |
static struct page *__read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
1591 |
pgoff_t index, |
1da177e4c Linux-2.6.12-rc2 |
1592 |
int (*filler)(void *,struct page*), |
0531b2aac mm: add new 'read... |
1593 1594 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
1595 |
{ |
eb2be1893 mm: buffered writ... |
1596 |
struct page *page; |
1da177e4c Linux-2.6.12-rc2 |
1597 1598 1599 1600 |
int err; repeat: page = find_get_page(mapping, index); if (!page) { |
0531b2aac mm: add new 'read... |
1601 |
page = __page_cache_alloc(gfp | __GFP_COLD); |
eb2be1893 mm: buffered writ... |
1602 1603 1604 1605 1606 1607 1608 |
if (!page) return ERR_PTR(-ENOMEM); err = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL); if (unlikely(err)) { page_cache_release(page); if (err == -EEXIST) goto repeat; |
1da177e4c Linux-2.6.12-rc2 |
1609 |
/* Presumably ENOMEM for radix tree node */ |
1da177e4c Linux-2.6.12-rc2 |
1610 1611 |
return ERR_PTR(err); } |
1da177e4c Linux-2.6.12-rc2 |
1612 1613 1614 1615 1616 1617 |
err = filler(data, page); if (err < 0) { page_cache_release(page); page = ERR_PTR(err); } } |
1da177e4c Linux-2.6.12-rc2 |
1618 1619 |
return page; } |
0531b2aac mm: add new 'read... |
1620 |
static struct page *do_read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
1621 |
pgoff_t index, |
1da177e4c Linux-2.6.12-rc2 |
1622 |
int (*filler)(void *,struct page*), |
0531b2aac mm: add new 'read... |
1623 1624 |
void *data, gfp_t gfp) |
1da177e4c Linux-2.6.12-rc2 |
1625 1626 1627 1628 1629 |
{ struct page *page; int err; retry: |
0531b2aac mm: add new 'read... |
1630 |
page = __read_cache_page(mapping, index, filler, data, gfp); |
1da177e4c Linux-2.6.12-rc2 |
1631 |
if (IS_ERR(page)) |
c855ff371 Fix a bad error c... |
1632 |
return page; |
1da177e4c Linux-2.6.12-rc2 |
1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 |
if (PageUptodate(page)) goto out; lock_page(page); if (!page->mapping) { unlock_page(page); page_cache_release(page); goto retry; } if (PageUptodate(page)) { unlock_page(page); goto out; } err = filler(data, page); if (err < 0) { page_cache_release(page); |
c855ff371 Fix a bad error c... |
1649 |
return ERR_PTR(err); |
1da177e4c Linux-2.6.12-rc2 |
1650 |
} |
c855ff371 Fix a bad error c... |
1651 |
out: |
6fe6900e1 mm: make read_cac... |
1652 1653 1654 |
mark_page_accessed(page); return page; } |
0531b2aac mm: add new 'read... |
1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 |
/** * read_cache_page_async - read into page cache, fill it if needed * @mapping: the page's address_space * @index: the page index * @filler: function to perform the read * @data: destination for read data * * Same as read_cache_page, but don't wait for page to become unlocked * after submitting it to the filler. * * Read into the page cache. If a page already exists, and PageUptodate() is * not set, try to fill the page but don't wait for it to become unlocked. * * If the page does not get brought uptodate, return -EIO. */ struct page *read_cache_page_async(struct address_space *mapping, pgoff_t index, int (*filler)(void *,struct page*), void *data) { return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); } |
6fe6900e1 mm: make read_cac... |
1678 |
EXPORT_SYMBOL(read_cache_page_async); |
0531b2aac mm: add new 'read... |
1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 |
static struct page *wait_on_page_read(struct page *page) { if (!IS_ERR(page)) { wait_on_page_locked(page); if (!PageUptodate(page)) { page_cache_release(page); page = ERR_PTR(-EIO); } } return page; } /** * read_cache_page_gfp - read into page cache, using specified page allocation flags. * @mapping: the page's address_space * @index: the page index * @gfp: the page allocator flags to use if allocating * * This is the same as "read_mapping_page(mapping, index, NULL)", but with * any new page allocations done using the specified allocation flags. Note * that the Radix tree operations will still use GFP_KERNEL, so you can't * expect to do this atomically or anything like that - but you can pass in * other page requirements. * * If the page does not get brought uptodate, return -EIO. */ struct page *read_cache_page_gfp(struct address_space *mapping, pgoff_t index, gfp_t gfp) { filler_t *filler = (filler_t *)mapping->a_ops->readpage; return wait_on_page_read(do_read_cache_page(mapping, index, filler, NULL, gfp)); } EXPORT_SYMBOL(read_cache_page_gfp); |
6fe6900e1 mm: make read_cac... |
1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 |
/** * read_cache_page - read into page cache, fill it if needed * @mapping: the page's address_space * @index: the page index * @filler: function to perform the read * @data: destination for read data * * Read into the page cache. If a page already exists, and PageUptodate() is * not set, try to fill the page then wait for it to become unlocked. * * If the page does not get brought uptodate, return -EIO. */ struct page *read_cache_page(struct address_space *mapping, |
57f6b96c0 filemap: convert ... |
1727 |
pgoff_t index, |
6fe6900e1 mm: make read_cac... |
1728 1729 1730 |
int (*filler)(void *,struct page*), void *data) { |
0531b2aac mm: add new 'read... |
1731 |
return wait_on_page_read(read_cache_page_async(mapping, index, filler, data)); |
1da177e4c Linux-2.6.12-rc2 |
1732 |
} |
1da177e4c Linux-2.6.12-rc2 |
1733 1734 1735 |
EXPORT_SYMBOL(read_cache_page); /* |
1da177e4c Linux-2.6.12-rc2 |
1736 1737 1738 1739 1740 |
* The logic we want is * * if suid or (sgid and xgrp) * remove privs */ |
01de85e05 [PATCH] Add lockl... |
1741 |
int should_remove_suid(struct dentry *dentry) |
1da177e4c Linux-2.6.12-rc2 |
1742 1743 1744 |
{ mode_t mode = dentry->d_inode->i_mode; int kill = 0; |
1da177e4c Linux-2.6.12-rc2 |
1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 |
/* suid always must be killed */ if (unlikely(mode & S_ISUID)) kill = ATTR_KILL_SUID; /* * sgid without any exec bits is just a mandatory locking mark; leave * it alone. If some exec bits are set, it's a real sgid; kill it. */ if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) kill |= ATTR_KILL_SGID; |
7f5ff766a kill suid bit onl... |
1756 |
if (unlikely(kill && !capable(CAP_FSETID) && S_ISREG(mode))) |
01de85e05 [PATCH] Add lockl... |
1757 |
return kill; |
1da177e4c Linux-2.6.12-rc2 |
1758 |
|
01de85e05 [PATCH] Add lockl... |
1759 1760 |
return 0; } |
d23a147bb [PATCH] Export sh... |
1761 |
EXPORT_SYMBOL(should_remove_suid); |
01de85e05 [PATCH] Add lockl... |
1762 |
|
7f3d4ee10 vfs: splice remov... |
1763 |
static int __remove_suid(struct dentry *dentry, int kill) |
01de85e05 [PATCH] Add lockl... |
1764 1765 1766 1767 1768 1769 |
{ struct iattr newattrs; newattrs.ia_valid = ATTR_FORCE | kill; return notify_change(dentry, &newattrs); } |
2f1936b87 [patch 3/5] vfs: ... |
1770 |
int file_remove_suid(struct file *file) |
01de85e05 [PATCH] Add lockl... |
1771 |
{ |
2f1936b87 [patch 3/5] vfs: ... |
1772 |
struct dentry *dentry = file->f_path.dentry; |
b53767719 Implement file po... |
1773 1774 1775 |
int killsuid = should_remove_suid(dentry); int killpriv = security_inode_need_killpriv(dentry); int error = 0; |
01de85e05 [PATCH] Add lockl... |
1776 |
|
b53767719 Implement file po... |
1777 1778 1779 1780 1781 1782 |
if (killpriv < 0) return killpriv; if (killpriv) error = security_inode_killpriv(dentry); if (!error && killsuid) error = __remove_suid(dentry, killsuid); |
01de85e05 [PATCH] Add lockl... |
1783 |
|
b53767719 Implement file po... |
1784 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
1785 |
} |
2f1936b87 [patch 3/5] vfs: ... |
1786 |
EXPORT_SYMBOL(file_remove_suid); |
1da177e4c Linux-2.6.12-rc2 |
1787 |
|
2f718ffc1 mm: buffered writ... |
1788 |
static size_t __iovec_copy_from_user_inatomic(char *vaddr, |
1da177e4c Linux-2.6.12-rc2 |
1789 1790 |
const struct iovec *iov, size_t base, size_t bytes) { |
f18005369 x86, mm: dont use... |
1791 |
size_t copied = 0, left = 0; |
1da177e4c Linux-2.6.12-rc2 |
1792 1793 1794 1795 1796 1797 |
while (bytes) { char __user *buf = iov->iov_base + base; int copy = min(bytes, iov->iov_len - base); base = 0; |
f18005369 x86, mm: dont use... |
1798 |
left = __copy_from_user_inatomic(vaddr, buf, copy); |
1da177e4c Linux-2.6.12-rc2 |
1799 1800 1801 1802 |
copied += copy; bytes -= copy; vaddr += copy; iov++; |
01408c493 [PATCH] Prepare f... |
1803 |
if (unlikely(left)) |
1da177e4c Linux-2.6.12-rc2 |
1804 |
break; |
1da177e4c Linux-2.6.12-rc2 |
1805 1806 1807 1808 1809 |
} return copied - left; } /* |
2f718ffc1 mm: buffered writ... |
1810 |
* Copy as much as we can into the page and return the number of bytes which |
af901ca18 tree-wide: fix as... |
1811 |
* were successfully copied. If a fault is encountered then return the number of |
2f718ffc1 mm: buffered writ... |
1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 |
* bytes which were copied. */ size_t iov_iter_copy_from_user_atomic(struct page *page, struct iov_iter *i, unsigned long offset, size_t bytes) { char *kaddr; size_t copied; BUG_ON(!in_atomic()); kaddr = kmap_atomic(page, KM_USER0); if (likely(i->nr_segs == 1)) { int left; char __user *buf = i->iov->iov_base + i->iov_offset; |
f18005369 x86, mm: dont use... |
1825 |
left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); |
2f718ffc1 mm: buffered writ... |
1826 1827 1828 1829 1830 1831 1832 1833 1834 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } kunmap_atomic(kaddr, KM_USER0); return copied; } |
89e107877 fs: new cont helpers |
1835 |
EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); |
2f718ffc1 mm: buffered writ... |
1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 |
/* * This has the same sideeffects and return value as * iov_iter_copy_from_user_atomic(). * The difference is that it attempts to resolve faults. * Page must not be locked. */ size_t iov_iter_copy_from_user(struct page *page, struct iov_iter *i, unsigned long offset, size_t bytes) { char *kaddr; size_t copied; kaddr = kmap(page); if (likely(i->nr_segs == 1)) { int left; char __user *buf = i->iov->iov_base + i->iov_offset; |
f18005369 x86, mm: dont use... |
1853 |
left = __copy_from_user(kaddr + offset, buf, bytes); |
2f718ffc1 mm: buffered writ... |
1854 1855 1856 1857 1858 1859 1860 1861 |
copied = bytes - left; } else { copied = __iovec_copy_from_user_inatomic(kaddr + offset, i->iov, i->iov_offset, bytes); } kunmap(page); return copied; } |
89e107877 fs: new cont helpers |
1862 |
EXPORT_SYMBOL(iov_iter_copy_from_user); |
2f718ffc1 mm: buffered writ... |
1863 |
|
f7009264c iov_iter_advance(... |
1864 |
void iov_iter_advance(struct iov_iter *i, size_t bytes) |
2f718ffc1 mm: buffered writ... |
1865 |
{ |
f7009264c iov_iter_advance(... |
1866 |
BUG_ON(i->count < bytes); |
2f718ffc1 mm: buffered writ... |
1867 1868 |
if (likely(i->nr_segs == 1)) { i->iov_offset += bytes; |
f7009264c iov_iter_advance(... |
1869 |
i->count -= bytes; |
2f718ffc1 mm: buffered writ... |
1870 1871 1872 |
} else { const struct iovec *iov = i->iov; size_t base = i->iov_offset; |
124d3b704 fix writev regres... |
1873 1874 |
/* * The !iov->iov_len check ensures we skip over unlikely |
f7009264c iov_iter_advance(... |
1875 |
* zero-length segments (without overruning the iovec). |
124d3b704 fix writev regres... |
1876 |
*/ |
94ad374a0 Fix off-by-one er... |
1877 |
while (bytes || unlikely(i->count && !iov->iov_len)) { |
f7009264c iov_iter_advance(... |
1878 |
int copy; |
2f718ffc1 mm: buffered writ... |
1879 |
|
f7009264c iov_iter_advance(... |
1880 1881 1882 |
copy = min(bytes, iov->iov_len - base); BUG_ON(!i->count || i->count < copy); i->count -= copy; |
2f718ffc1 mm: buffered writ... |
1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 |
bytes -= copy; base += copy; if (iov->iov_len == base) { iov++; base = 0; } } i->iov = iov; i->iov_offset = base; } } |
89e107877 fs: new cont helpers |
1894 |
EXPORT_SYMBOL(iov_iter_advance); |
2f718ffc1 mm: buffered writ... |
1895 |
|
afddba49d fs: introduce wri... |
1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 |
/* * Fault in the first iovec of the given iov_iter, to a maximum length * of bytes. Returns 0 on success, or non-zero if the memory could not be * accessed (ie. because it is an invalid address). * * writev-intensive code may want this to prefault several iovecs -- that * would be possible (callers must not rely on the fact that _only_ the * first iovec will be faulted with the current implementation). */ int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) |
2f718ffc1 mm: buffered writ... |
1906 |
{ |
2f718ffc1 mm: buffered writ... |
1907 |
char __user *buf = i->iov->iov_base + i->iov_offset; |
afddba49d fs: introduce wri... |
1908 1909 |
bytes = min(bytes, i->iov->iov_len - i->iov_offset); return fault_in_pages_readable(buf, bytes); |
2f718ffc1 mm: buffered writ... |
1910 |
} |
89e107877 fs: new cont helpers |
1911 |
EXPORT_SYMBOL(iov_iter_fault_in_readable); |
2f718ffc1 mm: buffered writ... |
1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 |
/* * Return the count of just the current iov_iter segment. */ size_t iov_iter_single_seg_count(struct iov_iter *i) { const struct iovec *iov = i->iov; if (i->nr_segs == 1) return i->count; else return min(i->count, iov->iov_len - i->iov_offset); } |
89e107877 fs: new cont helpers |
1924 |
EXPORT_SYMBOL(iov_iter_single_seg_count); |
2f718ffc1 mm: buffered writ... |
1925 1926 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1927 1928 |
* Performs necessary checks before doing a write * |
485bb99b4 [PATCH] kernel-do... |
1929 |
* Can adjust writing position or amount of bytes to write. |
1da177e4c Linux-2.6.12-rc2 |
1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 |
* Returns appropriate error code that caller should return or * zero in case that write should be allowed. */ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk) { struct inode *inode = file->f_mapping->host; unsigned long limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; if (unlikely(*pos < 0)) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 |
if (!isblk) { /* FIXME: this is for backwards compatibility with 2.4 */ if (file->f_flags & O_APPEND) *pos = i_size_read(inode); if (limit != RLIM_INFINITY) { if (*pos >= limit) { send_sig(SIGXFSZ, current, 0); return -EFBIG; } if (*count > limit - (typeof(limit))*pos) { *count = limit - (typeof(limit))*pos; } } } /* * LFS rule */ if (unlikely(*pos + *count > MAX_NON_LFS && !(file->f_flags & O_LARGEFILE))) { if (*pos >= MAX_NON_LFS) { |
1da177e4c Linux-2.6.12-rc2 |
1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 |
return -EFBIG; } if (*count > MAX_NON_LFS - (unsigned long)*pos) { *count = MAX_NON_LFS - (unsigned long)*pos; } } /* * Are we about to exceed the fs block limit ? * * If we have written data it becomes a short write. If we have * exceeded without writing data we send a signal and return EFBIG. * Linus frestrict idea will clean these up nicely.. */ if (likely(!isblk)) { if (unlikely(*pos >= inode->i_sb->s_maxbytes)) { if (*count || *pos > inode->i_sb->s_maxbytes) { |
1da177e4c Linux-2.6.12-rc2 |
1979 1980 1981 1982 1983 1984 1985 1986 |
return -EFBIG; } /* zero-length writes at ->s_maxbytes are OK */ } if (unlikely(*pos + *count > inode->i_sb->s_maxbytes)) *count = inode->i_sb->s_maxbytes - *pos; } else { |
9361401eb [PATCH] BLOCK: Ma... |
1987 |
#ifdef CONFIG_BLOCK |
1da177e4c Linux-2.6.12-rc2 |
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 |
loff_t isize; if (bdev_read_only(I_BDEV(inode))) return -EPERM; isize = i_size_read(inode); if (*pos >= isize) { if (*count || *pos > isize) return -ENOSPC; } if (*pos + *count > isize) *count = isize - *pos; |
9361401eb [PATCH] BLOCK: Ma... |
1999 2000 2001 |
#else return -EPERM; #endif |
1da177e4c Linux-2.6.12-rc2 |
2002 2003 2004 2005 |
} return 0; } EXPORT_SYMBOL(generic_write_checks); |
afddba49d fs: introduce wri... |
2006 2007 2008 2009 2010 |
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... |
2011 |
return aops->write_begin(file, mapping, pos, len, flags, |
afddba49d fs: introduce wri... |
2012 |
pagep, fsdata); |
afddba49d fs: introduce wri... |
2013 2014 2015 2016 2017 2018 2019 2020 |
} 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... |
2021 |
|
4e02ed4b4 fs: remove prepar... |
2022 2023 |
mark_page_accessed(page); return aops->write_end(file, mapping, pos, len, copied, page, fsdata); |
afddba49d fs: introduce wri... |
2024 2025 |
} EXPORT_SYMBOL(pagecache_write_end); |
1da177e4c Linux-2.6.12-rc2 |
2026 2027 2028 2029 2030 2031 2032 2033 2034 |
ssize_t generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, unsigned long *nr_segs, loff_t pos, loff_t *ppos, size_t count, size_t ocount) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; ssize_t written; |
a969e903a kill generic_file... |
2035 2036 |
size_t write_len; pgoff_t end; |
1da177e4c Linux-2.6.12-rc2 |
2037 2038 2039 |
if (count != ocount) *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); |
a969e903a kill generic_file... |
2040 2041 |
write_len = iov_length(iov, *nr_segs); end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; |
a969e903a kill generic_file... |
2042 |
|
48b47c561 mm: direct IO sta... |
2043 |
written = filemap_write_and_wait_range(mapping, pos, pos + write_len - 1); |
a969e903a kill generic_file... |
2044 2045 2046 2047 2048 2049 2050 |
if (written) goto out; /* * After a write we want buffered reads to be sure to go to disk to get * the new data. We invalidate clean cached page from the region we're * about to write. We do this *before* the write so that we can return |
6ccfa806a VFS: fix dio writ... |
2051 |
* without clobbering -EIOCBQUEUED from ->direct_IO(). |
a969e903a kill generic_file... |
2052 2053 2054 2055 |
*/ if (mapping->nrpages) { written = invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); |
6ccfa806a VFS: fix dio writ... |
2056 2057 2058 2059 2060 2061 2062 |
/* * If a page can not be invalidated, return 0 to fall back * to buffered write. */ if (written) { if (written == -EBUSY) return 0; |
a969e903a kill generic_file... |
2063 |
goto out; |
6ccfa806a VFS: fix dio writ... |
2064 |
} |
a969e903a kill generic_file... |
2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 |
} written = mapping->a_ops->direct_IO(WRITE, iocb, iov, pos, *nr_segs); /* * Finally, try again to invalidate clean pages which might have been * cached by non-direct readahead, or faulted in by get_user_pages() * if the source of the write was an mmap'ed region of the file * we're writing. Either one is a pretty crazy thing to do, * so we don't support it 100%. If this invalidation * fails, tough, the write still worked... */ if (mapping->nrpages) { invalidate_inode_pages2_range(mapping, pos >> PAGE_CACHE_SHIFT, end); } |
1da177e4c Linux-2.6.12-rc2 |
2081 2082 2083 2084 2085 2086 2087 2088 |
if (written > 0) { loff_t end = pos + written; if (end > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { i_size_write(inode, end); mark_inode_dirty(inode); } *ppos = end; } |
a969e903a kill generic_file... |
2089 |
out: |
1da177e4c Linux-2.6.12-rc2 |
2090 2091 2092 |
return written; } EXPORT_SYMBOL(generic_file_direct_write); |
eb2be1893 mm: buffered writ... |
2093 2094 2095 2096 |
/* * 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... |
2097 2098 |
struct page *grab_cache_page_write_begin(struct address_space *mapping, pgoff_t index, unsigned flags) |
eb2be1893 mm: buffered writ... |
2099 2100 2101 |
{ int status; struct page *page; |
54566b2c1 fs: symlink write... |
2102 2103 2104 |
gfp_t gfp_notmask = 0; if (flags & AOP_FLAG_NOFS) gfp_notmask = __GFP_FS; |
eb2be1893 mm: buffered writ... |
2105 2106 2107 2108 |
repeat: page = find_lock_page(mapping, index); if (likely(page)) return page; |
54566b2c1 fs: symlink write... |
2109 |
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~gfp_notmask); |
eb2be1893 mm: buffered writ... |
2110 2111 |
if (!page) return NULL; |
54566b2c1 fs: symlink write... |
2112 2113 |
status = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL & ~gfp_notmask); |
eb2be1893 mm: buffered writ... |
2114 2115 2116 2117 2118 2119 2120 2121 |
if (unlikely(status)) { page_cache_release(page); if (status == -EEXIST) goto repeat; return NULL; } return page; } |
54566b2c1 fs: symlink write... |
2122 |
EXPORT_SYMBOL(grab_cache_page_write_begin); |
eb2be1893 mm: buffered writ... |
2123 |
|
afddba49d fs: introduce wri... |
2124 2125 2126 2127 2128 2129 2130 |
static ssize_t generic_perform_write(struct file *file, struct iov_iter *i, loff_t pos) { struct address_space *mapping = file->f_mapping; const struct address_space_operations *a_ops = mapping->a_ops; long status = 0; ssize_t written = 0; |
674b892ed mm: restore KERNE... |
2131 2132 2133 2134 2135 2136 2137 |
unsigned int flags = 0; /* * Copies from kernel address space cannot fail (NFSD is a big user). */ if (segment_eq(get_fs(), KERNEL_DS)) flags |= AOP_FLAG_UNINTERRUPTIBLE; |
afddba49d fs: introduce wri... |
2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 |
do { struct page *page; pgoff_t index; /* Pagecache index for current page */ unsigned long offset; /* Offset into pagecache page */ unsigned long bytes; /* Bytes to write to page */ size_t copied; /* Bytes copied from user */ void *fsdata; offset = (pos & (PAGE_CACHE_SIZE - 1)); index = pos >> PAGE_CACHE_SHIFT; bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_count(i)); again: /* * Bring in the user page that we will copy from _first_. * Otherwise there's a nasty deadlock on copying from the * same page as we're writing to, without it being marked * up-to-date. * * Not only is this an optimisation, but it is also required * to check that the address is actually valid, when atomic * usercopies are used, below. */ if (unlikely(iov_iter_fault_in_readable(i, bytes))) { status = -EFAULT; break; } |
674b892ed mm: restore KERNE... |
2168 |
status = a_ops->write_begin(file, mapping, pos, bytes, flags, |
afddba49d fs: introduce wri... |
2169 2170 2171 2172 2173 2174 2175 2176 |
&page, &fsdata); if (unlikely(status)) break; pagefault_disable(); copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); pagefault_enable(); flush_dcache_page(page); |
c8236db9c mm: mark page acc... |
2177 |
mark_page_accessed(page); |
afddba49d fs: introduce wri... |
2178 2179 2180 2181 2182 2183 2184 |
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... |
2185 |
iov_iter_advance(i, copied); |
afddba49d fs: introduce wri... |
2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 |
if (unlikely(copied == 0)) { /* * If we were unable to copy any data at all, we must * fall back to a single segment length write. * * If we didn't fallback here, we could livelock * because not all segments in the iov can be copied at * once without a pagefault. */ bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_single_seg_count(i)); goto again; } |
afddba49d fs: introduce wri... |
2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 |
pos += copied; written += copied; balance_dirty_pages_ratelimited(mapping); } while (iov_iter_count(i)); return written ? written : status; } ssize_t generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos, loff_t *ppos, size_t count, ssize_t written) { struct file *file = iocb->ki_filp; |
afddba49d fs: introduce wri... |
2215 2216 2217 2218 |
ssize_t status; struct iov_iter i; iov_iter_init(&i, iov, nr_segs, count, written); |
4e02ed4b4 fs: remove prepar... |
2219 |
status = generic_perform_write(file, &i, pos); |
1da177e4c Linux-2.6.12-rc2 |
2220 |
|
1da177e4c Linux-2.6.12-rc2 |
2221 |
if (likely(status >= 0)) { |
afddba49d fs: introduce wri... |
2222 2223 |
written += status; *ppos = pos + status; |
1da177e4c Linux-2.6.12-rc2 |
2224 2225 |
} |
1da177e4c Linux-2.6.12-rc2 |
2226 2227 2228 |
return written ? written : status; } EXPORT_SYMBOL(generic_file_buffered_write); |
e4dd9de3c vfs: Export __gen... |
2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 |
/** * __generic_file_aio_write - write data to a file * @iocb: IO state structure (file, offset, etc.) * @iov: vector with data to write * @nr_segs: number of segments in the vector * @ppos: position where to write * * This function does all the work needed for actually writing data to a * file. It does all basic checks, removes SUID from the file, updates * modification times and calls proper subroutines depending on whether we * do direct IO or a standard buffered write. * * It expects i_mutex to be grabbed unless we work on a block device or similar * object which does not need locking at all. * * This function does *not* take care of syncing data in case of O_SYNC write. * A caller has to handle it. This is mainly due to the fact that we want to * avoid syncing under i_mutex. */ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t *ppos) |
1da177e4c Linux-2.6.12-rc2 |
2250 2251 |
{ struct file *file = iocb->ki_filp; |
fb5527e68 [PATCH] direct-io... |
2252 |
struct address_space * mapping = file->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
2253 2254 2255 |
size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
2256 2257 2258 2259 2260 |
loff_t pos; ssize_t written; ssize_t err; ocount = 0; |
0ceb33143 mm: move common s... |
2261 2262 2263 |
err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (err) return err; |
1da177e4c Linux-2.6.12-rc2 |
2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 |
count = ocount; pos = *ppos; vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); /* We can write back this queue in page reclaim */ current->backing_dev_info = mapping->backing_dev_info; written = 0; err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); if (err) goto out; if (count == 0) goto out; |
2f1936b87 [patch 3/5] vfs: ... |
2280 |
err = file_remove_suid(file); |
1da177e4c Linux-2.6.12-rc2 |
2281 2282 |
if (err) goto out; |
870f48179 [PATCH] replace i... |
2283 |
file_update_time(file); |
1da177e4c Linux-2.6.12-rc2 |
2284 2285 2286 |
/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { |
fb5527e68 [PATCH] direct-io... |
2287 2288 2289 2290 2291 |
loff_t endbyte; ssize_t written_buffered; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, ppos, count, ocount); |
1da177e4c Linux-2.6.12-rc2 |
2292 2293 2294 2295 2296 2297 2298 2299 |
if (written < 0 || written == count) goto out; /* * direct-io write to a hole: fall through to buffered I/O * for completing the rest of the request. */ pos += written; count -= written; |
fb5527e68 [PATCH] direct-io... |
2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 |
written_buffered = generic_file_buffered_write(iocb, iov, nr_segs, pos, ppos, count, written); /* * If generic_file_buffered_write() retuned a synchronous error * then we want to return the number of bytes which were * direct-written, or the error code if that was zero. Note * that this differs from normal direct-io semantics, which * will return -EFOO even if some bytes were written. */ if (written_buffered < 0) { err = written_buffered; goto out; } |
1da177e4c Linux-2.6.12-rc2 |
2314 |
|
fb5527e68 [PATCH] direct-io... |
2315 2316 2317 2318 2319 2320 |
/* * We need to ensure that the page cache pages are written to * disk and invalidated to preserve the expected O_DIRECT * semantics. */ endbyte = pos + written_buffered - written - 1; |
c05c4edd8 direct I/O fallba... |
2321 |
err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); |
fb5527e68 [PATCH] direct-io... |
2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 |
if (err == 0) { written = written_buffered; invalidate_mapping_pages(mapping, pos >> PAGE_CACHE_SHIFT, endbyte >> PAGE_CACHE_SHIFT); } else { /* * We don't know how much we wrote, so just return * the number of bytes which were direct-written */ } } else { written = generic_file_buffered_write(iocb, iov, nr_segs, pos, ppos, count, written); } |
1da177e4c Linux-2.6.12-rc2 |
2337 2338 2339 2340 |
out: current->backing_dev_info = NULL; return written ? written : err; } |
e4dd9de3c vfs: Export __gen... |
2341 |
EXPORT_SYMBOL(__generic_file_aio_write); |
e4dd9de3c vfs: Export __gen... |
2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 |
/** * generic_file_aio_write - write data to a file * @iocb: IO state structure * @iov: vector with data to write * @nr_segs: number of segments in the vector * @pos: position in file where to write * * This is a wrapper around __generic_file_aio_write() to be used by most * filesystems. It takes care of syncing the file in case of O_SYNC file * and acquires i_mutex as needed. */ |
027445c37 [PATCH] Vectorize... |
2353 2354 |
ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) |
1da177e4c Linux-2.6.12-rc2 |
2355 2356 |
{ struct file *file = iocb->ki_filp; |
148f948ba vfs: Introduce ne... |
2357 |
struct inode *inode = file->f_mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
2358 |
ssize_t ret; |
1da177e4c Linux-2.6.12-rc2 |
2359 2360 |
BUG_ON(iocb->ki_pos != pos); |
1b1dcc1b5 [PATCH] mutex sub... |
2361 |
mutex_lock(&inode->i_mutex); |
e4dd9de3c vfs: Export __gen... |
2362 |
ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); |
1b1dcc1b5 [PATCH] mutex sub... |
2363 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2364 |
|
148f948ba vfs: Introduce ne... |
2365 |
if (ret > 0 || ret == -EIOCBQUEUED) { |
1da177e4c Linux-2.6.12-rc2 |
2366 |
ssize_t err; |
148f948ba vfs: Introduce ne... |
2367 |
err = generic_write_sync(file, pos, ret); |
c7b50db21 vfs: Remove synci... |
2368 |
if (err < 0 && ret > 0) |
1da177e4c Linux-2.6.12-rc2 |
2369 2370 2371 2372 2373 |
ret = err; } return ret; } EXPORT_SYMBOL(generic_file_aio_write); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 |
/** * try_to_release_page() - release old fs-specific metadata on a page * * @page: the page which the kernel is trying to free * @gfp_mask: memory allocation flags (and I/O mode) * * The address_space is to try to release any data against the page * (presumably at page->private). If the release was successful, return `1'. * Otherwise return zero. * |
266cf658e FS-Cache: Recruit... |
2384 2385 2386 |
* 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... |
2387 |
* The @gfp_mask argument specifies whether I/O may be performed to release |
3f31fddfa jbd: fix race bet... |
2388 |
* this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS). |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2389 |
* |
cf9a2ae8d [PATCH] BLOCK: Mo... |
2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 |
*/ 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); |