Blame view
mm/shmem.c
71.6 KB
1da177e4c
|
1 2 3 4 5 6 7 8 |
/* * Resizable virtual memory filesystem for Linux. * * Copyright (C) 2000 Linus Torvalds. * 2000 Transmeta Corp. * 2000-2001 Christoph Rohland * 2000-2001 SAP AG * 2002 Red Hat Inc. |
0edd73b33
|
9 10 |
* Copyright (C) 2002-2005 Hugh Dickins. * Copyright (C) 2002-2005 VERITAS Software Corporation. |
1da177e4c
|
11 12 13 14 15 16 |
* Copyright (C) 2004 Andi Kleen, SuSE Labs * * Extended attribute support for tmpfs: * Copyright (c) 2004, Luke Kenneth Casson Leighton <lkcl@lkcl.net> * Copyright (c) 2004 Red Hat, Inc., James Morris <jmorris@redhat.com> * |
853ac43ab
|
17 18 19 |
* tiny-shmem: * Copyright (c) 2004, 2008 Matt Mackall <mpm@selenic.com> * |
1da177e4c
|
20 21 |
* This file is released under the GPL. */ |
853ac43ab
|
22 23 24 25 |
#include <linux/fs.h> #include <linux/init.h> #include <linux/vfs.h> #include <linux/mount.h> |
caefba174
|
26 |
#include <linux/pagemap.h> |
853ac43ab
|
27 28 29 |
#include <linux/file.h> #include <linux/mm.h> #include <linux/module.h> |
7e496299d
|
30 |
#include <linux/percpu_counter.h> |
853ac43ab
|
31 32 33 34 35 |
#include <linux/swap.h> static struct vfsmount *shm_mnt; #ifdef CONFIG_SHMEM |
1da177e4c
|
36 37 38 39 40 |
/* * This virtual memory filesystem is heavily based on the ramfs. It * extends ramfs by the ability to use swap and honor resource limits * which makes it a completely usable filesystem. */ |
39f0247d3
|
41 |
#include <linux/xattr.h> |
a56942551
|
42 |
#include <linux/exportfs.h> |
1c7c474c3
|
43 |
#include <linux/posix_acl.h> |
39f0247d3
|
44 |
#include <linux/generic_acl.h> |
1da177e4c
|
45 |
#include <linux/mman.h> |
1da177e4c
|
46 47 48 49 |
#include <linux/string.h> #include <linux/slab.h> #include <linux/backing-dev.h> #include <linux/shmem_fs.h> |
1da177e4c
|
50 |
#include <linux/writeback.h> |
1da177e4c
|
51 52 53 54 55 |
#include <linux/blkdev.h> #include <linux/security.h> #include <linux/swapops.h> #include <linux/mempolicy.h> #include <linux/namei.h> |
b00dc3ad7
|
56 |
#include <linux/ctype.h> |
304dbdb7a
|
57 |
#include <linux/migrate.h> |
c1f60a5a4
|
58 |
#include <linux/highmem.h> |
680d794ba
|
59 |
#include <linux/seq_file.h> |
925629278
|
60 |
#include <linux/magic.h> |
304dbdb7a
|
61 |
|
1da177e4c
|
62 63 64 |
#include <asm/uaccess.h> #include <asm/div64.h> #include <asm/pgtable.h> |
caefba174
|
65 66 67 68 69 70 71 72 73 74 75 76 |
/* * The maximum size of a shmem/tmpfs file is limited by the maximum size of * its triple-indirect swap vector - see illustration at shmem_swp_entry(). * * With 4kB page size, maximum file size is just over 2TB on a 32-bit kernel, * but one eighth of that on a 64-bit kernel. With 8kB page size, maximum * file size is just over 4TB on a 64-bit kernel, but 16TB on a 32-bit kernel, * MAX_LFS_FILESIZE being then more restrictive than swap vector layout. * * We use / and * instead of shifts in the definitions below, so that the swap * vector can be tested with small even values (e.g. 20) for ENTRIES_PER_PAGE. */ |
1da177e4c
|
77 |
#define ENTRIES_PER_PAGE (PAGE_CACHE_SIZE/sizeof(unsigned long)) |
61609d01c
|
78 |
#define ENTRIES_PER_PAGEPAGE ((unsigned long long)ENTRIES_PER_PAGE*ENTRIES_PER_PAGE) |
1da177e4c
|
79 |
|
caefba174
|
80 81 |
#define SHMSWP_MAX_INDEX (SHMEM_NR_DIRECT + (ENTRIES_PER_PAGEPAGE/2) * (ENTRIES_PER_PAGE+1)) #define SHMSWP_MAX_BYTES (SHMSWP_MAX_INDEX << PAGE_CACHE_SHIFT) |
1da177e4c
|
82 |
|
caefba174
|
83 84 85 86 |
#define SHMEM_MAX_BYTES min_t(unsigned long long, SHMSWP_MAX_BYTES, MAX_LFS_FILESIZE) #define SHMEM_MAX_INDEX ((unsigned long)((SHMEM_MAX_BYTES+1) >> PAGE_CACHE_SHIFT)) #define BLOCKS_PER_PAGE (PAGE_CACHE_SIZE/512) |
1da177e4c
|
87 88 89 90 91 92 93 94 95 96 97 |
#define VM_ACCT(size) (PAGE_CACHE_ALIGN(size) >> PAGE_SHIFT) /* info->flags needs VM_flags to handle pagein/truncate races efficiently */ #define SHMEM_PAGEIN VM_READ #define SHMEM_TRUNCATE VM_WRITE /* Definition to limit shmem_truncate's steps between cond_rescheds */ #define LATENCY_LIMIT 64 /* Pretend that each entry is of this size in directory's i_size */ #define BOGO_DIRENT_SIZE 20 |
1da177e4c
|
98 99 |
/* Flag allocation requirements to shmem_getpage and shmem_swp_alloc */ enum sgp_type { |
1da177e4c
|
100 101 |
SGP_READ, /* don't exceed i_size, don't allocate page */ SGP_CACHE, /* don't exceed i_size, may allocate page */ |
a0ee5ec52
|
102 |
SGP_DIRTY, /* like SGP_CACHE, but set new page dirty */ |
1da177e4c
|
103 104 |
SGP_WRITE, /* may exceed i_size, may allocate page */ }; |
b76db7354
|
105 |
#ifdef CONFIG_TMPFS |
680d794ba
|
106 107 108 109 110 111 112 113 114 |
static unsigned long shmem_default_max_blocks(void) { return totalram_pages / 2; } static unsigned long shmem_default_max_inodes(void) { return min(totalram_pages - totalhigh_pages, totalram_pages / 2); } |
b76db7354
|
115 |
#endif |
680d794ba
|
116 |
|
1da177e4c
|
117 118 |
static int shmem_getpage(struct inode *inode, unsigned long idx, struct page **pagep, enum sgp_type sgp, int *type); |
6daa0e286
|
119 |
static inline struct page *shmem_dir_alloc(gfp_t gfp_mask) |
1da177e4c
|
120 121 122 123 124 |
{ /* * The above definition of ENTRIES_PER_PAGE, and the use of * BLOCKS_PER_PAGE on indirect pages, assume PAGE_CACHE_SIZE: * might be reconsidered if it ever diverges from PAGE_SIZE. |
769848c03
|
125 |
* |
e12ba74d8
|
126 |
* Mobility flags are masked out as swap vectors cannot move |
1da177e4c
|
127 |
*/ |
e12ba74d8
|
128 |
return alloc_pages((gfp_mask & ~GFP_MOVABLE_MASK) | __GFP_ZERO, |
769848c03
|
129 |
PAGE_CACHE_SHIFT-PAGE_SHIFT); |
1da177e4c
|
130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 |
} static inline void shmem_dir_free(struct page *page) { __free_pages(page, PAGE_CACHE_SHIFT-PAGE_SHIFT); } static struct page **shmem_dir_map(struct page *page) { return (struct page **)kmap_atomic(page, KM_USER0); } static inline void shmem_dir_unmap(struct page **dir) { kunmap_atomic(dir, KM_USER0); } static swp_entry_t *shmem_swp_map(struct page *page) { return (swp_entry_t *)kmap_atomic(page, KM_USER1); } static inline void shmem_swp_balance_unmap(void) { /* * When passing a pointer to an i_direct entry, to code which * also handles indirect entries and so will shmem_swp_unmap, * we must arrange for the preempt count to remain in balance. * What kmap_atomic of a lowmem page does depends on config * and architecture, so pretend to kmap_atomic some lowmem page. */ (void) kmap_atomic(ZERO_PAGE(0), KM_USER1); } static inline void shmem_swp_unmap(swp_entry_t *entry) { kunmap_atomic(entry, KM_USER1); } static inline struct shmem_sb_info *SHMEM_SB(struct super_block *sb) { return sb->s_fs_info; } /* * shmem_file_setup pre-accounts the whole fixed size of a VM object, * for shared memory and for shared anonymous (/dev/zero) mappings * (unless MAP_NORESERVE and sysctl_overcommit_memory <= 1), * consistent with the pre-accounting of private mappings ... */ static inline int shmem_acct_size(unsigned long flags, loff_t size) { |
0b0a0806b
|
182 183 |
return (flags & VM_NORESERVE) ? 0 : security_vm_enough_memory_kern(VM_ACCT(size)); |
1da177e4c
|
184 185 186 187 |
} static inline void shmem_unacct_size(unsigned long flags, loff_t size) { |
0b0a0806b
|
188 |
if (!(flags & VM_NORESERVE)) |
1da177e4c
|
189 190 191 192 193 194 195 196 197 198 199 |
vm_unacct_memory(VM_ACCT(size)); } /* * ... whereas tmpfs objects are accounted incrementally as * pages are allocated, in order to allow huge sparse files. * shmem_getpage reports shmem_acct_block failure as -ENOSPC not -ENOMEM, * so that a failure on a sparse tmpfs mapping will give SIGBUS not OOM. */ static inline int shmem_acct_block(unsigned long flags) { |
0b0a0806b
|
200 201 |
return (flags & VM_NORESERVE) ? security_vm_enough_memory_kern(VM_ACCT(PAGE_CACHE_SIZE)) : 0; |
1da177e4c
|
202 203 204 205 |
} static inline void shmem_unacct_blocks(unsigned long flags, long pages) { |
0b0a0806b
|
206 |
if (flags & VM_NORESERVE) |
1da177e4c
|
207 208 |
vm_unacct_memory(pages * VM_ACCT(PAGE_CACHE_SIZE)); } |
759b9775c
|
209 |
static const struct super_operations shmem_ops; |
f5e54d6e5
|
210 |
static const struct address_space_operations shmem_aops; |
15ad7cdcf
|
211 |
static const struct file_operations shmem_file_operations; |
92e1d5be9
|
212 213 214 |
static const struct inode_operations shmem_inode_operations; static const struct inode_operations shmem_dir_inode_operations; static const struct inode_operations shmem_special_inode_operations; |
f0f37e2f7
|
215 |
static const struct vm_operations_struct shmem_vm_ops; |
1da177e4c
|
216 |
|
6c231b7ba
|
217 |
static struct backing_dev_info shmem_backing_dev_info __read_mostly = { |
1da177e4c
|
218 |
.ra_pages = 0, /* No readahead */ |
4f98a2fee
|
219 |
.capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK | BDI_CAP_SWAP_BACKED, |
1da177e4c
|
220 221 222 223 |
.unplug_io_fn = default_unplug_io_fn, }; static LIST_HEAD(shmem_swaplist); |
cb5f7b9a4
|
224 |
static DEFINE_MUTEX(shmem_swaplist_mutex); |
1da177e4c
|
225 226 227 228 |
static void shmem_free_blocks(struct inode *inode, long pages) { struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb); |
0edd73b33
|
229 |
if (sbinfo->max_blocks) { |
7e496299d
|
230 231 |
percpu_counter_add(&sbinfo->used_blocks, -pages); spin_lock(&inode->i_lock); |
1da177e4c
|
232 |
inode->i_blocks -= pages*BLOCKS_PER_PAGE; |
7e496299d
|
233 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
234 235 |
} } |
5b04c6890
|
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 |
static int shmem_reserve_inode(struct super_block *sb) { struct shmem_sb_info *sbinfo = SHMEM_SB(sb); if (sbinfo->max_inodes) { spin_lock(&sbinfo->stat_lock); if (!sbinfo->free_inodes) { spin_unlock(&sbinfo->stat_lock); return -ENOSPC; } sbinfo->free_inodes--; spin_unlock(&sbinfo->stat_lock); } return 0; } static void shmem_free_inode(struct super_block *sb) { struct shmem_sb_info *sbinfo = SHMEM_SB(sb); if (sbinfo->max_inodes) { spin_lock(&sbinfo->stat_lock); sbinfo->free_inodes++; spin_unlock(&sbinfo->stat_lock); } } |
467118102
|
260 |
/** |
1da177e4c
|
261 |
* shmem_recalc_inode - recalculate the size of an inode |
1da177e4c
|
262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 |
* @inode: inode to recalc * * We have to calculate the free blocks since the mm can drop * undirtied hole pages behind our back. * * But normally info->alloced == inode->i_mapping->nrpages + info->swapped * So mm freed is info->alloced - (inode->i_mapping->nrpages + info->swapped) * * It has to be called with the spinlock held. */ static void shmem_recalc_inode(struct inode *inode) { struct shmem_inode_info *info = SHMEM_I(inode); long freed; freed = info->alloced - info->swapped - inode->i_mapping->nrpages; if (freed > 0) { info->alloced -= freed; shmem_unacct_blocks(info->flags, freed); shmem_free_blocks(inode, freed); } } |
467118102
|
284 |
/** |
1da177e4c
|
285 |
* shmem_swp_entry - find the swap vector position in the info structure |
1da177e4c
|
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 |
* @info: info structure for the inode * @index: index of the page to find * @page: optional page to add to the structure. Has to be preset to * all zeros * * If there is no space allocated yet it will return NULL when * page is NULL, else it will use the page for the needed block, * setting it to NULL on return to indicate that it has been used. * * The swap vector is organized the following way: * * There are SHMEM_NR_DIRECT entries directly stored in the * shmem_inode_info structure. So small files do not need an addional * allocation. * * For pages with index > SHMEM_NR_DIRECT there is the pointer * i_indirect which points to a page which holds in the first half * doubly indirect blocks, in the second half triple indirect blocks: * * For an artificial ENTRIES_PER_PAGE = 4 this would lead to the * following layout (for SHMEM_NR_DIRECT == 16): * * i_indirect -> dir --> 16-19 * | +-> 20-23 * | * +-->dir2 --> 24-27 * | +-> 28-31 * | +-> 32-35 * | +-> 36-39 * | * +-->dir3 --> 40-43 * +-> 44-47 * +-> 48-51 * +-> 52-55 */ static swp_entry_t *shmem_swp_entry(struct shmem_inode_info *info, unsigned long index, struct page **page) { unsigned long offset; struct page **dir; struct page *subdir; if (index < SHMEM_NR_DIRECT) { shmem_swp_balance_unmap(); return info->i_direct+index; } if (!info->i_indirect) { if (page) { info->i_indirect = *page; *page = NULL; } return NULL; /* need another page */ } index -= SHMEM_NR_DIRECT; offset = index % ENTRIES_PER_PAGE; index /= ENTRIES_PER_PAGE; dir = shmem_dir_map(info->i_indirect); if (index >= ENTRIES_PER_PAGE/2) { index -= ENTRIES_PER_PAGE/2; dir += ENTRIES_PER_PAGE/2 + index/ENTRIES_PER_PAGE; index %= ENTRIES_PER_PAGE; subdir = *dir; if (!subdir) { if (page) { *dir = *page; *page = NULL; } shmem_dir_unmap(dir); return NULL; /* need another page */ } shmem_dir_unmap(dir); dir = shmem_dir_map(subdir); } dir += index; subdir = *dir; if (!subdir) { if (!page || !(subdir = *page)) { shmem_dir_unmap(dir); return NULL; /* need a page */ } *dir = subdir; *page = NULL; } shmem_dir_unmap(dir); return shmem_swp_map(subdir) + offset; } static void shmem_swp_set(struct shmem_inode_info *info, swp_entry_t *entry, unsigned long value) { long incdec = value? 1: -1; entry->val = value; info->swapped += incdec; |
4c21e2f24
|
381 382 383 384 |
if ((unsigned long)(entry - info->i_direct) >= SHMEM_NR_DIRECT) { struct page *page = kmap_atomic_to_page(entry); set_page_private(page, page_private(page) + incdec); } |
1da177e4c
|
385 |
} |
467118102
|
386 |
/** |
1da177e4c
|
387 |
* shmem_swp_alloc - get the position of the swap entry for the page. |
1da177e4c
|
388 389 390 |
* @info: info structure for the inode * @index: index of the page to find * @sgp: check and recheck i_size? skip allocation? |
467118102
|
391 392 |
* * If the entry does not exist, allocate it. |
1da177e4c
|
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 |
*/ static swp_entry_t *shmem_swp_alloc(struct shmem_inode_info *info, unsigned long index, enum sgp_type sgp) { struct inode *inode = &info->vfs_inode; struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb); struct page *page = NULL; swp_entry_t *entry; if (sgp != SGP_WRITE && ((loff_t) index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) return ERR_PTR(-EINVAL); while (!(entry = shmem_swp_entry(info, index, &page))) { if (sgp == SGP_READ) return shmem_swp_map(ZERO_PAGE(0)); /* |
7e496299d
|
409 |
* Test used_blocks against 1 less max_blocks, since we have 1 data |
1da177e4c
|
410 411 412 |
* page (and perhaps indirect index pages) yet to allocate: * a waste to allocate index if we cannot allocate data. */ |
0edd73b33
|
413 |
if (sbinfo->max_blocks) { |
7e496299d
|
414 |
if (percpu_counter_compare(&sbinfo->used_blocks, (sbinfo->max_blocks - 1)) > 0) |
1da177e4c
|
415 |
return ERR_PTR(-ENOSPC); |
7e496299d
|
416 417 |
percpu_counter_inc(&sbinfo->used_blocks); spin_lock(&inode->i_lock); |
1da177e4c
|
418 |
inode->i_blocks += BLOCKS_PER_PAGE; |
7e496299d
|
419 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
420 421 422 |
} spin_unlock(&info->lock); |
769848c03
|
423 |
page = shmem_dir_alloc(mapping_gfp_mask(inode->i_mapping)); |
1da177e4c
|
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 |
spin_lock(&info->lock); if (!page) { shmem_free_blocks(inode, 1); return ERR_PTR(-ENOMEM); } if (sgp != SGP_WRITE && ((loff_t) index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) { entry = ERR_PTR(-EINVAL); break; } if (info->next_index <= index) info->next_index = index + 1; } if (page) { /* another task gave its page, or truncated the file */ shmem_free_blocks(inode, 1); shmem_dir_free(page); } if (info->next_index <= index && !IS_ERR(entry)) info->next_index = index + 1; return entry; } |
467118102
|
447 |
/** |
1da177e4c
|
448 |
* shmem_free_swp - free some swap entries in a directory |
1ae700063
|
449 450 451 |
* @dir: pointer to the directory * @edir: pointer after last entry of the directory * @punch_lock: pointer to spinlock when needed for the holepunch case |
1da177e4c
|
452 |
*/ |
1ae700063
|
453 454 |
static int shmem_free_swp(swp_entry_t *dir, swp_entry_t *edir, spinlock_t *punch_lock) |
1da177e4c
|
455 |
{ |
1ae700063
|
456 |
spinlock_t *punch_unlock = NULL; |
1da177e4c
|
457 458 459 460 461 |
swp_entry_t *ptr; int freed = 0; for (ptr = dir; ptr < edir; ptr++) { if (ptr->val) { |
1ae700063
|
462 463 464 465 466 467 468 |
if (unlikely(punch_lock)) { punch_unlock = punch_lock; punch_lock = NULL; spin_lock(punch_unlock); if (!ptr->val) continue; } |
1da177e4c
|
469 470 471 472 473 |
free_swap_and_cache(*ptr); *ptr = (swp_entry_t){0}; freed++; } } |
1ae700063
|
474 475 |
if (punch_unlock) spin_unlock(punch_unlock); |
1da177e4c
|
476 477 |
return freed; } |
1ae700063
|
478 479 |
static int shmem_map_and_free_swp(struct page *subdir, int offset, int limit, struct page ***dir, spinlock_t *punch_lock) |
1da177e4c
|
480 481 482 483 484 485 486 487 488 |
{ swp_entry_t *ptr; int freed = 0; ptr = shmem_swp_map(subdir); for (; offset < limit; offset += LATENCY_LIMIT) { int size = limit - offset; if (size > LATENCY_LIMIT) size = LATENCY_LIMIT; |
1ae700063
|
489 490 |
freed += shmem_free_swp(ptr+offset, ptr+offset+size, punch_lock); |
1da177e4c
|
491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 |
if (need_resched()) { shmem_swp_unmap(ptr); if (*dir) { shmem_dir_unmap(*dir); *dir = NULL; } cond_resched(); ptr = shmem_swp_map(subdir); } } shmem_swp_unmap(ptr); return freed; } static void shmem_free_pages(struct list_head *next) { struct page *page; int freed = 0; do { page = container_of(next, struct page, lru); next = next->next; shmem_dir_free(page); freed++; if (freed >= LATENCY_LIMIT) { cond_resched(); freed = 0; } } while (next); } |
f6b3ec238
|
521 |
static void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) |
1da177e4c
|
522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 |
{ struct shmem_inode_info *info = SHMEM_I(inode); unsigned long idx; unsigned long size; unsigned long limit; unsigned long stage; unsigned long diroff; struct page **dir; struct page *topdir; struct page *middir; struct page *subdir; swp_entry_t *ptr; LIST_HEAD(pages_to_free); long nr_pages_to_free = 0; long nr_swaps_freed = 0; int offset; int freed; |
a2646d1e6
|
539 |
int punch_hole; |
1ae700063
|
540 541 |
spinlock_t *needs_lock; spinlock_t *punch_lock; |
a2646d1e6
|
542 |
unsigned long upper_limit; |
1da177e4c
|
543 544 |
inode->i_ctime = inode->i_mtime = CURRENT_TIME; |
f6b3ec238
|
545 |
idx = (start + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; |
1da177e4c
|
546 547 548 549 550 |
if (idx >= info->next_index) return; spin_lock(&info->lock); info->flags |= SHMEM_TRUNCATE; |
f6b3ec238
|
551 552 |
if (likely(end == (loff_t) -1)) { limit = info->next_index; |
a2646d1e6
|
553 |
upper_limit = SHMEM_MAX_INDEX; |
f6b3ec238
|
554 |
info->next_index = idx; |
1ae700063
|
555 |
needs_lock = NULL; |
a2646d1e6
|
556 |
punch_hole = 0; |
f6b3ec238
|
557 |
} else { |
a2646d1e6
|
558 559 560 561 562 563 564 565 |
if (end + 1 >= inode->i_size) { /* we may free a little more */ limit = (inode->i_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; upper_limit = SHMEM_MAX_INDEX; } else { limit = (end + 1) >> PAGE_CACHE_SHIFT; upper_limit = limit; } |
1ae700063
|
566 |
needs_lock = &info->lock; |
f6b3ec238
|
567 568 |
punch_hole = 1; } |
1da177e4c
|
569 |
topdir = info->i_indirect; |
f6b3ec238
|
570 |
if (topdir && idx <= SHMEM_NR_DIRECT && !punch_hole) { |
1da177e4c
|
571 572 573 574 575 576 577 578 579 580 581 |
info->i_indirect = NULL; nr_pages_to_free++; list_add(&topdir->lru, &pages_to_free); } spin_unlock(&info->lock); if (info->swapped && idx < SHMEM_NR_DIRECT) { ptr = info->i_direct; size = limit; if (size > SHMEM_NR_DIRECT) size = SHMEM_NR_DIRECT; |
1ae700063
|
582 |
nr_swaps_freed = shmem_free_swp(ptr+idx, ptr+size, needs_lock); |
1da177e4c
|
583 |
} |
92a3d03aa
|
584 585 586 587 588 |
/* * If there are no indirect blocks or we are punching a hole * below indirect blocks, nothing to be done. */ |
a2646d1e6
|
589 |
if (!topdir || limit <= SHMEM_NR_DIRECT) |
1da177e4c
|
590 |
goto done2; |
1ae700063
|
591 592 593 594 595 596 597 598 599 600 601 602 |
/* * The truncation case has already dropped info->lock, and we're safe * because i_size and next_index have already been lowered, preventing * access beyond. But in the punch_hole case, we still need to take * the lock when updating the swap directory, because there might be * racing accesses by shmem_getpage(SGP_CACHE), shmem_unuse_inode or * shmem_writepage. However, whenever we find we can remove a whole * directory page (not at the misaligned start or end of the range), * we first NULLify its pointer in the level above, and then have no * need to take the lock when updating its contents: needs_lock and * punch_lock (either pointing to info->lock or NULL) manage this. */ |
a2646d1e6
|
603 |
upper_limit -= SHMEM_NR_DIRECT; |
1da177e4c
|
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 |
limit -= SHMEM_NR_DIRECT; idx = (idx > SHMEM_NR_DIRECT)? (idx - SHMEM_NR_DIRECT): 0; offset = idx % ENTRIES_PER_PAGE; idx -= offset; dir = shmem_dir_map(topdir); stage = ENTRIES_PER_PAGEPAGE/2; if (idx < ENTRIES_PER_PAGEPAGE/2) { middir = topdir; diroff = idx/ENTRIES_PER_PAGE; } else { dir += ENTRIES_PER_PAGE/2; dir += (idx - ENTRIES_PER_PAGEPAGE/2)/ENTRIES_PER_PAGEPAGE; while (stage <= idx) stage += ENTRIES_PER_PAGEPAGE; middir = *dir; if (*dir) { diroff = ((idx - ENTRIES_PER_PAGEPAGE/2) % ENTRIES_PER_PAGEPAGE) / ENTRIES_PER_PAGE; |
a2646d1e6
|
623 |
if (!diroff && !offset && upper_limit >= stage) { |
1ae700063
|
624 625 626 627 628 629 630 |
if (needs_lock) { spin_lock(needs_lock); *dir = NULL; spin_unlock(needs_lock); needs_lock = NULL; } else *dir = NULL; |
1da177e4c
|
631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 |
nr_pages_to_free++; list_add(&middir->lru, &pages_to_free); } shmem_dir_unmap(dir); dir = shmem_dir_map(middir); } else { diroff = 0; offset = 0; idx = stage; } } for (; idx < limit; idx += ENTRIES_PER_PAGE, diroff++) { if (unlikely(idx == stage)) { shmem_dir_unmap(dir); dir = shmem_dir_map(topdir) + ENTRIES_PER_PAGE/2 + idx/ENTRIES_PER_PAGEPAGE; while (!*dir) { dir++; idx += ENTRIES_PER_PAGEPAGE; if (idx >= limit) goto done1; } stage = idx + ENTRIES_PER_PAGEPAGE; middir = *dir; |
1ae700063
|
656 657 |
if (punch_hole) needs_lock = &info->lock; |
a2646d1e6
|
658 |
if (upper_limit >= stage) { |
1ae700063
|
659 660 661 662 663 664 665 |
if (needs_lock) { spin_lock(needs_lock); *dir = NULL; spin_unlock(needs_lock); needs_lock = NULL; } else *dir = NULL; |
a2646d1e6
|
666 667 668 |
nr_pages_to_free++; list_add(&middir->lru, &pages_to_free); } |
1da177e4c
|
669 670 671 672 673 |
shmem_dir_unmap(dir); cond_resched(); dir = shmem_dir_map(middir); diroff = 0; } |
1ae700063
|
674 |
punch_lock = needs_lock; |
1da177e4c
|
675 |
subdir = dir[diroff]; |
1ae700063
|
676 677 678 679 680 681 682 683 684 685 686 687 |
if (subdir && !offset && upper_limit-idx >= ENTRIES_PER_PAGE) { if (needs_lock) { spin_lock(needs_lock); dir[diroff] = NULL; spin_unlock(needs_lock); punch_lock = NULL; } else dir[diroff] = NULL; nr_pages_to_free++; list_add(&subdir->lru, &pages_to_free); } if (subdir && page_private(subdir) /* has swap entries */) { |
1da177e4c
|
688 689 690 691 |
size = limit - idx; if (size > ENTRIES_PER_PAGE) size = ENTRIES_PER_PAGE; freed = shmem_map_and_free_swp(subdir, |
1ae700063
|
692 |
offset, size, &dir, punch_lock); |
1da177e4c
|
693 694 695 |
if (!dir) dir = shmem_dir_map(middir); nr_swaps_freed += freed; |
1ae700063
|
696 |
if (offset || punch_lock) { |
1da177e4c
|
697 |
spin_lock(&info->lock); |
1ae700063
|
698 699 |
set_page_private(subdir, page_private(subdir) - freed); |
1da177e4c
|
700 |
spin_unlock(&info->lock); |
1ae700063
|
701 702 |
} else BUG_ON(page_private(subdir) != freed); |
1da177e4c
|
703 |
} |
1ae700063
|
704 |
offset = 0; |
1da177e4c
|
705 706 707 708 709 710 711 |
} done1: shmem_dir_unmap(dir); done2: if (inode->i_mapping->nrpages && (info->flags & SHMEM_PAGEIN)) { /* * Call truncate_inode_pages again: racing shmem_unuse_inode |
3889e6e76
|
712 713 714 715 716 |
* may have swizzled a page in from swap since * truncate_pagecache or generic_delete_inode did it, before we * lowered next_index. Also, though shmem_getpage checks * i_size before adding to cache, no recheck after: so fix the * narrow window there too. |
16a100190
|
717 718 719 720 721 |
* * Recalling truncate_inode_pages_range and unmap_mapping_range * every time for punch_hole (which never got a chance to clear * SHMEM_PAGEIN at the start of vmtruncate_range) is expensive, * yet hardly ever necessary: try to optimize them out later. |
1da177e4c
|
722 |
*/ |
f6b3ec238
|
723 |
truncate_inode_pages_range(inode->i_mapping, start, end); |
16a100190
|
724 725 726 |
if (punch_hole) unmap_mapping_range(inode->i_mapping, start, end - start, 1); |
1da177e4c
|
727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 |
} spin_lock(&info->lock); info->flags &= ~SHMEM_TRUNCATE; info->swapped -= nr_swaps_freed; if (nr_pages_to_free) shmem_free_blocks(inode, nr_pages_to_free); shmem_recalc_inode(inode); spin_unlock(&info->lock); /* * Empty swap vector directory pages to be freed? */ if (!list_empty(&pages_to_free)) { pages_to_free.prev->next = NULL; shmem_free_pages(pages_to_free.next); } } static int shmem_notify_change(struct dentry *dentry, struct iattr *attr) { struct inode *inode = dentry->d_inode; |
af5a30d8c
|
749 |
loff_t newsize = attr->ia_size; |
1da177e4c
|
750 |
int error; |
db78b877f
|
751 752 753 |
error = inode_change_ok(inode, attr); if (error) return error; |
af5a30d8c
|
754 755 |
if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE) && newsize != inode->i_size) { |
3889e6e76
|
756 757 758 |
struct page *page = NULL; if (newsize < inode->i_size) { |
1da177e4c
|
759 760 761 762 763 764 765 |
/* * If truncating down to a partial page, then * if that page is already allocated, hold it * in memory until the truncation is over, so * truncate_partial_page cannnot miss it were * it assigned to swap. */ |
3889e6e76
|
766 |
if (newsize & (PAGE_CACHE_SIZE-1)) { |
1da177e4c
|
767 |
(void) shmem_getpage(inode, |
3889e6e76
|
768 |
newsize >> PAGE_CACHE_SHIFT, |
1da177e4c
|
769 |
&page, SGP_READ, NULL); |
d3602444e
|
770 771 |
if (page) unlock_page(page); |
1da177e4c
|
772 773 774 775 776 777 778 779 |
} /* * Reset SHMEM_PAGEIN flag so that shmem_truncate can * detect if any pages might have been added to cache * after truncate_inode_pages. But we needn't bother * if it's being fully truncated to zero-length: the * nrpages check is efficient enough in that case. */ |
3889e6e76
|
780 |
if (newsize) { |
1da177e4c
|
781 782 783 784 785 786 |
struct shmem_inode_info *info = SHMEM_I(inode); spin_lock(&info->lock); info->flags &= ~SHMEM_PAGEIN; spin_unlock(&info->lock); } } |
3889e6e76
|
787 |
|
2c27c65ed
|
788 789 |
/* XXX(truncate): truncate_setsize should be called last */ truncate_setsize(inode, newsize); |
3889e6e76
|
790 791 |
if (page) page_cache_release(page); |
3889e6e76
|
792 |
shmem_truncate_range(inode, newsize, (loff_t)-1); |
1da177e4c
|
793 |
} |
db78b877f
|
794 |
setattr_copy(inode, attr); |
39f0247d3
|
795 |
#ifdef CONFIG_TMPFS_POSIX_ACL |
db78b877f
|
796 |
if (attr->ia_valid & ATTR_MODE) |
1c7c474c3
|
797 |
error = generic_acl_chmod(inode); |
39f0247d3
|
798 |
#endif |
1da177e4c
|
799 800 |
return error; } |
1f895f75d
|
801 |
static void shmem_evict_inode(struct inode *inode) |
1da177e4c
|
802 |
{ |
1da177e4c
|
803 |
struct shmem_inode_info *info = SHMEM_I(inode); |
3889e6e76
|
804 |
if (inode->i_mapping->a_ops == &shmem_aops) { |
fef266580
|
805 |
truncate_inode_pages(inode->i_mapping, 0); |
1da177e4c
|
806 807 |
shmem_unacct_size(info->flags, inode->i_size); inode->i_size = 0; |
3889e6e76
|
808 |
shmem_truncate_range(inode, 0, (loff_t)-1); |
1da177e4c
|
809 |
if (!list_empty(&info->swaplist)) { |
cb5f7b9a4
|
810 |
mutex_lock(&shmem_swaplist_mutex); |
1da177e4c
|
811 |
list_del_init(&info->swaplist); |
cb5f7b9a4
|
812 |
mutex_unlock(&shmem_swaplist_mutex); |
1da177e4c
|
813 814 |
} } |
0edd73b33
|
815 |
BUG_ON(inode->i_blocks); |
5b04c6890
|
816 |
shmem_free_inode(inode->i_sb); |
1f895f75d
|
817 |
end_writeback(inode); |
1da177e4c
|
818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 |
} static inline int shmem_find_swp(swp_entry_t entry, swp_entry_t *dir, swp_entry_t *edir) { swp_entry_t *ptr; for (ptr = dir; ptr < edir; ptr++) { if (ptr->val == entry.val) return ptr - dir; } return -1; } static int shmem_unuse_inode(struct shmem_inode_info *info, swp_entry_t entry, struct page *page) { struct inode *inode; unsigned long idx; unsigned long size; unsigned long limit; unsigned long stage; struct page **dir; struct page *subdir; swp_entry_t *ptr; int offset; |
d9fe526a8
|
842 |
int error; |
1da177e4c
|
843 844 845 846 |
idx = 0; ptr = info->i_direct; spin_lock(&info->lock); |
1b1b32f2c
|
847 848 849 850 |
if (!info->swapped) { list_del_init(&info->swaplist); goto lost2; } |
1da177e4c
|
851 852 853 854 855 |
limit = info->next_index; size = limit; if (size > SHMEM_NR_DIRECT) size = SHMEM_NR_DIRECT; offset = shmem_find_swp(entry, ptr, ptr+size); |
2e0e26c76
|
856 |
if (offset >= 0) |
1da177e4c
|
857 |
goto found; |
1da177e4c
|
858 859 860 861 862 863 864 865 866 |
if (!info->i_indirect) goto lost2; dir = shmem_dir_map(info->i_indirect); stage = SHMEM_NR_DIRECT + ENTRIES_PER_PAGEPAGE/2; for (idx = SHMEM_NR_DIRECT; idx < limit; idx += ENTRIES_PER_PAGE, dir++) { if (unlikely(idx == stage)) { shmem_dir_unmap(dir-1); |
cb5f7b9a4
|
867 868 869 870 871 872 873 874 |
if (cond_resched_lock(&info->lock)) { /* check it has not been truncated */ if (limit > info->next_index) { limit = info->next_index; if (idx >= limit) goto lost2; } } |
1da177e4c
|
875 876 877 878 879 880 881 882 883 884 885 886 887 888 |
dir = shmem_dir_map(info->i_indirect) + ENTRIES_PER_PAGE/2 + idx/ENTRIES_PER_PAGEPAGE; while (!*dir) { dir++; idx += ENTRIES_PER_PAGEPAGE; if (idx >= limit) goto lost1; } stage = idx + ENTRIES_PER_PAGEPAGE; subdir = *dir; shmem_dir_unmap(dir); dir = shmem_dir_map(subdir); } subdir = *dir; |
4c21e2f24
|
889 |
if (subdir && page_private(subdir)) { |
1da177e4c
|
890 891 892 893 894 |
ptr = shmem_swp_map(subdir); size = limit - idx; if (size > ENTRIES_PER_PAGE) size = ENTRIES_PER_PAGE; offset = shmem_find_swp(entry, ptr, ptr+size); |
2e0e26c76
|
895 |
shmem_swp_unmap(ptr); |
1da177e4c
|
896 897 898 899 |
if (offset >= 0) { shmem_dir_unmap(dir); goto found; } |
1da177e4c
|
900 901 902 903 904 905 906 907 908 |
} } lost1: shmem_dir_unmap(dir-1); lost2: spin_unlock(&info->lock); return 0; found: idx += offset; |
2e0e26c76
|
909 910 |
inode = igrab(&info->vfs_inode); spin_unlock(&info->lock); |
1b1b32f2c
|
911 912 |
/* * Move _head_ to start search for next from here. |
1f895f75d
|
913 |
* But be careful: shmem_evict_inode checks list_empty without taking |
1b1b32f2c
|
914 915 916 917 918 919 |
* mutex, and there's an instant in list_move_tail when info->swaplist * would appear empty, if it were the only one on shmem_swaplist. We * could avoid doing it if inode NULL; or use this minor optimization. */ if (shmem_swaplist.next != &info->swaplist) list_move_tail(&shmem_swaplist, &info->swaplist); |
2e0e26c76
|
920 921 922 923 924 |
mutex_unlock(&shmem_swaplist_mutex); error = 1; if (!inode) goto out; |
d13d14430
|
925 |
/* |
b5a84319a
|
926 927 928 |
* Charge page using GFP_KERNEL while we can wait. * Charged back to the user(not to caller) when swap account is used. * add_to_page_cache() will be called with GFP_NOWAIT. |
d13d14430
|
929 |
*/ |
82369553d
|
930 |
error = mem_cgroup_cache_charge(page, current->mm, GFP_KERNEL); |
b409f9fcf
|
931 932 |
if (error) goto out; |
82369553d
|
933 |
error = radix_tree_preload(GFP_KERNEL); |
69029cd55
|
934 935 936 937 |
if (error) { mem_cgroup_uncharge_cache_page(page); goto out; } |
b409f9fcf
|
938 |
error = 1; |
2e0e26c76
|
939 940 941 |
spin_lock(&info->lock); ptr = shmem_swp_entry(info, idx, NULL); |
69029cd55
|
942 |
if (ptr && ptr->val == entry.val) { |
e286781d5
|
943 |
error = add_to_page_cache_locked(page, inode->i_mapping, |
b409f9fcf
|
944 |
idx, GFP_NOWAIT); |
69029cd55
|
945 946 947 |
/* does mem_cgroup_uncharge_cache_page on error */ } else /* we must compensate for our precharge above */ mem_cgroup_uncharge_cache_page(page); |
d9fe526a8
|
948 949 |
if (error == -EEXIST) { struct page *filepage = find_get_page(inode->i_mapping, idx); |
2e0e26c76
|
950 |
error = 1; |
d9fe526a8
|
951 952 953 954 955 956 957 958 959 960 961 |
if (filepage) { /* * There might be a more uptodate page coming down * from a stacked writepage: forget our swappage if so. */ if (PageUptodate(filepage)) error = 0; page_cache_release(filepage); } } if (!error) { |
73b1262fa
|
962 963 |
delete_from_swap_cache(page); set_page_dirty(page); |
1da177e4c
|
964 |
info->flags |= SHMEM_PAGEIN; |
2e0e26c76
|
965 966 967 |
shmem_swp_set(info, ptr, 0); swap_free(entry); error = 1; /* not an error, but entry was found */ |
1da177e4c
|
968 |
} |
2e0e26c76
|
969 970 |
if (ptr) shmem_swp_unmap(ptr); |
1da177e4c
|
971 |
spin_unlock(&info->lock); |
b409f9fcf
|
972 |
radix_tree_preload_end(); |
2e0e26c76
|
973 974 975 976 977 |
out: unlock_page(page); page_cache_release(page); iput(inode); /* allows for NULL */ return error; |
1da177e4c
|
978 979 980 981 982 983 984 985 986 987 |
} /* * shmem_unuse() search for an eventually swapped out shmem page. */ int shmem_unuse(swp_entry_t entry, struct page *page) { struct list_head *p, *next; struct shmem_inode_info *info; int found = 0; |
cb5f7b9a4
|
988 |
mutex_lock(&shmem_swaplist_mutex); |
1da177e4c
|
989 990 |
list_for_each_safe(p, next, &shmem_swaplist) { info = list_entry(p, struct shmem_inode_info, swaplist); |
1b1b32f2c
|
991 |
found = shmem_unuse_inode(info, entry, page); |
cb5f7b9a4
|
992 |
cond_resched(); |
2e0e26c76
|
993 994 |
if (found) goto out; |
1da177e4c
|
995 |
} |
cb5f7b9a4
|
996 |
mutex_unlock(&shmem_swaplist_mutex); |
aaa468653
|
997 998 999 1000 1001 1002 1003 1004 |
/* * Can some race bring us here? We've been holding page lock, * so I think not; but would rather try again later than BUG() */ unlock_page(page); page_cache_release(page); out: return (found < 0) ? found : 0; |
1da177e4c
|
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 |
} /* * Move the page from the page cache to the swap cache. */ static int shmem_writepage(struct page *page, struct writeback_control *wbc) { struct shmem_inode_info *info; swp_entry_t *entry, swap; struct address_space *mapping; unsigned long index; struct inode *inode; BUG_ON(!PageLocked(page)); |
1da177e4c
|
1019 1020 1021 1022 1023 1024 |
mapping = page->mapping; index = page->index; inode = mapping->host; info = SHMEM_I(inode); if (info->flags & VM_LOCKED) goto redirty; |
d9fe526a8
|
1025 |
if (!total_swap_pages) |
1da177e4c
|
1026 |
goto redirty; |
d9fe526a8
|
1027 1028 1029 1030 1031 |
/* * shmem_backing_dev_info's capabilities prevent regular writeback or * sync from ever calling shmem_writepage; but a stacking filesystem * may use the ->writepage of its underlying filesystem, in which case * tmpfs should write out to swap only in response to memory pressure, |
5b0830cb9
|
1032 1033 1034 |
* and not for the writeback threads or sync. However, in those cases, * we do still want to check if there's a redundant swappage to be * discarded. |
d9fe526a8
|
1035 1036 1037 1038 1039 |
*/ if (wbc->for_reclaim) swap = get_swap_page(); else swap.val = 0; |
1da177e4c
|
1040 |
spin_lock(&info->lock); |
1da177e4c
|
1041 1042 1043 1044 1045 |
if (index >= info->next_index) { BUG_ON(!(info->flags & SHMEM_TRUNCATE)); goto unlock; } entry = shmem_swp_entry(info, index, NULL); |
d9fe526a8
|
1046 1047 1048 1049 1050 1051 1052 1053 1054 |
if (entry->val) { /* * The more uptodate page coming down from a stacked * writepage should replace our old swappage. */ free_swap_and_cache(*entry); shmem_swp_set(info, entry, 0); } shmem_recalc_inode(inode); |
1da177e4c
|
1055 |
|
d9fe526a8
|
1056 |
if (swap.val && add_to_swap_cache(page, swap, GFP_ATOMIC) == 0) { |
73b1262fa
|
1057 |
remove_from_page_cache(page); |
1da177e4c
|
1058 1059 |
shmem_swp_set(info, entry, swap.val); shmem_swp_unmap(entry); |
1b1b32f2c
|
1060 1061 1062 1063 |
if (list_empty(&info->swaplist)) inode = igrab(inode); else inode = NULL; |
1da177e4c
|
1064 |
spin_unlock(&info->lock); |
aaa468653
|
1065 |
swap_shmem_alloc(swap); |
d9fe526a8
|
1066 |
BUG_ON(page_mapped(page)); |
73b1262fa
|
1067 |
page_cache_release(page); /* pagecache ref */ |
9fab5619b
|
1068 |
swap_writepage(page, wbc); |
1b1b32f2c
|
1069 1070 1071 1072 1073 1074 1075 |
if (inode) { mutex_lock(&shmem_swaplist_mutex); /* move instead of add in case we're racing */ list_move_tail(&info->swaplist, &shmem_swaplist); mutex_unlock(&shmem_swaplist_mutex); iput(inode); } |
1da177e4c
|
1076 1077 1078 1079 1080 1081 |
return 0; } shmem_swp_unmap(entry); unlock: spin_unlock(&info->lock); |
2ca4532a4
|
1082 1083 1084 1085 |
/* * add_to_swap_cache() doesn't return -EEXIST, so we can safely * clear SWAP_HAS_CACHE flag. */ |
cb4b86ba4
|
1086 |
swapcache_free(swap, NULL); |
1da177e4c
|
1087 1088 |
redirty: set_page_dirty(page); |
d9fe526a8
|
1089 1090 1091 1092 |
if (wbc->for_reclaim) return AOP_WRITEPAGE_ACTIVATE; /* Return with page locked */ unlock_page(page); return 0; |
1da177e4c
|
1093 1094 1095 |
} #ifdef CONFIG_NUMA |
680d794ba
|
1096 |
#ifdef CONFIG_TMPFS |
71fe804b6
|
1097 |
static void shmem_show_mpol(struct seq_file *seq, struct mempolicy *mpol) |
680d794ba
|
1098 |
{ |
095f1fc4e
|
1099 |
char buffer[64]; |
680d794ba
|
1100 |
|
71fe804b6
|
1101 |
if (!mpol || mpol->mode == MPOL_DEFAULT) |
095f1fc4e
|
1102 |
return; /* show nothing */ |
680d794ba
|
1103 |
|
71fe804b6
|
1104 |
mpol_to_str(buffer, sizeof(buffer), mpol, 1); |
095f1fc4e
|
1105 1106 |
seq_printf(seq, ",mpol=%s", buffer); |
680d794ba
|
1107 |
} |
71fe804b6
|
1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 |
static struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) { struct mempolicy *mpol = NULL; if (sbinfo->mpol) { spin_lock(&sbinfo->stat_lock); /* prevent replace/use races */ mpol = sbinfo->mpol; mpol_get(mpol); spin_unlock(&sbinfo->stat_lock); } return mpol; } |
680d794ba
|
1120 |
#endif /* CONFIG_TMPFS */ |
02098feaa
|
1121 1122 |
static struct page *shmem_swapin(swp_entry_t entry, gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) |
1da177e4c
|
1123 |
{ |
52cd3b074
|
1124 |
struct mempolicy mpol, *spol; |
1da177e4c
|
1125 |
struct vm_area_struct pvma; |
c4cc6d07b
|
1126 |
struct page *page; |
1da177e4c
|
1127 |
|
52cd3b074
|
1128 1129 |
spol = mpol_cond_copy(&mpol, mpol_shared_policy_lookup(&info->policy, idx)); |
1da177e4c
|
1130 |
/* Create a pseudo vma that just contains the policy */ |
c4cc6d07b
|
1131 |
pvma.vm_start = 0; |
1da177e4c
|
1132 |
pvma.vm_pgoff = idx; |
c4cc6d07b
|
1133 |
pvma.vm_ops = NULL; |
52cd3b074
|
1134 |
pvma.vm_policy = spol; |
02098feaa
|
1135 |
page = swapin_readahead(entry, gfp, &pvma, 0); |
1da177e4c
|
1136 1137 |
return page; } |
02098feaa
|
1138 1139 |
static struct page *shmem_alloc_page(gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) |
1da177e4c
|
1140 1141 |
{ struct vm_area_struct pvma; |
1da177e4c
|
1142 |
|
c4cc6d07b
|
1143 1144 |
/* Create a pseudo vma that just contains the policy */ pvma.vm_start = 0; |
1da177e4c
|
1145 |
pvma.vm_pgoff = idx; |
c4cc6d07b
|
1146 1147 |
pvma.vm_ops = NULL; pvma.vm_policy = mpol_shared_policy_lookup(&info->policy, idx); |
52cd3b074
|
1148 1149 1150 1151 1152 |
/* * alloc_page_vma() will drop the shared policy reference */ return alloc_page_vma(gfp, &pvma, 0); |
1da177e4c
|
1153 |
} |
680d794ba
|
1154 1155 |
#else /* !CONFIG_NUMA */ #ifdef CONFIG_TMPFS |
71fe804b6
|
1156 |
static inline void shmem_show_mpol(struct seq_file *seq, struct mempolicy *p) |
680d794ba
|
1157 1158 1159 |
{ } #endif /* CONFIG_TMPFS */ |
02098feaa
|
1160 1161 |
static inline struct page *shmem_swapin(swp_entry_t entry, gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) |
1da177e4c
|
1162 |
{ |
02098feaa
|
1163 |
return swapin_readahead(entry, gfp, NULL, 0); |
1da177e4c
|
1164 |
} |
02098feaa
|
1165 1166 |
static inline struct page *shmem_alloc_page(gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) |
1da177e4c
|
1167 |
{ |
e84e2e132
|
1168 |
return alloc_page(gfp); |
1da177e4c
|
1169 |
} |
680d794ba
|
1170 |
#endif /* CONFIG_NUMA */ |
1da177e4c
|
1171 |
|
71fe804b6
|
1172 1173 1174 1175 1176 1177 |
#if !defined(CONFIG_NUMA) || !defined(CONFIG_TMPFS) static inline struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) { return NULL; } #endif |
1da177e4c
|
1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 |
/* * shmem_getpage - either get the page from swap or allocate a new one * * If we allocate a new one we do not mark it dirty. That's up to the * vm. If we swap it in we mark it dirty since we also free the swap * entry since a page cannot live in both the swap and page cache */ static int shmem_getpage(struct inode *inode, unsigned long idx, struct page **pagep, enum sgp_type sgp, int *type) { struct address_space *mapping = inode->i_mapping; struct shmem_inode_info *info = SHMEM_I(inode); struct shmem_sb_info *sbinfo; struct page *filepage = *pagep; struct page *swappage; |
ff36b8016
|
1193 |
struct page *prealloc_page = NULL; |
1da177e4c
|
1194 1195 |
swp_entry_t *entry; swp_entry_t swap; |
02098feaa
|
1196 |
gfp_t gfp; |
1da177e4c
|
1197 1198 1199 1200 |
int error; if (idx >= SHMEM_MAX_INDEX) return -EFBIG; |
54cb8821d
|
1201 1202 |
if (type) |
83c54070e
|
1203 |
*type = 0; |
54cb8821d
|
1204 |
|
1da177e4c
|
1205 1206 1207 1208 |
/* * Normally, filepage is NULL on entry, and either found * uptodate immediately, or allocated and zeroed, or read * in under swappage, which is then assigned to filepage. |
5402b976a
|
1209 |
* But shmem_readpage (required for splice) passes in a locked |
ae9764164
|
1210 1211 |
* filepage, which may be found not uptodate by other callers * too, and may need to be copied from the swappage read in. |
1da177e4c
|
1212 1213 1214 1215 1216 1217 |
*/ repeat: if (!filepage) filepage = find_lock_page(mapping, idx); if (filepage && PageUptodate(filepage)) goto done; |
02098feaa
|
1218 |
gfp = mapping_gfp_mask(mapping); |
b409f9fcf
|
1219 1220 1221 1222 1223 1224 1225 1226 1227 |
if (!filepage) { /* * Try to preload while we can wait, to not make a habit of * draining atomic reserves; but don't latch on to this cpu. */ error = radix_tree_preload(gfp & ~__GFP_HIGHMEM); if (error) goto failed; radix_tree_preload_end(); |
ff36b8016
|
1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 |
if (sgp != SGP_READ && !prealloc_page) { /* We don't care if this fails */ prealloc_page = shmem_alloc_page(gfp, info, idx); if (prealloc_page) { if (mem_cgroup_cache_charge(prealloc_page, current->mm, GFP_KERNEL)) { page_cache_release(prealloc_page); prealloc_page = NULL; } } } |
b409f9fcf
|
1239 |
} |
ff36b8016
|
1240 |
error = 0; |
1da177e4c
|
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 |
spin_lock(&info->lock); shmem_recalc_inode(inode); entry = shmem_swp_alloc(info, idx, sgp); if (IS_ERR(entry)) { spin_unlock(&info->lock); error = PTR_ERR(entry); goto failed; } swap = *entry; if (swap.val) { /* Look it up and read it in.. */ swappage = lookup_swap_cache(swap); if (!swappage) { shmem_swp_unmap(entry); |
1da177e4c
|
1257 |
/* here we actually do the io */ |
83c54070e
|
1258 |
if (type && !(*type & VM_FAULT_MAJOR)) { |
f8891e5e1
|
1259 |
__count_vm_event(PGMAJFAULT); |
83c54070e
|
1260 |
*type |= VM_FAULT_MAJOR; |
1da177e4c
|
1261 |
} |
f8891e5e1
|
1262 |
spin_unlock(&info->lock); |
02098feaa
|
1263 |
swappage = shmem_swapin(swap, gfp, info, idx); |
1da177e4c
|
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 |
if (!swappage) { spin_lock(&info->lock); entry = shmem_swp_alloc(info, idx, sgp); if (IS_ERR(entry)) error = PTR_ERR(entry); else { if (entry->val == swap.val) error = -ENOMEM; shmem_swp_unmap(entry); } spin_unlock(&info->lock); if (error) goto failed; goto repeat; } wait_on_page_locked(swappage); page_cache_release(swappage); goto repeat; } /* We have to do this with page locked to prevent races */ |
529ae9aaa
|
1285 |
if (!trylock_page(swappage)) { |
1da177e4c
|
1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 |
shmem_swp_unmap(entry); spin_unlock(&info->lock); wait_on_page_locked(swappage); page_cache_release(swappage); goto repeat; } if (PageWriteback(swappage)) { shmem_swp_unmap(entry); spin_unlock(&info->lock); wait_on_page_writeback(swappage); unlock_page(swappage); page_cache_release(swappage); goto repeat; } if (!PageUptodate(swappage)) { shmem_swp_unmap(entry); spin_unlock(&info->lock); unlock_page(swappage); page_cache_release(swappage); error = -EIO; goto failed; } if (filepage) { shmem_swp_set(info, entry, 0); shmem_swp_unmap(entry); delete_from_swap_cache(swappage); spin_unlock(&info->lock); copy_highpage(filepage, swappage); unlock_page(swappage); page_cache_release(swappage); flush_dcache_page(filepage); SetPageUptodate(filepage); set_page_dirty(filepage); swap_free(swap); |
e286781d5
|
1321 1322 |
} else if (!(error = add_to_page_cache_locked(swappage, mapping, idx, GFP_NOWAIT))) { |
1da177e4c
|
1323 1324 1325 |
info->flags |= SHMEM_PAGEIN; shmem_swp_set(info, entry, 0); shmem_swp_unmap(entry); |
73b1262fa
|
1326 |
delete_from_swap_cache(swappage); |
1da177e4c
|
1327 1328 |
spin_unlock(&info->lock); filepage = swappage; |
73b1262fa
|
1329 |
set_page_dirty(filepage); |
1da177e4c
|
1330 1331 1332 1333 |
swap_free(swap); } else { shmem_swp_unmap(entry); spin_unlock(&info->lock); |
82369553d
|
1334 |
if (error == -ENOMEM) { |
ae3abae64
|
1335 1336 1337 1338 1339 1340 |
/* * reclaim from proper memory cgroup and * call memcg's OOM if needed. */ error = mem_cgroup_shmem_charge_fallback( swappage, |
b5a84319a
|
1341 |
current->mm, |
c9b0ed514
|
1342 |
gfp); |
b5a84319a
|
1343 1344 1345 |
if (error) { unlock_page(swappage); page_cache_release(swappage); |
82369553d
|
1346 |
goto failed; |
b5a84319a
|
1347 |
} |
82369553d
|
1348 |
} |
b5a84319a
|
1349 1350 |
unlock_page(swappage); page_cache_release(swappage); |
1da177e4c
|
1351 1352 1353 1354 1355 1356 |
goto repeat; } } else if (sgp == SGP_READ && !filepage) { shmem_swp_unmap(entry); filepage = find_get_page(mapping, idx); if (filepage && |
529ae9aaa
|
1357 |
(!PageUptodate(filepage) || !trylock_page(filepage))) { |
1da177e4c
|
1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 |
spin_unlock(&info->lock); wait_on_page_locked(filepage); page_cache_release(filepage); filepage = NULL; goto repeat; } spin_unlock(&info->lock); } else { shmem_swp_unmap(entry); sbinfo = SHMEM_SB(inode->i_sb); |
0edd73b33
|
1368 |
if (sbinfo->max_blocks) { |
7e496299d
|
1369 |
if ((percpu_counter_compare(&sbinfo->used_blocks, sbinfo->max_blocks) > 0) || |
1da177e4c
|
1370 |
shmem_acct_block(info->flags)) { |
1da177e4c
|
1371 1372 1373 1374 |
spin_unlock(&info->lock); error = -ENOSPC; goto failed; } |
7e496299d
|
1375 1376 |
percpu_counter_inc(&sbinfo->used_blocks); spin_lock(&inode->i_lock); |
1da177e4c
|
1377 |
inode->i_blocks += BLOCKS_PER_PAGE; |
7e496299d
|
1378 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1379 1380 1381 1382 1383 1384 1385 |
} else if (shmem_acct_block(info->flags)) { spin_unlock(&info->lock); error = -ENOSPC; goto failed; } if (!filepage) { |
69029cd55
|
1386 |
int ret; |
ff36b8016
|
1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 |
if (!prealloc_page) { spin_unlock(&info->lock); filepage = shmem_alloc_page(gfp, info, idx); if (!filepage) { shmem_unacct_blocks(info->flags, 1); shmem_free_blocks(inode, 1); error = -ENOMEM; goto failed; } SetPageSwapBacked(filepage); |
1da177e4c
|
1397 |
|
ff36b8016
|
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 |
/* * Precharge page while we can wait, compensate * after */ error = mem_cgroup_cache_charge(filepage, current->mm, GFP_KERNEL); if (error) { page_cache_release(filepage); shmem_unacct_blocks(info->flags, 1); shmem_free_blocks(inode, 1); filepage = NULL; goto failed; } spin_lock(&info->lock); } else { filepage = prealloc_page; prealloc_page = NULL; SetPageSwapBacked(filepage); |
82369553d
|
1417 |
} |
1da177e4c
|
1418 1419 1420 1421 1422 1423 1424 |
entry = shmem_swp_alloc(info, idx, sgp); if (IS_ERR(entry)) error = PTR_ERR(entry); else { swap = *entry; shmem_swp_unmap(entry); } |
69029cd55
|
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 |
ret = error || swap.val; if (ret) mem_cgroup_uncharge_cache_page(filepage); else ret = add_to_page_cache_lru(filepage, mapping, idx, GFP_NOWAIT); /* * At add_to_page_cache_lru() failure, uncharge will * be done automatically. */ if (ret) { |
1da177e4c
|
1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 |
spin_unlock(&info->lock); page_cache_release(filepage); shmem_unacct_blocks(info->flags, 1); shmem_free_blocks(inode, 1); filepage = NULL; if (error) goto failed; goto repeat; } info->flags |= SHMEM_PAGEIN; } info->alloced++; spin_unlock(&info->lock); |
e84e2e132
|
1450 |
clear_highpage(filepage); |
1da177e4c
|
1451 1452 |
flush_dcache_page(filepage); SetPageUptodate(filepage); |
a0ee5ec52
|
1453 1454 |
if (sgp == SGP_DIRTY) set_page_dirty(filepage); |
1da177e4c
|
1455 1456 |
} done: |
d3602444e
|
1457 |
*pagep = filepage; |
ff36b8016
|
1458 1459 |
error = 0; goto out; |
1da177e4c
|
1460 1461 1462 1463 1464 1465 |
failed: if (*pagep != filepage) { unlock_page(filepage); page_cache_release(filepage); } |
ff36b8016
|
1466 1467 1468 1469 1470 |
out: if (prealloc_page) { mem_cgroup_uncharge_cache_page(prealloc_page); page_cache_release(prealloc_page); } |
1da177e4c
|
1471 1472 |
return error; } |
d0217ac04
|
1473 |
static int shmem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
1da177e4c
|
1474 |
{ |
d3ac7f892
|
1475 |
struct inode *inode = vma->vm_file->f_path.dentry->d_inode; |
1da177e4c
|
1476 |
int error; |
d0217ac04
|
1477 |
int ret; |
1da177e4c
|
1478 |
|
d0217ac04
|
1479 1480 |
if (((loff_t)vmf->pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) return VM_FAULT_SIGBUS; |
d00806b18
|
1481 |
|
27d54b398
|
1482 |
error = shmem_getpage(inode, vmf->pgoff, &vmf->page, SGP_CACHE, &ret); |
d0217ac04
|
1483 1484 |
if (error) return ((error == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS); |
1da177e4c
|
1485 |
|
83c54070e
|
1486 |
return ret | VM_FAULT_LOCKED; |
1da177e4c
|
1487 |
} |
1da177e4c
|
1488 |
#ifdef CONFIG_NUMA |
d8dc74f21
|
1489 |
static int shmem_set_policy(struct vm_area_struct *vma, struct mempolicy *new) |
1da177e4c
|
1490 |
{ |
d3ac7f892
|
1491 |
struct inode *i = vma->vm_file->f_path.dentry->d_inode; |
1da177e4c
|
1492 1493 |
return mpol_set_shared_policy(&SHMEM_I(i)->policy, vma, new); } |
d8dc74f21
|
1494 1495 |
static struct mempolicy *shmem_get_policy(struct vm_area_struct *vma, unsigned long addr) |
1da177e4c
|
1496 |
{ |
d3ac7f892
|
1497 |
struct inode *i = vma->vm_file->f_path.dentry->d_inode; |
1da177e4c
|
1498 1499 1500 1501 1502 1503 1504 1505 1506 |
unsigned long idx; idx = ((addr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; return mpol_shared_policy_lookup(&SHMEM_I(i)->policy, idx); } #endif int shmem_lock(struct file *file, int lock, struct user_struct *user) { |
d3ac7f892
|
1507 |
struct inode *inode = file->f_path.dentry->d_inode; |
1da177e4c
|
1508 1509 1510 1511 1512 1513 1514 1515 |
struct shmem_inode_info *info = SHMEM_I(inode); int retval = -ENOMEM; spin_lock(&info->lock); if (lock && !(info->flags & VM_LOCKED)) { if (!user_shm_lock(inode->i_size, user)) goto out_nomem; info->flags |= VM_LOCKED; |
89e004ea5
|
1516 |
mapping_set_unevictable(file->f_mapping); |
1da177e4c
|
1517 1518 1519 1520 |
} if (!lock && (info->flags & VM_LOCKED) && user) { user_shm_unlock(inode->i_size, user); info->flags &= ~VM_LOCKED; |
89e004ea5
|
1521 1522 |
mapping_clear_unevictable(file->f_mapping); scan_mapping_unevictable_pages(file->f_mapping); |
1da177e4c
|
1523 1524 |
} retval = 0; |
89e004ea5
|
1525 |
|
1da177e4c
|
1526 1527 1528 1529 |
out_nomem: spin_unlock(&info->lock); return retval; } |
9b83a6a85
|
1530 |
static int shmem_mmap(struct file *file, struct vm_area_struct *vma) |
1da177e4c
|
1531 1532 1533 |
{ file_accessed(file); vma->vm_ops = &shmem_vm_ops; |
d0217ac04
|
1534 |
vma->vm_flags |= VM_CAN_NONLINEAR; |
1da177e4c
|
1535 1536 |
return 0; } |
454abafe9
|
1537 1538 |
static struct inode *shmem_get_inode(struct super_block *sb, const struct inode *dir, int mode, dev_t dev, unsigned long flags) |
1da177e4c
|
1539 1540 1541 1542 |
{ struct inode *inode; struct shmem_inode_info *info; struct shmem_sb_info *sbinfo = SHMEM_SB(sb); |
5b04c6890
|
1543 1544 |
if (shmem_reserve_inode(sb)) return NULL; |
1da177e4c
|
1545 1546 1547 |
inode = new_inode(sb); if (inode) { |
454abafe9
|
1548 |
inode_init_owner(inode, dir, mode); |
1da177e4c
|
1549 |
inode->i_blocks = 0; |
1da177e4c
|
1550 1551 |
inode->i_mapping->backing_dev_info = &shmem_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; |
91828a405
|
1552 |
inode->i_generation = get_seconds(); |
1da177e4c
|
1553 1554 1555 |
info = SHMEM_I(inode); memset(info, 0, (char *)inode - (char *)info); spin_lock_init(&info->lock); |
0b0a0806b
|
1556 |
info->flags = flags & VM_NORESERVE; |
1da177e4c
|
1557 |
INIT_LIST_HEAD(&info->swaplist); |
72c04902d
|
1558 |
cache_no_acl(inode); |
1da177e4c
|
1559 1560 1561 |
switch (mode & S_IFMT) { default: |
39f0247d3
|
1562 |
inode->i_op = &shmem_special_inode_operations; |
1da177e4c
|
1563 1564 1565 |
init_special_inode(inode, mode, dev); break; case S_IFREG: |
14fcc23fd
|
1566 |
inode->i_mapping->a_ops = &shmem_aops; |
1da177e4c
|
1567 1568 |
inode->i_op = &shmem_inode_operations; inode->i_fop = &shmem_file_operations; |
71fe804b6
|
1569 1570 |
mpol_shared_policy_init(&info->policy, shmem_get_sbmpol(sbinfo)); |
1da177e4c
|
1571 1572 |
break; case S_IFDIR: |
d8c76e6f4
|
1573 |
inc_nlink(inode); |
1da177e4c
|
1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 |
/* Some things misbehave if size == 0 on a directory */ inode->i_size = 2 * BOGO_DIRENT_SIZE; inode->i_op = &shmem_dir_inode_operations; inode->i_fop = &simple_dir_operations; break; case S_IFLNK: /* * Must not load anything in the rbtree, * mpol_free_shared_policy will not be called. */ |
71fe804b6
|
1584 |
mpol_shared_policy_init(&info->policy, NULL); |
1da177e4c
|
1585 1586 |
break; } |
5b04c6890
|
1587 1588 |
} else shmem_free_inode(sb); |
1da177e4c
|
1589 1590 1591 1592 |
return inode; } #ifdef CONFIG_TMPFS |
92e1d5be9
|
1593 1594 |
static const struct inode_operations shmem_symlink_inode_operations; static const struct inode_operations shmem_symlink_inline_operations; |
1da177e4c
|
1595 1596 |
/* |
800d15a53
|
1597 |
* Normally tmpfs avoids the use of shmem_readpage and shmem_write_begin; |
ae9764164
|
1598 1599 |
* but providing them allows a tmpfs file to be used for splice, sendfile, and * below the loop driver, in the generic fashion that many filesystems support. |
1da177e4c
|
1600 |
*/ |
ae9764164
|
1601 1602 1603 1604 1605 1606 1607 |
static int shmem_readpage(struct file *file, struct page *page) { struct inode *inode = page->mapping->host; int error = shmem_getpage(inode, page->index, &page, SGP_CACHE, NULL); unlock_page(page); return error; } |
1da177e4c
|
1608 |
static int |
800d15a53
|
1609 1610 1611 |
shmem_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) |
1da177e4c
|
1612 |
{ |
800d15a53
|
1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 |
struct inode *inode = mapping->host; pgoff_t index = pos >> PAGE_CACHE_SHIFT; *pagep = NULL; return shmem_getpage(inode, index, pagep, SGP_WRITE, NULL); } static int shmem_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata) { struct inode *inode = mapping->host; |
d3602444e
|
1625 1626 |
if (pos + copied > inode->i_size) i_size_write(inode, pos + copied); |
800d15a53
|
1627 |
set_page_dirty(page); |
6746aff74
|
1628 |
unlock_page(page); |
800d15a53
|
1629 |
page_cache_release(page); |
800d15a53
|
1630 |
return copied; |
1da177e4c
|
1631 |
} |
1da177e4c
|
1632 1633 |
static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_t *desc, read_actor_t actor) { |
d3ac7f892
|
1634 |
struct inode *inode = filp->f_path.dentry->d_inode; |
1da177e4c
|
1635 1636 |
struct address_space *mapping = inode->i_mapping; unsigned long index, offset; |
a0ee5ec52
|
1637 1638 1639 1640 1641 1642 1643 1644 1645 |
enum sgp_type sgp = SGP_READ; /* * Might this read be for a stacking filesystem? Then when reading * holes of a sparse file, we actually need to allocate those pages, * and even mark them dirty, so it cannot exceed the max_blocks limit. */ if (segment_eq(get_fs(), KERNEL_DS)) sgp = SGP_DIRTY; |
1da177e4c
|
1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 |
index = *ppos >> PAGE_CACHE_SHIFT; offset = *ppos & ~PAGE_CACHE_MASK; for (;;) { struct page *page = NULL; unsigned long end_index, nr, ret; loff_t i_size = i_size_read(inode); end_index = i_size >> PAGE_CACHE_SHIFT; if (index > end_index) break; if (index == end_index) { nr = i_size & ~PAGE_CACHE_MASK; if (nr <= offset) break; } |
a0ee5ec52
|
1663 |
desc->error = shmem_getpage(inode, index, &page, sgp, NULL); |
1da177e4c
|
1664 1665 1666 1667 1668 |
if (desc->error) { if (desc->error == -EINVAL) desc->error = 0; break; } |
d3602444e
|
1669 1670 |
if (page) unlock_page(page); |
1da177e4c
|
1671 1672 1673 |
/* * We must evaluate after, since reads (unlike writes) |
1b1dcc1b5
|
1674 |
* are called without i_mutex protection against truncate |
1da177e4c
|
1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 |
*/ nr = PAGE_CACHE_SIZE; i_size = i_size_read(inode); end_index = i_size >> PAGE_CACHE_SHIFT; if (index == end_index) { nr = i_size & ~PAGE_CACHE_MASK; if (nr <= offset) { if (page) page_cache_release(page); break; } } nr -= offset; if (page) { /* * 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); /* * Mark the page accessed if we read the beginning. */ if (!offset) mark_page_accessed(page); |
b5810039a
|
1702 |
} else { |
1da177e4c
|
1703 |
page = ZERO_PAGE(0); |
b5810039a
|
1704 1705 |
page_cache_get(page); } |
1da177e4c
|
1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 |
/* * 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; page_cache_release(page); if (ret != nr || !desc->count) break; cond_resched(); } *ppos = ((loff_t) index << PAGE_CACHE_SHIFT) + offset; file_accessed(filp); } |
bcd78e496
|
1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 |
static ssize_t shmem_file_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { struct file *filp = iocb->ki_filp; ssize_t retval; unsigned long seg; size_t count; loff_t *ppos = &iocb->ki_pos; retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; for (seg = 0; seg < nr_segs; seg++) { read_descriptor_t desc; 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_shmem_file_read(filp, ppos, &desc, file_read_actor); retval += desc.written; if (desc.error) { retval = retval ?: desc.error; break; } if (desc.count > 0) break; } return retval; |
1da177e4c
|
1764 |
} |
726c33422
|
1765 |
static int shmem_statfs(struct dentry *dentry, struct kstatfs *buf) |
1da177e4c
|
1766 |
{ |
726c33422
|
1767 |
struct shmem_sb_info *sbinfo = SHMEM_SB(dentry->d_sb); |
1da177e4c
|
1768 1769 1770 1771 |
buf->f_type = TMPFS_MAGIC; buf->f_bsize = PAGE_CACHE_SIZE; buf->f_namelen = NAME_MAX; |
0edd73b33
|
1772 |
if (sbinfo->max_blocks) { |
1da177e4c
|
1773 |
buf->f_blocks = sbinfo->max_blocks; |
7e496299d
|
1774 1775 |
buf->f_bavail = buf->f_bfree = sbinfo->max_blocks - percpu_counter_sum(&sbinfo->used_blocks); |
0edd73b33
|
1776 1777 |
} if (sbinfo->max_inodes) { |
1da177e4c
|
1778 1779 |
buf->f_files = sbinfo->max_inodes; buf->f_ffree = sbinfo->free_inodes; |
1da177e4c
|
1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 |
} /* else leave those fields 0 like simple_statfs */ return 0; } /* * File creation. Allocate an inode, and we're done.. */ static int shmem_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) { |
0b0a0806b
|
1791 |
struct inode *inode; |
1da177e4c
|
1792 |
int error = -ENOSPC; |
454abafe9
|
1793 |
inode = shmem_get_inode(dir->i_sb, dir, mode, dev, VM_NORESERVE); |
1da177e4c
|
1794 |
if (inode) { |
570bc1c2e
|
1795 1796 1797 1798 1799 1800 1801 |
error = security_inode_init_security(inode, dir, NULL, NULL, NULL); if (error) { if (error != -EOPNOTSUPP) { iput(inode); return error; } |
39f0247d3
|
1802 |
} |
1c7c474c3
|
1803 1804 |
#ifdef CONFIG_TMPFS_POSIX_ACL error = generic_acl_init(inode, dir); |
39f0247d3
|
1805 1806 1807 |
if (error) { iput(inode); return error; |
570bc1c2e
|
1808 |
} |
718deb6b6
|
1809 1810 |
#else error = 0; |
1c7c474c3
|
1811 |
#endif |
1da177e4c
|
1812 1813 1814 1815 |
dir->i_size += BOGO_DIRENT_SIZE; dir->i_ctime = dir->i_mtime = CURRENT_TIME; d_instantiate(dentry, inode); dget(dentry); /* Extra count - pin the dentry in core */ |
1da177e4c
|
1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 |
} return error; } static int shmem_mkdir(struct inode *dir, struct dentry *dentry, int mode) { int error; if ((error = shmem_mknod(dir, dentry, mode | S_IFDIR, 0))) return error; |
d8c76e6f4
|
1826 |
inc_nlink(dir); |
1da177e4c
|
1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 |
return 0; } static int shmem_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) { return shmem_mknod(dir, dentry, mode | S_IFREG, 0); } /* * Link a file.. */ static int shmem_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) { struct inode *inode = old_dentry->d_inode; |
5b04c6890
|
1842 |
int ret; |
1da177e4c
|
1843 1844 1845 1846 1847 1848 |
/* * No ordinary (disk based) filesystem counts links as inodes; * but each new link needs a new dentry, pinning lowmem, and * tmpfs dentries cannot be pruned until they are unlinked. */ |
5b04c6890
|
1849 1850 1851 |
ret = shmem_reserve_inode(inode->i_sb); if (ret) goto out; |
1da177e4c
|
1852 1853 1854 |
dir->i_size += BOGO_DIRENT_SIZE; inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME; |
d8c76e6f4
|
1855 |
inc_nlink(inode); |
1da177e4c
|
1856 1857 1858 |
atomic_inc(&inode->i_count); /* New dentry reference */ dget(dentry); /* Extra pinning count for the created dentry */ d_instantiate(dentry, inode); |
5b04c6890
|
1859 1860 |
out: return ret; |
1da177e4c
|
1861 1862 1863 1864 1865 |
} static int shmem_unlink(struct inode *dir, struct dentry *dentry) { struct inode *inode = dentry->d_inode; |
5b04c6890
|
1866 1867 |
if (inode->i_nlink > 1 && !S_ISDIR(inode->i_mode)) shmem_free_inode(inode->i_sb); |
1da177e4c
|
1868 1869 1870 |
dir->i_size -= BOGO_DIRENT_SIZE; inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME; |
9a53c3a78
|
1871 |
drop_nlink(inode); |
1da177e4c
|
1872 1873 1874 1875 1876 1877 1878 1879 |
dput(dentry); /* Undo the count from "create" - this does all the work */ return 0; } static int shmem_rmdir(struct inode *dir, struct dentry *dentry) { if (!simple_empty(dentry)) return -ENOTEMPTY; |
9a53c3a78
|
1880 1881 |
drop_nlink(dentry->d_inode); drop_nlink(dir); |
1da177e4c
|
1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 |
return shmem_unlink(dir, dentry); } /* * The VFS layer already does all the dentry stuff for rename, * we just have to decrement the usage count for the target if * it exists so that the VFS layer correctly free's it when it * gets overwritten. */ static int shmem_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry) { struct inode *inode = old_dentry->d_inode; int they_are_dirs = S_ISDIR(inode->i_mode); if (!simple_empty(new_dentry)) return -ENOTEMPTY; if (new_dentry->d_inode) { (void) shmem_unlink(new_dir, new_dentry); if (they_are_dirs) |
9a53c3a78
|
1902 |
drop_nlink(old_dir); |
1da177e4c
|
1903 |
} else if (they_are_dirs) { |
9a53c3a78
|
1904 |
drop_nlink(old_dir); |
d8c76e6f4
|
1905 |
inc_nlink(new_dir); |
1da177e4c
|
1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 |
} old_dir->i_size -= BOGO_DIRENT_SIZE; new_dir->i_size += BOGO_DIRENT_SIZE; old_dir->i_ctime = old_dir->i_mtime = new_dir->i_ctime = new_dir->i_mtime = inode->i_ctime = CURRENT_TIME; return 0; } static int shmem_symlink(struct inode *dir, struct dentry *dentry, const char *symname) { int error; int len; struct inode *inode; struct page *page = NULL; char *kaddr; struct shmem_inode_info *info; len = strlen(symname) + 1; if (len > PAGE_CACHE_SIZE) return -ENAMETOOLONG; |
454abafe9
|
1928 |
inode = shmem_get_inode(dir->i_sb, dir, S_IFLNK|S_IRWXUGO, 0, VM_NORESERVE); |
1da177e4c
|
1929 1930 |
if (!inode) return -ENOSPC; |
570bc1c2e
|
1931 1932 1933 1934 1935 1936 1937 1938 1939 |
error = security_inode_init_security(inode, dir, NULL, NULL, NULL); if (error) { if (error != -EOPNOTSUPP) { iput(inode); return error; } error = 0; } |
1da177e4c
|
1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 |
info = SHMEM_I(inode); inode->i_size = len-1; if (len <= (char *)inode - (char *)info) { /* do it inline */ memcpy(info, symname, len); inode->i_op = &shmem_symlink_inline_operations; } else { error = shmem_getpage(inode, 0, &page, SGP_WRITE, NULL); if (error) { iput(inode); return error; } |
14fcc23fd
|
1952 |
inode->i_mapping->a_ops = &shmem_aops; |
1da177e4c
|
1953 1954 1955 1956 1957 |
inode->i_op = &shmem_symlink_inode_operations; kaddr = kmap_atomic(page, KM_USER0); memcpy(kaddr, symname, len); kunmap_atomic(kaddr, KM_USER0); set_page_dirty(page); |
6746aff74
|
1958 |
unlock_page(page); |
1da177e4c
|
1959 1960 |
page_cache_release(page); } |
1da177e4c
|
1961 1962 1963 1964 1965 1966 |
dir->i_size += BOGO_DIRENT_SIZE; dir->i_ctime = dir->i_mtime = CURRENT_TIME; d_instantiate(dentry, inode); dget(dentry); return 0; } |
cc314eef0
|
1967 |
static void *shmem_follow_link_inline(struct dentry *dentry, struct nameidata *nd) |
1da177e4c
|
1968 1969 |
{ nd_set_link(nd, (char *)SHMEM_I(dentry->d_inode)); |
cc314eef0
|
1970 |
return NULL; |
1da177e4c
|
1971 |
} |
cc314eef0
|
1972 |
static void *shmem_follow_link(struct dentry *dentry, struct nameidata *nd) |
1da177e4c
|
1973 1974 1975 1976 |
{ struct page *page = NULL; int res = shmem_getpage(dentry->d_inode, 0, &page, SGP_READ, NULL); nd_set_link(nd, res ? ERR_PTR(res) : kmap(page)); |
d3602444e
|
1977 1978 |
if (page) unlock_page(page); |
cc314eef0
|
1979 |
return page; |
1da177e4c
|
1980 |
} |
cc314eef0
|
1981 |
static void shmem_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie) |
1da177e4c
|
1982 1983 |
{ if (!IS_ERR(nd_get_link(nd))) { |
cc314eef0
|
1984 |
struct page *page = cookie; |
1da177e4c
|
1985 1986 1987 |
kunmap(page); mark_page_accessed(page); page_cache_release(page); |
1da177e4c
|
1988 1989 |
} } |
92e1d5be9
|
1990 |
static const struct inode_operations shmem_symlink_inline_operations = { |
1da177e4c
|
1991 1992 |
.readlink = generic_readlink, .follow_link = shmem_follow_link_inline, |
1da177e4c
|
1993 |
}; |
92e1d5be9
|
1994 |
static const struct inode_operations shmem_symlink_inode_operations = { |
1da177e4c
|
1995 1996 1997 |
.readlink = generic_readlink, .follow_link = shmem_follow_link, .put_link = shmem_put_link, |
1da177e4c
|
1998 |
}; |
39f0247d3
|
1999 |
#ifdef CONFIG_TMPFS_POSIX_ACL |
467118102
|
2000 |
/* |
39f0247d3
|
2001 2002 2003 2004 2005 |
* Superblocks without xattr inode operations will get security.* xattr * support from the VFS "for free". As soon as we have any other xattrs * like ACLs, we also need to implement the security.* handlers at * filesystem level, though. */ |
431547b3c
|
2006 |
static size_t shmem_xattr_security_list(struct dentry *dentry, char *list, |
39f0247d3
|
2007 |
size_t list_len, const char *name, |
431547b3c
|
2008 |
size_t name_len, int handler_flags) |
39f0247d3
|
2009 |
{ |
431547b3c
|
2010 |
return security_inode_listsecurity(dentry->d_inode, list, list_len); |
39f0247d3
|
2011 |
} |
431547b3c
|
2012 2013 |
static int shmem_xattr_security_get(struct dentry *dentry, const char *name, void *buffer, size_t size, int handler_flags) |
39f0247d3
|
2014 2015 2016 |
{ if (strcmp(name, "") == 0) return -EINVAL; |
431547b3c
|
2017 |
return xattr_getsecurity(dentry->d_inode, name, buffer, size); |
39f0247d3
|
2018 |
} |
431547b3c
|
2019 2020 |
static int shmem_xattr_security_set(struct dentry *dentry, const char *name, const void *value, size_t size, int flags, int handler_flags) |
39f0247d3
|
2021 2022 2023 |
{ if (strcmp(name, "") == 0) return -EINVAL; |
431547b3c
|
2024 2025 |
return security_inode_setsecurity(dentry->d_inode, name, value, size, flags); |
39f0247d3
|
2026 |
} |
bb4354538
|
2027 |
static const struct xattr_handler shmem_xattr_security_handler = { |
39f0247d3
|
2028 2029 2030 2031 2032 |
.prefix = XATTR_SECURITY_PREFIX, .list = shmem_xattr_security_list, .get = shmem_xattr_security_get, .set = shmem_xattr_security_set, }; |
bb4354538
|
2033 |
static const struct xattr_handler *shmem_xattr_handlers[] = { |
1c7c474c3
|
2034 2035 |
&generic_acl_access_handler, &generic_acl_default_handler, |
39f0247d3
|
2036 2037 2038 2039 |
&shmem_xattr_security_handler, NULL }; #endif |
91828a405
|
2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 |
static struct dentry *shmem_get_parent(struct dentry *child) { return ERR_PTR(-ESTALE); } static int shmem_match(struct inode *ino, void *vfh) { __u32 *fh = vfh; __u64 inum = fh[2]; inum = (inum << 32) | fh[1]; return ino->i_ino == inum && fh[0] == ino->i_generation; } |
480b116c9
|
2052 2053 |
static struct dentry *shmem_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len, int fh_type) |
91828a405
|
2054 |
{ |
91828a405
|
2055 |
struct inode *inode; |
480b116c9
|
2056 2057 2058 2059 2060 2061 |
struct dentry *dentry = NULL; u64 inum = fid->raw[2]; inum = (inum << 32) | fid->raw[1]; if (fh_len < 3) return NULL; |
91828a405
|
2062 |
|
480b116c9
|
2063 2064 |
inode = ilookup5(sb, (unsigned long)(inum + fid->raw[0]), shmem_match, fid->raw); |
91828a405
|
2065 |
if (inode) { |
480b116c9
|
2066 |
dentry = d_find_alias(inode); |
91828a405
|
2067 2068 |
iput(inode); } |
480b116c9
|
2069 |
return dentry; |
91828a405
|
2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 |
} static int shmem_encode_fh(struct dentry *dentry, __u32 *fh, int *len, int connectable) { struct inode *inode = dentry->d_inode; if (*len < 3) return 255; if (hlist_unhashed(&inode->i_hash)) { /* Unfortunately insert_inode_hash is not idempotent, * so as we hash inodes here rather than at creation * time, we need a lock to ensure we only try * to do it once */ static DEFINE_SPINLOCK(lock); spin_lock(&lock); if (hlist_unhashed(&inode->i_hash)) __insert_inode_hash(inode, inode->i_ino + inode->i_generation); spin_unlock(&lock); } fh[0] = inode->i_generation; fh[1] = inode->i_ino; fh[2] = ((__u64)inode->i_ino) >> 32; *len = 3; return 1; } |
396551644
|
2101 |
static const struct export_operations shmem_export_ops = { |
91828a405
|
2102 |
.get_parent = shmem_get_parent, |
91828a405
|
2103 |
.encode_fh = shmem_encode_fh, |
480b116c9
|
2104 |
.fh_to_dentry = shmem_fh_to_dentry, |
91828a405
|
2105 |
}; |
680d794ba
|
2106 2107 |
static int shmem_parse_options(char *options, struct shmem_sb_info *sbinfo, bool remount) |
1da177e4c
|
2108 2109 |
{ char *this_char, *value, *rest; |
b00dc3ad7
|
2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 |
while (options != NULL) { this_char = options; for (;;) { /* * NUL-terminate this option: unfortunately, * mount options form a comma-separated list, * but mpol's nodelist may also contain commas. */ options = strchr(options, ','); if (options == NULL) break; options++; if (!isdigit(*options)) { options[-1] = '\0'; break; } } |
1da177e4c
|
2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 |
if (!*this_char) continue; if ((value = strchr(this_char,'=')) != NULL) { *value++ = 0; } else { printk(KERN_ERR "tmpfs: No value for mount option '%s' ", this_char); return 1; } if (!strcmp(this_char,"size")) { unsigned long long size; size = memparse(value,&rest); if (*rest == '%') { size <<= PAGE_SHIFT; size *= totalram_pages; do_div(size, 100); rest++; } if (*rest) goto bad_val; |
680d794ba
|
2150 2151 |
sbinfo->max_blocks = DIV_ROUND_UP(size, PAGE_CACHE_SIZE); |
1da177e4c
|
2152 |
} else if (!strcmp(this_char,"nr_blocks")) { |
680d794ba
|
2153 |
sbinfo->max_blocks = memparse(value, &rest); |
1da177e4c
|
2154 2155 2156 |
if (*rest) goto bad_val; } else if (!strcmp(this_char,"nr_inodes")) { |
680d794ba
|
2157 |
sbinfo->max_inodes = memparse(value, &rest); |
1da177e4c
|
2158 2159 2160 |
if (*rest) goto bad_val; } else if (!strcmp(this_char,"mode")) { |
680d794ba
|
2161 |
if (remount) |
1da177e4c
|
2162 |
continue; |
680d794ba
|
2163 |
sbinfo->mode = simple_strtoul(value, &rest, 8) & 07777; |
1da177e4c
|
2164 2165 2166 |
if (*rest) goto bad_val; } else if (!strcmp(this_char,"uid")) { |
680d794ba
|
2167 |
if (remount) |
1da177e4c
|
2168 |
continue; |
680d794ba
|
2169 |
sbinfo->uid = simple_strtoul(value, &rest, 0); |
1da177e4c
|
2170 2171 2172 |
if (*rest) goto bad_val; } else if (!strcmp(this_char,"gid")) { |
680d794ba
|
2173 |
if (remount) |
1da177e4c
|
2174 |
continue; |
680d794ba
|
2175 |
sbinfo->gid = simple_strtoul(value, &rest, 0); |
1da177e4c
|
2176 2177 |
if (*rest) goto bad_val; |
7339ff830
|
2178 |
} else if (!strcmp(this_char,"mpol")) { |
71fe804b6
|
2179 |
if (mpol_parse_str(value, &sbinfo->mpol, 1)) |
7339ff830
|
2180 |
goto bad_val; |
1da177e4c
|
2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 |
} else { printk(KERN_ERR "tmpfs: Bad mount option %s ", this_char); return 1; } } return 0; bad_val: printk(KERN_ERR "tmpfs: Bad value '%s' for mount option '%s' ", value, this_char); return 1; } static int shmem_remount_fs(struct super_block *sb, int *flags, char *data) { struct shmem_sb_info *sbinfo = SHMEM_SB(sb); |
680d794ba
|
2201 |
struct shmem_sb_info config = *sbinfo; |
0edd73b33
|
2202 2203 |
unsigned long inodes; int error = -EINVAL; |
680d794ba
|
2204 |
if (shmem_parse_options(data, &config, true)) |
0edd73b33
|
2205 |
return error; |
1da177e4c
|
2206 |
|
0edd73b33
|
2207 |
spin_lock(&sbinfo->stat_lock); |
0edd73b33
|
2208 |
inodes = sbinfo->max_inodes - sbinfo->free_inodes; |
7e496299d
|
2209 |
if (percpu_counter_compare(&sbinfo->used_blocks, config.max_blocks) > 0) |
0edd73b33
|
2210 |
goto out; |
680d794ba
|
2211 |
if (config.max_inodes < inodes) |
0edd73b33
|
2212 2213 2214 2215 2216 2217 2218 |
goto out; /* * Those tests also disallow limited->unlimited while any are in * use, so i_blocks will always be zero when max_blocks is zero; * but we must separately disallow unlimited->limited, because * in that case we have no record of how much is already in use. */ |
680d794ba
|
2219 |
if (config.max_blocks && !sbinfo->max_blocks) |
0edd73b33
|
2220 |
goto out; |
680d794ba
|
2221 |
if (config.max_inodes && !sbinfo->max_inodes) |
0edd73b33
|
2222 2223 2224 |
goto out; error = 0; |
680d794ba
|
2225 |
sbinfo->max_blocks = config.max_blocks; |
680d794ba
|
2226 2227 |
sbinfo->max_inodes = config.max_inodes; sbinfo->free_inodes = config.max_inodes - inodes; |
71fe804b6
|
2228 2229 2230 |
mpol_put(sbinfo->mpol); sbinfo->mpol = config.mpol; /* transfers initial ref */ |
0edd73b33
|
2231 2232 2233 |
out: spin_unlock(&sbinfo->stat_lock); return error; |
1da177e4c
|
2234 |
} |
680d794ba
|
2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 |
static int shmem_show_options(struct seq_file *seq, struct vfsmount *vfs) { struct shmem_sb_info *sbinfo = SHMEM_SB(vfs->mnt_sb); if (sbinfo->max_blocks != shmem_default_max_blocks()) seq_printf(seq, ",size=%luk", sbinfo->max_blocks << (PAGE_CACHE_SHIFT - 10)); if (sbinfo->max_inodes != shmem_default_max_inodes()) seq_printf(seq, ",nr_inodes=%lu", sbinfo->max_inodes); if (sbinfo->mode != (S_IRWXUGO | S_ISVTX)) seq_printf(seq, ",mode=%03o", sbinfo->mode); if (sbinfo->uid != 0) seq_printf(seq, ",uid=%u", sbinfo->uid); if (sbinfo->gid != 0) seq_printf(seq, ",gid=%u", sbinfo->gid); |
71fe804b6
|
2251 |
shmem_show_mpol(seq, sbinfo->mpol); |
680d794ba
|
2252 2253 2254 |
return 0; } #endif /* CONFIG_TMPFS */ |
1da177e4c
|
2255 2256 2257 |
static void shmem_put_super(struct super_block *sb) { |
602586a83
|
2258 2259 2260 2261 |
struct shmem_sb_info *sbinfo = SHMEM_SB(sb); percpu_counter_destroy(&sbinfo->used_blocks); kfree(sbinfo); |
1da177e4c
|
2262 2263 |
sb->s_fs_info = NULL; } |
2b2af54a5
|
2264 |
int shmem_fill_super(struct super_block *sb, void *data, int silent) |
1da177e4c
|
2265 2266 2267 |
{ struct inode *inode; struct dentry *root; |
0edd73b33
|
2268 |
struct shmem_sb_info *sbinfo; |
680d794ba
|
2269 2270 2271 |
int err = -ENOMEM; /* Round up to L1_CACHE_BYTES to resist false sharing */ |
425fbf047
|
2272 |
sbinfo = kzalloc(max((int)sizeof(struct shmem_sb_info), |
680d794ba
|
2273 2274 2275 |
L1_CACHE_BYTES), GFP_KERNEL); if (!sbinfo) return -ENOMEM; |
680d794ba
|
2276 |
sbinfo->mode = S_IRWXUGO | S_ISVTX; |
76aac0e9a
|
2277 2278 |
sbinfo->uid = current_fsuid(); sbinfo->gid = current_fsgid(); |
680d794ba
|
2279 |
sb->s_fs_info = sbinfo; |
1da177e4c
|
2280 |
|
0edd73b33
|
2281 |
#ifdef CONFIG_TMPFS |
1da177e4c
|
2282 2283 2284 2285 2286 2287 |
/* * Per default we only allow half of the physical ram per * tmpfs instance, limiting inodes to one per page of lowmem; * but the internal instance is left unlimited. */ if (!(sb->s_flags & MS_NOUSER)) { |
680d794ba
|
2288 2289 2290 2291 2292 2293 |
sbinfo->max_blocks = shmem_default_max_blocks(); sbinfo->max_inodes = shmem_default_max_inodes(); if (shmem_parse_options(data, sbinfo, false)) { err = -EINVAL; goto failed; } |
1da177e4c
|
2294 |
} |
91828a405
|
2295 |
sb->s_export_op = &shmem_export_ops; |
1da177e4c
|
2296 2297 2298 |
#else sb->s_flags |= MS_NOUSER; #endif |
0edd73b33
|
2299 |
spin_lock_init(&sbinfo->stat_lock); |
602586a83
|
2300 2301 |
if (percpu_counter_init(&sbinfo->used_blocks, 0)) goto failed; |
680d794ba
|
2302 |
sbinfo->free_inodes = sbinfo->max_inodes; |
0edd73b33
|
2303 |
|
1da177e4c
|
2304 2305 2306 2307 2308 |
sb->s_maxbytes = SHMEM_MAX_BYTES; sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; sb->s_magic = TMPFS_MAGIC; sb->s_op = &shmem_ops; |
cfd95a9cf
|
2309 |
sb->s_time_gran = 1; |
39f0247d3
|
2310 2311 2312 2313 |
#ifdef CONFIG_TMPFS_POSIX_ACL sb->s_xattr = shmem_xattr_handlers; sb->s_flags |= MS_POSIXACL; #endif |
0edd73b33
|
2314 |
|
454abafe9
|
2315 |
inode = shmem_get_inode(sb, NULL, S_IFDIR | sbinfo->mode, 0, VM_NORESERVE); |
1da177e4c
|
2316 2317 |
if (!inode) goto failed; |
680d794ba
|
2318 2319 |
inode->i_uid = sbinfo->uid; inode->i_gid = sbinfo->gid; |
1da177e4c
|
2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 |
root = d_alloc_root(inode); if (!root) goto failed_iput; sb->s_root = root; return 0; failed_iput: iput(inode); failed: shmem_put_super(sb); return err; } |
fcc234f88
|
2332 |
static struct kmem_cache *shmem_inode_cachep; |
1da177e4c
|
2333 2334 2335 2336 |
static struct inode *shmem_alloc_inode(struct super_block *sb) { struct shmem_inode_info *p; |
e94b17660
|
2337 |
p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL); |
1da177e4c
|
2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 |
if (!p) return NULL; return &p->vfs_inode; } static void shmem_destroy_inode(struct inode *inode) { if ((inode->i_mode & S_IFMT) == S_IFREG) { /* only struct inode is valid if it's an inline symlink */ mpol_free_shared_policy(&SHMEM_I(inode)->policy); } kmem_cache_free(shmem_inode_cachep, SHMEM_I(inode)); } |
51cc50685
|
2351 |
static void init_once(void *foo) |
1da177e4c
|
2352 2353 |
{ struct shmem_inode_info *p = (struct shmem_inode_info *) foo; |
a35afb830
|
2354 |
inode_init_once(&p->vfs_inode); |
1da177e4c
|
2355 2356 2357 2358 2359 2360 |
} static int init_inodecache(void) { shmem_inode_cachep = kmem_cache_create("shmem_inode_cache", sizeof(struct shmem_inode_info), |
040b5c6f9
|
2361 |
0, SLAB_PANIC, init_once); |
1da177e4c
|
2362 2363 2364 2365 2366 |
return 0; } static void destroy_inodecache(void) { |
1a1d92c10
|
2367 |
kmem_cache_destroy(shmem_inode_cachep); |
1da177e4c
|
2368 |
} |
f5e54d6e5
|
2369 |
static const struct address_space_operations shmem_aops = { |
1da177e4c
|
2370 |
.writepage = shmem_writepage, |
767193253
|
2371 |
.set_page_dirty = __set_page_dirty_no_writeback, |
1da177e4c
|
2372 |
#ifdef CONFIG_TMPFS |
ae9764164
|
2373 |
.readpage = shmem_readpage, |
800d15a53
|
2374 2375 |
.write_begin = shmem_write_begin, .write_end = shmem_write_end, |
1da177e4c
|
2376 |
#endif |
304dbdb7a
|
2377 |
.migratepage = migrate_page, |
aa261f549
|
2378 |
.error_remove_page = generic_error_remove_page, |
1da177e4c
|
2379 |
}; |
15ad7cdcf
|
2380 |
static const struct file_operations shmem_file_operations = { |
1da177e4c
|
2381 2382 2383 |
.mmap = shmem_mmap, #ifdef CONFIG_TMPFS .llseek = generic_file_llseek, |
bcd78e496
|
2384 |
.read = do_sync_read, |
5402b976a
|
2385 |
.write = do_sync_write, |
bcd78e496
|
2386 |
.aio_read = shmem_file_aio_read, |
5402b976a
|
2387 |
.aio_write = generic_file_aio_write, |
1b061d924
|
2388 |
.fsync = noop_fsync, |
ae9764164
|
2389 2390 |
.splice_read = generic_file_splice_read, .splice_write = generic_file_splice_write, |
1da177e4c
|
2391 2392 |
#endif }; |
92e1d5be9
|
2393 |
static const struct inode_operations shmem_inode_operations = { |
1da177e4c
|
2394 |
.setattr = shmem_notify_change, |
f6b3ec238
|
2395 |
.truncate_range = shmem_truncate_range, |
39f0247d3
|
2396 2397 2398 2399 2400 |
#ifdef CONFIG_TMPFS_POSIX_ACL .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = generic_listxattr, .removexattr = generic_removexattr, |
1c7c474c3
|
2401 |
.check_acl = generic_check_acl, |
39f0247d3
|
2402 |
#endif |
1da177e4c
|
2403 |
}; |
92e1d5be9
|
2404 |
static const struct inode_operations shmem_dir_inode_operations = { |
1da177e4c
|
2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 |
#ifdef CONFIG_TMPFS .create = shmem_create, .lookup = simple_lookup, .link = shmem_link, .unlink = shmem_unlink, .symlink = shmem_symlink, .mkdir = shmem_mkdir, .rmdir = shmem_rmdir, .mknod = shmem_mknod, .rename = shmem_rename, |
1da177e4c
|
2415 |
#endif |
39f0247d3
|
2416 2417 2418 2419 2420 2421 |
#ifdef CONFIG_TMPFS_POSIX_ACL .setattr = shmem_notify_change, .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = generic_listxattr, .removexattr = generic_removexattr, |
1c7c474c3
|
2422 |
.check_acl = generic_check_acl, |
39f0247d3
|
2423 2424 |
#endif }; |
92e1d5be9
|
2425 |
static const struct inode_operations shmem_special_inode_operations = { |
39f0247d3
|
2426 2427 2428 2429 2430 2431 |
#ifdef CONFIG_TMPFS_POSIX_ACL .setattr = shmem_notify_change, .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = generic_listxattr, .removexattr = generic_removexattr, |
1c7c474c3
|
2432 |
.check_acl = generic_check_acl, |
39f0247d3
|
2433 |
#endif |
1da177e4c
|
2434 |
}; |
759b9775c
|
2435 |
static const struct super_operations shmem_ops = { |
1da177e4c
|
2436 2437 2438 2439 2440 |
.alloc_inode = shmem_alloc_inode, .destroy_inode = shmem_destroy_inode, #ifdef CONFIG_TMPFS .statfs = shmem_statfs, .remount_fs = shmem_remount_fs, |
680d794ba
|
2441 |
.show_options = shmem_show_options, |
1da177e4c
|
2442 |
#endif |
1f895f75d
|
2443 |
.evict_inode = shmem_evict_inode, |
1da177e4c
|
2444 2445 2446 |
.drop_inode = generic_delete_inode, .put_super = shmem_put_super, }; |
f0f37e2f7
|
2447 |
static const struct vm_operations_struct shmem_vm_ops = { |
54cb8821d
|
2448 |
.fault = shmem_fault, |
1da177e4c
|
2449 2450 2451 2452 2453 |
#ifdef CONFIG_NUMA .set_policy = shmem_set_policy, .get_policy = shmem_get_policy, #endif }; |
454e2398b
|
2454 2455 |
static int shmem_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, struct vfsmount *mnt) |
1da177e4c
|
2456 |
{ |
454e2398b
|
2457 |
return get_sb_nodev(fs_type, flags, data, shmem_fill_super, mnt); |
1da177e4c
|
2458 2459 2460 2461 2462 2463 2464 2465 |
} static struct file_system_type tmpfs_fs_type = { .owner = THIS_MODULE, .name = "tmpfs", .get_sb = shmem_get_sb, .kill_sb = kill_litter_super, }; |
1da177e4c
|
2466 |
|
2b2af54a5
|
2467 |
int __init init_tmpfs(void) |
1da177e4c
|
2468 2469 |
{ int error; |
e0bf68dde
|
2470 2471 2472 |
error = bdi_init(&shmem_backing_dev_info); if (error) goto out4; |
1da177e4c
|
2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 |
error = init_inodecache(); if (error) goto out3; error = register_filesystem(&tmpfs_fs_type); if (error) { printk(KERN_ERR "Could not register tmpfs "); goto out2; } |
95dc112a5
|
2483 |
|
1f5ce9e93
|
2484 |
shm_mnt = vfs_kern_mount(&tmpfs_fs_type, MS_NOUSER, |
1da177e4c
|
2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 |
tmpfs_fs_type.name, NULL); if (IS_ERR(shm_mnt)) { error = PTR_ERR(shm_mnt); printk(KERN_ERR "Could not kern_mount tmpfs "); goto out1; } return 0; out1: unregister_filesystem(&tmpfs_fs_type); out2: destroy_inodecache(); out3: |
e0bf68dde
|
2499 2500 |
bdi_destroy(&shmem_backing_dev_info); out4: |
1da177e4c
|
2501 2502 2503 |
shm_mnt = ERR_PTR(error); return error; } |
853ac43ab
|
2504 |
|
87946a722
|
2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 |
#ifdef CONFIG_CGROUP_MEM_RES_CTLR /** * mem_cgroup_get_shmem_target - find a page or entry assigned to the shmem file * @inode: the inode to be searched * @pgoff: the offset to be searched * @pagep: the pointer for the found page to be stored * @ent: the pointer for the found swap entry to be stored * * If a page is found, refcount of it is incremented. Callers should handle * these refcount. */ void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent) { swp_entry_t entry = { .val = 0 }, *ptr; struct page *page = NULL; struct shmem_inode_info *info = SHMEM_I(inode); if ((pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) goto out; spin_lock(&info->lock); ptr = shmem_swp_entry(info, pgoff, NULL); #ifdef CONFIG_SWAP if (ptr && ptr->val) { entry.val = ptr->val; page = find_get_page(&swapper_space, entry.val); } else #endif page = find_get_page(inode->i_mapping, pgoff); if (ptr) shmem_swp_unmap(ptr); spin_unlock(&info->lock); out: *pagep = page; *ent = entry; } #endif |
853ac43ab
|
2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 |
#else /* !CONFIG_SHMEM */ /* * tiny-shmem: simple shmemfs and tmpfs using ramfs code * * This is intended for small system where the benefits of the full * shmem code (swap-backed and resource-limited) are outweighed by * their complexity. On systems without swap this code should be * effectively equivalent, but much lighter weight. */ #include <linux/ramfs.h> static struct file_system_type tmpfs_fs_type = { .name = "tmpfs", .get_sb = ramfs_get_sb, .kill_sb = kill_litter_super, }; |
2b2af54a5
|
2561 |
int __init init_tmpfs(void) |
853ac43ab
|
2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 |
{ BUG_ON(register_filesystem(&tmpfs_fs_type) != 0); shm_mnt = kern_mount(&tmpfs_fs_type); BUG_ON(IS_ERR(shm_mnt)); return 0; } int shmem_unuse(swp_entry_t entry, struct page *page) { return 0; } |
3f96b79ad
|
2575 2576 2577 2578 |
int shmem_lock(struct file *file, int lock, struct user_struct *user) { return 0; } |
87946a722
|
2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 |
#ifdef CONFIG_CGROUP_MEM_RES_CTLR /** * mem_cgroup_get_shmem_target - find a page or entry assigned to the shmem file * @inode: the inode to be searched * @pgoff: the offset to be searched * @pagep: the pointer for the found page to be stored * @ent: the pointer for the found swap entry to be stored * * If a page is found, refcount of it is incremented. Callers should handle * these refcount. */ void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent) { struct page *page = NULL; if ((pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) goto out; page = find_get_page(inode->i_mapping, pgoff); out: *pagep = page; *ent = (swp_entry_t){ .val = 0 }; } #endif |
0b0a0806b
|
2603 2604 |
#define shmem_vm_ops generic_file_vm_ops #define shmem_file_operations ramfs_file_operations |
454abafe9
|
2605 |
#define shmem_get_inode(sb, dir, mode, dev, flags) ramfs_get_inode(sb, dir, mode, dev) |
0b0a0806b
|
2606 2607 |
#define shmem_acct_size(flags, size) 0 #define shmem_unacct_size(flags, size) do {} while (0) |
caefba174
|
2608 |
#define SHMEM_MAX_BYTES MAX_LFS_FILESIZE |
853ac43ab
|
2609 2610 2611 2612 |
#endif /* CONFIG_SHMEM */ /* common code */ |
1da177e4c
|
2613 |
|
467118102
|
2614 |
/** |
1da177e4c
|
2615 |
* shmem_file_setup - get an unlinked file living in tmpfs |
1da177e4c
|
2616 2617 |
* @name: name for dentry (to be seen in /proc/<pid>/maps * @size: size to be set for the file |
0b0a0806b
|
2618 |
* @flags: VM_NORESERVE suppresses pre-accounting of the entire object size |
1da177e4c
|
2619 |
*/ |
168f5ac66
|
2620 |
struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags) |
1da177e4c
|
2621 2622 2623 2624 |
{ int error; struct file *file; struct inode *inode; |
2c48b9c45
|
2625 2626 |
struct path path; struct dentry *root; |
1da177e4c
|
2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 |
struct qstr this; if (IS_ERR(shm_mnt)) return (void *)shm_mnt; if (size < 0 || size > SHMEM_MAX_BYTES) return ERR_PTR(-EINVAL); if (shmem_acct_size(flags, size)) return ERR_PTR(-ENOMEM); error = -ENOMEM; this.name = name; this.len = strlen(name); this.hash = 0; /* will go */ root = shm_mnt->mnt_root; |
2c48b9c45
|
2643 2644 |
path.dentry = d_alloc(root, &this); if (!path.dentry) |
1da177e4c
|
2645 |
goto put_memory; |
2c48b9c45
|
2646 |
path.mnt = mntget(shm_mnt); |
1da177e4c
|
2647 |
|
1da177e4c
|
2648 |
error = -ENOSPC; |
454abafe9
|
2649 |
inode = shmem_get_inode(root->d_sb, NULL, S_IFREG | S_IRWXUGO, 0, flags); |
1da177e4c
|
2650 |
if (!inode) |
4b42af81f
|
2651 |
goto put_dentry; |
1da177e4c
|
2652 |
|
2c48b9c45
|
2653 |
d_instantiate(path.dentry, inode); |
1da177e4c
|
2654 2655 |
inode->i_size = size; inode->i_nlink = 0; /* It is unlinked */ |
853ac43ab
|
2656 2657 2658 |
#ifndef CONFIG_MMU error = ramfs_nommu_expand_for_mapping(inode, size); if (error) |
4b42af81f
|
2659 |
goto put_dentry; |
853ac43ab
|
2660 |
#endif |
4b42af81f
|
2661 2662 |
error = -ENFILE; |
2c48b9c45
|
2663 |
file = alloc_file(&path, FMODE_WRITE | FMODE_READ, |
4b42af81f
|
2664 2665 2666 |
&shmem_file_operations); if (!file) goto put_dentry; |
1da177e4c
|
2667 |
return file; |
1da177e4c
|
2668 |
put_dentry: |
2c48b9c45
|
2669 |
path_put(&path); |
1da177e4c
|
2670 2671 2672 2673 |
put_memory: shmem_unacct_size(flags, size); return ERR_PTR(error); } |
395e0ddc4
|
2674 |
EXPORT_SYMBOL_GPL(shmem_file_setup); |
1da177e4c
|
2675 |
|
467118102
|
2676 |
/** |
1da177e4c
|
2677 |
* shmem_zero_setup - setup a shared anonymous mapping |
1da177e4c
|
2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 |
* @vma: the vma to be mmapped is prepared by do_mmap_pgoff */ int shmem_zero_setup(struct vm_area_struct *vma) { struct file *file; loff_t size = vma->vm_end - vma->vm_start; file = shmem_file_setup("dev/zero", size, vma->vm_flags); if (IS_ERR(file)) return PTR_ERR(file); if (vma->vm_file) fput(vma->vm_file); vma->vm_file = file; vma->vm_ops = &shmem_vm_ops; return 0; } |