Blame view
fs/super.c
34.9 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
/* * linux/fs/super.c * * Copyright (C) 1991, 1992 Linus Torvalds * * super.c contains code to handle: - mount structures * - super-block tables * - filesystem drivers list * - mount system call * - umount system call * - ustat system call * * GK 2/5/95 - Changed to support mounting the root fs via NFS * * Added kerneld support: Jacques Gelinas and Bjorn Ekwall * Added change_root: Werner Almesberger & Hans Lermen, Feb '96 * Added options to /proc/mounts: |
96de0e252
|
18 |
* Torbjörn Lindh (torbjorn.lindh@gopta.se), April 14, 1996. |
1da177e4c
|
19 20 21 |
* Added devfs support: Richard Gooch <rgooch@atnf.csiro.au>, 13-JAN-1998 * Heavily rewritten for 'one fs - one tree' dcache architecture. AV, Mar 2000 */ |
630d9c472
|
22 |
#include <linux/export.h> |
1da177e4c
|
23 |
#include <linux/slab.h> |
1da177e4c
|
24 |
#include <linux/blkdev.h> |
1da177e4c
|
25 26 |
#include <linux/mount.h> #include <linux/security.h> |
1da177e4c
|
27 28 |
#include <linux/writeback.h> /* for the emergency remount stuff */ #include <linux/idr.h> |
353ab6e97
|
29 |
#include <linux/mutex.h> |
5477d0fac
|
30 |
#include <linux/backing-dev.h> |
ceb5bdc2d
|
31 |
#include <linux/rculist_bl.h> |
c515e1fd3
|
32 |
#include <linux/cleancache.h> |
404015308
|
33 |
#include <linux/fsnotify.h> |
5accdf82b
|
34 |
#include <linux/lockdep.h> |
6d59e7f58
|
35 |
#include "internal.h" |
1da177e4c
|
36 |
|
1da177e4c
|
37 38 |
LIST_HEAD(super_blocks); DEFINE_SPINLOCK(sb_lock); |
5accdf82b
|
39 40 41 42 43 |
static char *sb_writers_name[SB_FREEZE_LEVELS] = { "sb_writers", "sb_pagefaults", "sb_internal", }; |
b0d40c92a
|
44 45 46 47 48 49 50 |
/* * One thing we have to be careful of with a per-sb shrinker is that we don't * drop the last active reference to the superblock from within the shrinker. * If that happens we could trigger unregistering the shrinker from within the * shrinker path and that leads to deadlock on the shrinker_rwsem. Hence we * take a passive reference to the superblock to avoid this from occurring. */ |
0a234c6dc
|
51 52 |
static unsigned long super_cache_scan(struct shrinker *shrink, struct shrink_control *sc) |
b0d40c92a
|
53 54 |
{ struct super_block *sb; |
0a234c6dc
|
55 56 57 58 59 |
long fs_objects = 0; long total_objects; long freed = 0; long dentries; long inodes; |
b0d40c92a
|
60 61 62 63 64 65 66 |
sb = container_of(shrink, struct super_block, s_shrink); /* * Deadlock avoidance. We may hold various FS locks, and we don't want * to recurse into the FS that called us in clear_inode() and friends.. */ |
0a234c6dc
|
67 68 |
if (!(sc->gfp_mask & __GFP_FS)) return SHRINK_STOP; |
b0d40c92a
|
69 70 |
if (!grab_super_passive(sb)) |
0a234c6dc
|
71 |
return SHRINK_STOP; |
b0d40c92a
|
72 |
|
d04079039
|
73 |
if (sb->s_op->nr_cached_objects) |
9b17c6238
|
74 |
fs_objects = sb->s_op->nr_cached_objects(sb, sc->nid); |
0e1fdafd9
|
75 |
|
9b17c6238
|
76 77 |
inodes = list_lru_count_node(&sb->s_inode_lru, sc->nid); dentries = list_lru_count_node(&sb->s_dentry_lru, sc->nid); |
f60415675
|
78 |
total_objects = dentries + inodes + fs_objects + 1; |
475d0db74
|
79 80 |
if (!total_objects) total_objects = 1; |
0e1fdafd9
|
81 |
|
0a234c6dc
|
82 |
/* proportion the scan between the caches */ |
f60415675
|
83 |
dentries = mult_frac(sc->nr_to_scan, dentries, total_objects); |
bc3b14cb2
|
84 |
inodes = mult_frac(sc->nr_to_scan, inodes, total_objects); |
b0d40c92a
|
85 |
|
0a234c6dc
|
86 87 88 89 |
/* * prune the dcache first as the icache is pinned by it, then * prune the icache, followed by the filesystem specific caches */ |
9b17c6238
|
90 91 |
freed = prune_dcache_sb(sb, dentries, sc->nid); freed += prune_icache_sb(sb, inodes, sc->nid); |
0a234c6dc
|
92 93 94 95 |
if (fs_objects) { fs_objects = mult_frac(sc->nr_to_scan, fs_objects, total_objects); |
9b17c6238
|
96 97 |
freed += sb->s_op->free_cached_objects(sb, fs_objects, sc->nid); |
b0d40c92a
|
98 |
} |
0a234c6dc
|
99 100 101 102 103 104 105 106 107 108 109 |
drop_super(sb); return freed; } static unsigned long super_cache_count(struct shrinker *shrink, struct shrink_control *sc) { struct super_block *sb; long total_objects = 0; sb = container_of(shrink, struct super_block, s_shrink); |
d23da150a
|
110 111 112 113 114 115 116 117 |
/* * Don't call grab_super_passive as it is a potential * scalability bottleneck. The counts could get updated * between super_cache_count and super_cache_scan anyway. * Call to super_cache_count with shrinker_rwsem held * ensures the safety of call to list_lru_count_node() and * s_op->nr_cached_objects(). */ |
0a234c6dc
|
118 |
if (sb->s_op && sb->s_op->nr_cached_objects) |
9b17c6238
|
119 120 |
total_objects = sb->s_op->nr_cached_objects(sb, sc->nid); |
0a234c6dc
|
121 |
|
9b17c6238
|
122 123 124 125 |
total_objects += list_lru_count_node(&sb->s_dentry_lru, sc->nid); total_objects += list_lru_count_node(&sb->s_inode_lru, sc->nid); |
0a234c6dc
|
126 |
|
55f841ce9
|
127 |
total_objects = vfs_pressure_ratio(total_objects); |
0e1fdafd9
|
128 |
return total_objects; |
b0d40c92a
|
129 |
} |
7eb5e8826
|
130 131 132 133 134 135 136 |
/** * destroy_super - frees a superblock * @s: superblock to free * * Frees a superblock. */ static void destroy_super(struct super_block *s) |
5accdf82b
|
137 138 |
{ int i; |
7eb5e8826
|
139 140 |
list_lru_destroy(&s->s_dentry_lru); list_lru_destroy(&s->s_inode_lru); |
5accdf82b
|
141 142 |
for (i = 0; i < SB_FREEZE_LEVELS; i++) percpu_counter_destroy(&s->s_writers.counter[i]); |
7eb5e8826
|
143 144 145 146 |
security_sb_free(s); WARN_ON(!list_empty(&s->s_mounts)); kfree(s->s_subtype); kfree(s->s_options); |
e2fec7c35
|
147 |
kfree_rcu(s, rcu); |
5accdf82b
|
148 |
} |
1da177e4c
|
149 150 |
/** * alloc_super - create new superblock |
fe2bbc483
|
151 |
* @type: filesystem type superblock should belong to |
9249e17fe
|
152 |
* @flags: the mount flags |
1da177e4c
|
153 154 155 156 |
* * Allocates and initializes a new &struct super_block. alloc_super() * returns a pointer new superblock or %NULL if allocation had failed. */ |
9249e17fe
|
157 |
static struct super_block *alloc_super(struct file_system_type *type, int flags) |
1da177e4c
|
158 |
{ |
11b0b5abb
|
159 |
struct super_block *s = kzalloc(sizeof(struct super_block), GFP_USER); |
b87221de6
|
160 |
static const struct super_operations default_op; |
7eb5e8826
|
161 162 163 164 |
int i; if (!s) return NULL; |
1da177e4c
|
165 |
|
b5bd856a0
|
166 |
INIT_LIST_HEAD(&s->s_mounts); |
7eb5e8826
|
167 168 |
if (security_sb_alloc(s)) goto fail; |
7b7a8665e
|
169 |
|
7eb5e8826
|
170 |
for (i = 0; i < SB_FREEZE_LEVELS; i++) { |
908c7f194
|
171 172 |
if (percpu_counter_init(&s->s_writers.counter[i], 0, GFP_KERNEL) < 0) |
7eb5e8826
|
173 174 175 |
goto fail; lockdep_init_map(&s->s_writers.lock_map[i], sb_writers_name[i], &type->s_writers_key[i], 0); |
1da177e4c
|
176 |
} |
7eb5e8826
|
177 178 179 180 181 182 183 184 185 186 187 188 |
init_waitqueue_head(&s->s_writers.wait); init_waitqueue_head(&s->s_writers.wait_unfrozen); s->s_flags = flags; s->s_bdi = &default_backing_dev_info; INIT_HLIST_NODE(&s->s_instances); INIT_HLIST_BL_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); if (list_lru_init(&s->s_dentry_lru)) goto fail; if (list_lru_init(&s->s_inode_lru)) goto fail; |
7eb5e8826
|
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 |
init_rwsem(&s->s_umount); lockdep_set_class(&s->s_umount, &type->s_umount_key); /* * sget() can have s_umount recursion. * * When it cannot find a suitable sb, it allocates a new * one (this one), and tries again to find a suitable old * one. * * In case that succeeds, it will acquire the s_umount * lock of the old one. Since these are clearly distrinct * locks, and this object isn't exposed yet, there's no * risk of deadlocks. * * Annotate this by putting this lock in a different * subclass. */ down_write_nested(&s->s_umount, SINGLE_DEPTH_NESTING); s->s_count = 1; atomic_set(&s->s_active, 1); mutex_init(&s->s_vfs_rename_mutex); lockdep_set_class(&s->s_vfs_rename_mutex, &type->s_vfs_rename_key); mutex_init(&s->s_dquot.dqio_mutex); mutex_init(&s->s_dquot.dqonoff_mutex); |
7eb5e8826
|
213 214 215 216 217 218 219 220 221 222 |
s->s_maxbytes = MAX_NON_LFS; s->s_op = &default_op; s->s_time_gran = 1000000000; s->cleancache_poolid = -1; s->s_shrink.seeks = DEFAULT_SEEKS; s->s_shrink.scan_objects = super_cache_scan; s->s_shrink.count_objects = super_cache_count; s->s_shrink.batch = 1024; s->s_shrink.flags = SHRINKER_NUMA_AWARE; |
1da177e4c
|
223 |
return s; |
5ca302c8e
|
224 |
|
7eb5e8826
|
225 226 227 |
fail: destroy_super(s); return NULL; |
1da177e4c
|
228 229 230 231 232 |
} /* Superblock refcounting */ /* |
35cf7ba0b
|
233 |
* Drop a superblock's refcount. The caller must hold sb_lock. |
1da177e4c
|
234 |
*/ |
f47ec3f28
|
235 |
static void __put_super(struct super_block *sb) |
1da177e4c
|
236 |
{ |
1da177e4c
|
237 |
if (!--sb->s_count) { |
551de6f34
|
238 |
list_del_init(&sb->s_list); |
1da177e4c
|
239 |
destroy_super(sb); |
1da177e4c
|
240 |
} |
1da177e4c
|
241 242 243 244 245 246 247 248 249 |
} /** * put_super - drop a temporary reference to superblock * @sb: superblock in question * * Drops a temporary reference, frees superblock if there's no * references left. */ |
f47ec3f28
|
250 |
static void put_super(struct super_block *sb) |
1da177e4c
|
251 252 253 254 255 256 257 258 |
{ spin_lock(&sb_lock); __put_super(sb); spin_unlock(&sb_lock); } /** |
1712ac8fd
|
259 |
* deactivate_locked_super - drop an active reference to superblock |
1da177e4c
|
260 261 |
* @s: superblock to deactivate * |
1712ac8fd
|
262 263 |
* Drops an active reference to superblock, converting it into a temprory * one if there is no other active references left. In that case we |
1da177e4c
|
264 265 |
* tell fs driver to shut it down and drop the temporary reference we * had just acquired. |
1712ac8fd
|
266 267 |
* * Caller holds exclusive lock on superblock; that lock is released. |
1da177e4c
|
268 |
*/ |
1712ac8fd
|
269 |
void deactivate_locked_super(struct super_block *s) |
1da177e4c
|
270 271 |
{ struct file_system_type *fs = s->s_type; |
b20bd1a5e
|
272 |
if (atomic_dec_and_test(&s->s_active)) { |
3167760f8
|
273 |
cleancache_invalidate_fs(s); |
b0d40c92a
|
274 |
unregister_shrinker(&s->s_shrink); |
28f2cd4f6
|
275 |
fs->kill_sb(s); |
f5e1dd345
|
276 |
|
1da177e4c
|
277 278 |
put_filesystem(fs); put_super(s); |
1712ac8fd
|
279 280 |
} else { up_write(&s->s_umount); |
1da177e4c
|
281 282 |
} } |
1712ac8fd
|
283 |
EXPORT_SYMBOL(deactivate_locked_super); |
1da177e4c
|
284 285 |
/** |
1712ac8fd
|
286 |
* deactivate_super - drop an active reference to superblock |
74dbbdd7f
|
287 288 |
* @s: superblock to deactivate * |
1712ac8fd
|
289 290 291 |
* Variant of deactivate_locked_super(), except that superblock is *not* * locked by caller. If we are going to drop the final active reference, * lock will be acquired prior to that. |
74dbbdd7f
|
292 |
*/ |
1712ac8fd
|
293 |
void deactivate_super(struct super_block *s) |
74dbbdd7f
|
294 |
{ |
1712ac8fd
|
295 296 297 |
if (!atomic_add_unless(&s->s_active, -1, 1)) { down_write(&s->s_umount); deactivate_locked_super(s); |
74dbbdd7f
|
298 299 |
} } |
1712ac8fd
|
300 |
EXPORT_SYMBOL(deactivate_super); |
74dbbdd7f
|
301 302 |
/** |
1da177e4c
|
303 304 305 306 307 308 309 310 |
* grab_super - acquire an active reference * @s: reference we are trying to make active * * Tries to acquire an active reference. grab_super() is used when we * had just found a superblock in super_blocks or fs_type->fs_supers * and want to turn it into a full-blown active reference. grab_super() * is called with sb_lock held and drops it. Returns 1 in case of * success, 0 if we had failed (superblock contents was already dead or |
acfec9a5a
|
311 312 313 |
* dying when grab_super() had been called). Note that this is only * called for superblocks not in rundown mode (== ones still on ->fs_supers * of their type), so increment of ->s_count is OK here. |
1da177e4c
|
314 |
*/ |
9c4dbee79
|
315 |
static int grab_super(struct super_block *s) __releases(sb_lock) |
1da177e4c
|
316 317 318 319 |
{ s->s_count++; spin_unlock(&sb_lock); down_write(&s->s_umount); |
acfec9a5a
|
320 321 322 323 |
if ((s->s_flags & MS_BORN) && atomic_inc_not_zero(&s->s_active)) { put_super(s); return 1; } |
1da177e4c
|
324 325 |
up_write(&s->s_umount); put_super(s); |
1da177e4c
|
326 327 |
return 0; } |
cf9a2ae8d
|
328 |
/* |
12ad3ab66
|
329 |
* grab_super_passive - acquire a passive reference |
331cbdeed
|
330 |
* @sb: reference we are trying to grab |
12ad3ab66
|
331 332 333 334 335 336 337 338 339 340 341 342 |
* * Tries to acquire a passive reference. This is used in places where we * cannot take an active reference but we need to ensure that the * superblock does not go away while we are working on it. It returns * false if a reference was not gained, and returns true with the s_umount * lock held in read mode if a reference is gained. On successful return, * the caller must drop the s_umount lock and the passive reference when * done. */ bool grab_super_passive(struct super_block *sb) { spin_lock(&sb_lock); |
a5166169f
|
343 |
if (hlist_unhashed(&sb->s_instances)) { |
12ad3ab66
|
344 345 346 347 348 349 350 351 |
spin_unlock(&sb_lock); return false; } sb->s_count++; spin_unlock(&sb_lock); if (down_read_trylock(&sb->s_umount)) { |
dabe0dc19
|
352 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
12ad3ab66
|
353 354 355 356 357 358 359 |
return true; up_read(&sb->s_umount); } put_super(sb); return false; } |
1da177e4c
|
360 361 362 363 364 365 366 367 368 |
/** * generic_shutdown_super - common helper for ->kill_sb() * @sb: superblock to kill * * generic_shutdown_super() does all fs-independent work on superblock * shutdown. Typical ->kill_sb() should pick all fs-specific objects * that need destruction out of superblock, call generic_shutdown_super() * and release aforementioned objects. Note: dentries and inodes _are_ * taken care of and do not need specific handling. |
c636ebdb1
|
369 370 371 372 |
* * Upon calling this function, the filesystem may no longer alter or * rearrange the set of dentries belonging to this super_block, nor may it * change the attachments of dentries to inodes. |
1da177e4c
|
373 374 375 |
*/ void generic_shutdown_super(struct super_block *sb) { |
ee9b6d61a
|
376 |
const struct super_operations *sop = sb->s_op; |
1da177e4c
|
377 |
|
c636ebdb1
|
378 379 |
if (sb->s_root) { shrink_dcache_for_umount(sb); |
60b0680fa
|
380 |
sync_filesystem(sb); |
1da177e4c
|
381 |
sb->s_flags &= ~MS_ACTIVE; |
efaee1920
|
382 |
|
63997e98a
|
383 384 385 |
fsnotify_unmount_inodes(&sb->s_inodes); evict_inodes(sb); |
1da177e4c
|
386 |
|
7b7a8665e
|
387 388 389 390 |
if (sb->s_dio_done_wq) { destroy_workqueue(sb->s_dio_done_wq); sb->s_dio_done_wq = NULL; } |
1da177e4c
|
391 392 |
if (sop->put_super) sop->put_super(sb); |
63997e98a
|
393 |
if (!list_empty(&sb->s_inodes)) { |
7b4fe29e0
|
394 395 396 397 |
printk("VFS: Busy inodes after unmount of %s. " "Self-destruct in 5 seconds. Have a nice day... ", sb->s_id); |
1da177e4c
|
398 |
} |
1da177e4c
|
399 400 401 |
} spin_lock(&sb_lock); /* should be initialized for __put_super_and_need_restart() */ |
a5166169f
|
402 |
hlist_del_init(&sb->s_instances); |
1da177e4c
|
403 404 405 406 407 408 409 410 411 412 413 |
spin_unlock(&sb_lock); up_write(&sb->s_umount); } EXPORT_SYMBOL(generic_shutdown_super); /** * sget - find or create a superblock * @type: filesystem type superblock should belong to * @test: comparison callback * @set: setup callback |
9249e17fe
|
414 |
* @flags: mount flags |
1da177e4c
|
415 416 417 418 419 |
* @data: argument to each of them */ struct super_block *sget(struct file_system_type *type, int (*test)(struct super_block *,void *), int (*set)(struct super_block *,void *), |
9249e17fe
|
420 |
int flags, |
1da177e4c
|
421 422 423 |
void *data) { struct super_block *s = NULL; |
d47301271
|
424 |
struct super_block *old; |
1da177e4c
|
425 426 427 428 |
int err; retry: spin_lock(&sb_lock); |
d47301271
|
429 |
if (test) { |
b67bfe0d4
|
430 |
hlist_for_each_entry(old, &type->fs_supers, s_instances) { |
d47301271
|
431 432 433 434 |
if (!test(old, data)) continue; if (!grab_super(old)) goto retry; |
a3cfbb53b
|
435 436 |
if (s) { up_write(&s->s_umount); |
d47301271
|
437 |
destroy_super(s); |
7a4dec538
|
438 |
s = NULL; |
a3cfbb53b
|
439 |
} |
d47301271
|
440 441 |
return old; } |
1da177e4c
|
442 443 444 |
} if (!s) { spin_unlock(&sb_lock); |
9249e17fe
|
445 |
s = alloc_super(type, flags); |
1da177e4c
|
446 447 448 449 450 451 452 453 |
if (!s) return ERR_PTR(-ENOMEM); goto retry; } err = set(s, data); if (err) { spin_unlock(&sb_lock); |
a3cfbb53b
|
454 |
up_write(&s->s_umount); |
1da177e4c
|
455 456 457 458 459 460 |
destroy_super(s); return ERR_PTR(err); } s->s_type = type; strlcpy(s->s_id, type->name, sizeof(s->s_id)); list_add_tail(&s->s_list, &super_blocks); |
a5166169f
|
461 |
hlist_add_head(&s->s_instances, &type->fs_supers); |
1da177e4c
|
462 463 |
spin_unlock(&sb_lock); get_filesystem(type); |
b0d40c92a
|
464 |
register_shrinker(&s->s_shrink); |
1da177e4c
|
465 466 467 468 469 470 471 472 473 474 475 476 |
return s; } EXPORT_SYMBOL(sget); void drop_super(struct super_block *sb) { up_read(&sb->s_umount); put_super(sb); } EXPORT_SYMBOL(drop_super); |
e50047533
|
477 |
/** |
01a05b337
|
478 479 480 481 482 483 484 485 486 |
* iterate_supers - call function for all active superblocks * @f: function to call * @arg: argument to pass to it * * Scans the superblock list and calls given function, passing it * locked superblock and given argument. */ void iterate_supers(void (*f)(struct super_block *, void *), void *arg) { |
dca332528
|
487 |
struct super_block *sb, *p = NULL; |
01a05b337
|
488 489 |
spin_lock(&sb_lock); |
dca332528
|
490 |
list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
491 |
if (hlist_unhashed(&sb->s_instances)) |
01a05b337
|
492 493 494 495 496 |
continue; sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); |
dabe0dc19
|
497 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
01a05b337
|
498 499 500 501 |
f(sb, arg); up_read(&sb->s_umount); spin_lock(&sb_lock); |
dca332528
|
502 503 504 |
if (p) __put_super(p); p = sb; |
01a05b337
|
505 |
} |
dca332528
|
506 507 |
if (p) __put_super(p); |
01a05b337
|
508 509 510 511 |
spin_unlock(&sb_lock); } /** |
43e15cdbe
|
512 513 514 515 516 517 518 519 520 521 522 523 524 525 |
* iterate_supers_type - call function for superblocks of given type * @type: fs type * @f: function to call * @arg: argument to pass to it * * Scans the superblock list and calls given function, passing it * locked superblock and given argument. */ void iterate_supers_type(struct file_system_type *type, void (*f)(struct super_block *, void *), void *arg) { struct super_block *sb, *p = NULL; spin_lock(&sb_lock); |
b67bfe0d4
|
526 |
hlist_for_each_entry(sb, &type->fs_supers, s_instances) { |
43e15cdbe
|
527 528 529 530 |
sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); |
dabe0dc19
|
531 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
43e15cdbe
|
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 |
f(sb, arg); up_read(&sb->s_umount); spin_lock(&sb_lock); if (p) __put_super(p); p = sb; } if (p) __put_super(p); spin_unlock(&sb_lock); } EXPORT_SYMBOL(iterate_supers_type); /** |
1da177e4c
|
548 549 550 551 552 553 |
* get_super - get the superblock of a device * @bdev: device to get the superblock for * * Scans the superblock list and finds the superblock of the file system * mounted on the device given. %NULL is returned if no match is found. */ |
df40c01a9
|
554 |
struct super_block *get_super(struct block_device *bdev) |
1da177e4c
|
555 |
{ |
618f06362
|
556 |
struct super_block *sb; |
1da177e4c
|
557 558 |
if (!bdev) return NULL; |
618f06362
|
559 |
|
1da177e4c
|
560 |
spin_lock(&sb_lock); |
618f06362
|
561 562 |
rescan: list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
563 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
564 |
continue; |
618f06362
|
565 566 |
if (sb->s_bdev == bdev) { sb->s_count++; |
1da177e4c
|
567 |
spin_unlock(&sb_lock); |
618f06362
|
568 |
down_read(&sb->s_umount); |
df40c01a9
|
569 |
/* still alive? */ |
dabe0dc19
|
570 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
618f06362
|
571 572 |
return sb; up_read(&sb->s_umount); |
df40c01a9
|
573 |
/* nope, got unmounted */ |
618f06362
|
574 |
spin_lock(&sb_lock); |
df40c01a9
|
575 576 |
__put_super(sb); goto rescan; |
1da177e4c
|
577 578 579 580 581 582 583 |
} } spin_unlock(&sb_lock); return NULL; } EXPORT_SYMBOL(get_super); |
4504230a7
|
584 585 |
/** |
6b6dc836a
|
586 587 588 589 590 591 592 593 594 595 596 597 |
* get_super_thawed - get thawed superblock of a device * @bdev: device to get the superblock for * * Scans the superblock list and finds the superblock of the file system * mounted on the device. The superblock is returned once it is thawed * (or immediately if it was not frozen). %NULL is returned if no match * is found. */ struct super_block *get_super_thawed(struct block_device *bdev) { while (1) { struct super_block *s = get_super(bdev); |
5accdf82b
|
598 |
if (!s || s->s_writers.frozen == SB_UNFROZEN) |
6b6dc836a
|
599 600 |
return s; up_read(&s->s_umount); |
5accdf82b
|
601 602 |
wait_event(s->s_writers.wait_unfrozen, s->s_writers.frozen == SB_UNFROZEN); |
6b6dc836a
|
603 604 605 606 607 608 |
put_super(s); } } EXPORT_SYMBOL(get_super_thawed); /** |
4504230a7
|
609 610 611 612 613 |
* get_active_super - get an active reference to the superblock of a device * @bdev: device to get the superblock for * * Scans the superblock list and finds the superblock of the file system * mounted on the device given. Returns the superblock with an active |
d3f214730
|
614 |
* reference or %NULL if none was found. |
4504230a7
|
615 616 617 618 619 620 621 |
*/ struct super_block *get_active_super(struct block_device *bdev) { struct super_block *sb; if (!bdev) return NULL; |
1494583de
|
622 |
restart: |
4504230a7
|
623 624 |
spin_lock(&sb_lock); list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
625 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
626 |
continue; |
1494583de
|
627 |
if (sb->s_bdev == bdev) { |
acfec9a5a
|
628 |
if (!grab_super(sb)) |
1494583de
|
629 |
goto restart; |
acfec9a5a
|
630 631 |
up_write(&sb->s_umount); return sb; |
1494583de
|
632 |
} |
4504230a7
|
633 634 635 636 |
} spin_unlock(&sb_lock); return NULL; } |
1da177e4c
|
637 |
|
df40c01a9
|
638 |
struct super_block *user_get_super(dev_t dev) |
1da177e4c
|
639 |
{ |
618f06362
|
640 |
struct super_block *sb; |
1da177e4c
|
641 |
|
1da177e4c
|
642 |
spin_lock(&sb_lock); |
618f06362
|
643 644 |
rescan: list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
645 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
646 |
continue; |
618f06362
|
647 648 |
if (sb->s_dev == dev) { sb->s_count++; |
1da177e4c
|
649 |
spin_unlock(&sb_lock); |
618f06362
|
650 |
down_read(&sb->s_umount); |
df40c01a9
|
651 |
/* still alive? */ |
dabe0dc19
|
652 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
618f06362
|
653 654 |
return sb; up_read(&sb->s_umount); |
df40c01a9
|
655 |
/* nope, got unmounted */ |
618f06362
|
656 |
spin_lock(&sb_lock); |
df40c01a9
|
657 658 |
__put_super(sb); goto rescan; |
1da177e4c
|
659 660 661 662 663 |
} } spin_unlock(&sb_lock); return NULL; } |
1da177e4c
|
664 |
/** |
1da177e4c
|
665 666 667 668 669 670 671 672 673 674 675 |
* do_remount_sb - asks filesystem to change mount options. * @sb: superblock in question * @flags: numeric part of options * @data: the rest of options * @force: whether or not to force the change * * Alters the mount options of a mounted file system. */ int do_remount_sb(struct super_block *sb, int flags, void *data, int force) { int retval; |
c79d967de
|
676 |
int remount_ro; |
4504230a7
|
677 |
|
5accdf82b
|
678 |
if (sb->s_writers.frozen != SB_UNFROZEN) |
4504230a7
|
679 |
return -EBUSY; |
9361401eb
|
680 |
#ifdef CONFIG_BLOCK |
1da177e4c
|
681 682 |
if (!(flags & MS_RDONLY) && bdev_read_only(sb->s_bdev)) return -EACCES; |
9361401eb
|
683 |
#endif |
4504230a7
|
684 |
|
d208bbdda
|
685 |
remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY); |
d208bbdda
|
686 |
|
0aec09d04
|
687 688 689 |
if (remount_ro) { if (sb->s_pins.first) { up_write(&sb->s_umount); |
8fa1f1c2b
|
690 |
sb_pin_kill(sb); |
0aec09d04
|
691 692 693 694 695 696 697 698 699 |
down_write(&sb->s_umount); if (!sb->s_root) return 0; if (sb->s_writers.frozen != SB_UNFROZEN) return -EBUSY; remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY); } } shrink_dcache_sb(sb); |
1da177e4c
|
700 701 |
/* If we are remounting RDONLY and current sb is read/write, make sure there are no rw files opened */ |
d208bbdda
|
702 |
if (remount_ro) { |
4ed5e82fe
|
703 |
if (force) { |
eee5cc270
|
704 705 |
sb->s_readonly_remount = 1; smp_wmb(); |
4ed5e82fe
|
706 707 708 709 |
} else { retval = sb_prepare_remount_readonly(sb); if (retval) return retval; |
4ed5e82fe
|
710 |
} |
1da177e4c
|
711 712 713 |
} if (sb->s_op->remount_fs) { |
1da177e4c
|
714 |
retval = sb->s_op->remount_fs(sb, &flags, data); |
2833eb2b4
|
715 716 |
if (retval) { if (!force) |
4ed5e82fe
|
717 |
goto cancel_readonly; |
2833eb2b4
|
718 719 720 721 722 |
/* If forced remount, go ahead despite any errors */ WARN(1, "forced remount of a %s fs returned %i ", sb->s_type->name, retval); } |
1da177e4c
|
723 724 |
} sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK); |
4ed5e82fe
|
725 726 727 |
/* Needs to be ordered wrt mnt_is_readonly() */ smp_wmb(); sb->s_readonly_remount = 0; |
c79d967de
|
728 |
|
d208bbdda
|
729 730 731 732 733 734 735 736 737 738 |
/* * Some filesystems modify their metadata via some other path than the * bdev buffer cache (eg. use a private mapping, or directories in * pagecache, etc). Also file data modifications go via their own * mappings. So If we try to mount readonly then copy the filesystem * from bdev, we could get stale data, so invalidate it to give a best * effort at coherency. */ if (remount_ro && sb->s_bdev) invalidate_bdev(sb->s_bdev); |
1da177e4c
|
739 |
return 0; |
4ed5e82fe
|
740 741 742 743 |
cancel_readonly: sb->s_readonly_remount = 0; return retval; |
1da177e4c
|
744 |
} |
a2a9537ac
|
745 |
static void do_emergency_remount(struct work_struct *work) |
1da177e4c
|
746 |
{ |
dca332528
|
747 |
struct super_block *sb, *p = NULL; |
1da177e4c
|
748 749 |
spin_lock(&sb_lock); |
dca332528
|
750 |
list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
751 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
752 |
continue; |
1da177e4c
|
753 754 |
sb->s_count++; spin_unlock(&sb_lock); |
443b94baa
|
755 |
down_write(&sb->s_umount); |
dabe0dc19
|
756 757 |
if (sb->s_root && sb->s_bdev && (sb->s_flags & MS_BORN) && !(sb->s_flags & MS_RDONLY)) { |
1da177e4c
|
758 |
/* |
1da177e4c
|
759 760 |
* What lock protects sb->s_flags?? */ |
1da177e4c
|
761 |
do_remount_sb(sb, MS_RDONLY, NULL, 1); |
1da177e4c
|
762 |
} |
443b94baa
|
763 |
up_write(&sb->s_umount); |
1da177e4c
|
764 |
spin_lock(&sb_lock); |
dca332528
|
765 766 767 |
if (p) __put_super(p); p = sb; |
1da177e4c
|
768 |
} |
dca332528
|
769 770 |
if (p) __put_super(p); |
1da177e4c
|
771 |
spin_unlock(&sb_lock); |
a2a9537ac
|
772 |
kfree(work); |
1da177e4c
|
773 774 775 776 777 778 |
printk("Emergency Remount complete "); } void emergency_remount(void) { |
a2a9537ac
|
779 780 781 782 783 784 785 |
struct work_struct *work; work = kmalloc(sizeof(*work), GFP_ATOMIC); if (work) { INIT_WORK(work, do_emergency_remount); schedule_work(work); } |
1da177e4c
|
786 787 788 789 790 791 |
} /* * Unnamed block devices are dummy devices used by virtual * filesystems which don't use real block-devices. -- jrs */ |
ad76cbc63
|
792 |
static DEFINE_IDA(unnamed_dev_ida); |
1da177e4c
|
793 |
static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */ |
a2a4dc494
|
794 795 796 797 |
/* Many userspace utilities consider an FSID of 0 invalid. * Always return at least 1 from get_anon_bdev. */ static int unnamed_dev_start = 1; |
1da177e4c
|
798 |
|
0ee5dc676
|
799 |
int get_anon_bdev(dev_t *p) |
1da177e4c
|
800 801 802 803 804 |
{ int dev; int error; retry: |
ad76cbc63
|
805 |
if (ida_pre_get(&unnamed_dev_ida, GFP_ATOMIC) == 0) |
1da177e4c
|
806 807 |
return -ENOMEM; spin_lock(&unnamed_dev_lock); |
c63e09ecc
|
808 |
error = ida_get_new_above(&unnamed_dev_ida, unnamed_dev_start, &dev); |
f21f62208
|
809 810 |
if (!error) unnamed_dev_start = dev + 1; |
1da177e4c
|
811 812 813 814 815 816 |
spin_unlock(&unnamed_dev_lock); if (error == -EAGAIN) /* We raced and lost with another CPU. */ goto retry; else if (error) return -EAGAIN; |
e8c8d1bc0
|
817 |
if (dev == (1 << MINORBITS)) { |
1da177e4c
|
818 |
spin_lock(&unnamed_dev_lock); |
ad76cbc63
|
819 |
ida_remove(&unnamed_dev_ida, dev); |
f21f62208
|
820 821 |
if (unnamed_dev_start > dev) unnamed_dev_start = dev; |
1da177e4c
|
822 823 824 |
spin_unlock(&unnamed_dev_lock); return -EMFILE; } |
0ee5dc676
|
825 |
*p = MKDEV(0, dev & MINORMASK); |
1da177e4c
|
826 827 |
return 0; } |
0ee5dc676
|
828 |
EXPORT_SYMBOL(get_anon_bdev); |
1da177e4c
|
829 |
|
0ee5dc676
|
830 |
void free_anon_bdev(dev_t dev) |
1da177e4c
|
831 |
{ |
0ee5dc676
|
832 |
int slot = MINOR(dev); |
1da177e4c
|
833 |
spin_lock(&unnamed_dev_lock); |
ad76cbc63
|
834 |
ida_remove(&unnamed_dev_ida, slot); |
c63e09ecc
|
835 836 |
if (slot < unnamed_dev_start) unnamed_dev_start = slot; |
1da177e4c
|
837 838 |
spin_unlock(&unnamed_dev_lock); } |
0ee5dc676
|
839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 |
EXPORT_SYMBOL(free_anon_bdev); int set_anon_super(struct super_block *s, void *data) { int error = get_anon_bdev(&s->s_dev); if (!error) s->s_bdi = &noop_backing_dev_info; return error; } EXPORT_SYMBOL(set_anon_super); void kill_anon_super(struct super_block *sb) { dev_t dev = sb->s_dev; generic_shutdown_super(sb); free_anon_bdev(dev); } |
1da177e4c
|
857 858 |
EXPORT_SYMBOL(kill_anon_super); |
1da177e4c
|
859 860 861 862 863 864 865 866 |
void kill_litter_super(struct super_block *sb) { if (sb->s_root) d_genocide(sb->s_root); kill_anon_super(sb); } EXPORT_SYMBOL(kill_litter_super); |
909e6d947
|
867 868 869 870 871 872 873 874 875 876 |
static int ns_test_super(struct super_block *sb, void *data) { return sb->s_fs_info == data; } static int ns_set_super(struct super_block *sb, void *data) { sb->s_fs_info = data; return set_anon_super(sb, NULL); } |
ceefda693
|
877 878 |
struct dentry *mount_ns(struct file_system_type *fs_type, int flags, void *data, int (*fill_super)(struct super_block *, void *, int)) |
909e6d947
|
879 880 |
{ struct super_block *sb; |
9249e17fe
|
881 |
sb = sget(fs_type, ns_test_super, ns_set_super, flags, data); |
909e6d947
|
882 |
if (IS_ERR(sb)) |
ceefda693
|
883 |
return ERR_CAST(sb); |
909e6d947
|
884 885 886 |
if (!sb->s_root) { int err; |
909e6d947
|
887 888 |
err = fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) { |
74dbbdd7f
|
889 |
deactivate_locked_super(sb); |
ceefda693
|
890 |
return ERR_PTR(err); |
909e6d947
|
891 892 893 894 |
} sb->s_flags |= MS_ACTIVE; } |
ceefda693
|
895 |
return dget(sb->s_root); |
909e6d947
|
896 |
} |
ceefda693
|
897 |
EXPORT_SYMBOL(mount_ns); |
909e6d947
|
898 |
|
9361401eb
|
899 |
#ifdef CONFIG_BLOCK |
1da177e4c
|
900 901 902 903 |
static int set_bdev_super(struct super_block *s, void *data) { s->s_bdev = data; s->s_dev = s->s_bdev->bd_dev; |
32a88aa1b
|
904 905 906 907 908 909 |
/* * We set the bdi here to the queue backing, file systems can * overwrite this in ->fill_super() */ s->s_bdi = &bdev_get_queue(s->s_bdev)->backing_dev_info; |
1da177e4c
|
910 911 912 913 914 915 916 |
return 0; } static int test_bdev_super(struct super_block *s, void *data) { return (void *)s->s_bdev == data; } |
152a08366
|
917 |
struct dentry *mount_bdev(struct file_system_type *fs_type, |
1da177e4c
|
918 |
int flags, const char *dev_name, void *data, |
152a08366
|
919 |
int (*fill_super)(struct super_block *, void *, int)) |
1da177e4c
|
920 921 922 |
{ struct block_device *bdev; struct super_block *s; |
d4d776299
|
923 |
fmode_t mode = FMODE_READ | FMODE_EXCL; |
1da177e4c
|
924 |
int error = 0; |
30c40d2c0
|
925 926 |
if (!(flags & MS_RDONLY)) mode |= FMODE_WRITE; |
d4d776299
|
927 |
bdev = blkdev_get_by_path(dev_name, mode, fs_type); |
1da177e4c
|
928 |
if (IS_ERR(bdev)) |
152a08366
|
929 |
return ERR_CAST(bdev); |
1da177e4c
|
930 931 932 933 934 935 |
/* * once the super is inserted into the list by sget, s_umount * will protect the lockfs code from trying to start a snapshot * while we are mounting */ |
4fadd7bb2
|
936 937 938 939 940 941 |
mutex_lock(&bdev->bd_fsfreeze_mutex); if (bdev->bd_fsfreeze_count > 0) { mutex_unlock(&bdev->bd_fsfreeze_mutex); error = -EBUSY; goto error_bdev; } |
9249e17fe
|
942 943 |
s = sget(fs_type, test_bdev_super, set_bdev_super, flags | MS_NOSEC, bdev); |
4fadd7bb2
|
944 |
mutex_unlock(&bdev->bd_fsfreeze_mutex); |
1da177e4c
|
945 |
if (IS_ERR(s)) |
454e2398b
|
946 |
goto error_s; |
1da177e4c
|
947 948 949 |
if (s->s_root) { if ((flags ^ s->s_flags) & MS_RDONLY) { |
74dbbdd7f
|
950 |
deactivate_locked_super(s); |
454e2398b
|
951 952 |
error = -EBUSY; goto error_bdev; |
1da177e4c
|
953 |
} |
454e2398b
|
954 |
|
4f331f01b
|
955 956 |
/* * s_umount nests inside bd_mutex during |
e525fd89d
|
957 958 959 960 |
* __invalidate_device(). blkdev_put() acquires * bd_mutex and can't be called under s_umount. Drop * s_umount temporarily. This is safe as we're * holding an active reference. |
4f331f01b
|
961 962 |
*/ up_write(&s->s_umount); |
d4d776299
|
963 |
blkdev_put(bdev, mode); |
4f331f01b
|
964 |
down_write(&s->s_umount); |
1da177e4c
|
965 966 |
} else { char b[BDEVNAME_SIZE]; |
30c40d2c0
|
967 |
s->s_mode = mode; |
1da177e4c
|
968 |
strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); |
e78c9a004
|
969 |
sb_set_blocksize(s, block_size(bdev)); |
9b04c997b
|
970 |
error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); |
1da177e4c
|
971 |
if (error) { |
74dbbdd7f
|
972 |
deactivate_locked_super(s); |
454e2398b
|
973 |
goto error; |
fa675765a
|
974 |
} |
454e2398b
|
975 976 |
s->s_flags |= MS_ACTIVE; |
87d8fe1ee
|
977 |
bdev->bd_super = s; |
1da177e4c
|
978 |
} |
152a08366
|
979 |
return dget(s->s_root); |
1da177e4c
|
980 |
|
454e2398b
|
981 982 983 |
error_s: error = PTR_ERR(s); error_bdev: |
d4d776299
|
984 |
blkdev_put(bdev, mode); |
454e2398b
|
985 |
error: |
152a08366
|
986 987 988 |
return ERR_PTR(error); } EXPORT_SYMBOL(mount_bdev); |
1da177e4c
|
989 990 991 |
void kill_block_super(struct super_block *sb) { struct block_device *bdev = sb->s_bdev; |
30c40d2c0
|
992 |
fmode_t mode = sb->s_mode; |
1da177e4c
|
993 |
|
ddbaaf302
|
994 |
bdev->bd_super = NULL; |
1da177e4c
|
995 996 |
generic_shutdown_super(sb); sync_blockdev(bdev); |
d4d776299
|
997 |
WARN_ON_ONCE(!(mode & FMODE_EXCL)); |
e525fd89d
|
998 |
blkdev_put(bdev, mode | FMODE_EXCL); |
1da177e4c
|
999 1000 1001 |
} EXPORT_SYMBOL(kill_block_super); |
9361401eb
|
1002 |
#endif |
1da177e4c
|
1003 |
|
3c26ff6e4
|
1004 |
struct dentry *mount_nodev(struct file_system_type *fs_type, |
1da177e4c
|
1005 |
int flags, void *data, |
3c26ff6e4
|
1006 |
int (*fill_super)(struct super_block *, void *, int)) |
1da177e4c
|
1007 1008 |
{ int error; |
9249e17fe
|
1009 |
struct super_block *s = sget(fs_type, NULL, set_anon_super, flags, NULL); |
1da177e4c
|
1010 1011 |
if (IS_ERR(s)) |
3c26ff6e4
|
1012 |
return ERR_CAST(s); |
1da177e4c
|
1013 |
|
9b04c997b
|
1014 |
error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); |
1da177e4c
|
1015 |
if (error) { |
74dbbdd7f
|
1016 |
deactivate_locked_super(s); |
3c26ff6e4
|
1017 |
return ERR_PTR(error); |
1da177e4c
|
1018 1019 |
} s->s_flags |= MS_ACTIVE; |
3c26ff6e4
|
1020 |
return dget(s->s_root); |
1da177e4c
|
1021 |
} |
3c26ff6e4
|
1022 |
EXPORT_SYMBOL(mount_nodev); |
1da177e4c
|
1023 1024 1025 1026 |
static int compare_single(struct super_block *s, void *p) { return 1; } |
fc14f2fef
|
1027 |
struct dentry *mount_single(struct file_system_type *fs_type, |
1da177e4c
|
1028 |
int flags, void *data, |
fc14f2fef
|
1029 |
int (*fill_super)(struct super_block *, void *, int)) |
1da177e4c
|
1030 1031 1032 |
{ struct super_block *s; int error; |
9249e17fe
|
1033 |
s = sget(fs_type, compare_single, set_anon_super, flags, NULL); |
1da177e4c
|
1034 |
if (IS_ERR(s)) |
fc14f2fef
|
1035 |
return ERR_CAST(s); |
1da177e4c
|
1036 |
if (!s->s_root) { |
9b04c997b
|
1037 |
error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); |
1da177e4c
|
1038 |
if (error) { |
74dbbdd7f
|
1039 |
deactivate_locked_super(s); |
fc14f2fef
|
1040 |
return ERR_PTR(error); |
1da177e4c
|
1041 1042 |
} s->s_flags |= MS_ACTIVE; |
9329d1bea
|
1043 1044 |
} else { do_remount_sb(s, flags, data, 0); |
1da177e4c
|
1045 |
} |
fc14f2fef
|
1046 1047 1048 |
return dget(s->s_root); } EXPORT_SYMBOL(mount_single); |
9d412a43c
|
1049 1050 |
struct dentry * mount_fs(struct file_system_type *type, int flags, const char *name, void *data) |
1da177e4c
|
1051 |
{ |
c96e41e92
|
1052 |
struct dentry *root; |
9d412a43c
|
1053 |
struct super_block *sb; |
1da177e4c
|
1054 |
char *secdata = NULL; |
9d412a43c
|
1055 |
int error = -ENOMEM; |
8089352a1
|
1056 |
|
e00075298
|
1057 |
if (data && !(type->fs_flags & FS_BINARY_MOUNTDATA)) { |
1da177e4c
|
1058 |
secdata = alloc_secdata(); |
454e2398b
|
1059 |
if (!secdata) |
9d412a43c
|
1060 |
goto out; |
1da177e4c
|
1061 |
|
e00075298
|
1062 |
error = security_sb_copy_data(data, secdata); |
454e2398b
|
1063 |
if (error) |
1da177e4c
|
1064 |
goto out_free_secdata; |
1da177e4c
|
1065 |
} |
1a102ff92
|
1066 1067 1068 1069 |
root = type->mount(type, flags, name, data); if (IS_ERR(root)) { error = PTR_ERR(root); goto out_free_secdata; |
c96e41e92
|
1070 |
} |
9d412a43c
|
1071 1072 1073 |
sb = root->d_sb; BUG_ON(!sb); WARN_ON(!sb->s_bdi); |
6c5103890
|
1074 |
WARN_ON(sb->s_bdi == &default_backing_dev_info); |
9d412a43c
|
1075 |
sb->s_flags |= MS_BORN; |
454e2398b
|
1076 |
|
9d412a43c
|
1077 |
error = security_sb_kern_mount(sb, flags, secdata); |
5129a469a
|
1078 1079 |
if (error) goto out_sb; |
454e2398b
|
1080 |
|
42cb56ae2
|
1081 1082 1083 1084 |
/* * filesystems should never set s_maxbytes larger than MAX_LFS_FILESIZE * but s_maxbytes was an unsigned long long for many releases. Throw * this warning for a little while to try and catch filesystems that |
4358b5678
|
1085 |
* violate this rule. |
42cb56ae2
|
1086 |
*/ |
9d412a43c
|
1087 1088 1089 |
WARN((sb->s_maxbytes < 0), "%s set sb->s_maxbytes to " "negative value (%lld) ", type->name, sb->s_maxbytes); |
42cb56ae2
|
1090 |
|
9d412a43c
|
1091 |
up_write(&sb->s_umount); |
8680e22f2
|
1092 |
free_secdata(secdata); |
9d412a43c
|
1093 |
return root; |
1da177e4c
|
1094 |
out_sb: |
9d412a43c
|
1095 1096 |
dput(root); deactivate_locked_super(sb); |
1da177e4c
|
1097 1098 |
out_free_secdata: free_secdata(secdata); |
1da177e4c
|
1099 |
out: |
454e2398b
|
1100 |
return ERR_PTR(error); |
1da177e4c
|
1101 |
} |
5accdf82b
|
1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 |
/* * This is an internal function, please use sb_end_{write,pagefault,intwrite} * instead. */ void __sb_end_write(struct super_block *sb, int level) { percpu_counter_dec(&sb->s_writers.counter[level-1]); /* * Make sure s_writers are updated before we wake up waiters in * freeze_super(). */ smp_mb(); if (waitqueue_active(&sb->s_writers.wait)) wake_up(&sb->s_writers.wait); rwsem_release(&sb->s_writers.lock_map[level-1], 1, _RET_IP_); } EXPORT_SYMBOL(__sb_end_write); #ifdef CONFIG_LOCKDEP /* * We want lockdep to tell us about possible deadlocks with freezing but * it's it bit tricky to properly instrument it. Getting a freeze protection * works as getting a read lock but there are subtle problems. XFS for example * gets freeze protection on internal level twice in some cases, which is OK * only because we already hold a freeze protection also on higher level. Due * to these cases we have to tell lockdep we are doing trylock when we * already hold a freeze protection for a higher freeze level. */ static void acquire_freeze_lock(struct super_block *sb, int level, bool trylock, unsigned long ip) { int i; if (!trylock) { for (i = 0; i < level - 1; i++) if (lock_is_held(&sb->s_writers.lock_map[i])) { trylock = true; break; } } rwsem_acquire_read(&sb->s_writers.lock_map[level-1], 0, trylock, ip); } #endif /* * This is an internal function, please use sb_start_{write,pagefault,intwrite} * instead. */ int __sb_start_write(struct super_block *sb, int level, bool wait) { retry: if (unlikely(sb->s_writers.frozen >= level)) { if (!wait) return 0; wait_event(sb->s_writers.wait_unfrozen, sb->s_writers.frozen < level); } #ifdef CONFIG_LOCKDEP acquire_freeze_lock(sb, level, !wait, _RET_IP_); #endif percpu_counter_inc(&sb->s_writers.counter[level-1]); /* * Make sure counter is updated before we check for frozen. * freeze_super() first sets frozen and then checks the counter. */ smp_mb(); if (unlikely(sb->s_writers.frozen >= level)) { __sb_end_write(sb, level); goto retry; } return 1; } EXPORT_SYMBOL(__sb_start_write); /** * sb_wait_write - wait until all writers to given file system finish * @sb: the super for which we wait * @level: type of writers we wait for (normal vs page fault) * * This function waits until there are no writers of given type to given file * system. Caller of this function should make sure there can be no new writers * of type @level before calling this function. Otherwise this function can * livelock. */ static void sb_wait_write(struct super_block *sb, int level) { s64 writers; /* * We just cycle-through lockdep here so that it does not complain * about returning with lock to userspace */ rwsem_acquire(&sb->s_writers.lock_map[level-1], 0, 0, _THIS_IP_); rwsem_release(&sb->s_writers.lock_map[level-1], 1, _THIS_IP_); do { DEFINE_WAIT(wait); /* * We use a barrier in prepare_to_wait() to separate setting * of frozen and checking of the counter */ prepare_to_wait(&sb->s_writers.wait, &wait, TASK_UNINTERRUPTIBLE); writers = percpu_counter_sum(&sb->s_writers.counter[level-1]); if (writers) schedule(); finish_wait(&sb->s_writers.wait, &wait); } while (writers); } |
18e9e5104
|
1215 |
/** |
7000d3c42
|
1216 1217 |
* freeze_super - lock the filesystem and force it into a consistent state * @sb: the super to lock |
18e9e5104
|
1218 1219 1220 1221 |
* * Syncs the super to make sure the filesystem is consistent and calls the fs's * freeze_fs. Subsequent calls to this without first thawing the fs will return * -EBUSY. |
5accdf82b
|
1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 |
* * During this function, sb->s_writers.frozen goes through these values: * * SB_UNFROZEN: File system is normal, all writes progress as usual. * * SB_FREEZE_WRITE: The file system is in the process of being frozen. New * writes should be blocked, though page faults are still allowed. We wait for * all writes to complete and then proceed to the next stage. * * SB_FREEZE_PAGEFAULT: Freezing continues. Now also page faults are blocked * but internal fs threads can still modify the filesystem (although they * should not dirty new pages or inodes), writeback can run etc. After waiting * for all running page faults we sync the filesystem which will clean all * dirty pages and inodes (no new dirty pages or inodes can be created when * sync is running). * * SB_FREEZE_FS: The file system is frozen. Now all internal sources of fs * modification are blocked (e.g. XFS preallocation truncation on inode * reclaim). This is usually implemented by blocking new transactions for * filesystems that have them and need this additional guard. After all * internal writers are finished we call ->freeze_fs() to finish filesystem * freezing. Then we transition to SB_FREEZE_COMPLETE state. This state is * mostly auxiliary for filesystems to verify they do not modify frozen fs. * * sb->s_writers.frozen is protected by sb->s_umount. |
18e9e5104
|
1247 1248 1249 1250 1251 1252 1253 |
*/ int freeze_super(struct super_block *sb) { int ret; atomic_inc(&sb->s_active); down_write(&sb->s_umount); |
5accdf82b
|
1254 |
if (sb->s_writers.frozen != SB_UNFROZEN) { |
18e9e5104
|
1255 1256 1257 |
deactivate_locked_super(sb); return -EBUSY; } |
dabe0dc19
|
1258 1259 1260 1261 |
if (!(sb->s_flags & MS_BORN)) { up_write(&sb->s_umount); return 0; /* sic - it's "nothing to do" */ } |
18e9e5104
|
1262 |
if (sb->s_flags & MS_RDONLY) { |
5accdf82b
|
1263 1264 |
/* Nothing to do really... */ sb->s_writers.frozen = SB_FREEZE_COMPLETE; |
18e9e5104
|
1265 1266 1267 |
up_write(&sb->s_umount); return 0; } |
5accdf82b
|
1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 |
/* From now on, no new normal writers can start */ sb->s_writers.frozen = SB_FREEZE_WRITE; smp_wmb(); /* Release s_umount to preserve sb_start_write -> s_umount ordering */ up_write(&sb->s_umount); sb_wait_write(sb, SB_FREEZE_WRITE); /* Now we go and block page faults... */ down_write(&sb->s_umount); sb->s_writers.frozen = SB_FREEZE_PAGEFAULT; |
18e9e5104
|
1280 |
smp_wmb(); |
5accdf82b
|
1281 1282 1283 |
sb_wait_write(sb, SB_FREEZE_PAGEFAULT); /* All writers are done so after syncing there won't be dirty data */ |
18e9e5104
|
1284 |
sync_filesystem(sb); |
5accdf82b
|
1285 1286 |
/* Now wait for internal filesystem counter */ sb->s_writers.frozen = SB_FREEZE_FS; |
18e9e5104
|
1287 |
smp_wmb(); |
5accdf82b
|
1288 |
sb_wait_write(sb, SB_FREEZE_FS); |
18e9e5104
|
1289 |
|
18e9e5104
|
1290 1291 1292 1293 1294 1295 |
if (sb->s_op->freeze_fs) { ret = sb->s_op->freeze_fs(sb); if (ret) { printk(KERN_ERR "VFS:Filesystem freeze failed "); |
5accdf82b
|
1296 |
sb->s_writers.frozen = SB_UNFROZEN; |
e1616300a
|
1297 |
smp_wmb(); |
5accdf82b
|
1298 |
wake_up(&sb->s_writers.wait_unfrozen); |
18e9e5104
|
1299 1300 1301 1302 |
deactivate_locked_super(sb); return ret; } } |
5accdf82b
|
1303 1304 1305 1306 1307 |
/* * This is just for debugging purposes so that fs can warn if it * sees write activity when frozen is set to SB_FREEZE_COMPLETE. */ sb->s_writers.frozen = SB_FREEZE_COMPLETE; |
18e9e5104
|
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 |
up_write(&sb->s_umount); return 0; } EXPORT_SYMBOL(freeze_super); /** * thaw_super -- unlock filesystem * @sb: the super to thaw * * Unlocks the filesystem and marks it writeable again after freeze_super(). */ int thaw_super(struct super_block *sb) { int error; down_write(&sb->s_umount); |
5accdf82b
|
1324 |
if (sb->s_writers.frozen == SB_UNFROZEN) { |
18e9e5104
|
1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 |
up_write(&sb->s_umount); return -EINVAL; } if (sb->s_flags & MS_RDONLY) goto out; if (sb->s_op->unfreeze_fs) { error = sb->s_op->unfreeze_fs(sb); if (error) { printk(KERN_ERR "VFS:Filesystem thaw failed "); |
18e9e5104
|
1338 1339 1340 1341 1342 1343 |
up_write(&sb->s_umount); return error; } } out: |
5accdf82b
|
1344 |
sb->s_writers.frozen = SB_UNFROZEN; |
18e9e5104
|
1345 |
smp_wmb(); |
5accdf82b
|
1346 |
wake_up(&sb->s_writers.wait_unfrozen); |
18e9e5104
|
1347 1348 1349 1350 1351 |
deactivate_locked_super(sb); return 0; } EXPORT_SYMBOL(thaw_super); |