Blame view
fs/super.c
35.3 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 25 |
#include <linux/acct.h> #include <linux/blkdev.h> |
1da177e4c
|
26 27 |
#include <linux/mount.h> #include <linux/security.h> |
1da177e4c
|
28 29 |
#include <linux/writeback.h> /* for the emergency remount stuff */ #include <linux/idr.h> |
353ab6e97
|
30 |
#include <linux/mutex.h> |
5477d0fac
|
31 |
#include <linux/backing-dev.h> |
ceb5bdc2d
|
32 |
#include <linux/rculist_bl.h> |
c515e1fd3
|
33 |
#include <linux/cleancache.h> |
404015308
|
34 |
#include <linux/fsnotify.h> |
5accdf82b
|
35 |
#include <linux/lockdep.h> |
6d59e7f58
|
36 |
#include "internal.h" |
1da177e4c
|
37 |
|
1da177e4c
|
38 39 |
LIST_HEAD(super_blocks); DEFINE_SPINLOCK(sb_lock); |
5accdf82b
|
40 41 42 43 44 |
static char *sb_writers_name[SB_FREEZE_LEVELS] = { "sb_writers", "sb_pagefaults", "sb_internal", }; |
b0d40c92a
|
45 46 47 48 49 50 51 52 53 54 |
/* * 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. */ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) { struct super_block *sb; |
0e1fdafd9
|
55 56 |
int fs_objects = 0; int total_objects; |
b0d40c92a
|
57 58 59 60 61 62 63 64 65 66 67 |
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.. */ if (sc->nr_to_scan && !(sc->gfp_mask & __GFP_FS)) return -1; if (!grab_super_passive(sb)) |
8e125cd85
|
68 |
return -1; |
b0d40c92a
|
69 |
|
0e1fdafd9
|
70 71 72 73 74 |
if (sb->s_op && sb->s_op->nr_cached_objects) fs_objects = sb->s_op->nr_cached_objects(sb); total_objects = sb->s_nr_dentry_unused + sb->s_nr_inodes_unused + fs_objects + 1; |
b0d40c92a
|
75 |
if (sc->nr_to_scan) { |
0e1fdafd9
|
76 77 |
int dentries; int inodes; |
b0d40c92a
|
78 |
|
0e1fdafd9
|
79 80 81 82 83 84 85 86 87 88 89 90 91 92 |
/* proportion the scan between the caches */ dentries = (sc->nr_to_scan * sb->s_nr_dentry_unused) / total_objects; inodes = (sc->nr_to_scan * sb->s_nr_inodes_unused) / total_objects; if (fs_objects) fs_objects = (sc->nr_to_scan * fs_objects) / total_objects; /* * prune the dcache first as the icache is pinned by it, then * prune the icache, followed by the filesystem specific caches */ prune_dcache_sb(sb, dentries); prune_icache_sb(sb, inodes); |
b0d40c92a
|
93 |
|
0e1fdafd9
|
94 95 96 97 98 99 |
if (fs_objects && sb->s_op->free_cached_objects) { sb->s_op->free_cached_objects(sb, fs_objects); fs_objects = sb->s_op->nr_cached_objects(sb); } total_objects = sb->s_nr_dentry_unused + sb->s_nr_inodes_unused + fs_objects; |
b0d40c92a
|
100 |
} |
0e1fdafd9
|
101 |
total_objects = (total_objects / 100) * sysctl_vfs_cache_pressure; |
b0d40c92a
|
102 |
drop_super(sb); |
0e1fdafd9
|
103 |
return total_objects; |
b0d40c92a
|
104 |
} |
5accdf82b
|
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 |
static int init_sb_writers(struct super_block *s, struct file_system_type *type) { int err; int i; for (i = 0; i < SB_FREEZE_LEVELS; i++) { err = percpu_counter_init(&s->s_writers.counter[i], 0); if (err < 0) goto err_out; lockdep_init_map(&s->s_writers.lock_map[i], sb_writers_name[i], &type->s_writers_key[i], 0); } init_waitqueue_head(&s->s_writers.wait); init_waitqueue_head(&s->s_writers.wait_unfrozen); return 0; err_out: while (--i >= 0) percpu_counter_destroy(&s->s_writers.counter[i]); return err; } static void destroy_sb_writers(struct super_block *s) { int i; for (i = 0; i < SB_FREEZE_LEVELS; i++) percpu_counter_destroy(&s->s_writers.counter[i]); } |
1da177e4c
|
133 134 |
/** * alloc_super - create new superblock |
fe2bbc483
|
135 |
* @type: filesystem type superblock should belong to |
9249e17fe
|
136 |
* @flags: the mount flags |
1da177e4c
|
137 138 139 140 |
* * Allocates and initializes a new &struct super_block. alloc_super() * returns a pointer new superblock or %NULL if allocation had failed. */ |
9249e17fe
|
141 |
static struct super_block *alloc_super(struct file_system_type *type, int flags) |
1da177e4c
|
142 |
{ |
11b0b5abb
|
143 |
struct super_block *s = kzalloc(sizeof(struct super_block), GFP_USER); |
b87221de6
|
144 |
static const struct super_operations default_op; |
1da177e4c
|
145 146 |
if (s) { |
1da177e4c
|
147 |
if (security_sb_alloc(s)) { |
5accdf82b
|
148 149 150 151 |
/* * We cannot call security_sb_free() without * security_sb_alloc() succeeding. So bail out manually */ |
1da177e4c
|
152 153 154 155 |
kfree(s); s = NULL; goto out; } |
6416ccb78
|
156 157 |
#ifdef CONFIG_SMP s->s_files = alloc_percpu(struct list_head); |
5accdf82b
|
158 159 160 |
if (!s->s_files) goto err_out; else { |
6416ccb78
|
161 162 163 164 165 166 |
int i; for_each_possible_cpu(i) INIT_LIST_HEAD(per_cpu_ptr(s->s_files, i)); } #else |
1da177e4c
|
167 |
INIT_LIST_HEAD(&s->s_files); |
6416ccb78
|
168 |
#endif |
5accdf82b
|
169 170 |
if (init_sb_writers(s, type)) goto err_out; |
9249e17fe
|
171 |
s->s_flags = flags; |
95f28604a
|
172 |
s->s_bdi = &default_backing_dev_info; |
a5166169f
|
173 |
INIT_HLIST_NODE(&s->s_instances); |
ceb5bdc2d
|
174 |
INIT_HLIST_BL_HEAD(&s->s_anon); |
1da177e4c
|
175 |
INIT_LIST_HEAD(&s->s_inodes); |
da3bbdd46
|
176 |
INIT_LIST_HEAD(&s->s_dentry_lru); |
98b745c64
|
177 |
INIT_LIST_HEAD(&s->s_inode_lru); |
09cc9fc7a
|
178 |
spin_lock_init(&s->s_inode_lru_lock); |
39f7c4db1
|
179 |
INIT_LIST_HEAD(&s->s_mounts); |
1da177e4c
|
180 |
init_rwsem(&s->s_umount); |
7892f2f48
|
181 |
mutex_init(&s->s_lock); |
897c6ff95
|
182 |
lockdep_set_class(&s->s_umount, &type->s_umount_key); |
cf5162499
|
183 184 185 186 187 188 |
/* * The locking rules for s_lock are up to the * filesystem. For example ext3fs has different * lock ordering than usbfs: */ lockdep_set_class(&s->s_lock, &type->s_lock_key); |
ada723dcd
|
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 |
/* * 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); |
b20bd1a5e
|
205 |
s->s_count = 1; |
1da177e4c
|
206 |
atomic_set(&s->s_active, 1); |
a11f3a057
|
207 |
mutex_init(&s->s_vfs_rename_mutex); |
51ee049e7
|
208 |
lockdep_set_class(&s->s_vfs_rename_mutex, &type->s_vfs_rename_key); |
d3be915fc
|
209 210 |
mutex_init(&s->s_dquot.dqio_mutex); mutex_init(&s->s_dquot.dqonoff_mutex); |
1da177e4c
|
211 |
init_rwsem(&s->s_dquot.dqptr_sem); |
1da177e4c
|
212 |
s->s_maxbytes = MAX_NON_LFS; |
1da177e4c
|
213 214 |
s->s_op = &default_op; s->s_time_gran = 1000000000; |
c515e1fd3
|
215 |
s->cleancache_poolid = -1; |
b0d40c92a
|
216 217 218 |
s->s_shrink.seeks = DEFAULT_SEEKS; s->s_shrink.shrink = prune_super; |
8ab47664d
|
219 |
s->s_shrink.batch = 1024; |
1da177e4c
|
220 221 222 |
} out: return s; |
5accdf82b
|
223 224 225 226 227 228 229 230 231 232 |
err_out: security_sb_free(s); #ifdef CONFIG_SMP if (s->s_files) free_percpu(s->s_files); #endif destroy_sb_writers(s); kfree(s); s = NULL; goto out; |
1da177e4c
|
233 234 235 236 237 238 239 240 241 242 |
} /** * destroy_super - frees a superblock * @s: superblock to free * * Frees a superblock. */ static inline void destroy_super(struct super_block *s) { |
6416ccb78
|
243 244 245 |
#ifdef CONFIG_SMP free_percpu(s->s_files); #endif |
5accdf82b
|
246 |
destroy_sb_writers(s); |
1da177e4c
|
247 |
security_sb_free(s); |
39f7c4db1
|
248 |
WARN_ON(!list_empty(&s->s_mounts)); |
79c0b2df7
|
249 |
kfree(s->s_subtype); |
b3b304a23
|
250 |
kfree(s->s_options); |
1da177e4c
|
251 252 253 254 255 256 |
kfree(s); } /* Superblock refcounting */ /* |
35cf7ba0b
|
257 |
* Drop a superblock's refcount. The caller must hold sb_lock. |
1da177e4c
|
258 |
*/ |
f47ec3f28
|
259 |
static void __put_super(struct super_block *sb) |
1da177e4c
|
260 |
{ |
1da177e4c
|
261 |
if (!--sb->s_count) { |
551de6f34
|
262 |
list_del_init(&sb->s_list); |
1da177e4c
|
263 |
destroy_super(sb); |
1da177e4c
|
264 |
} |
1da177e4c
|
265 266 267 268 269 270 271 272 273 |
} /** * 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
|
274 |
static void put_super(struct super_block *sb) |
1da177e4c
|
275 276 277 278 279 280 281 282 |
{ spin_lock(&sb_lock); __put_super(sb); spin_unlock(&sb_lock); } /** |
1712ac8fd
|
283 |
* deactivate_locked_super - drop an active reference to superblock |
1da177e4c
|
284 285 |
* @s: superblock to deactivate * |
1712ac8fd
|
286 287 |
* 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
|
288 289 |
* tell fs driver to shut it down and drop the temporary reference we * had just acquired. |
1712ac8fd
|
290 291 |
* * Caller holds exclusive lock on superblock; that lock is released. |
1da177e4c
|
292 |
*/ |
1712ac8fd
|
293 |
void deactivate_locked_super(struct super_block *s) |
1da177e4c
|
294 295 |
{ struct file_system_type *fs = s->s_type; |
b20bd1a5e
|
296 |
if (atomic_dec_and_test(&s->s_active)) { |
3167760f8
|
297 |
cleancache_invalidate_fs(s); |
1da177e4c
|
298 |
fs->kill_sb(s); |
b0d40c92a
|
299 300 301 |
/* caches are now gone, we can safely kill the shrinker now */ unregister_shrinker(&s->s_shrink); |
d863b50ab
|
302 303 304 305 306 |
/* * We need to call rcu_barrier so all the delayed rcu free * inodes are flushed before we release the fs module. */ rcu_barrier(); |
1da177e4c
|
307 308 |
put_filesystem(fs); put_super(s); |
1712ac8fd
|
309 310 |
} else { up_write(&s->s_umount); |
1da177e4c
|
311 312 |
} } |
1712ac8fd
|
313 |
EXPORT_SYMBOL(deactivate_locked_super); |
1da177e4c
|
314 315 |
/** |
1712ac8fd
|
316 |
* deactivate_super - drop an active reference to superblock |
74dbbdd7f
|
317 318 |
* @s: superblock to deactivate * |
1712ac8fd
|
319 320 321 |
* 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
|
322 |
*/ |
1712ac8fd
|
323 |
void deactivate_super(struct super_block *s) |
74dbbdd7f
|
324 |
{ |
1712ac8fd
|
325 326 327 |
if (!atomic_add_unless(&s->s_active, -1, 1)) { down_write(&s->s_umount); deactivate_locked_super(s); |
74dbbdd7f
|
328 329 |
} } |
1712ac8fd
|
330 |
EXPORT_SYMBOL(deactivate_super); |
74dbbdd7f
|
331 332 |
/** |
1da177e4c
|
333 334 335 336 337 338 339 340 341 342 |
* 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 * dying when grab_super() had been called). */ |
9c4dbee79
|
343 |
static int grab_super(struct super_block *s) __releases(sb_lock) |
1da177e4c
|
344 |
{ |
b20bd1a5e
|
345 346 |
if (atomic_inc_not_zero(&s->s_active)) { spin_unlock(&sb_lock); |
b20bd1a5e
|
347 348 349 |
return 1; } /* it's going away */ |
1da177e4c
|
350 351 |
s->s_count++; spin_unlock(&sb_lock); |
1712ac8fd
|
352 |
/* wait for it to die */ |
1da177e4c
|
353 |
down_write(&s->s_umount); |
1da177e4c
|
354 355 |
up_write(&s->s_umount); put_super(s); |
1da177e4c
|
356 357 |
return 0; } |
cf9a2ae8d
|
358 |
/* |
12ad3ab66
|
359 |
* grab_super_passive - acquire a passive reference |
331cbdeed
|
360 |
* @sb: reference we are trying to grab |
12ad3ab66
|
361 362 363 364 365 366 367 368 369 370 371 372 |
* * 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
|
373 |
if (hlist_unhashed(&sb->s_instances)) { |
12ad3ab66
|
374 375 376 377 378 379 380 381 |
spin_unlock(&sb_lock); return false; } sb->s_count++; spin_unlock(&sb_lock); if (down_read_trylock(&sb->s_umount)) { |
dabe0dc19
|
382 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
12ad3ab66
|
383 384 385 386 387 388 389 390 391 |
return true; up_read(&sb->s_umount); } put_super(sb); return false; } /* |
914e26379
|
392 393 394 395 |
* Superblock locking. We really ought to get rid of these two. */ void lock_super(struct super_block * sb) { |
914e26379
|
396 397 398 399 400 |
mutex_lock(&sb->s_lock); } void unlock_super(struct super_block * sb) { |
914e26379
|
401 402 403 404 405 |
mutex_unlock(&sb->s_lock); } EXPORT_SYMBOL(lock_super); EXPORT_SYMBOL(unlock_super); |
1da177e4c
|
406 407 408 409 410 411 412 413 414 |
/** * 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
|
415 416 417 418 |
* * 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
|
419 420 421 |
*/ void generic_shutdown_super(struct super_block *sb) { |
ee9b6d61a
|
422 |
const struct super_operations *sop = sb->s_op; |
1da177e4c
|
423 |
|
c636ebdb1
|
424 425 |
if (sb->s_root) { shrink_dcache_for_umount(sb); |
60b0680fa
|
426 |
sync_filesystem(sb); |
1da177e4c
|
427 |
sb->s_flags &= ~MS_ACTIVE; |
efaee1920
|
428 |
|
63997e98a
|
429 430 431 |
fsnotify_unmount_inodes(&sb->s_inodes); evict_inodes(sb); |
1da177e4c
|
432 |
|
1da177e4c
|
433 434 |
if (sop->put_super) sop->put_super(sb); |
63997e98a
|
435 |
if (!list_empty(&sb->s_inodes)) { |
7b4fe29e0
|
436 437 438 439 |
printk("VFS: Busy inodes after unmount of %s. " "Self-destruct in 5 seconds. Have a nice day... ", sb->s_id); |
1da177e4c
|
440 |
} |
1da177e4c
|
441 442 443 |
} spin_lock(&sb_lock); /* should be initialized for __put_super_and_need_restart() */ |
a5166169f
|
444 |
hlist_del_init(&sb->s_instances); |
1da177e4c
|
445 446 447 448 449 450 451 452 453 454 455 |
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
|
456 |
* @flags: mount flags |
1da177e4c
|
457 458 459 460 461 |
* @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
|
462 |
int flags, |
1da177e4c
|
463 464 465 |
void *data) { struct super_block *s = NULL; |
a5166169f
|
466 |
struct hlist_node *node; |
d47301271
|
467 |
struct super_block *old; |
1da177e4c
|
468 469 470 471 |
int err; retry: spin_lock(&sb_lock); |
d47301271
|
472 |
if (test) { |
a5166169f
|
473 |
hlist_for_each_entry(old, node, &type->fs_supers, s_instances) { |
d47301271
|
474 475 476 477 |
if (!test(old, data)) continue; if (!grab_super(old)) goto retry; |
a3cfbb53b
|
478 479 |
if (s) { up_write(&s->s_umount); |
d47301271
|
480 |
destroy_super(s); |
7a4dec538
|
481 |
s = NULL; |
a3cfbb53b
|
482 |
} |
d3f214730
|
483 |
down_write(&old->s_umount); |
7a4dec538
|
484 485 486 487 |
if (unlikely(!(old->s_flags & MS_BORN))) { deactivate_locked_super(old); goto retry; } |
d47301271
|
488 489 |
return old; } |
1da177e4c
|
490 491 492 |
} if (!s) { spin_unlock(&sb_lock); |
9249e17fe
|
493 |
s = alloc_super(type, flags); |
1da177e4c
|
494 495 496 497 498 499 500 501 |
if (!s) return ERR_PTR(-ENOMEM); goto retry; } err = set(s, data); if (err) { spin_unlock(&sb_lock); |
a3cfbb53b
|
502 |
up_write(&s->s_umount); |
1da177e4c
|
503 504 505 506 507 508 |
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
|
509 |
hlist_add_head(&s->s_instances, &type->fs_supers); |
1da177e4c
|
510 511 |
spin_unlock(&sb_lock); get_filesystem(type); |
b0d40c92a
|
512 |
register_shrinker(&s->s_shrink); |
1da177e4c
|
513 514 515 516 517 518 519 520 521 522 523 524 |
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
|
525 |
/** |
01a05b337
|
526 527 528 529 530 531 532 533 534 |
* 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
|
535 |
struct super_block *sb, *p = NULL; |
01a05b337
|
536 537 |
spin_lock(&sb_lock); |
dca332528
|
538 |
list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
539 |
if (hlist_unhashed(&sb->s_instances)) |
01a05b337
|
540 541 542 543 544 |
continue; sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); |
dabe0dc19
|
545 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
01a05b337
|
546 547 548 549 |
f(sb, arg); up_read(&sb->s_umount); spin_lock(&sb_lock); |
dca332528
|
550 551 552 |
if (p) __put_super(p); p = sb; |
01a05b337
|
553 |
} |
dca332528
|
554 555 |
if (p) __put_super(p); |
01a05b337
|
556 557 558 559 |
spin_unlock(&sb_lock); } /** |
43e15cdbe
|
560 561 562 563 564 565 566 567 568 569 570 571 |
* 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; |
a5166169f
|
572 |
struct hlist_node *node; |
43e15cdbe
|
573 574 |
spin_lock(&sb_lock); |
a5166169f
|
575 |
hlist_for_each_entry(sb, node, &type->fs_supers, s_instances) { |
43e15cdbe
|
576 577 578 579 |
sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); |
dabe0dc19
|
580 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
43e15cdbe
|
581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 |
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
|
597 598 599 600 601 602 |
* 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
|
603 |
struct super_block *get_super(struct block_device *bdev) |
1da177e4c
|
604 |
{ |
618f06362
|
605 |
struct super_block *sb; |
1da177e4c
|
606 607 |
if (!bdev) return NULL; |
618f06362
|
608 |
|
1da177e4c
|
609 |
spin_lock(&sb_lock); |
618f06362
|
610 611 |
rescan: list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
612 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
613 |
continue; |
618f06362
|
614 615 |
if (sb->s_bdev == bdev) { sb->s_count++; |
1da177e4c
|
616 |
spin_unlock(&sb_lock); |
618f06362
|
617 |
down_read(&sb->s_umount); |
df40c01a9
|
618 |
/* still alive? */ |
dabe0dc19
|
619 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
618f06362
|
620 621 |
return sb; up_read(&sb->s_umount); |
df40c01a9
|
622 |
/* nope, got unmounted */ |
618f06362
|
623 |
spin_lock(&sb_lock); |
df40c01a9
|
624 625 |
__put_super(sb); goto rescan; |
1da177e4c
|
626 627 628 629 630 631 632 |
} } spin_unlock(&sb_lock); return NULL; } EXPORT_SYMBOL(get_super); |
4504230a7
|
633 634 |
/** |
6b6dc836a
|
635 636 637 638 639 640 641 642 643 644 645 646 |
* 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
|
647 |
if (!s || s->s_writers.frozen == SB_UNFROZEN) |
6b6dc836a
|
648 649 |
return s; up_read(&s->s_umount); |
5accdf82b
|
650 651 |
wait_event(s->s_writers.wait_unfrozen, s->s_writers.frozen == SB_UNFROZEN); |
6b6dc836a
|
652 653 654 655 656 657 |
put_super(s); } } EXPORT_SYMBOL(get_super_thawed); /** |
4504230a7
|
658 659 660 661 662 |
* 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
|
663 |
* reference or %NULL if none was found. |
4504230a7
|
664 665 666 667 668 669 670 |
*/ struct super_block *get_active_super(struct block_device *bdev) { struct super_block *sb; if (!bdev) return NULL; |
1494583de
|
671 |
restart: |
4504230a7
|
672 673 |
spin_lock(&sb_lock); list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
674 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
675 |
continue; |
1494583de
|
676 677 678 679 680 681 |
if (sb->s_bdev == bdev) { if (grab_super(sb)) /* drops sb_lock */ return sb; else goto restart; } |
4504230a7
|
682 683 684 685 |
} spin_unlock(&sb_lock); return NULL; } |
1da177e4c
|
686 |
|
df40c01a9
|
687 |
struct super_block *user_get_super(dev_t dev) |
1da177e4c
|
688 |
{ |
618f06362
|
689 |
struct super_block *sb; |
1da177e4c
|
690 |
|
1da177e4c
|
691 |
spin_lock(&sb_lock); |
618f06362
|
692 693 |
rescan: list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
694 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
695 |
continue; |
618f06362
|
696 697 |
if (sb->s_dev == dev) { sb->s_count++; |
1da177e4c
|
698 |
spin_unlock(&sb_lock); |
618f06362
|
699 |
down_read(&sb->s_umount); |
df40c01a9
|
700 |
/* still alive? */ |
dabe0dc19
|
701 |
if (sb->s_root && (sb->s_flags & MS_BORN)) |
618f06362
|
702 703 |
return sb; up_read(&sb->s_umount); |
df40c01a9
|
704 |
/* nope, got unmounted */ |
618f06362
|
705 |
spin_lock(&sb_lock); |
df40c01a9
|
706 707 |
__put_super(sb); goto rescan; |
1da177e4c
|
708 709 710 711 712 |
} } spin_unlock(&sb_lock); return NULL; } |
1da177e4c
|
713 |
/** |
1da177e4c
|
714 715 716 717 718 719 720 721 722 723 724 |
* 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
|
725 |
int remount_ro; |
4504230a7
|
726 |
|
5accdf82b
|
727 |
if (sb->s_writers.frozen != SB_UNFROZEN) |
4504230a7
|
728 |
return -EBUSY; |
9361401eb
|
729 |
#ifdef CONFIG_BLOCK |
1da177e4c
|
730 731 |
if (!(flags & MS_RDONLY) && bdev_read_only(sb->s_bdev)) return -EACCES; |
9361401eb
|
732 |
#endif |
4504230a7
|
733 |
|
1da177e4c
|
734 735 736 |
if (flags & MS_RDONLY) acct_auto_close(sb); shrink_dcache_sb(sb); |
60b0680fa
|
737 |
sync_filesystem(sb); |
1da177e4c
|
738 |
|
d208bbdda
|
739 |
remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY); |
d208bbdda
|
740 |
|
1da177e4c
|
741 742 |
/* If we are remounting RDONLY and current sb is read/write, make sure there are no rw files opened */ |
d208bbdda
|
743 |
if (remount_ro) { |
4ed5e82fe
|
744 |
if (force) { |
1da177e4c
|
745 |
mark_files_ro(sb); |
4ed5e82fe
|
746 747 748 749 |
} else { retval = sb_prepare_remount_readonly(sb); if (retval) return retval; |
4ed5e82fe
|
750 |
} |
1da177e4c
|
751 752 753 |
} if (sb->s_op->remount_fs) { |
1da177e4c
|
754 |
retval = sb->s_op->remount_fs(sb, &flags, data); |
2833eb2b4
|
755 756 |
if (retval) { if (!force) |
4ed5e82fe
|
757 |
goto cancel_readonly; |
2833eb2b4
|
758 759 760 761 762 |
/* If forced remount, go ahead despite any errors */ WARN(1, "forced remount of a %s fs returned %i ", sb->s_type->name, retval); } |
1da177e4c
|
763 764 |
} sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK); |
4ed5e82fe
|
765 766 767 |
/* Needs to be ordered wrt mnt_is_readonly() */ smp_wmb(); sb->s_readonly_remount = 0; |
c79d967de
|
768 |
|
d208bbdda
|
769 770 771 772 773 774 775 776 777 778 |
/* * 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
|
779 |
return 0; |
4ed5e82fe
|
780 781 782 783 |
cancel_readonly: sb->s_readonly_remount = 0; return retval; |
1da177e4c
|
784 |
} |
a2a9537ac
|
785 |
static void do_emergency_remount(struct work_struct *work) |
1da177e4c
|
786 |
{ |
dca332528
|
787 |
struct super_block *sb, *p = NULL; |
1da177e4c
|
788 789 |
spin_lock(&sb_lock); |
dca332528
|
790 |
list_for_each_entry(sb, &super_blocks, s_list) { |
a5166169f
|
791 |
if (hlist_unhashed(&sb->s_instances)) |
551de6f34
|
792 |
continue; |
1da177e4c
|
793 794 |
sb->s_count++; spin_unlock(&sb_lock); |
443b94baa
|
795 |
down_write(&sb->s_umount); |
dabe0dc19
|
796 797 |
if (sb->s_root && sb->s_bdev && (sb->s_flags & MS_BORN) && !(sb->s_flags & MS_RDONLY)) { |
1da177e4c
|
798 |
/* |
1da177e4c
|
799 800 |
* What lock protects sb->s_flags?? */ |
1da177e4c
|
801 |
do_remount_sb(sb, MS_RDONLY, NULL, 1); |
1da177e4c
|
802 |
} |
443b94baa
|
803 |
up_write(&sb->s_umount); |
1da177e4c
|
804 |
spin_lock(&sb_lock); |
dca332528
|
805 806 807 |
if (p) __put_super(p); p = sb; |
1da177e4c
|
808 |
} |
dca332528
|
809 810 |
if (p) __put_super(p); |
1da177e4c
|
811 |
spin_unlock(&sb_lock); |
a2a9537ac
|
812 |
kfree(work); |
1da177e4c
|
813 814 815 816 817 818 |
printk("Emergency Remount complete "); } void emergency_remount(void) { |
a2a9537ac
|
819 820 821 822 823 824 825 |
struct work_struct *work; work = kmalloc(sizeof(*work), GFP_ATOMIC); if (work) { INIT_WORK(work, do_emergency_remount); schedule_work(work); } |
1da177e4c
|
826 827 828 829 830 831 |
} /* * Unnamed block devices are dummy devices used by virtual * filesystems which don't use real block-devices. -- jrs */ |
ad76cbc63
|
832 |
static DEFINE_IDA(unnamed_dev_ida); |
1da177e4c
|
833 |
static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */ |
c63e09ecc
|
834 |
static int unnamed_dev_start = 0; /* don't bother trying below it */ |
1da177e4c
|
835 |
|
0ee5dc676
|
836 |
int get_anon_bdev(dev_t *p) |
1da177e4c
|
837 838 839 840 841 |
{ int dev; int error; retry: |
ad76cbc63
|
842 |
if (ida_pre_get(&unnamed_dev_ida, GFP_ATOMIC) == 0) |
1da177e4c
|
843 844 |
return -ENOMEM; spin_lock(&unnamed_dev_lock); |
c63e09ecc
|
845 |
error = ida_get_new_above(&unnamed_dev_ida, unnamed_dev_start, &dev); |
f21f62208
|
846 847 |
if (!error) unnamed_dev_start = dev + 1; |
1da177e4c
|
848 849 850 851 852 853 854 855 856 |
spin_unlock(&unnamed_dev_lock); if (error == -EAGAIN) /* We raced and lost with another CPU. */ goto retry; else if (error) return -EAGAIN; if ((dev & MAX_ID_MASK) == (1 << MINORBITS)) { spin_lock(&unnamed_dev_lock); |
ad76cbc63
|
857 |
ida_remove(&unnamed_dev_ida, dev); |
f21f62208
|
858 859 |
if (unnamed_dev_start > dev) unnamed_dev_start = dev; |
1da177e4c
|
860 861 862 |
spin_unlock(&unnamed_dev_lock); return -EMFILE; } |
0ee5dc676
|
863 |
*p = MKDEV(0, dev & MINORMASK); |
1da177e4c
|
864 865 |
return 0; } |
0ee5dc676
|
866 |
EXPORT_SYMBOL(get_anon_bdev); |
1da177e4c
|
867 |
|
0ee5dc676
|
868 |
void free_anon_bdev(dev_t dev) |
1da177e4c
|
869 |
{ |
0ee5dc676
|
870 |
int slot = MINOR(dev); |
1da177e4c
|
871 |
spin_lock(&unnamed_dev_lock); |
ad76cbc63
|
872 |
ida_remove(&unnamed_dev_ida, slot); |
c63e09ecc
|
873 874 |
if (slot < unnamed_dev_start) unnamed_dev_start = slot; |
1da177e4c
|
875 876 |
spin_unlock(&unnamed_dev_lock); } |
0ee5dc676
|
877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 |
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
|
895 896 |
EXPORT_SYMBOL(kill_anon_super); |
1da177e4c
|
897 898 899 900 901 902 903 904 |
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
|
905 906 907 908 909 910 911 912 913 914 |
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
|
915 916 |
struct dentry *mount_ns(struct file_system_type *fs_type, int flags, void *data, int (*fill_super)(struct super_block *, void *, int)) |
909e6d947
|
917 918 |
{ struct super_block *sb; |
9249e17fe
|
919 |
sb = sget(fs_type, ns_test_super, ns_set_super, flags, data); |
909e6d947
|
920 |
if (IS_ERR(sb)) |
ceefda693
|
921 |
return ERR_CAST(sb); |
909e6d947
|
922 923 924 |
if (!sb->s_root) { int err; |
909e6d947
|
925 926 |
err = fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) { |
74dbbdd7f
|
927 |
deactivate_locked_super(sb); |
ceefda693
|
928 |
return ERR_PTR(err); |
909e6d947
|
929 930 931 932 |
} sb->s_flags |= MS_ACTIVE; } |
ceefda693
|
933 |
return dget(sb->s_root); |
909e6d947
|
934 |
} |
ceefda693
|
935 |
EXPORT_SYMBOL(mount_ns); |
909e6d947
|
936 |
|
9361401eb
|
937 |
#ifdef CONFIG_BLOCK |
1da177e4c
|
938 939 940 941 |
static int set_bdev_super(struct super_block *s, void *data) { s->s_bdev = data; s->s_dev = s->s_bdev->bd_dev; |
32a88aa1b
|
942 943 944 945 946 947 |
/* * 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
|
948 949 950 951 952 953 954 |
return 0; } static int test_bdev_super(struct super_block *s, void *data) { return (void *)s->s_bdev == data; } |
152a08366
|
955 |
struct dentry *mount_bdev(struct file_system_type *fs_type, |
1da177e4c
|
956 |
int flags, const char *dev_name, void *data, |
152a08366
|
957 |
int (*fill_super)(struct super_block *, void *, int)) |
1da177e4c
|
958 959 960 |
{ struct block_device *bdev; struct super_block *s; |
d4d776299
|
961 |
fmode_t mode = FMODE_READ | FMODE_EXCL; |
1da177e4c
|
962 |
int error = 0; |
30c40d2c0
|
963 964 |
if (!(flags & MS_RDONLY)) mode |= FMODE_WRITE; |
d4d776299
|
965 |
bdev = blkdev_get_by_path(dev_name, mode, fs_type); |
1da177e4c
|
966 |
if (IS_ERR(bdev)) |
152a08366
|
967 |
return ERR_CAST(bdev); |
1da177e4c
|
968 969 970 971 972 973 |
/* * 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
|
974 975 976 977 978 979 |
mutex_lock(&bdev->bd_fsfreeze_mutex); if (bdev->bd_fsfreeze_count > 0) { mutex_unlock(&bdev->bd_fsfreeze_mutex); error = -EBUSY; goto error_bdev; } |
9249e17fe
|
980 981 |
s = sget(fs_type, test_bdev_super, set_bdev_super, flags | MS_NOSEC, bdev); |
4fadd7bb2
|
982 |
mutex_unlock(&bdev->bd_fsfreeze_mutex); |
1da177e4c
|
983 |
if (IS_ERR(s)) |
454e2398b
|
984 |
goto error_s; |
1da177e4c
|
985 986 987 |
if (s->s_root) { if ((flags ^ s->s_flags) & MS_RDONLY) { |
74dbbdd7f
|
988 |
deactivate_locked_super(s); |
454e2398b
|
989 990 |
error = -EBUSY; goto error_bdev; |
1da177e4c
|
991 |
} |
454e2398b
|
992 |
|
4f331f01b
|
993 994 |
/* * s_umount nests inside bd_mutex during |
e525fd89d
|
995 996 997 998 |
* __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
|
999 1000 |
*/ up_write(&s->s_umount); |
d4d776299
|
1001 |
blkdev_put(bdev, mode); |
4f331f01b
|
1002 |
down_write(&s->s_umount); |
1da177e4c
|
1003 1004 |
} else { char b[BDEVNAME_SIZE]; |
30c40d2c0
|
1005 |
s->s_mode = mode; |
1da177e4c
|
1006 |
strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); |
e78c9a004
|
1007 |
sb_set_blocksize(s, block_size(bdev)); |
9b04c997b
|
1008 |
error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); |
1da177e4c
|
1009 |
if (error) { |
74dbbdd7f
|
1010 |
deactivate_locked_super(s); |
454e2398b
|
1011 |
goto error; |
fa675765a
|
1012 |
} |
454e2398b
|
1013 1014 |
s->s_flags |= MS_ACTIVE; |
87d8fe1ee
|
1015 |
bdev->bd_super = s; |
1da177e4c
|
1016 |
} |
152a08366
|
1017 |
return dget(s->s_root); |
1da177e4c
|
1018 |
|
454e2398b
|
1019 1020 1021 |
error_s: error = PTR_ERR(s); error_bdev: |
d4d776299
|
1022 |
blkdev_put(bdev, mode); |
454e2398b
|
1023 |
error: |
152a08366
|
1024 1025 1026 |
return ERR_PTR(error); } EXPORT_SYMBOL(mount_bdev); |
1da177e4c
|
1027 1028 1029 |
void kill_block_super(struct super_block *sb) { struct block_device *bdev = sb->s_bdev; |
30c40d2c0
|
1030 |
fmode_t mode = sb->s_mode; |
1da177e4c
|
1031 |
|
ddbaaf302
|
1032 |
bdev->bd_super = NULL; |
1da177e4c
|
1033 1034 |
generic_shutdown_super(sb); sync_blockdev(bdev); |
d4d776299
|
1035 |
WARN_ON_ONCE(!(mode & FMODE_EXCL)); |
e525fd89d
|
1036 |
blkdev_put(bdev, mode | FMODE_EXCL); |
1da177e4c
|
1037 1038 1039 |
} EXPORT_SYMBOL(kill_block_super); |
9361401eb
|
1040 |
#endif |
1da177e4c
|
1041 |
|
3c26ff6e4
|
1042 |
struct dentry *mount_nodev(struct file_system_type *fs_type, |
1da177e4c
|
1043 |
int flags, void *data, |
3c26ff6e4
|
1044 |
int (*fill_super)(struct super_block *, void *, int)) |
1da177e4c
|
1045 1046 |
{ int error; |
9249e17fe
|
1047 |
struct super_block *s = sget(fs_type, NULL, set_anon_super, flags, NULL); |
1da177e4c
|
1048 1049 |
if (IS_ERR(s)) |
3c26ff6e4
|
1050 |
return ERR_CAST(s); |
1da177e4c
|
1051 |
|
9b04c997b
|
1052 |
error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); |
1da177e4c
|
1053 |
if (error) { |
74dbbdd7f
|
1054 |
deactivate_locked_super(s); |
3c26ff6e4
|
1055 |
return ERR_PTR(error); |
1da177e4c
|
1056 1057 |
} s->s_flags |= MS_ACTIVE; |
3c26ff6e4
|
1058 |
return dget(s->s_root); |
1da177e4c
|
1059 |
} |
3c26ff6e4
|
1060 |
EXPORT_SYMBOL(mount_nodev); |
1da177e4c
|
1061 1062 1063 1064 |
static int compare_single(struct super_block *s, void *p) { return 1; } |
fc14f2fef
|
1065 |
struct dentry *mount_single(struct file_system_type *fs_type, |
1da177e4c
|
1066 |
int flags, void *data, |
fc14f2fef
|
1067 |
int (*fill_super)(struct super_block *, void *, int)) |
1da177e4c
|
1068 1069 1070 |
{ struct super_block *s; int error; |
9249e17fe
|
1071 |
s = sget(fs_type, compare_single, set_anon_super, flags, NULL); |
1da177e4c
|
1072 |
if (IS_ERR(s)) |
fc14f2fef
|
1073 |
return ERR_CAST(s); |
1da177e4c
|
1074 |
if (!s->s_root) { |
9b04c997b
|
1075 |
error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); |
1da177e4c
|
1076 |
if (error) { |
74dbbdd7f
|
1077 |
deactivate_locked_super(s); |
fc14f2fef
|
1078 |
return ERR_PTR(error); |
1da177e4c
|
1079 1080 |
} s->s_flags |= MS_ACTIVE; |
9329d1bea
|
1081 1082 |
} else { do_remount_sb(s, flags, data, 0); |
1da177e4c
|
1083 |
} |
fc14f2fef
|
1084 1085 1086 |
return dget(s->s_root); } EXPORT_SYMBOL(mount_single); |
9d412a43c
|
1087 1088 |
struct dentry * mount_fs(struct file_system_type *type, int flags, const char *name, void *data) |
1da177e4c
|
1089 |
{ |
c96e41e92
|
1090 |
struct dentry *root; |
9d412a43c
|
1091 |
struct super_block *sb; |
1da177e4c
|
1092 |
char *secdata = NULL; |
9d412a43c
|
1093 |
int error = -ENOMEM; |
8089352a1
|
1094 |
|
e00075298
|
1095 |
if (data && !(type->fs_flags & FS_BINARY_MOUNTDATA)) { |
1da177e4c
|
1096 |
secdata = alloc_secdata(); |
454e2398b
|
1097 |
if (!secdata) |
9d412a43c
|
1098 |
goto out; |
1da177e4c
|
1099 |
|
e00075298
|
1100 |
error = security_sb_copy_data(data, secdata); |
454e2398b
|
1101 |
if (error) |
1da177e4c
|
1102 |
goto out_free_secdata; |
1da177e4c
|
1103 |
} |
1a102ff92
|
1104 1105 1106 1107 |
root = type->mount(type, flags, name, data); if (IS_ERR(root)) { error = PTR_ERR(root); goto out_free_secdata; |
c96e41e92
|
1108 |
} |
9d412a43c
|
1109 1110 1111 |
sb = root->d_sb; BUG_ON(!sb); WARN_ON(!sb->s_bdi); |
6c5103890
|
1112 |
WARN_ON(sb->s_bdi == &default_backing_dev_info); |
9d412a43c
|
1113 |
sb->s_flags |= MS_BORN; |
454e2398b
|
1114 |
|
9d412a43c
|
1115 |
error = security_sb_kern_mount(sb, flags, secdata); |
5129a469a
|
1116 1117 |
if (error) goto out_sb; |
454e2398b
|
1118 |
|
42cb56ae2
|
1119 1120 1121 1122 |
/* * 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
|
1123 |
* violate this rule. |
42cb56ae2
|
1124 |
*/ |
9d412a43c
|
1125 1126 1127 |
WARN((sb->s_maxbytes < 0), "%s set sb->s_maxbytes to " "negative value (%lld) ", type->name, sb->s_maxbytes); |
42cb56ae2
|
1128 |
|
9d412a43c
|
1129 |
up_write(&sb->s_umount); |
8680e22f2
|
1130 |
free_secdata(secdata); |
9d412a43c
|
1131 |
return root; |
1da177e4c
|
1132 |
out_sb: |
9d412a43c
|
1133 1134 |
dput(root); deactivate_locked_super(sb); |
1da177e4c
|
1135 1136 |
out_free_secdata: free_secdata(secdata); |
1da177e4c
|
1137 |
out: |
454e2398b
|
1138 |
return ERR_PTR(error); |
1da177e4c
|
1139 |
} |
5accdf82b
|
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 1215 1216 1217 1218 1219 1220 1221 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 1247 1248 1249 1250 1251 1252 |
/* * 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
|
1253 |
/** |
7000d3c42
|
1254 1255 |
* freeze_super - lock the filesystem and force it into a consistent state * @sb: the super to lock |
18e9e5104
|
1256 1257 1258 1259 |
* * 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
|
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 |
* * 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
|
1285 1286 1287 1288 1289 1290 1291 |
*/ int freeze_super(struct super_block *sb) { int ret; atomic_inc(&sb->s_active); down_write(&sb->s_umount); |
5accdf82b
|
1292 |
if (sb->s_writers.frozen != SB_UNFROZEN) { |
18e9e5104
|
1293 1294 1295 |
deactivate_locked_super(sb); return -EBUSY; } |
dabe0dc19
|
1296 1297 1298 1299 |
if (!(sb->s_flags & MS_BORN)) { up_write(&sb->s_umount); return 0; /* sic - it's "nothing to do" */ } |
18e9e5104
|
1300 |
if (sb->s_flags & MS_RDONLY) { |
5accdf82b
|
1301 1302 |
/* Nothing to do really... */ sb->s_writers.frozen = SB_FREEZE_COMPLETE; |
18e9e5104
|
1303 1304 1305 |
up_write(&sb->s_umount); return 0; } |
5accdf82b
|
1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 |
/* 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
|
1318 |
smp_wmb(); |
5accdf82b
|
1319 1320 1321 |
sb_wait_write(sb, SB_FREEZE_PAGEFAULT); /* All writers are done so after syncing there won't be dirty data */ |
18e9e5104
|
1322 |
sync_filesystem(sb); |
5accdf82b
|
1323 1324 |
/* Now wait for internal filesystem counter */ sb->s_writers.frozen = SB_FREEZE_FS; |
18e9e5104
|
1325 |
smp_wmb(); |
5accdf82b
|
1326 |
sb_wait_write(sb, SB_FREEZE_FS); |
18e9e5104
|
1327 |
|
18e9e5104
|
1328 1329 1330 1331 1332 1333 |
if (sb->s_op->freeze_fs) { ret = sb->s_op->freeze_fs(sb); if (ret) { printk(KERN_ERR "VFS:Filesystem freeze failed "); |
5accdf82b
|
1334 |
sb->s_writers.frozen = SB_UNFROZEN; |
e1616300a
|
1335 |
smp_wmb(); |
5accdf82b
|
1336 |
wake_up(&sb->s_writers.wait_unfrozen); |
18e9e5104
|
1337 1338 1339 1340 |
deactivate_locked_super(sb); return ret; } } |
5accdf82b
|
1341 1342 1343 1344 1345 |
/* * 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
|
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 |
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
|
1362 |
if (sb->s_writers.frozen == SB_UNFROZEN) { |
18e9e5104
|
1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 |
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
|
1376 1377 1378 1379 1380 1381 |
up_write(&sb->s_umount); return error; } } out: |
5accdf82b
|
1382 |
sb->s_writers.frozen = SB_UNFROZEN; |
18e9e5104
|
1383 |
smp_wmb(); |
5accdf82b
|
1384 |
wake_up(&sb->s_writers.wait_unfrozen); |
18e9e5104
|
1385 1386 1387 1388 1389 |
deactivate_locked_super(sb); return 0; } EXPORT_SYMBOL(thaw_super); |