Blame view
fs/namespace.c
65.4 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 |
/* * linux/fs/namespace.c * * (C) Copyright Al Viro 2000, 2001 * Released under GPL v2. * * Based on code from fs/super.c, copyright Linus Torvalds and others. * Heavily rewritten. */ |
1da177e4c
|
10 11 12 |
#include <linux/syscalls.h> #include <linux/slab.h> #include <linux/sched.h> |
99b7db7b8
|
13 14 |
#include <linux/spinlock.h> #include <linux/percpu.h> |
1da177e4c
|
15 |
#include <linux/init.h> |
15a67dd8c
|
16 |
#include <linux/kernel.h> |
1da177e4c
|
17 |
#include <linux/acct.h> |
16f7e0fe2
|
18 |
#include <linux/capability.h> |
3d733633a
|
19 |
#include <linux/cpumask.h> |
1da177e4c
|
20 |
#include <linux/module.h> |
f20a9ead0
|
21 |
#include <linux/sysfs.h> |
1da177e4c
|
22 |
#include <linux/seq_file.h> |
6b3286ed1
|
23 |
#include <linux/mnt_namespace.h> |
1da177e4c
|
24 |
#include <linux/namei.h> |
b43f3cbd2
|
25 |
#include <linux/nsproxy.h> |
1da177e4c
|
26 27 |
#include <linux/security.h> #include <linux/mount.h> |
07f3f05c1
|
28 |
#include <linux/ramfs.h> |
13f14b4d8
|
29 |
#include <linux/log2.h> |
73cd49ecd
|
30 |
#include <linux/idr.h> |
5ad4e53bd
|
31 |
#include <linux/fs_struct.h> |
2504c5d63
|
32 |
#include <linux/fsnotify.h> |
1da177e4c
|
33 34 |
#include <asm/uaccess.h> #include <asm/unistd.h> |
07b20889e
|
35 |
#include "pnode.h" |
948730b0e
|
36 |
#include "internal.h" |
1da177e4c
|
37 |
|
13f14b4d8
|
38 39 |
#define HASH_SHIFT ilog2(PAGE_SIZE / sizeof(struct list_head)) #define HASH_SIZE (1UL << HASH_SHIFT) |
5addc5dd8
|
40 |
static int event; |
73cd49ecd
|
41 |
static DEFINE_IDA(mnt_id_ida); |
719f5d7f0
|
42 |
static DEFINE_IDA(mnt_group_ida); |
99b7db7b8
|
43 |
static DEFINE_SPINLOCK(mnt_id_lock); |
f21f62208
|
44 45 |
static int mnt_id_start = 0; static int mnt_group_start = 1; |
1da177e4c
|
46 |
|
fa3536cc1
|
47 |
static struct list_head *mount_hashtable __read_mostly; |
e18b890bb
|
48 |
static struct kmem_cache *mnt_cache __read_mostly; |
390c68436
|
49 |
static struct rw_semaphore namespace_sem; |
1da177e4c
|
50 |
|
f87fd4c2a
|
51 |
/* /sys/fs */ |
00d266662
|
52 53 |
struct kobject *fs_kobj; EXPORT_SYMBOL_GPL(fs_kobj); |
f87fd4c2a
|
54 |
|
99b7db7b8
|
55 56 57 58 59 60 61 62 63 |
/* * vfsmount lock may be taken for read to prevent changes to the * vfsmount hash, ie. during mountpoint lookups or walking back * up the tree. * * It should be taken for write in all cases where the vfsmount * tree or hash is modified or when a vfsmount structure is modified. */ DEFINE_BRLOCK(vfsmount_lock); |
1da177e4c
|
64 65 |
static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry) { |
b58fed8b1
|
66 67 |
unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES); tmp += ((unsigned long)dentry / L1_CACHE_BYTES); |
13f14b4d8
|
68 69 |
tmp = tmp + (tmp >> HASH_SHIFT); return tmp & (HASH_SIZE - 1); |
1da177e4c
|
70 |
} |
3d733633a
|
71 |
#define MNT_WRITER_UNDERFLOW_LIMIT -(1<<16) |
99b7db7b8
|
72 73 74 75 |
/* * allocation is serialized by namespace_sem, but we need the spinlock to * serialize with freeing. */ |
73cd49ecd
|
76 77 78 79 80 81 |
static int mnt_alloc_id(struct vfsmount *mnt) { int res; retry: ida_pre_get(&mnt_id_ida, GFP_KERNEL); |
99b7db7b8
|
82 |
spin_lock(&mnt_id_lock); |
f21f62208
|
83 84 85 |
res = ida_get_new_above(&mnt_id_ida, mnt_id_start, &mnt->mnt_id); if (!res) mnt_id_start = mnt->mnt_id + 1; |
99b7db7b8
|
86 |
spin_unlock(&mnt_id_lock); |
73cd49ecd
|
87 88 89 90 91 92 93 94 |
if (res == -EAGAIN) goto retry; return res; } static void mnt_free_id(struct vfsmount *mnt) { |
f21f62208
|
95 |
int id = mnt->mnt_id; |
99b7db7b8
|
96 |
spin_lock(&mnt_id_lock); |
f21f62208
|
97 98 99 |
ida_remove(&mnt_id_ida, id); if (mnt_id_start > id) mnt_id_start = id; |
99b7db7b8
|
100 |
spin_unlock(&mnt_id_lock); |
73cd49ecd
|
101 |
} |
719f5d7f0
|
102 103 104 105 106 107 108 |
/* * Allocate a new peer group ID * * mnt_group_ida is protected by namespace_sem */ static int mnt_alloc_group_id(struct vfsmount *mnt) { |
f21f62208
|
109 |
int res; |
719f5d7f0
|
110 111 |
if (!ida_pre_get(&mnt_group_ida, GFP_KERNEL)) return -ENOMEM; |
f21f62208
|
112 113 114 115 116 117 118 |
res = ida_get_new_above(&mnt_group_ida, mnt_group_start, &mnt->mnt_group_id); if (!res) mnt_group_start = mnt->mnt_group_id + 1; return res; |
719f5d7f0
|
119 120 121 122 123 124 125 |
} /* * Release a peer group ID */ void mnt_release_group_id(struct vfsmount *mnt) { |
f21f62208
|
126 127 128 129 |
int id = mnt->mnt_group_id; ida_remove(&mnt_group_ida, id); if (mnt_group_start > id) mnt_group_start = id; |
719f5d7f0
|
130 131 |
mnt->mnt_group_id = 0; } |
b3e19d924
|
132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 |
/* * vfsmount lock must be held for read */ static inline void mnt_add_count(struct vfsmount *mnt, int n) { #ifdef CONFIG_SMP this_cpu_add(mnt->mnt_pcp->mnt_count, n); #else preempt_disable(); mnt->mnt_count += n; preempt_enable(); #endif } static inline void mnt_set_count(struct vfsmount *mnt, int n) { #ifdef CONFIG_SMP this_cpu_write(mnt->mnt_pcp->mnt_count, n); #else mnt->mnt_count = n; #endif } /* * vfsmount lock must be held for read */ static inline void mnt_inc_count(struct vfsmount *mnt) { mnt_add_count(mnt, 1); } /* * vfsmount lock must be held for read */ static inline void mnt_dec_count(struct vfsmount *mnt) { mnt_add_count(mnt, -1); } /* * vfsmount lock must be held for write */ unsigned int mnt_get_count(struct vfsmount *mnt) { #ifdef CONFIG_SMP |
f03c65993
|
177 |
unsigned int count = 0; |
b3e19d924
|
178 179 180 181 182 183 184 185 186 187 188 |
int cpu; for_each_possible_cpu(cpu) { count += per_cpu_ptr(mnt->mnt_pcp, cpu)->mnt_count; } return count; #else return mnt->mnt_count; #endif } |
9d412a43c
|
189 |
static struct vfsmount *alloc_vfsmnt(const char *name) |
1da177e4c
|
190 |
{ |
c37622296
|
191 |
struct vfsmount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL); |
1da177e4c
|
192 |
if (mnt) { |
73cd49ecd
|
193 194 195 |
int err; err = mnt_alloc_id(mnt); |
88b387824
|
196 197 198 199 200 201 202 |
if (err) goto out_free_cache; if (name) { mnt->mnt_devname = kstrdup(name, GFP_KERNEL); if (!mnt->mnt_devname) goto out_free_id; |
73cd49ecd
|
203 |
} |
b3e19d924
|
204 205 206 207 |
#ifdef CONFIG_SMP mnt->mnt_pcp = alloc_percpu(struct mnt_pcp); if (!mnt->mnt_pcp) goto out_free_devname; |
f03c65993
|
208 |
this_cpu_add(mnt->mnt_pcp->mnt_count, 1); |
b3e19d924
|
209 210 211 212 |
#else mnt->mnt_count = 1; mnt->mnt_writers = 0; #endif |
1da177e4c
|
213 214 215 216 |
INIT_LIST_HEAD(&mnt->mnt_hash); INIT_LIST_HEAD(&mnt->mnt_child); INIT_LIST_HEAD(&mnt->mnt_mounts); INIT_LIST_HEAD(&mnt->mnt_list); |
55e700b92
|
217 |
INIT_LIST_HEAD(&mnt->mnt_expire); |
03e06e68f
|
218 |
INIT_LIST_HEAD(&mnt->mnt_share); |
a58b0eb8e
|
219 220 |
INIT_LIST_HEAD(&mnt->mnt_slave_list); INIT_LIST_HEAD(&mnt->mnt_slave); |
2504c5d63
|
221 222 223 |
#ifdef CONFIG_FSNOTIFY INIT_HLIST_HEAD(&mnt->mnt_fsnotify_marks); #endif |
1da177e4c
|
224 225 |
} return mnt; |
88b387824
|
226 |
|
d3ef3d735
|
227 228 229 230 |
#ifdef CONFIG_SMP out_free_devname: kfree(mnt->mnt_devname); #endif |
88b387824
|
231 232 233 234 235 |
out_free_id: mnt_free_id(mnt); out_free_cache: kmem_cache_free(mnt_cache, mnt); return NULL; |
1da177e4c
|
236 |
} |
8366025eb
|
237 238 239 240 241 242 243 244 |
/* * Most r/o checks on a fs are for operations that take * discrete amounts of time, like a write() or unlink(). * We must keep track of when those operations start * (for permission checks) and when they end, so that * we can determine when writes are able to occur to * a filesystem. */ |
3d733633a
|
245 246 247 248 249 250 251 252 253 254 255 256 257 |
/* * __mnt_is_readonly: check whether a mount is read-only * @mnt: the mount to check for its write status * * This shouldn't be used directly ouside of the VFS. * It does not guarantee that the filesystem will stay * r/w, just that it is right *now*. This can not and * should not be used in place of IS_RDONLY(inode). * mnt_want/drop_write() will _keep_ the filesystem * r/w. */ int __mnt_is_readonly(struct vfsmount *mnt) { |
2e4b7fcd9
|
258 259 260 261 262 |
if (mnt->mnt_flags & MNT_READONLY) return 1; if (mnt->mnt_sb->s_flags & MS_RDONLY) return 1; return 0; |
3d733633a
|
263 264 |
} EXPORT_SYMBOL_GPL(__mnt_is_readonly); |
c6653a838
|
265 |
static inline void mnt_inc_writers(struct vfsmount *mnt) |
d3ef3d735
|
266 267 |
{ #ifdef CONFIG_SMP |
b3e19d924
|
268 |
this_cpu_inc(mnt->mnt_pcp->mnt_writers); |
d3ef3d735
|
269 270 271 272 |
#else mnt->mnt_writers++; #endif } |
3d733633a
|
273 |
|
c6653a838
|
274 |
static inline void mnt_dec_writers(struct vfsmount *mnt) |
3d733633a
|
275 |
{ |
d3ef3d735
|
276 |
#ifdef CONFIG_SMP |
b3e19d924
|
277 |
this_cpu_dec(mnt->mnt_pcp->mnt_writers); |
d3ef3d735
|
278 279 280 |
#else mnt->mnt_writers--; #endif |
3d733633a
|
281 |
} |
3d733633a
|
282 |
|
c6653a838
|
283 |
static unsigned int mnt_get_writers(struct vfsmount *mnt) |
3d733633a
|
284 |
{ |
d3ef3d735
|
285 286 |
#ifdef CONFIG_SMP unsigned int count = 0; |
3d733633a
|
287 |
int cpu; |
3d733633a
|
288 289 |
for_each_possible_cpu(cpu) { |
b3e19d924
|
290 |
count += per_cpu_ptr(mnt->mnt_pcp, cpu)->mnt_writers; |
3d733633a
|
291 |
} |
3d733633a
|
292 |
|
d3ef3d735
|
293 294 295 296 |
return count; #else return mnt->mnt_writers; #endif |
3d733633a
|
297 298 299 300 301 302 303 304 305 306 |
} /* * Most r/o checks on a fs are for operations that take * discrete amounts of time, like a write() or unlink(). * We must keep track of when those operations start * (for permission checks) and when they end, so that * we can determine when writes are able to occur to * a filesystem. */ |
8366025eb
|
307 308 309 310 311 312 313 314 315 316 317 318 |
/** * mnt_want_write - get write access to a mount * @mnt: the mount on which to take a write * * This tells the low-level filesystem that a write is * about to be performed to it, and makes sure that * writes are allowed before returning success. When * the write operation is finished, mnt_drop_write() * must be called. This is effectively a refcount. */ int mnt_want_write(struct vfsmount *mnt) { |
3d733633a
|
319 |
int ret = 0; |
3d733633a
|
320 |
|
d3ef3d735
|
321 |
preempt_disable(); |
c6653a838
|
322 |
mnt_inc_writers(mnt); |
d3ef3d735
|
323 |
/* |
c6653a838
|
324 |
* The store to mnt_inc_writers must be visible before we pass |
d3ef3d735
|
325 326 327 328 329 330 331 332 333 334 335 336 |
* MNT_WRITE_HOLD loop below, so that the slowpath can see our * incremented count after it has set MNT_WRITE_HOLD. */ smp_mb(); while (mnt->mnt_flags & MNT_WRITE_HOLD) cpu_relax(); /* * After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will * be set to match its requirements. So we must not load that until * MNT_WRITE_HOLD is cleared. */ smp_rmb(); |
3d733633a
|
337 |
if (__mnt_is_readonly(mnt)) { |
c6653a838
|
338 |
mnt_dec_writers(mnt); |
3d733633a
|
339 340 341 |
ret = -EROFS; goto out; } |
3d733633a
|
342 |
out: |
d3ef3d735
|
343 |
preempt_enable(); |
3d733633a
|
344 |
return ret; |
8366025eb
|
345 346 347 348 |
} EXPORT_SYMBOL_GPL(mnt_want_write); /** |
96029c4e0
|
349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 |
* mnt_clone_write - get write access to a mount * @mnt: the mount on which to take a write * * This is effectively like mnt_want_write, except * it must only be used to take an extra write reference * on a mountpoint that we already know has a write reference * on it. This allows some optimisation. * * After finished, mnt_drop_write must be called as usual to * drop the reference. */ int mnt_clone_write(struct vfsmount *mnt) { /* superblock may be r/o */ if (__mnt_is_readonly(mnt)) return -EROFS; preempt_disable(); |
c6653a838
|
366 |
mnt_inc_writers(mnt); |
96029c4e0
|
367 368 369 370 371 372 373 374 375 376 377 378 379 380 |
preempt_enable(); return 0; } EXPORT_SYMBOL_GPL(mnt_clone_write); /** * mnt_want_write_file - get write access to a file's mount * @file: the file who's mount on which to take a write * * This is like mnt_want_write, but it takes a file and can * do some optimisations if the file is open for write already */ int mnt_want_write_file(struct file *file) { |
2d8dd38a5
|
381 382 |
struct inode *inode = file->f_dentry->d_inode; if (!(file->f_mode & FMODE_WRITE) || special_file(inode->i_mode)) |
96029c4e0
|
383 384 385 386 387 388 389 |
return mnt_want_write(file->f_path.mnt); else return mnt_clone_write(file->f_path.mnt); } EXPORT_SYMBOL_GPL(mnt_want_write_file); /** |
8366025eb
|
390 391 392 393 394 395 396 397 398 |
* mnt_drop_write - give up write access to a mount * @mnt: the mount on which to give up write access * * Tells the low-level filesystem that we are done * performing writes to it. Must be matched with * mnt_want_write() call above. */ void mnt_drop_write(struct vfsmount *mnt) { |
d3ef3d735
|
399 |
preempt_disable(); |
c6653a838
|
400 |
mnt_dec_writers(mnt); |
d3ef3d735
|
401 |
preempt_enable(); |
8366025eb
|
402 403 |
} EXPORT_SYMBOL_GPL(mnt_drop_write); |
2e4b7fcd9
|
404 |
static int mnt_make_readonly(struct vfsmount *mnt) |
8366025eb
|
405 |
{ |
3d733633a
|
406 |
int ret = 0; |
99b7db7b8
|
407 |
br_write_lock(vfsmount_lock); |
d3ef3d735
|
408 |
mnt->mnt_flags |= MNT_WRITE_HOLD; |
3d733633a
|
409 |
/* |
d3ef3d735
|
410 411 |
* After storing MNT_WRITE_HOLD, we'll read the counters. This store * should be visible before we do. |
3d733633a
|
412 |
*/ |
d3ef3d735
|
413 |
smp_mb(); |
3d733633a
|
414 |
/* |
d3ef3d735
|
415 416 417 418 419 420 421 422 423 424 425 426 427 428 |
* With writers on hold, if this value is zero, then there are * definitely no active writers (although held writers may subsequently * increment the count, they'll have to wait, and decrement it after * seeing MNT_READONLY). * * It is OK to have counter incremented on one CPU and decremented on * another: the sum will add up correctly. The danger would be when we * sum up each counter, if we read a counter before it is incremented, * but then read another CPU's count which it has been subsequently * decremented from -- we would see more decrements than we should. * MNT_WRITE_HOLD protects against this scenario, because * mnt_want_write first increments count, then smp_mb, then spins on * MNT_WRITE_HOLD, so it can't be decremented by another CPU while * we're counting up here. |
3d733633a
|
429 |
*/ |
c6653a838
|
430 |
if (mnt_get_writers(mnt) > 0) |
d3ef3d735
|
431 432 |
ret = -EBUSY; else |
2e4b7fcd9
|
433 |
mnt->mnt_flags |= MNT_READONLY; |
d3ef3d735
|
434 435 436 437 438 439 |
/* * MNT_READONLY must become visible before ~MNT_WRITE_HOLD, so writers * that become unheld will see MNT_READONLY. */ smp_wmb(); mnt->mnt_flags &= ~MNT_WRITE_HOLD; |
99b7db7b8
|
440 |
br_write_unlock(vfsmount_lock); |
3d733633a
|
441 |
return ret; |
8366025eb
|
442 |
} |
8366025eb
|
443 |
|
2e4b7fcd9
|
444 445 |
static void __mnt_unmake_readonly(struct vfsmount *mnt) { |
99b7db7b8
|
446 |
br_write_lock(vfsmount_lock); |
2e4b7fcd9
|
447 |
mnt->mnt_flags &= ~MNT_READONLY; |
99b7db7b8
|
448 |
br_write_unlock(vfsmount_lock); |
2e4b7fcd9
|
449 |
} |
9d412a43c
|
450 |
static void free_vfsmnt(struct vfsmount *mnt) |
1da177e4c
|
451 452 |
{ kfree(mnt->mnt_devname); |
73cd49ecd
|
453 |
mnt_free_id(mnt); |
d3ef3d735
|
454 |
#ifdef CONFIG_SMP |
b3e19d924
|
455 |
free_percpu(mnt->mnt_pcp); |
d3ef3d735
|
456 |
#endif |
1da177e4c
|
457 458 459 460 |
kmem_cache_free(mnt_cache, mnt); } /* |
a05964f39
|
461 462 |
* find the first or last mount at @dentry on vfsmount @mnt depending on * @dir. If @dir is set return the first mount else return the last mount. |
99b7db7b8
|
463 |
* vfsmount_lock must be held for read or write. |
1da177e4c
|
464 |
*/ |
a05964f39
|
465 466 |
struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry, int dir) |
1da177e4c
|
467 |
{ |
b58fed8b1
|
468 469 |
struct list_head *head = mount_hashtable + hash(mnt, dentry); struct list_head *tmp = head; |
1da177e4c
|
470 |
struct vfsmount *p, *found = NULL; |
1da177e4c
|
471 |
for (;;) { |
a05964f39
|
472 |
tmp = dir ? tmp->next : tmp->prev; |
1da177e4c
|
473 474 475 476 477 |
p = NULL; if (tmp == head) break; p = list_entry(tmp, struct vfsmount, mnt_hash); if (p->mnt_parent == mnt && p->mnt_mountpoint == dentry) { |
a05964f39
|
478 |
found = p; |
1da177e4c
|
479 480 481 |
break; } } |
1da177e4c
|
482 483 |
return found; } |
a05964f39
|
484 485 486 487 |
/* * lookup_mnt increments the ref count before returning * the vfsmount struct. */ |
1c755af4d
|
488 |
struct vfsmount *lookup_mnt(struct path *path) |
a05964f39
|
489 490 |
{ struct vfsmount *child_mnt; |
99b7db7b8
|
491 492 |
br_read_lock(vfsmount_lock); |
1c755af4d
|
493 |
if ((child_mnt = __lookup_mnt(path->mnt, path->dentry, 1))) |
a05964f39
|
494 |
mntget(child_mnt); |
99b7db7b8
|
495 |
br_read_unlock(vfsmount_lock); |
a05964f39
|
496 497 |
return child_mnt; } |
1da177e4c
|
498 499 |
static inline int check_mnt(struct vfsmount *mnt) { |
6b3286ed1
|
500 |
return mnt->mnt_ns == current->nsproxy->mnt_ns; |
1da177e4c
|
501 |
} |
99b7db7b8
|
502 503 504 |
/* * vfsmount lock must be held for write */ |
6b3286ed1
|
505 |
static void touch_mnt_namespace(struct mnt_namespace *ns) |
5addc5dd8
|
506 507 508 509 510 511 |
{ if (ns) { ns->event = ++event; wake_up_interruptible(&ns->poll); } } |
99b7db7b8
|
512 513 514 |
/* * vfsmount lock must be held for write */ |
6b3286ed1
|
515 |
static void __touch_mnt_namespace(struct mnt_namespace *ns) |
5addc5dd8
|
516 517 518 519 520 521 |
{ if (ns && ns->event != event) { ns->event = event; wake_up_interruptible(&ns->poll); } } |
99b7db7b8
|
522 |
/* |
5f57cbcc0
|
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 |
* Clear dentry's mounted state if it has no remaining mounts. * vfsmount_lock must be held for write. */ static void dentry_reset_mounted(struct vfsmount *mnt, struct dentry *dentry) { unsigned u; for (u = 0; u < HASH_SIZE; u++) { struct vfsmount *p; list_for_each_entry(p, &mount_hashtable[u], mnt_hash) { if (p->mnt_mountpoint == dentry) return; } } spin_lock(&dentry->d_lock); dentry->d_flags &= ~DCACHE_MOUNTED; spin_unlock(&dentry->d_lock); } /* |
99b7db7b8
|
544 545 |
* vfsmount lock must be held for write */ |
1a3906895
|
546 |
static void detach_mnt(struct vfsmount *mnt, struct path *old_path) |
1da177e4c
|
547 |
{ |
1a3906895
|
548 549 |
old_path->dentry = mnt->mnt_mountpoint; old_path->mnt = mnt->mnt_parent; |
1da177e4c
|
550 551 552 553 |
mnt->mnt_parent = mnt; mnt->mnt_mountpoint = mnt->mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); |
5f57cbcc0
|
554 |
dentry_reset_mounted(old_path->mnt, old_path->dentry); |
1da177e4c
|
555 |
} |
99b7db7b8
|
556 557 558 |
/* * vfsmount lock must be held for write */ |
b90fa9ae8
|
559 560 561 562 563 |
void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, struct vfsmount *child_mnt) { child_mnt->mnt_parent = mntget(mnt); child_mnt->mnt_mountpoint = dget(dentry); |
5f57cbcc0
|
564 565 566 |
spin_lock(&dentry->d_lock); dentry->d_flags |= DCACHE_MOUNTED; spin_unlock(&dentry->d_lock); |
b90fa9ae8
|
567 |
} |
99b7db7b8
|
568 569 570 |
/* * vfsmount lock must be held for write */ |
1a3906895
|
571 |
static void attach_mnt(struct vfsmount *mnt, struct path *path) |
1da177e4c
|
572 |
{ |
1a3906895
|
573 |
mnt_set_mountpoint(path->mnt, path->dentry, mnt); |
b90fa9ae8
|
574 |
list_add_tail(&mnt->mnt_hash, mount_hashtable + |
1a3906895
|
575 576 |
hash(path->mnt, path->dentry)); list_add_tail(&mnt->mnt_child, &path->mnt->mnt_mounts); |
b90fa9ae8
|
577 |
} |
7e3d0eb0b
|
578 579 580 581 582 583 584 585 586 587 588 589 590 591 |
static inline void __mnt_make_longterm(struct vfsmount *mnt) { #ifdef CONFIG_SMP atomic_inc(&mnt->mnt_longterm); #endif } /* needs vfsmount lock for write */ static inline void __mnt_make_shortterm(struct vfsmount *mnt) { #ifdef CONFIG_SMP atomic_dec(&mnt->mnt_longterm); #endif } |
b90fa9ae8
|
592 |
/* |
99b7db7b8
|
593 |
* vfsmount lock must be held for write |
b90fa9ae8
|
594 595 596 597 598 599 |
*/ static void commit_tree(struct vfsmount *mnt) { struct vfsmount *parent = mnt->mnt_parent; struct vfsmount *m; LIST_HEAD(head); |
6b3286ed1
|
600 |
struct mnt_namespace *n = parent->mnt_ns; |
b90fa9ae8
|
601 602 603 604 |
BUG_ON(parent == mnt); list_add_tail(&head, &mnt->mnt_list); |
f03c65993
|
605 |
list_for_each_entry(m, &head, mnt_list) { |
6b3286ed1
|
606 |
m->mnt_ns = n; |
7e3d0eb0b
|
607 |
__mnt_make_longterm(m); |
f03c65993
|
608 |
} |
b90fa9ae8
|
609 610 611 612 613 |
list_splice(&head, n->list.prev); list_add_tail(&mnt->mnt_hash, mount_hashtable + hash(parent, mnt->mnt_mountpoint)); list_add_tail(&mnt->mnt_child, &parent->mnt_mounts); |
6b3286ed1
|
614 |
touch_mnt_namespace(n); |
1da177e4c
|
615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 |
} static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) { struct list_head *next = p->mnt_mounts.next; if (next == &p->mnt_mounts) { while (1) { if (p == root) return NULL; next = p->mnt_child.next; if (next != &p->mnt_parent->mnt_mounts) break; p = p->mnt_parent; } } return list_entry(next, struct vfsmount, mnt_child); } |
9676f0c63
|
632 633 634 635 636 637 638 639 640 |
static struct vfsmount *skip_mnt_tree(struct vfsmount *p) { struct list_head *prev = p->mnt_mounts.prev; while (prev != &p->mnt_mounts) { p = list_entry(prev, struct vfsmount, mnt_child); prev = p->mnt_mounts.prev; } return p; } |
9d412a43c
|
641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 |
struct vfsmount * vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void *data) { struct vfsmount *mnt; struct dentry *root; if (!type) return ERR_PTR(-ENODEV); mnt = alloc_vfsmnt(name); if (!mnt) return ERR_PTR(-ENOMEM); if (flags & MS_KERNMOUNT) mnt->mnt_flags = MNT_INTERNAL; root = mount_fs(type, flags, name, data); if (IS_ERR(root)) { free_vfsmnt(mnt); return ERR_CAST(root); } mnt->mnt_root = root; mnt->mnt_sb = root->d_sb; mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; return mnt; } EXPORT_SYMBOL_GPL(vfs_kern_mount); |
36341f645
|
670 671 |
static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, int flag) |
1da177e4c
|
672 673 674 675 676 |
{ struct super_block *sb = old->mnt_sb; struct vfsmount *mnt = alloc_vfsmnt(old->mnt_devname); if (mnt) { |
719f5d7f0
|
677 678 679 680 681 682 683 684 685 686 |
if (flag & (CL_SLAVE | CL_PRIVATE)) mnt->mnt_group_id = 0; /* not a peer of original */ else mnt->mnt_group_id = old->mnt_group_id; if ((flag & CL_MAKE_SHARED) && !mnt->mnt_group_id) { int err = mnt_alloc_group_id(mnt); if (err) goto out_free; } |
be1a16a0a
|
687 |
mnt->mnt_flags = old->mnt_flags & ~MNT_WRITE_HOLD; |
1da177e4c
|
688 689 690 691 692 |
atomic_inc(&sb->s_active); mnt->mnt_sb = sb; mnt->mnt_root = dget(root); mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; |
b90fa9ae8
|
693 |
|
5afe00221
|
694 695 696 697 |
if (flag & CL_SLAVE) { list_add(&mnt->mnt_slave, &old->mnt_slave_list); mnt->mnt_master = old; CLEAR_MNT_SHARED(mnt); |
8aec08094
|
698 |
} else if (!(flag & CL_PRIVATE)) { |
796a6b521
|
699 |
if ((flag & CL_MAKE_SHARED) || IS_MNT_SHARED(old)) |
5afe00221
|
700 701 702 703 704 |
list_add(&mnt->mnt_share, &old->mnt_share); if (IS_MNT_SLAVE(old)) list_add(&mnt->mnt_slave, &old->mnt_slave); mnt->mnt_master = old->mnt_master; } |
b90fa9ae8
|
705 706 |
if (flag & CL_MAKE_SHARED) set_mnt_shared(mnt); |
1da177e4c
|
707 708 709 |
/* stick the duplicate mount on the same expiry list * as the original if that was on one */ |
36341f645
|
710 |
if (flag & CL_EXPIRE) { |
36341f645
|
711 712 |
if (!list_empty(&old->mnt_expire)) list_add(&mnt->mnt_expire, &old->mnt_expire); |
36341f645
|
713 |
} |
1da177e4c
|
714 715 |
} return mnt; |
719f5d7f0
|
716 717 718 719 |
out_free: free_vfsmnt(mnt); return NULL; |
1da177e4c
|
720 |
} |
b3e19d924
|
721 |
static inline void mntfree(struct vfsmount *mnt) |
1da177e4c
|
722 723 |
{ struct super_block *sb = mnt->mnt_sb; |
b3e19d924
|
724 |
|
3d733633a
|
725 |
/* |
3d733633a
|
726 727 728 729 730 |
* This probably indicates that somebody messed * up a mnt_want/drop_write() pair. If this * happens, the filesystem was probably unable * to make r/w->r/o transitions. */ |
d3ef3d735
|
731 |
/* |
b3e19d924
|
732 733 |
* The locking used to deal with mnt_count decrement provides barriers, * so mnt_get_writers() below is safe. |
d3ef3d735
|
734 |
*/ |
c6653a838
|
735 |
WARN_ON(mnt_get_writers(mnt)); |
ca9c726ee
|
736 |
fsnotify_vfsmount_delete(mnt); |
1da177e4c
|
737 738 739 740 |
dput(mnt->mnt_root); free_vfsmnt(mnt); deactivate_super(sb); } |
f03c65993
|
741 |
static void mntput_no_expire(struct vfsmount *mnt) |
b3e19d924
|
742 |
{ |
b3e19d924
|
743 |
put_again: |
f03c65993
|
744 745 746 747 |
#ifdef CONFIG_SMP br_read_lock(vfsmount_lock); if (likely(atomic_read(&mnt->mnt_longterm))) { mnt_dec_count(mnt); |
b3e19d924
|
748 |
br_read_unlock(vfsmount_lock); |
f03c65993
|
749 |
return; |
b3e19d924
|
750 |
} |
f03c65993
|
751 |
br_read_unlock(vfsmount_lock); |
b3e19d924
|
752 |
|
99b7db7b8
|
753 |
br_write_lock(vfsmount_lock); |
f03c65993
|
754 |
mnt_dec_count(mnt); |
b3e19d924
|
755 |
if (mnt_get_count(mnt)) { |
99b7db7b8
|
756 757 758 |
br_write_unlock(vfsmount_lock); return; } |
b3e19d924
|
759 |
#else |
b3e19d924
|
760 761 |
mnt_dec_count(mnt); if (likely(mnt_get_count(mnt))) |
99b7db7b8
|
762 |
return; |
b3e19d924
|
763 |
br_write_lock(vfsmount_lock); |
f03c65993
|
764 |
#endif |
b3e19d924
|
765 766 767 768 769 770 |
if (unlikely(mnt->mnt_pinned)) { mnt_add_count(mnt, mnt->mnt_pinned + 1); mnt->mnt_pinned = 0; br_write_unlock(vfsmount_lock); acct_auto_close_mnt(mnt); goto put_again; |
7b7b1ace2
|
771 |
} |
99b7db7b8
|
772 |
br_write_unlock(vfsmount_lock); |
b3e19d924
|
773 774 |
mntfree(mnt); } |
b3e19d924
|
775 776 777 778 779 780 781 |
void mntput(struct vfsmount *mnt) { if (mnt) { /* avoid cacheline pingpong, hope gcc doesn't get "smart" */ if (unlikely(mnt->mnt_expiry_mark)) mnt->mnt_expiry_mark = 0; |
f03c65993
|
782 |
mntput_no_expire(mnt); |
b3e19d924
|
783 784 785 786 787 788 789 790 791 792 793 |
} } EXPORT_SYMBOL(mntput); struct vfsmount *mntget(struct vfsmount *mnt) { if (mnt) mnt_inc_count(mnt); return mnt; } EXPORT_SYMBOL(mntget); |
7b7b1ace2
|
794 795 |
void mnt_pin(struct vfsmount *mnt) { |
99b7db7b8
|
796 |
br_write_lock(vfsmount_lock); |
7b7b1ace2
|
797 |
mnt->mnt_pinned++; |
99b7db7b8
|
798 |
br_write_unlock(vfsmount_lock); |
7b7b1ace2
|
799 |
} |
7b7b1ace2
|
800 801 802 803 |
EXPORT_SYMBOL(mnt_pin); void mnt_unpin(struct vfsmount *mnt) { |
99b7db7b8
|
804 |
br_write_lock(vfsmount_lock); |
7b7b1ace2
|
805 |
if (mnt->mnt_pinned) { |
b3e19d924
|
806 |
mnt_inc_count(mnt); |
7b7b1ace2
|
807 808 |
mnt->mnt_pinned--; } |
99b7db7b8
|
809 |
br_write_unlock(vfsmount_lock); |
7b7b1ace2
|
810 |
} |
7b7b1ace2
|
811 |
EXPORT_SYMBOL(mnt_unpin); |
1da177e4c
|
812 |
|
b3b304a23
|
813 814 815 816 817 818 819 820 821 822 823 824 825 826 |
static inline void mangle(struct seq_file *m, const char *s) { seq_escape(m, s, " \t \\"); } /* * Simple .show_options callback for filesystems which don't want to * implement more complex mount option showing. * * See also save_mount_options(). */ int generic_show_options(struct seq_file *m, struct vfsmount *mnt) { |
2a32cebd6
|
827 828 829 830 |
const char *options; rcu_read_lock(); options = rcu_dereference(mnt->mnt_sb->s_options); |
b3b304a23
|
831 832 833 834 835 |
if (options != NULL && options[0]) { seq_putc(m, ','); mangle(m, options); } |
2a32cebd6
|
836 |
rcu_read_unlock(); |
b3b304a23
|
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 |
return 0; } EXPORT_SYMBOL(generic_show_options); /* * If filesystem uses generic_show_options(), this function should be * called from the fill_super() callback. * * The .remount_fs callback usually needs to be handled in a special * way, to make sure, that previous options are not overwritten if the * remount fails. * * Also note, that if the filesystem's .remount_fs function doesn't * reset all options to their default value, but changes only newly * given options, then the displayed options will not reflect reality * any more. */ void save_mount_options(struct super_block *sb, char *options) { |
2a32cebd6
|
857 858 |
BUG_ON(sb->s_options); rcu_assign_pointer(sb->s_options, kstrdup(options, GFP_KERNEL)); |
b3b304a23
|
859 860 |
} EXPORT_SYMBOL(save_mount_options); |
2a32cebd6
|
861 862 863 864 865 866 867 868 869 870 |
void replace_mount_options(struct super_block *sb, char *options) { char *old = sb->s_options; rcu_assign_pointer(sb->s_options, options); if (old) { synchronize_rcu(); kfree(old); } } EXPORT_SYMBOL(replace_mount_options); |
a1a2c409b
|
871 |
#ifdef CONFIG_PROC_FS |
1da177e4c
|
872 873 874 |
/* iterator */ static void *m_start(struct seq_file *m, loff_t *pos) { |
a1a2c409b
|
875 |
struct proc_mounts *p = m->private; |
1da177e4c
|
876 |
|
390c68436
|
877 |
down_read(&namespace_sem); |
a1a2c409b
|
878 |
return seq_list_start(&p->ns->list, *pos); |
1da177e4c
|
879 880 881 882 |
} static void *m_next(struct seq_file *m, void *v, loff_t *pos) { |
a1a2c409b
|
883 |
struct proc_mounts *p = m->private; |
b0765fb85
|
884 |
|
a1a2c409b
|
885 |
return seq_list_next(v, &p->ns->list, pos); |
1da177e4c
|
886 887 888 889 |
} static void m_stop(struct seq_file *m, void *v) { |
390c68436
|
890 |
up_read(&namespace_sem); |
1da177e4c
|
891 |
} |
9f5596af4
|
892 893 894 895 |
int mnt_had_events(struct proc_mounts *p) { struct mnt_namespace *ns = p->ns; int res = 0; |
99b7db7b8
|
896 |
br_read_lock(vfsmount_lock); |
f15146380
|
897 898 |
if (p->m.poll_event != ns->event) { p->m.poll_event = ns->event; |
9f5596af4
|
899 900 |
res = 1; } |
99b7db7b8
|
901 |
br_read_unlock(vfsmount_lock); |
9f5596af4
|
902 903 904 |
return res; } |
2d4d4864a
|
905 906 907 908 |
struct proc_fs_info { int flag; const char *str; }; |
2069f4578
|
909 |
static int show_sb_opts(struct seq_file *m, struct super_block *sb) |
1da177e4c
|
910 |
{ |
2d4d4864a
|
911 |
static const struct proc_fs_info fs_info[] = { |
1da177e4c
|
912 913 914 |
{ MS_SYNCHRONOUS, ",sync" }, { MS_DIRSYNC, ",dirsync" }, { MS_MANDLOCK, ",mand" }, |
1da177e4c
|
915 916 |
{ 0, NULL } }; |
2d4d4864a
|
917 918 919 920 921 922 |
const struct proc_fs_info *fs_infop; for (fs_infop = fs_info; fs_infop->flag; fs_infop++) { if (sb->s_flags & fs_infop->flag) seq_puts(m, fs_infop->str); } |
2069f4578
|
923 924 |
return security_sb_show_options(m, sb); |
2d4d4864a
|
925 926 927 928 929 |
} static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt) { static const struct proc_fs_info mnt_info[] = { |
1da177e4c
|
930 931 932 |
{ MNT_NOSUID, ",nosuid" }, { MNT_NODEV, ",nodev" }, { MNT_NOEXEC, ",noexec" }, |
fc33a7bb9
|
933 934 |
{ MNT_NOATIME, ",noatime" }, { MNT_NODIRATIME, ",nodiratime" }, |
47ae32d6a
|
935 |
{ MNT_RELATIME, ",relatime" }, |
1da177e4c
|
936 937 |
{ 0, NULL } }; |
2d4d4864a
|
938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 |
const struct proc_fs_info *fs_infop; for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) { if (mnt->mnt_flags & fs_infop->flag) seq_puts(m, fs_infop->str); } } static void show_type(struct seq_file *m, struct super_block *sb) { mangle(m, sb->s_type->name); if (sb->s_subtype && sb->s_subtype[0]) { seq_putc(m, '.'); mangle(m, sb->s_subtype); } } static int show_vfsmnt(struct seq_file *m, void *v) { struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); int err = 0; |
c32c2f63a
|
959 |
struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; |
1da177e4c
|
960 |
|
c7f404b40
|
961 962 963 964 965 966 967 |
if (mnt->mnt_sb->s_op->show_devname) { err = mnt->mnt_sb->s_op->show_devname(m, mnt); if (err) goto out; } else { mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); } |
1da177e4c
|
968 |
seq_putc(m, ' '); |
c32c2f63a
|
969 970 |
seq_path(m, &mnt_path, " \t \\"); |
1da177e4c
|
971 |
seq_putc(m, ' '); |
2d4d4864a
|
972 |
show_type(m, mnt->mnt_sb); |
2e4b7fcd9
|
973 |
seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw"); |
2069f4578
|
974 975 976 |
err = show_sb_opts(m, mnt->mnt_sb); if (err) goto out; |
2d4d4864a
|
977 |
show_mnt_opts(m, mnt); |
1da177e4c
|
978 979 980 981 |
if (mnt->mnt_sb->s_op->show_options) err = mnt->mnt_sb->s_op->show_options(m, mnt); seq_puts(m, " 0 0 "); |
2069f4578
|
982 |
out: |
1da177e4c
|
983 984 |
return err; } |
a1a2c409b
|
985 |
const struct seq_operations mounts_op = { |
1da177e4c
|
986 987 988 989 990 |
.start = m_start, .next = m_next, .stop = m_stop, .show = show_vfsmnt }; |
2d4d4864a
|
991 992 993 994 995 996 997 998 999 1000 1001 |
static int show_mountinfo(struct seq_file *m, void *v) { struct proc_mounts *p = m->private; struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); struct super_block *sb = mnt->mnt_sb; struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; struct path root = p->root; int err = 0; seq_printf(m, "%i %i %u:%u ", mnt->mnt_id, mnt->mnt_parent->mnt_id, MAJOR(sb->s_dev), MINOR(sb->s_dev)); |
c7f404b40
|
1002 1003 1004 1005 1006 1007 1008 |
if (sb->s_op->show_path) err = sb->s_op->show_path(m, mnt); else seq_dentry(m, mnt->mnt_root, " \t \\"); if (err) goto out; |
2d4d4864a
|
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 |
seq_putc(m, ' '); seq_path_root(m, &mnt_path, &root, " \t \\"); if (root.mnt != p->root.mnt || root.dentry != p->root.dentry) { /* * Mountpoint is outside root, discard that one. Ugly, * but less so than trying to do that in iterator in a * race-free way (due to renames). */ return SEQ_SKIP; } seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw"); show_mnt_opts(m, mnt); /* Tagged fields ("foo:X" or "bar") */ if (IS_MNT_SHARED(mnt)) seq_printf(m, " shared:%i", mnt->mnt_group_id); |
97e7e0f71
|
1026 1027 1028 1029 1030 1031 1032 |
if (IS_MNT_SLAVE(mnt)) { int master = mnt->mnt_master->mnt_group_id; int dom = get_dominating_id(mnt, &p->root); seq_printf(m, " master:%i", master); if (dom && dom != master) seq_printf(m, " propagate_from:%i", dom); } |
2d4d4864a
|
1033 1034 1035 1036 1037 1038 1039 |
if (IS_MNT_UNBINDABLE(mnt)) seq_puts(m, " unbindable"); /* Filesystem specific data */ seq_puts(m, " - "); show_type(m, sb); seq_putc(m, ' '); |
c7f404b40
|
1040 1041 1042 1043 1044 1045 |
if (sb->s_op->show_devname) err = sb->s_op->show_devname(m, mnt); else mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); if (err) goto out; |
2d4d4864a
|
1046 |
seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw"); |
2069f4578
|
1047 1048 1049 |
err = show_sb_opts(m, sb); if (err) goto out; |
2d4d4864a
|
1050 1051 1052 1053 |
if (sb->s_op->show_options) err = sb->s_op->show_options(m, mnt); seq_putc(m, ' '); |
2069f4578
|
1054 |
out: |
2d4d4864a
|
1055 1056 1057 1058 1059 1060 1061 1062 1063 |
return err; } const struct seq_operations mountinfo_op = { .start = m_start, .next = m_next, .stop = m_stop, .show = show_mountinfo, }; |
b4629fe2f
|
1064 1065 |
static int show_vfsstat(struct seq_file *m, void *v) { |
b0765fb85
|
1066 |
struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); |
c32c2f63a
|
1067 |
struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; |
b4629fe2f
|
1068 1069 1070 |
int err = 0; /* device */ |
c7f404b40
|
1071 1072 1073 1074 1075 1076 1077 1078 1079 |
if (mnt->mnt_sb->s_op->show_devname) { err = mnt->mnt_sb->s_op->show_devname(m, mnt); } else { if (mnt->mnt_devname) { seq_puts(m, "device "); mangle(m, mnt->mnt_devname); } else seq_puts(m, "no device"); } |
b4629fe2f
|
1080 1081 1082 |
/* mount point */ seq_puts(m, " mounted on "); |
c32c2f63a
|
1083 1084 |
seq_path(m, &mnt_path, " \t \\"); |
b4629fe2f
|
1085 1086 1087 1088 |
seq_putc(m, ' '); /* file system type */ seq_puts(m, "with fstype "); |
2d4d4864a
|
1089 |
show_type(m, mnt->mnt_sb); |
b4629fe2f
|
1090 1091 1092 1093 |
/* optional statistics */ if (mnt->mnt_sb->s_op->show_stats) { seq_putc(m, ' '); |
c7f404b40
|
1094 1095 |
if (!err) err = mnt->mnt_sb->s_op->show_stats(m, mnt); |
b4629fe2f
|
1096 1097 1098 1099 1100 1101 |
} seq_putc(m, ' '); return err; } |
a1a2c409b
|
1102 |
const struct seq_operations mountstats_op = { |
b4629fe2f
|
1103 1104 1105 1106 1107 |
.start = m_start, .next = m_next, .stop = m_stop, .show = show_vfsstat, }; |
a1a2c409b
|
1108 |
#endif /* CONFIG_PROC_FS */ |
b4629fe2f
|
1109 |
|
1da177e4c
|
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 |
/** * may_umount_tree - check if a mount tree is busy * @mnt: root of mount tree * * This is called to check if a tree of mounts has any * open files, pwds, chroots or sub mounts that are * busy. */ int may_umount_tree(struct vfsmount *mnt) { |
36341f645
|
1120 1121 1122 |
int actual_refs = 0; int minimum_refs = 0; struct vfsmount *p; |
1da177e4c
|
1123 |
|
b3e19d924
|
1124 1125 |
/* write lock needed for mnt_get_count */ br_write_lock(vfsmount_lock); |
36341f645
|
1126 |
for (p = mnt; p; p = next_mnt(p, mnt)) { |
b3e19d924
|
1127 |
actual_refs += mnt_get_count(p); |
1da177e4c
|
1128 |
minimum_refs += 2; |
1da177e4c
|
1129 |
} |
b3e19d924
|
1130 |
br_write_unlock(vfsmount_lock); |
1da177e4c
|
1131 1132 |
if (actual_refs > minimum_refs) |
e3474a8eb
|
1133 |
return 0; |
1da177e4c
|
1134 |
|
e3474a8eb
|
1135 |
return 1; |
1da177e4c
|
1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 |
} EXPORT_SYMBOL(may_umount_tree); /** * may_umount - check if a mount point is busy * @mnt: root of mount * * This is called to check if a mount point has any * open files, pwds, chroots or sub mounts. If the * mount has sub mounts this will return busy * regardless of whether the sub mounts are busy. * * Doesn't take quota and stuff into account. IOW, in some cases it will * give false negatives. The main reason why it's here is that we need * a non-destructive way to look for easily umountable filesystems. */ int may_umount(struct vfsmount *mnt) { |
e3474a8eb
|
1155 |
int ret = 1; |
8ad08d8a0
|
1156 |
down_read(&namespace_sem); |
b3e19d924
|
1157 |
br_write_lock(vfsmount_lock); |
a05964f39
|
1158 |
if (propagate_mount_busy(mnt, 2)) |
e3474a8eb
|
1159 |
ret = 0; |
b3e19d924
|
1160 |
br_write_unlock(vfsmount_lock); |
8ad08d8a0
|
1161 |
up_read(&namespace_sem); |
a05964f39
|
1162 |
return ret; |
1da177e4c
|
1163 1164 1165 |
} EXPORT_SYMBOL(may_umount); |
b90fa9ae8
|
1166 |
void release_mounts(struct list_head *head) |
70fbcdf4d
|
1167 1168 |
{ struct vfsmount *mnt; |
bf066c7db
|
1169 |
while (!list_empty(head)) { |
b5e618181
|
1170 |
mnt = list_first_entry(head, struct vfsmount, mnt_hash); |
70fbcdf4d
|
1171 1172 1173 1174 |
list_del_init(&mnt->mnt_hash); if (mnt->mnt_parent != mnt) { struct dentry *dentry; struct vfsmount *m; |
99b7db7b8
|
1175 1176 |
br_write_lock(vfsmount_lock); |
70fbcdf4d
|
1177 1178 1179 1180 |
dentry = mnt->mnt_mountpoint; m = mnt->mnt_parent; mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; |
7c4b93d82
|
1181 |
m->mnt_ghosts--; |
99b7db7b8
|
1182 |
br_write_unlock(vfsmount_lock); |
70fbcdf4d
|
1183 1184 1185 |
dput(dentry); mntput(m); } |
f03c65993
|
1186 |
mntput(mnt); |
70fbcdf4d
|
1187 1188 |
} } |
99b7db7b8
|
1189 1190 1191 1192 |
/* * vfsmount lock must be held for write * namespace_sem must be held for write */ |
a05964f39
|
1193 |
void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) |
1da177e4c
|
1194 |
{ |
7b8a53fd8
|
1195 |
LIST_HEAD(tmp_list); |
1da177e4c
|
1196 |
struct vfsmount *p; |
1da177e4c
|
1197 |
|
1bfba4e8e
|
1198 |
for (p = mnt; p; p = next_mnt(p, mnt)) |
7b8a53fd8
|
1199 |
list_move(&p->mnt_hash, &tmp_list); |
1da177e4c
|
1200 |
|
a05964f39
|
1201 |
if (propagate) |
7b8a53fd8
|
1202 |
propagate_umount(&tmp_list); |
a05964f39
|
1203 |
|
7b8a53fd8
|
1204 |
list_for_each_entry(p, &tmp_list, mnt_hash) { |
70fbcdf4d
|
1205 1206 |
list_del_init(&p->mnt_expire); list_del_init(&p->mnt_list); |
6b3286ed1
|
1207 1208 |
__touch_mnt_namespace(p->mnt_ns); p->mnt_ns = NULL; |
7e3d0eb0b
|
1209 |
__mnt_make_shortterm(p); |
70fbcdf4d
|
1210 |
list_del_init(&p->mnt_child); |
7c4b93d82
|
1211 1212 |
if (p->mnt_parent != p) { p->mnt_parent->mnt_ghosts++; |
5f57cbcc0
|
1213 |
dentry_reset_mounted(p->mnt_parent, p->mnt_mountpoint); |
7c4b93d82
|
1214 |
} |
a05964f39
|
1215 |
change_mnt_propagation(p, MS_PRIVATE); |
1da177e4c
|
1216 |
} |
7b8a53fd8
|
1217 |
list_splice(&tmp_list, kill); |
1da177e4c
|
1218 |
} |
c35038bec
|
1219 |
static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts); |
1da177e4c
|
1220 1221 |
static int do_umount(struct vfsmount *mnt, int flags) { |
b58fed8b1
|
1222 |
struct super_block *sb = mnt->mnt_sb; |
1da177e4c
|
1223 |
int retval; |
70fbcdf4d
|
1224 |
LIST_HEAD(umount_list); |
1da177e4c
|
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 |
retval = security_sb_umount(mnt, flags); if (retval) return retval; /* * Allow userspace to request a mountpoint be expired rather than * unmounting unconditionally. Unmount only happens if: * (1) the mark is already set (the mark is cleared by mntput()) * (2) the usage count == 1 [parent vfsmount] + 1 [sys_umount] */ if (flags & MNT_EXPIRE) { |
6ac08c39a
|
1237 |
if (mnt == current->fs->root.mnt || |
1da177e4c
|
1238 1239 |
flags & (MNT_FORCE | MNT_DETACH)) return -EINVAL; |
b3e19d924
|
1240 1241 1242 1243 1244 1245 |
/* * probably don't strictly need the lock here if we examined * all race cases, but it's a slowpath. */ br_write_lock(vfsmount_lock); if (mnt_get_count(mnt) != 2) { |
bf9faa2aa
|
1246 |
br_write_unlock(vfsmount_lock); |
1da177e4c
|
1247 |
return -EBUSY; |
b3e19d924
|
1248 1249 |
} br_write_unlock(vfsmount_lock); |
1da177e4c
|
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 |
if (!xchg(&mnt->mnt_expiry_mark, 1)) return -EAGAIN; } /* * If we may have to abort operations to get out of this * mount, and they will themselves hold resources we must * allow the fs to do things. In the Unix tradition of * 'Gee thats tricky lets do it in userspace' the umount_begin * might fail to complete on the first run through as other tasks * must return, and the like. Thats for the mount program to worry * about for the moment. */ |
42faad996
|
1264 |
if (flags & MNT_FORCE && sb->s_op->umount_begin) { |
42faad996
|
1265 |
sb->s_op->umount_begin(sb); |
42faad996
|
1266 |
} |
1da177e4c
|
1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 |
/* * No sense to grab the lock for this test, but test itself looks * somewhat bogus. Suggestions for better replacement? * Ho-hum... In principle, we might treat that as umount + switch * to rootfs. GC would eventually take care of the old vfsmount. * Actually it makes sense, especially if rootfs would contain a * /reboot - static binary that would close all descriptors and * call reboot(9). Then init(8) could umount root and exec /reboot. */ |
6ac08c39a
|
1277 |
if (mnt == current->fs->root.mnt && !(flags & MNT_DETACH)) { |
1da177e4c
|
1278 1279 1280 1281 1282 |
/* * Special case for "unmounting" root ... * we just try to remount it readonly. */ down_write(&sb->s_umount); |
4aa98cf76
|
1283 |
if (!(sb->s_flags & MS_RDONLY)) |
1da177e4c
|
1284 |
retval = do_remount_sb(sb, MS_RDONLY, NULL, 0); |
1da177e4c
|
1285 1286 1287 |
up_write(&sb->s_umount); return retval; } |
390c68436
|
1288 |
down_write(&namespace_sem); |
99b7db7b8
|
1289 |
br_write_lock(vfsmount_lock); |
5addc5dd8
|
1290 |
event++; |
1da177e4c
|
1291 |
|
c35038bec
|
1292 1293 |
if (!(flags & MNT_DETACH)) shrink_submounts(mnt, &umount_list); |
1da177e4c
|
1294 |
retval = -EBUSY; |
a05964f39
|
1295 |
if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) { |
1da177e4c
|
1296 |
if (!list_empty(&mnt->mnt_list)) |
a05964f39
|
1297 |
umount_tree(mnt, 1, &umount_list); |
1da177e4c
|
1298 1299 |
retval = 0; } |
99b7db7b8
|
1300 |
br_write_unlock(vfsmount_lock); |
390c68436
|
1301 |
up_write(&namespace_sem); |
70fbcdf4d
|
1302 |
release_mounts(&umount_list); |
1da177e4c
|
1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 |
return retval; } /* * Now umount can handle mount points as well as block devices. * This is important for filesystems which use unnamed block devices. * * We now support a flag for forced unmount like the other 'big iron' * unixes. Our API is identical to OSF/1 to avoid making a mess of AMD */ |
bdc480e3b
|
1313 |
SYSCALL_DEFINE2(umount, char __user *, name, int, flags) |
1da177e4c
|
1314 |
{ |
2d8f30380
|
1315 |
struct path path; |
1da177e4c
|
1316 |
int retval; |
db1f05bb8
|
1317 |
int lookup_flags = 0; |
1da177e4c
|
1318 |
|
db1f05bb8
|
1319 1320 1321 1322 1323 1324 1325 |
if (flags & ~(MNT_FORCE | MNT_DETACH | MNT_EXPIRE | UMOUNT_NOFOLLOW)) return -EINVAL; if (!(flags & UMOUNT_NOFOLLOW)) lookup_flags |= LOOKUP_FOLLOW; retval = user_path_at(AT_FDCWD, name, lookup_flags, &path); |
1da177e4c
|
1326 1327 1328 |
if (retval) goto out; retval = -EINVAL; |
2d8f30380
|
1329 |
if (path.dentry != path.mnt->mnt_root) |
1da177e4c
|
1330 |
goto dput_and_out; |
2d8f30380
|
1331 |
if (!check_mnt(path.mnt)) |
1da177e4c
|
1332 1333 1334 1335 1336 |
goto dput_and_out; retval = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto dput_and_out; |
2d8f30380
|
1337 |
retval = do_umount(path.mnt, flags); |
1da177e4c
|
1338 |
dput_and_out: |
429731b15
|
1339 |
/* we mustn't call path_put() as that would clear mnt_expiry_mark */ |
2d8f30380
|
1340 1341 |
dput(path.dentry); mntput_no_expire(path.mnt); |
1da177e4c
|
1342 1343 1344 1345 1346 1347 1348 |
out: return retval; } #ifdef __ARCH_WANT_SYS_OLDUMOUNT /* |
b58fed8b1
|
1349 |
* The 2.0 compatible umount. No flags. |
1da177e4c
|
1350 |
*/ |
bdc480e3b
|
1351 |
SYSCALL_DEFINE1(oldumount, char __user *, name) |
1da177e4c
|
1352 |
{ |
b58fed8b1
|
1353 |
return sys_umount(name, 0); |
1da177e4c
|
1354 1355 1356 |
} #endif |
2d92ab3c6
|
1357 |
static int mount_is_safe(struct path *path) |
1da177e4c
|
1358 1359 1360 1361 1362 |
{ if (capable(CAP_SYS_ADMIN)) return 0; return -EPERM; #ifdef notyet |
2d92ab3c6
|
1363 |
if (S_ISLNK(path->dentry->d_inode->i_mode)) |
1da177e4c
|
1364 |
return -EPERM; |
2d92ab3c6
|
1365 |
if (path->dentry->d_inode->i_mode & S_ISVTX) { |
da9592ede
|
1366 |
if (current_uid() != path->dentry->d_inode->i_uid) |
1da177e4c
|
1367 1368 |
return -EPERM; } |
2d92ab3c6
|
1369 |
if (inode_permission(path->dentry->d_inode, MAY_WRITE)) |
1da177e4c
|
1370 1371 1372 1373 |
return -EPERM; return 0; #endif } |
b90fa9ae8
|
1374 |
struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, |
36341f645
|
1375 |
int flag) |
1da177e4c
|
1376 1377 |
{ struct vfsmount *res, *p, *q, *r, *s; |
1a3906895
|
1378 |
struct path path; |
1da177e4c
|
1379 |
|
9676f0c63
|
1380 1381 |
if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) return NULL; |
36341f645
|
1382 |
res = q = clone_mnt(mnt, dentry, flag); |
1da177e4c
|
1383 1384 1385 1386 1387 |
if (!q) goto Enomem; q->mnt_mountpoint = mnt->mnt_mountpoint; p = mnt; |
fdadd65fb
|
1388 |
list_for_each_entry(r, &mnt->mnt_mounts, mnt_child) { |
7ec02ef15
|
1389 |
if (!is_subdir(r->mnt_mountpoint, dentry)) |
1da177e4c
|
1390 1391 1392 |
continue; for (s = r; s; s = next_mnt(s, r)) { |
9676f0c63
|
1393 1394 1395 1396 |
if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(s)) { s = skip_mnt_tree(s); continue; } |
1da177e4c
|
1397 1398 1399 1400 1401 |
while (p != s->mnt_parent) { p = p->mnt_parent; q = q->mnt_parent; } p = s; |
1a3906895
|
1402 1403 |
path.mnt = q; path.dentry = p->mnt_mountpoint; |
36341f645
|
1404 |
q = clone_mnt(p, p->mnt_root, flag); |
1da177e4c
|
1405 1406 |
if (!q) goto Enomem; |
99b7db7b8
|
1407 |
br_write_lock(vfsmount_lock); |
1da177e4c
|
1408 |
list_add_tail(&q->mnt_list, &res->mnt_list); |
1a3906895
|
1409 |
attach_mnt(q, &path); |
99b7db7b8
|
1410 |
br_write_unlock(vfsmount_lock); |
1da177e4c
|
1411 1412 1413 |
} } return res; |
b58fed8b1
|
1414 |
Enomem: |
1da177e4c
|
1415 |
if (res) { |
70fbcdf4d
|
1416 |
LIST_HEAD(umount_list); |
99b7db7b8
|
1417 |
br_write_lock(vfsmount_lock); |
a05964f39
|
1418 |
umount_tree(res, 0, &umount_list); |
99b7db7b8
|
1419 |
br_write_unlock(vfsmount_lock); |
70fbcdf4d
|
1420 |
release_mounts(&umount_list); |
1da177e4c
|
1421 1422 1423 |
} return NULL; } |
589ff870e
|
1424 |
struct vfsmount *collect_mounts(struct path *path) |
8aec08094
|
1425 1426 |
{ struct vfsmount *tree; |
1a60a2807
|
1427 |
down_write(&namespace_sem); |
589ff870e
|
1428 |
tree = copy_tree(path->mnt, path->dentry, CL_COPY_ALL | CL_PRIVATE); |
1a60a2807
|
1429 |
up_write(&namespace_sem); |
8aec08094
|
1430 1431 1432 1433 1434 1435 |
return tree; } void drop_collected_mounts(struct vfsmount *mnt) { LIST_HEAD(umount_list); |
1a60a2807
|
1436 |
down_write(&namespace_sem); |
99b7db7b8
|
1437 |
br_write_lock(vfsmount_lock); |
8aec08094
|
1438 |
umount_tree(mnt, 0, &umount_list); |
99b7db7b8
|
1439 |
br_write_unlock(vfsmount_lock); |
1a60a2807
|
1440 |
up_write(&namespace_sem); |
8aec08094
|
1441 1442 |
release_mounts(&umount_list); } |
1f707137b
|
1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 |
int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg, struct vfsmount *root) { struct vfsmount *mnt; int res = f(root, arg); if (res) return res; list_for_each_entry(mnt, &root->mnt_list, mnt_list) { res = f(mnt, arg); if (res) return res; } return 0; } |
719f5d7f0
|
1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 |
static void cleanup_group_ids(struct vfsmount *mnt, struct vfsmount *end) { struct vfsmount *p; for (p = mnt; p != end; p = next_mnt(p, mnt)) { if (p->mnt_group_id && !IS_MNT_SHARED(p)) mnt_release_group_id(p); } } static int invent_group_ids(struct vfsmount *mnt, bool recurse) { struct vfsmount *p; for (p = mnt; p; p = recurse ? next_mnt(p, mnt) : NULL) { if (!p->mnt_group_id && !IS_MNT_SHARED(p)) { int err = mnt_alloc_group_id(p); if (err) { cleanup_group_ids(mnt, p); return err; } } } return 0; } |
b90fa9ae8
|
1483 1484 |
/* * @source_mnt : mount tree to be attached |
214444032
|
1485 1486 1487 1488 |
* @nd : place the mount tree @source_mnt is attached * @parent_nd : if non-null, detach the source_mnt from its parent and * store the parent mount and mountpoint dentry. * (done when source_mnt is moved) |
b90fa9ae8
|
1489 1490 1491 |
* * NOTE: in the table below explains the semantics when a source mount * of a given type is attached to a destination mount of a given type. |
9676f0c63
|
1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 |
* --------------------------------------------------------------------------- * | BIND MOUNT OPERATION | * |************************************************************************** * | source-->| shared | private | slave | unbindable | * | dest | | | | | * | | | | | | | * | v | | | | | * |************************************************************************** * | shared | shared (++) | shared (+) | shared(+++)| invalid | * | | | | | | * |non-shared| shared (+) | private | slave (*) | invalid | * *************************************************************************** |
b90fa9ae8
|
1504 1505 1506 1507 1508 1509 1510 1511 1512 |
* A bind operation clones the source mount and mounts the clone on the * destination mount. * * (++) the cloned mount is propagated to all the mounts in the propagation * tree of the destination mount and the cloned mount is added to * the peer group of the source mount. * (+) the cloned mount is created under the destination mount and is marked * as shared. The cloned mount is added to the peer group of the source * mount. |
5afe00221
|
1513 1514 1515 1516 1517 1518 1519 |
* (+++) the mount is propagated to all the mounts in the propagation tree * of the destination mount and the cloned mount is made slave * of the same master as that of the source mount. The cloned mount * is marked as 'shared and slave'. * (*) the cloned mount is made a slave of the same master as that of the * source mount. * |
9676f0c63
|
1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 |
* --------------------------------------------------------------------------- * | MOVE MOUNT OPERATION | * |************************************************************************** * | source-->| shared | private | slave | unbindable | * | dest | | | | | * | | | | | | | * | v | | | | | * |************************************************************************** * | shared | shared (+) | shared (+) | shared(+++) | invalid | * | | | | | | * |non-shared| shared (+*) | private | slave (*) | unbindable | * *************************************************************************** |
5afe00221
|
1532 1533 1534 |
* * (+) the mount is moved to the destination. And is then propagated to * all the mounts in the propagation tree of the destination mount. |
214444032
|
1535 |
* (+*) the mount is moved to the destination. |
5afe00221
|
1536 1537 1538 1539 |
* (+++) the mount is moved to the destination and is then propagated to * all the mounts belonging to the destination mount's propagation tree. * the mount is marked as 'shared and slave'. * (*) the mount continues to be a slave at the new location. |
b90fa9ae8
|
1540 1541 1542 1543 1544 1545 1546 |
* * if the source mount is a tree, the operations explained above is * applied to each mount in the tree. * Must be called without spinlocks held, since this function can sleep * in allocations. */ static int attach_recursive_mnt(struct vfsmount *source_mnt, |
1a3906895
|
1547 |
struct path *path, struct path *parent_path) |
b90fa9ae8
|
1548 1549 |
{ LIST_HEAD(tree_list); |
1a3906895
|
1550 1551 |
struct vfsmount *dest_mnt = path->mnt; struct dentry *dest_dentry = path->dentry; |
b90fa9ae8
|
1552 |
struct vfsmount *child, *p; |
719f5d7f0
|
1553 |
int err; |
b90fa9ae8
|
1554 |
|
719f5d7f0
|
1555 1556 1557 1558 1559 1560 1561 1562 |
if (IS_MNT_SHARED(dest_mnt)) { err = invent_group_ids(source_mnt, true); if (err) goto out; } err = propagate_mnt(dest_mnt, dest_dentry, source_mnt, &tree_list); if (err) goto out_cleanup_ids; |
b90fa9ae8
|
1563 |
|
99b7db7b8
|
1564 |
br_write_lock(vfsmount_lock); |
df1a1ad29
|
1565 |
|
b90fa9ae8
|
1566 1567 1568 1569 |
if (IS_MNT_SHARED(dest_mnt)) { for (p = source_mnt; p; p = next_mnt(p, source_mnt)) set_mnt_shared(p); } |
1a3906895
|
1570 1571 1572 |
if (parent_path) { detach_mnt(source_mnt, parent_path); attach_mnt(source_mnt, path); |
e5d67f071
|
1573 |
touch_mnt_namespace(parent_path->mnt->mnt_ns); |
214444032
|
1574 1575 1576 1577 |
} else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); commit_tree(source_mnt); } |
b90fa9ae8
|
1578 1579 1580 1581 1582 |
list_for_each_entry_safe(child, p, &tree_list, mnt_hash) { list_del_init(&child->mnt_hash); commit_tree(child); } |
99b7db7b8
|
1583 |
br_write_unlock(vfsmount_lock); |
b90fa9ae8
|
1584 |
return 0; |
719f5d7f0
|
1585 1586 1587 1588 1589 1590 |
out_cleanup_ids: if (IS_MNT_SHARED(dest_mnt)) cleanup_group_ids(source_mnt, NULL); out: return err; |
b90fa9ae8
|
1591 |
} |
b12cea919
|
1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 |
static int lock_mount(struct path *path) { struct vfsmount *mnt; retry: mutex_lock(&path->dentry->d_inode->i_mutex); if (unlikely(cant_mount(path->dentry))) { mutex_unlock(&path->dentry->d_inode->i_mutex); return -ENOENT; } down_write(&namespace_sem); mnt = lookup_mnt(path); if (likely(!mnt)) return 0; up_write(&namespace_sem); mutex_unlock(&path->dentry->d_inode->i_mutex); path_put(path); path->mnt = mnt; path->dentry = dget(mnt->mnt_root); goto retry; } static void unlock_mount(struct path *path) { up_write(&namespace_sem); mutex_unlock(&path->dentry->d_inode->i_mutex); } |
8c3ee42e8
|
1618 |
static int graft_tree(struct vfsmount *mnt, struct path *path) |
1da177e4c
|
1619 |
{ |
1da177e4c
|
1620 1621 |
if (mnt->mnt_sb->s_flags & MS_NOUSER) return -EINVAL; |
8c3ee42e8
|
1622 |
if (S_ISDIR(path->dentry->d_inode->i_mode) != |
1da177e4c
|
1623 1624 |
S_ISDIR(mnt->mnt_root->d_inode->i_mode)) return -ENOTDIR; |
b12cea919
|
1625 1626 |
if (d_unlinked(path->dentry)) return -ENOENT; |
1da177e4c
|
1627 |
|
b12cea919
|
1628 |
return attach_recursive_mnt(mnt, path, NULL); |
1da177e4c
|
1629 1630 1631 |
} /* |
7a2e8a8fa
|
1632 1633 1634 1635 1636 |
* Sanity check the flags to change_mnt_propagation. */ static int flags_to_propagation_type(int flags) { |
7c6e984df
|
1637 |
int type = flags & ~(MS_REC | MS_SILENT); |
7a2e8a8fa
|
1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 |
/* Fail if any non-propagation flags are set */ if (type & ~(MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) return 0; /* Only one propagation flag should be set */ if (!is_power_of_2(type)) return 0; return type; } /* |
07b20889e
|
1649 1650 |
* recursively change the type of the mountpoint. */ |
0a0d8a467
|
1651 |
static int do_change_type(struct path *path, int flag) |
07b20889e
|
1652 |
{ |
2d92ab3c6
|
1653 |
struct vfsmount *m, *mnt = path->mnt; |
07b20889e
|
1654 |
int recurse = flag & MS_REC; |
7a2e8a8fa
|
1655 |
int type; |
719f5d7f0
|
1656 |
int err = 0; |
07b20889e
|
1657 |
|
ee6f95829
|
1658 1659 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
2d92ab3c6
|
1660 |
if (path->dentry != path->mnt->mnt_root) |
07b20889e
|
1661 |
return -EINVAL; |
7a2e8a8fa
|
1662 1663 1664 |
type = flags_to_propagation_type(flag); if (!type) return -EINVAL; |
07b20889e
|
1665 |
down_write(&namespace_sem); |
719f5d7f0
|
1666 1667 1668 1669 1670 |
if (type == MS_SHARED) { err = invent_group_ids(mnt, recurse); if (err) goto out_unlock; } |
99b7db7b8
|
1671 |
br_write_lock(vfsmount_lock); |
07b20889e
|
1672 1673 |
for (m = mnt; m; m = (recurse ? next_mnt(m, mnt) : NULL)) change_mnt_propagation(m, type); |
99b7db7b8
|
1674 |
br_write_unlock(vfsmount_lock); |
719f5d7f0
|
1675 1676 |
out_unlock: |
07b20889e
|
1677 |
up_write(&namespace_sem); |
719f5d7f0
|
1678 |
return err; |
07b20889e
|
1679 1680 1681 |
} /* |
1da177e4c
|
1682 1683 |
* do loopback mount. */ |
0a0d8a467
|
1684 |
static int do_loopback(struct path *path, char *old_name, |
2dafe1c4d
|
1685 |
int recurse) |
1da177e4c
|
1686 |
{ |
b12cea919
|
1687 |
LIST_HEAD(umount_list); |
2d92ab3c6
|
1688 |
struct path old_path; |
1da177e4c
|
1689 |
struct vfsmount *mnt = NULL; |
2d92ab3c6
|
1690 |
int err = mount_is_safe(path); |
1da177e4c
|
1691 1692 1693 1694 |
if (err) return err; if (!old_name || !*old_name) return -EINVAL; |
815d405ce
|
1695 |
err = kern_path(old_name, LOOKUP_FOLLOW|LOOKUP_AUTOMOUNT, &old_path); |
1da177e4c
|
1696 1697 |
if (err) return err; |
b12cea919
|
1698 1699 1700 |
err = lock_mount(path); if (err) goto out; |
1da177e4c
|
1701 |
err = -EINVAL; |
2d92ab3c6
|
1702 |
if (IS_MNT_UNBINDABLE(old_path.mnt)) |
b12cea919
|
1703 |
goto out2; |
9676f0c63
|
1704 |
|
2d92ab3c6
|
1705 |
if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) |
b12cea919
|
1706 |
goto out2; |
1da177e4c
|
1707 |
|
ccd48bc7f
|
1708 1709 |
err = -ENOMEM; if (recurse) |
2d92ab3c6
|
1710 |
mnt = copy_tree(old_path.mnt, old_path.dentry, 0); |
ccd48bc7f
|
1711 |
else |
2d92ab3c6
|
1712 |
mnt = clone_mnt(old_path.mnt, old_path.dentry, 0); |
ccd48bc7f
|
1713 1714 |
if (!mnt) |
b12cea919
|
1715 |
goto out2; |
ccd48bc7f
|
1716 |
|
2d92ab3c6
|
1717 |
err = graft_tree(mnt, path); |
ccd48bc7f
|
1718 |
if (err) { |
99b7db7b8
|
1719 |
br_write_lock(vfsmount_lock); |
a05964f39
|
1720 |
umount_tree(mnt, 0, &umount_list); |
99b7db7b8
|
1721 |
br_write_unlock(vfsmount_lock); |
5b83d2c5c
|
1722 |
} |
b12cea919
|
1723 1724 1725 |
out2: unlock_mount(path); release_mounts(&umount_list); |
ccd48bc7f
|
1726 |
out: |
2d92ab3c6
|
1727 |
path_put(&old_path); |
1da177e4c
|
1728 1729 |
return err; } |
2e4b7fcd9
|
1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 |
static int change_mount_flags(struct vfsmount *mnt, int ms_flags) { int error = 0; int readonly_request = 0; if (ms_flags & MS_RDONLY) readonly_request = 1; if (readonly_request == __mnt_is_readonly(mnt)) return 0; if (readonly_request) error = mnt_make_readonly(mnt); else __mnt_unmake_readonly(mnt); return error; } |
1da177e4c
|
1746 1747 1748 1749 1750 |
/* * change filesystem flags. dir should be a physical root of filesystem. * If you've mounted a non-root directory somewhere and want to do remount * on it - tough luck. */ |
0a0d8a467
|
1751 |
static int do_remount(struct path *path, int flags, int mnt_flags, |
1da177e4c
|
1752 1753 1754 |
void *data) { int err; |
2d92ab3c6
|
1755 |
struct super_block *sb = path->mnt->mnt_sb; |
1da177e4c
|
1756 1757 1758 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
2d92ab3c6
|
1759 |
if (!check_mnt(path->mnt)) |
1da177e4c
|
1760 |
return -EINVAL; |
2d92ab3c6
|
1761 |
if (path->dentry != path->mnt->mnt_root) |
1da177e4c
|
1762 |
return -EINVAL; |
ff36fe2c8
|
1763 1764 1765 |
err = security_sb_remount(sb, data); if (err) return err; |
1da177e4c
|
1766 |
down_write(&sb->s_umount); |
2e4b7fcd9
|
1767 |
if (flags & MS_BIND) |
2d92ab3c6
|
1768 |
err = change_mount_flags(path->mnt, flags); |
4aa98cf76
|
1769 |
else |
2e4b7fcd9
|
1770 |
err = do_remount_sb(sb, flags, data, 0); |
7b43a79f3
|
1771 |
if (!err) { |
99b7db7b8
|
1772 |
br_write_lock(vfsmount_lock); |
495d6c9c6
|
1773 |
mnt_flags |= path->mnt->mnt_flags & MNT_PROPAGATION_MASK; |
2d92ab3c6
|
1774 |
path->mnt->mnt_flags = mnt_flags; |
99b7db7b8
|
1775 |
br_write_unlock(vfsmount_lock); |
7b43a79f3
|
1776 |
} |
1da177e4c
|
1777 |
up_write(&sb->s_umount); |
0e55a7cca
|
1778 |
if (!err) { |
99b7db7b8
|
1779 |
br_write_lock(vfsmount_lock); |
0e55a7cca
|
1780 |
touch_mnt_namespace(path->mnt->mnt_ns); |
99b7db7b8
|
1781 |
br_write_unlock(vfsmount_lock); |
0e55a7cca
|
1782 |
} |
1da177e4c
|
1783 1784 |
return err; } |
9676f0c63
|
1785 1786 1787 1788 1789 1790 1791 1792 1793 |
static inline int tree_contains_unbindable(struct vfsmount *mnt) { struct vfsmount *p; for (p = mnt; p; p = next_mnt(p, mnt)) { if (IS_MNT_UNBINDABLE(p)) return 1; } return 0; } |
0a0d8a467
|
1794 |
static int do_move_mount(struct path *path, char *old_name) |
1da177e4c
|
1795 |
{ |
2d92ab3c6
|
1796 |
struct path old_path, parent_path; |
1da177e4c
|
1797 1798 1799 1800 1801 1802 |
struct vfsmount *p; int err = 0; if (!capable(CAP_SYS_ADMIN)) return -EPERM; if (!old_name || !*old_name) return -EINVAL; |
2d92ab3c6
|
1803 |
err = kern_path(old_name, LOOKUP_FOLLOW, &old_path); |
1da177e4c
|
1804 1805 |
if (err) return err; |
b12cea919
|
1806 |
err = lock_mount(path); |
cc53ce53c
|
1807 1808 |
if (err < 0) goto out; |
1da177e4c
|
1809 |
err = -EINVAL; |
2d92ab3c6
|
1810 |
if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) |
1da177e4c
|
1811 |
goto out1; |
f3da392e9
|
1812 |
if (d_unlinked(path->dentry)) |
214444032
|
1813 |
goto out1; |
1da177e4c
|
1814 1815 |
err = -EINVAL; |
2d92ab3c6
|
1816 |
if (old_path.dentry != old_path.mnt->mnt_root) |
214444032
|
1817 |
goto out1; |
1da177e4c
|
1818 |
|
2d92ab3c6
|
1819 |
if (old_path.mnt == old_path.mnt->mnt_parent) |
214444032
|
1820 |
goto out1; |
1da177e4c
|
1821 |
|
2d92ab3c6
|
1822 1823 |
if (S_ISDIR(path->dentry->d_inode->i_mode) != S_ISDIR(old_path.dentry->d_inode->i_mode)) |
214444032
|
1824 1825 1826 1827 |
goto out1; /* * Don't move a mount residing in a shared parent. */ |
2d92ab3c6
|
1828 1829 |
if (old_path.mnt->mnt_parent && IS_MNT_SHARED(old_path.mnt->mnt_parent)) |
214444032
|
1830 |
goto out1; |
9676f0c63
|
1831 1832 1833 1834 |
/* * Don't move a mount tree containing unbindable mounts to a destination * mount which is shared. */ |
2d92ab3c6
|
1835 1836 |
if (IS_MNT_SHARED(path->mnt) && tree_contains_unbindable(old_path.mnt)) |
9676f0c63
|
1837 |
goto out1; |
1da177e4c
|
1838 |
err = -ELOOP; |
2d92ab3c6
|
1839 1840 |
for (p = path->mnt; p->mnt_parent != p; p = p->mnt_parent) if (p == old_path.mnt) |
214444032
|
1841 |
goto out1; |
1da177e4c
|
1842 |
|
2d92ab3c6
|
1843 |
err = attach_recursive_mnt(old_path.mnt, path, &parent_path); |
4ac913785
|
1844 |
if (err) |
214444032
|
1845 |
goto out1; |
1da177e4c
|
1846 1847 1848 |
/* if the mount is moved, it should no longer be expire * automatically */ |
2d92ab3c6
|
1849 |
list_del_init(&old_path.mnt->mnt_expire); |
1da177e4c
|
1850 |
out1: |
b12cea919
|
1851 |
unlock_mount(path); |
1da177e4c
|
1852 |
out: |
1da177e4c
|
1853 |
if (!err) |
1a3906895
|
1854 |
path_put(&parent_path); |
2d92ab3c6
|
1855 |
path_put(&old_path); |
1da177e4c
|
1856 1857 |
return err; } |
9d412a43c
|
1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 |
static struct vfsmount *fs_set_subtype(struct vfsmount *mnt, const char *fstype) { int err; const char *subtype = strchr(fstype, '.'); if (subtype) { subtype++; err = -EINVAL; if (!subtype[0]) goto err; } else subtype = ""; mnt->mnt_sb->s_subtype = kstrdup(subtype, GFP_KERNEL); err = -ENOMEM; if (!mnt->mnt_sb->s_subtype) goto err; return mnt; err: mntput(mnt); return ERR_PTR(err); } struct vfsmount * do_kern_mount(const char *fstype, int flags, const char *name, void *data) { struct file_system_type *type = get_fs_type(fstype); struct vfsmount *mnt; if (!type) return ERR_PTR(-ENODEV); mnt = vfs_kern_mount(type, flags, name, data); if (!IS_ERR(mnt) && (type->fs_flags & FS_HAS_SUBTYPE) && !mnt->mnt_sb->s_subtype) mnt = fs_set_subtype(mnt, fstype); put_filesystem(type); return mnt; } EXPORT_SYMBOL_GPL(do_kern_mount); /* * add a mount into a namespace's mount tree */ static int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flags) { int err; mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL); |
b12cea919
|
1905 1906 1907 |
err = lock_mount(path); if (err) return err; |
9d412a43c
|
1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 |
err = -EINVAL; if (!(mnt_flags & MNT_SHRINKABLE) && !check_mnt(path->mnt)) goto unlock; /* Refuse the same filesystem on the same mount point */ err = -EBUSY; if (path->mnt->mnt_sb == newmnt->mnt_sb && path->mnt->mnt_root == path->dentry) goto unlock; err = -EINVAL; if (S_ISLNK(newmnt->mnt_root->d_inode->i_mode)) goto unlock; newmnt->mnt_flags = mnt_flags; err = graft_tree(newmnt, path); unlock: |
b12cea919
|
1927 |
unlock_mount(path); |
9d412a43c
|
1928 1929 |
return err; } |
b1e75df45
|
1930 |
|
1da177e4c
|
1931 1932 1933 1934 |
/* * create a new mount for userspace and request it to be added into the * namespace's tree */ |
0a0d8a467
|
1935 |
static int do_new_mount(struct path *path, char *type, int flags, |
1da177e4c
|
1936 1937 1938 |
int mnt_flags, char *name, void *data) { struct vfsmount *mnt; |
15f9a3f3e
|
1939 |
int err; |
1da177e4c
|
1940 |
|
eca6f534e
|
1941 |
if (!type) |
1da177e4c
|
1942 1943 1944 1945 1946 1947 1948 1949 1950 |
return -EINVAL; /* we need capabilities... */ if (!capable(CAP_SYS_ADMIN)) return -EPERM; mnt = do_kern_mount(type, flags, name, data); if (IS_ERR(mnt)) return PTR_ERR(mnt); |
15f9a3f3e
|
1951 1952 1953 1954 |
err = do_add_mount(mnt, path, mnt_flags); if (err) mntput(mnt); return err; |
1da177e4c
|
1955 |
} |
19a167af7
|
1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 |
int finish_automount(struct vfsmount *m, struct path *path) { int err; /* The new mount record should have at least 2 refs to prevent it being * expired before we get a chance to add it */ BUG_ON(mnt_get_count(m) < 2); if (m->mnt_sb == path->mnt->mnt_sb && m->mnt_root == path->dentry) { |
b1e75df45
|
1966 1967 |
err = -ELOOP; goto fail; |
19a167af7
|
1968 |
} |
19a167af7
|
1969 |
err = do_add_mount(m, path, path->mnt->mnt_flags | MNT_SHRINKABLE); |
b1e75df45
|
1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 |
if (!err) return 0; fail: /* remove m from any expiration list it may be on */ if (!list_empty(&m->mnt_expire)) { down_write(&namespace_sem); br_write_lock(vfsmount_lock); list_del_init(&m->mnt_expire); br_write_unlock(vfsmount_lock); up_write(&namespace_sem); |
19a167af7
|
1980 |
} |
b1e75df45
|
1981 1982 |
mntput(m); mntput(m); |
19a167af7
|
1983 1984 |
return err; } |
ea5b778a8
|
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 |
/** * mnt_set_expiry - Put a mount on an expiration list * @mnt: The mount to list. * @expiry_list: The list to add the mount to. */ void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list) { down_write(&namespace_sem); br_write_lock(vfsmount_lock); list_add_tail(&mnt->mnt_expire, expiry_list); br_write_unlock(vfsmount_lock); up_write(&namespace_sem); } EXPORT_SYMBOL(mnt_set_expiry); /* |
1da177e4c
|
2003 2004 2005 2006 2007 2008 |
* process a list of expirable mountpoints with the intent of discarding any * mountpoints that aren't in use and haven't been touched since last we came * here */ void mark_mounts_for_expiry(struct list_head *mounts) { |
1da177e4c
|
2009 2010 |
struct vfsmount *mnt, *next; LIST_HEAD(graveyard); |
bcc5c7d2b
|
2011 |
LIST_HEAD(umounts); |
1da177e4c
|
2012 2013 2014 |
if (list_empty(mounts)) return; |
bcc5c7d2b
|
2015 |
down_write(&namespace_sem); |
99b7db7b8
|
2016 |
br_write_lock(vfsmount_lock); |
1da177e4c
|
2017 2018 2019 2020 2021 2022 2023 |
/* extract from the expiration list every vfsmount that matches the * following criteria: * - only referenced by its parent vfsmount * - still marked for expiry (marked on the last call here; marks are * cleared by mntput()) */ |
55e700b92
|
2024 |
list_for_each_entry_safe(mnt, next, mounts, mnt_expire) { |
1da177e4c
|
2025 |
if (!xchg(&mnt->mnt_expiry_mark, 1) || |
bcc5c7d2b
|
2026 |
propagate_mount_busy(mnt, 1)) |
1da177e4c
|
2027 |
continue; |
55e700b92
|
2028 |
list_move(&mnt->mnt_expire, &graveyard); |
1da177e4c
|
2029 |
} |
bcc5c7d2b
|
2030 2031 2032 2033 2034 |
while (!list_empty(&graveyard)) { mnt = list_first_entry(&graveyard, struct vfsmount, mnt_expire); touch_mnt_namespace(mnt->mnt_ns); umount_tree(mnt, 1, &umounts); } |
99b7db7b8
|
2035 |
br_write_unlock(vfsmount_lock); |
bcc5c7d2b
|
2036 2037 2038 |
up_write(&namespace_sem); release_mounts(&umounts); |
5528f911b
|
2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 |
} EXPORT_SYMBOL_GPL(mark_mounts_for_expiry); /* * Ripoff of 'select_parent()' * * search the list of submounts for a given mountpoint, and move any * shrinkable submounts to the 'graveyard' list. */ static int select_submounts(struct vfsmount *parent, struct list_head *graveyard) { struct vfsmount *this_parent = parent; struct list_head *next; int found = 0; repeat: next = this_parent->mnt_mounts.next; resume: while (next != &this_parent->mnt_mounts) { struct list_head *tmp = next; struct vfsmount *mnt = list_entry(tmp, struct vfsmount, mnt_child); next = tmp->next; if (!(mnt->mnt_flags & MNT_SHRINKABLE)) |
1da177e4c
|
2064 |
continue; |
5528f911b
|
2065 2066 2067 2068 2069 2070 2071 |
/* * Descend a level if the d_mounts list is non-empty. */ if (!list_empty(&mnt->mnt_mounts)) { this_parent = mnt; goto repeat; } |
1da177e4c
|
2072 |
|
5528f911b
|
2073 |
if (!propagate_mount_busy(mnt, 1)) { |
5528f911b
|
2074 2075 2076 |
list_move_tail(&mnt->mnt_expire, graveyard); found++; } |
1da177e4c
|
2077 |
} |
5528f911b
|
2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 |
/* * All done at this level ... ascend and resume the search */ if (this_parent != parent) { next = this_parent->mnt_child.next; this_parent = this_parent->mnt_parent; goto resume; } return found; } /* * process a list of expirable mountpoints with the intent of discarding any * submounts of a specific parent mountpoint |
99b7db7b8
|
2092 2093 |
* * vfsmount_lock must be held for write |
5528f911b
|
2094 |
*/ |
c35038bec
|
2095 |
static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts) |
5528f911b
|
2096 2097 |
{ LIST_HEAD(graveyard); |
c35038bec
|
2098 |
struct vfsmount *m; |
5528f911b
|
2099 |
|
5528f911b
|
2100 |
/* extract submounts of 'mountpoint' from the expiration list */ |
c35038bec
|
2101 |
while (select_submounts(mnt, &graveyard)) { |
bcc5c7d2b
|
2102 |
while (!list_empty(&graveyard)) { |
c35038bec
|
2103 |
m = list_first_entry(&graveyard, struct vfsmount, |
bcc5c7d2b
|
2104 |
mnt_expire); |
afef80b3d
|
2105 2106 |
touch_mnt_namespace(m->mnt_ns); umount_tree(m, 1, umounts); |
bcc5c7d2b
|
2107 2108 |
} } |
1da177e4c
|
2109 |
} |
1da177e4c
|
2110 2111 2112 2113 2114 2115 |
/* * Some copy_from_user() implementations do not return the exact number of * bytes remaining to copy on a fault. But copy_mount_options() requires that. * Note that this function differs from copy_from_user() in that it will oops * on bad values of `to', rather than returning a short copy. */ |
b58fed8b1
|
2116 2117 |
static long exact_copy_from_user(void *to, const void __user * from, unsigned long n) |
1da177e4c
|
2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 |
{ char *t = to; const char __user *f = from; char c; if (!access_ok(VERIFY_READ, from, n)) return n; while (n) { if (__get_user(c, f)) { memset(t, 0, n); break; } *t++ = c; f++; n--; } return n; } |
b58fed8b1
|
2137 |
int copy_mount_options(const void __user * data, unsigned long *where) |
1da177e4c
|
2138 2139 2140 2141 |
{ int i; unsigned long page; unsigned long size; |
b58fed8b1
|
2142 |
|
1da177e4c
|
2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 |
*where = 0; if (!data) return 0; if (!(page = __get_free_page(GFP_KERNEL))) return -ENOMEM; /* We only care that *some* data at the address the user * gave us is valid. Just in case, we'll zero * the remainder of the page. */ /* copy_from_user cannot cross TASK_SIZE ! */ size = TASK_SIZE - (unsigned long)data; if (size > PAGE_SIZE) size = PAGE_SIZE; i = size - exact_copy_from_user((void *)page, data, size); if (!i) { |
b58fed8b1
|
2161 |
free_page(page); |
1da177e4c
|
2162 2163 2164 2165 2166 2167 2168 |
return -EFAULT; } if (i != PAGE_SIZE) memset((char *)page + i, 0, PAGE_SIZE - i); *where = page; return 0; } |
eca6f534e
|
2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 |
int copy_mount_string(const void __user *data, char **where) { char *tmp; if (!data) { *where = NULL; return 0; } tmp = strndup_user(data, PAGE_SIZE); if (IS_ERR(tmp)) return PTR_ERR(tmp); *where = tmp; return 0; } |
1da177e4c
|
2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 |
/* * Flags is a 32-bit value that allows up to 31 non-fs dependent flags to * be given to the mount() call (ie: read-only, no-dev, no-suid etc). * * data is a (void *) that can point to any structure up to * PAGE_SIZE-1 bytes, which can contain arbitrary fs-dependent * information (or be NULL). * * Pre-0.97 versions of mount() didn't have a flags word. * When the flags word was introduced its top half was required * to have the magic value 0xC0ED, and this remained so until 2.4.0-test9. * Therefore, if this magic number is present, it carries no information * and must be discarded. */ |
b58fed8b1
|
2199 |
long do_mount(char *dev_name, char *dir_name, char *type_page, |
1da177e4c
|
2200 2201 |
unsigned long flags, void *data_page) { |
2d92ab3c6
|
2202 |
struct path path; |
1da177e4c
|
2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 |
int retval = 0; int mnt_flags = 0; /* Discard magic */ if ((flags & MS_MGC_MSK) == MS_MGC_VAL) flags &= ~MS_MGC_MSK; /* Basic sanity checks */ if (!dir_name || !*dir_name || !memchr(dir_name, 0, PAGE_SIZE)) return -EINVAL; |
1da177e4c
|
2214 2215 2216 |
if (data_page) ((char *)data_page)[PAGE_SIZE - 1] = 0; |
a27ab9f26
|
2217 2218 2219 2220 2221 2222 2223 2224 2225 |
/* ... and get the mountpoint */ retval = kern_path(dir_name, LOOKUP_FOLLOW, &path); if (retval) return retval; retval = security_sb_mount(dev_name, &path, type_page, flags, data_page); if (retval) goto dput_out; |
613cbe3d4
|
2226 2227 2228 |
/* Default to relatime unless overriden */ if (!(flags & MS_NOATIME)) mnt_flags |= MNT_RELATIME; |
0a1c01c94
|
2229 |
|
1da177e4c
|
2230 2231 2232 2233 2234 2235 2236 |
/* Separate the per-mountpoint flags */ if (flags & MS_NOSUID) mnt_flags |= MNT_NOSUID; if (flags & MS_NODEV) mnt_flags |= MNT_NODEV; if (flags & MS_NOEXEC) mnt_flags |= MNT_NOEXEC; |
fc33a7bb9
|
2237 2238 2239 2240 |
if (flags & MS_NOATIME) mnt_flags |= MNT_NOATIME; if (flags & MS_NODIRATIME) mnt_flags |= MNT_NODIRATIME; |
d0adde574
|
2241 2242 |
if (flags & MS_STRICTATIME) mnt_flags &= ~(MNT_RELATIME | MNT_NOATIME); |
2e4b7fcd9
|
2243 2244 |
if (flags & MS_RDONLY) mnt_flags |= MNT_READONLY; |
fc33a7bb9
|
2245 |
|
7a4dec538
|
2246 |
flags &= ~(MS_NOSUID | MS_NOEXEC | MS_NODEV | MS_ACTIVE | MS_BORN | |
d0adde574
|
2247 2248 |
MS_NOATIME | MS_NODIRATIME | MS_RELATIME| MS_KERNMOUNT | MS_STRICTATIME); |
1da177e4c
|
2249 |
|
1da177e4c
|
2250 |
if (flags & MS_REMOUNT) |
2d92ab3c6
|
2251 |
retval = do_remount(&path, flags & ~MS_REMOUNT, mnt_flags, |
1da177e4c
|
2252 2253 |
data_page); else if (flags & MS_BIND) |
2d92ab3c6
|
2254 |
retval = do_loopback(&path, dev_name, flags & MS_REC); |
9676f0c63
|
2255 |
else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) |
2d92ab3c6
|
2256 |
retval = do_change_type(&path, flags); |
1da177e4c
|
2257 |
else if (flags & MS_MOVE) |
2d92ab3c6
|
2258 |
retval = do_move_mount(&path, dev_name); |
1da177e4c
|
2259 |
else |
2d92ab3c6
|
2260 |
retval = do_new_mount(&path, type_page, flags, mnt_flags, |
1da177e4c
|
2261 2262 |
dev_name, data_page); dput_out: |
2d92ab3c6
|
2263 |
path_put(&path); |
1da177e4c
|
2264 2265 |
return retval; } |
cf8d2c11c
|
2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 |
static struct mnt_namespace *alloc_mnt_ns(void) { struct mnt_namespace *new_ns; new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL); if (!new_ns) return ERR_PTR(-ENOMEM); atomic_set(&new_ns->count, 1); new_ns->root = NULL; INIT_LIST_HEAD(&new_ns->list); init_waitqueue_head(&new_ns->poll); new_ns->event = 0; return new_ns; } |
f03c65993
|
2280 2281 |
void mnt_make_longterm(struct vfsmount *mnt) { |
7e3d0eb0b
|
2282 |
__mnt_make_longterm(mnt); |
f03c65993
|
2283 2284 2285 2286 |
} void mnt_make_shortterm(struct vfsmount *mnt) { |
7e3d0eb0b
|
2287 |
#ifdef CONFIG_SMP |
f03c65993
|
2288 2289 2290 2291 2292 |
if (atomic_add_unless(&mnt->mnt_longterm, -1, 1)) return; br_write_lock(vfsmount_lock); atomic_dec(&mnt->mnt_longterm); br_write_unlock(vfsmount_lock); |
7e3d0eb0b
|
2293 |
#endif |
f03c65993
|
2294 |
} |
741a29513
|
2295 2296 2297 2298 |
/* * Allocate a new namespace structure and populate it with contents * copied from the namespace of the passed in task structure. */ |
e3222c4ec
|
2299 |
static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, |
6b3286ed1
|
2300 |
struct fs_struct *fs) |
1da177e4c
|
2301 |
{ |
6b3286ed1
|
2302 |
struct mnt_namespace *new_ns; |
7f2da1e7d
|
2303 |
struct vfsmount *rootmnt = NULL, *pwdmnt = NULL; |
1da177e4c
|
2304 |
struct vfsmount *p, *q; |
cf8d2c11c
|
2305 2306 2307 |
new_ns = alloc_mnt_ns(); if (IS_ERR(new_ns)) return new_ns; |
1da177e4c
|
2308 |
|
390c68436
|
2309 |
down_write(&namespace_sem); |
1da177e4c
|
2310 |
/* First pass: copy the tree topology */ |
6b3286ed1
|
2311 |
new_ns->root = copy_tree(mnt_ns->root, mnt_ns->root->mnt_root, |
9676f0c63
|
2312 |
CL_COPY_ALL | CL_EXPIRE); |
1da177e4c
|
2313 |
if (!new_ns->root) { |
390c68436
|
2314 |
up_write(&namespace_sem); |
1da177e4c
|
2315 |
kfree(new_ns); |
5cc4a0341
|
2316 |
return ERR_PTR(-ENOMEM); |
1da177e4c
|
2317 |
} |
99b7db7b8
|
2318 |
br_write_lock(vfsmount_lock); |
1da177e4c
|
2319 |
list_add_tail(&new_ns->list, &new_ns->root->mnt_list); |
99b7db7b8
|
2320 |
br_write_unlock(vfsmount_lock); |
1da177e4c
|
2321 2322 2323 2324 2325 2326 |
/* * Second pass: switch the tsk->fs->* elements and mark new vfsmounts * as belonging to new namespace. We have already acquired a private * fs_struct, so tsk->fs->lock is not needed. */ |
6b3286ed1
|
2327 |
p = mnt_ns->root; |
1da177e4c
|
2328 2329 |
q = new_ns->root; while (p) { |
6b3286ed1
|
2330 |
q->mnt_ns = new_ns; |
7e3d0eb0b
|
2331 |
__mnt_make_longterm(q); |
1da177e4c
|
2332 |
if (fs) { |
6ac08c39a
|
2333 |
if (p == fs->root.mnt) { |
f03c65993
|
2334 |
fs->root.mnt = mntget(q); |
7e3d0eb0b
|
2335 |
__mnt_make_longterm(q); |
f03c65993
|
2336 |
mnt_make_shortterm(p); |
1da177e4c
|
2337 |
rootmnt = p; |
1da177e4c
|
2338 |
} |
6ac08c39a
|
2339 |
if (p == fs->pwd.mnt) { |
f03c65993
|
2340 |
fs->pwd.mnt = mntget(q); |
7e3d0eb0b
|
2341 |
__mnt_make_longterm(q); |
f03c65993
|
2342 |
mnt_make_shortterm(p); |
1da177e4c
|
2343 |
pwdmnt = p; |
1da177e4c
|
2344 |
} |
1da177e4c
|
2345 |
} |
6b3286ed1
|
2346 |
p = next_mnt(p, mnt_ns->root); |
1da177e4c
|
2347 2348 |
q = next_mnt(q, new_ns->root); } |
390c68436
|
2349 |
up_write(&namespace_sem); |
1da177e4c
|
2350 |
|
1da177e4c
|
2351 |
if (rootmnt) |
f03c65993
|
2352 |
mntput(rootmnt); |
1da177e4c
|
2353 |
if (pwdmnt) |
f03c65993
|
2354 |
mntput(pwdmnt); |
1da177e4c
|
2355 |
|
741a29513
|
2356 2357 |
return new_ns; } |
213dd266d
|
2358 |
struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns, |
e3222c4ec
|
2359 |
struct fs_struct *new_fs) |
741a29513
|
2360 |
{ |
6b3286ed1
|
2361 |
struct mnt_namespace *new_ns; |
741a29513
|
2362 |
|
e3222c4ec
|
2363 |
BUG_ON(!ns); |
6b3286ed1
|
2364 |
get_mnt_ns(ns); |
741a29513
|
2365 2366 |
if (!(flags & CLONE_NEWNS)) |
e3222c4ec
|
2367 |
return ns; |
741a29513
|
2368 |
|
e3222c4ec
|
2369 |
new_ns = dup_mnt_ns(ns, new_fs); |
741a29513
|
2370 |
|
6b3286ed1
|
2371 |
put_mnt_ns(ns); |
e3222c4ec
|
2372 |
return new_ns; |
1da177e4c
|
2373 |
} |
cf8d2c11c
|
2374 2375 2376 2377 |
/** * create_mnt_ns - creates a private namespace and adds a root filesystem * @mnt: pointer to the new root filesystem mountpoint */ |
a2770d86b
|
2378 |
struct mnt_namespace *create_mnt_ns(struct vfsmount *mnt) |
cf8d2c11c
|
2379 2380 2381 2382 2383 2384 |
{ struct mnt_namespace *new_ns; new_ns = alloc_mnt_ns(); if (!IS_ERR(new_ns)) { mnt->mnt_ns = new_ns; |
7e3d0eb0b
|
2385 |
__mnt_make_longterm(mnt); |
cf8d2c11c
|
2386 2387 2388 2389 2390 |
new_ns->root = mnt; list_add(&new_ns->list, &new_ns->root->mnt_list); } return new_ns; } |
a2770d86b
|
2391 |
EXPORT_SYMBOL(create_mnt_ns); |
cf8d2c11c
|
2392 |
|
bdc480e3b
|
2393 2394 |
SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, char __user *, type, unsigned long, flags, void __user *, data) |
1da177e4c
|
2395 |
{ |
eca6f534e
|
2396 2397 2398 2399 |
int ret; char *kernel_type; char *kernel_dir; char *kernel_dev; |
1da177e4c
|
2400 |
unsigned long data_page; |
1da177e4c
|
2401 |
|
eca6f534e
|
2402 2403 2404 |
ret = copy_mount_string(type, &kernel_type); if (ret < 0) goto out_type; |
1da177e4c
|
2405 |
|
eca6f534e
|
2406 2407 2408 2409 2410 |
kernel_dir = getname(dir_name); if (IS_ERR(kernel_dir)) { ret = PTR_ERR(kernel_dir); goto out_dir; } |
1da177e4c
|
2411 |
|
eca6f534e
|
2412 2413 2414 |
ret = copy_mount_string(dev_name, &kernel_dev); if (ret < 0) goto out_dev; |
1da177e4c
|
2415 |
|
eca6f534e
|
2416 2417 2418 |
ret = copy_mount_options(data, &data_page); if (ret < 0) goto out_data; |
1da177e4c
|
2419 |
|
eca6f534e
|
2420 2421 |
ret = do_mount(kernel_dev, kernel_dir, kernel_type, flags, (void *) data_page); |
1da177e4c
|
2422 |
|
eca6f534e
|
2423 2424 2425 2426 2427 2428 2429 2430 2431 |
free_page(data_page); out_data: kfree(kernel_dev); out_dev: putname(kernel_dir); out_dir: kfree(kernel_type); out_type: return ret; |
1da177e4c
|
2432 2433 2434 |
} /* |
1da177e4c
|
2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 |
* pivot_root Semantics: * Moves the root file system of the current process to the directory put_old, * makes new_root as the new root file system of the current process, and sets * root/cwd of all processes which had them on the current root to new_root. * * Restrictions: * The new_root and put_old must be directories, and must not be on the * same file system as the current process root. The put_old must be * underneath new_root, i.e. adding a non-zero number of /.. to the string * pointed to by put_old must yield the same directory as new_root. No other * file system may be mounted on put_old. After all, new_root is a mountpoint. * |
4a0d11fae
|
2447 2448 2449 2450 |
* Also, the current root cannot be on the 'rootfs' (initial ramfs) filesystem. * See Documentation/filesystems/ramfs-rootfs-initramfs.txt for alternatives * in this situation. * |
1da177e4c
|
2451 2452 2453 2454 2455 2456 2457 2458 |
* Notes: * - we don't move root/cwd if they are not at the root (reason: if something * cared enough to change them, it's probably wrong to force them elsewhere) * - it's okay to pick a root that isn't the root of a file system, e.g. * /nfs/my_root where /nfs is the mount point. It must be a mountpoint, * though, so you may need to say mount --bind /nfs/my_root /nfs/my_root * first. */ |
3480b2574
|
2459 2460 |
SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, const char __user *, put_old) |
1da177e4c
|
2461 2462 |
{ struct vfsmount *tmp; |
2d8f30380
|
2463 |
struct path new, old, parent_path, root_parent, root; |
1da177e4c
|
2464 2465 2466 2467 |
int error; if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
2d8f30380
|
2468 |
error = user_path_dir(new_root, &new); |
1da177e4c
|
2469 2470 |
if (error) goto out0; |
1da177e4c
|
2471 |
|
2d8f30380
|
2472 |
error = user_path_dir(put_old, &old); |
1da177e4c
|
2473 2474 |
if (error) goto out1; |
2d8f30380
|
2475 |
error = security_sb_pivotroot(&old, &new); |
b12cea919
|
2476 2477 |
if (error) goto out2; |
1da177e4c
|
2478 |
|
f7ad3c6be
|
2479 |
get_fs_root(current->fs, &root); |
b12cea919
|
2480 2481 2482 |
error = lock_mount(&old); if (error) goto out3; |
1da177e4c
|
2483 |
error = -EINVAL; |
2d8f30380
|
2484 2485 |
if (IS_MNT_SHARED(old.mnt) || IS_MNT_SHARED(new.mnt->mnt_parent) || |
8c3ee42e8
|
2486 |
IS_MNT_SHARED(root.mnt->mnt_parent)) |
b12cea919
|
2487 |
goto out4; |
27cb1572e
|
2488 |
if (!check_mnt(root.mnt) || !check_mnt(new.mnt)) |
b12cea919
|
2489 |
goto out4; |
1da177e4c
|
2490 |
error = -ENOENT; |
f3da392e9
|
2491 |
if (d_unlinked(new.dentry)) |
b12cea919
|
2492 |
goto out4; |
f3da392e9
|
2493 |
if (d_unlinked(old.dentry)) |
b12cea919
|
2494 |
goto out4; |
1da177e4c
|
2495 |
error = -EBUSY; |
2d8f30380
|
2496 2497 |
if (new.mnt == root.mnt || old.mnt == root.mnt) |
b12cea919
|
2498 |
goto out4; /* loop, on the same file system */ |
1da177e4c
|
2499 |
error = -EINVAL; |
8c3ee42e8
|
2500 |
if (root.mnt->mnt_root != root.dentry) |
b12cea919
|
2501 |
goto out4; /* not a mountpoint */ |
8c3ee42e8
|
2502 |
if (root.mnt->mnt_parent == root.mnt) |
b12cea919
|
2503 |
goto out4; /* not attached */ |
2d8f30380
|
2504 |
if (new.mnt->mnt_root != new.dentry) |
b12cea919
|
2505 |
goto out4; /* not a mountpoint */ |
2d8f30380
|
2506 |
if (new.mnt->mnt_parent == new.mnt) |
b12cea919
|
2507 |
goto out4; /* not attached */ |
4ac913785
|
2508 |
/* make sure we can reach put_old from new_root */ |
2d8f30380
|
2509 |
tmp = old.mnt; |
2d8f30380
|
2510 |
if (tmp != new.mnt) { |
1da177e4c
|
2511 2512 |
for (;;) { if (tmp->mnt_parent == tmp) |
b12cea919
|
2513 |
goto out4; /* already mounted on put_old */ |
2d8f30380
|
2514 |
if (tmp->mnt_parent == new.mnt) |
1da177e4c
|
2515 2516 2517 |
break; tmp = tmp->mnt_parent; } |
2d8f30380
|
2518 |
if (!is_subdir(tmp->mnt_mountpoint, new.dentry)) |
b12cea919
|
2519 |
goto out4; |
2d8f30380
|
2520 |
} else if (!is_subdir(old.dentry, new.dentry)) |
b12cea919
|
2521 |
goto out4; |
27cb1572e
|
2522 |
br_write_lock(vfsmount_lock); |
2d8f30380
|
2523 |
detach_mnt(new.mnt, &parent_path); |
8c3ee42e8
|
2524 |
detach_mnt(root.mnt, &root_parent); |
4ac913785
|
2525 |
/* mount old root on put_old */ |
2d8f30380
|
2526 |
attach_mnt(root.mnt, &old); |
4ac913785
|
2527 |
/* mount new_root on / */ |
2d8f30380
|
2528 |
attach_mnt(new.mnt, &root_parent); |
6b3286ed1
|
2529 |
touch_mnt_namespace(current->nsproxy->mnt_ns); |
99b7db7b8
|
2530 |
br_write_unlock(vfsmount_lock); |
2d8f30380
|
2531 |
chroot_fs_refs(&root, &new); |
1da177e4c
|
2532 |
error = 0; |
b12cea919
|
2533 2534 2535 2536 2537 2538 2539 |
out4: unlock_mount(&old); if (!error) { path_put(&root_parent); path_put(&parent_path); } out3: |
8c3ee42e8
|
2540 |
path_put(&root); |
b12cea919
|
2541 |
out2: |
2d8f30380
|
2542 |
path_put(&old); |
1da177e4c
|
2543 |
out1: |
2d8f30380
|
2544 |
path_put(&new); |
1da177e4c
|
2545 |
out0: |
1da177e4c
|
2546 |
return error; |
1da177e4c
|
2547 2548 2549 2550 2551 |
} static void __init init_mount_tree(void) { struct vfsmount *mnt; |
6b3286ed1
|
2552 |
struct mnt_namespace *ns; |
ac748a09f
|
2553 |
struct path root; |
1da177e4c
|
2554 2555 2556 2557 |
mnt = do_kern_mount("rootfs", 0, "rootfs", NULL); if (IS_ERR(mnt)) panic("Can't create rootfs"); |
b3e19d924
|
2558 |
|
3b22edc57
|
2559 2560 |
ns = create_mnt_ns(mnt); if (IS_ERR(ns)) |
1da177e4c
|
2561 |
panic("Can't allocate initial namespace"); |
6b3286ed1
|
2562 2563 2564 |
init_task.nsproxy->mnt_ns = ns; get_mnt_ns(ns); |
ac748a09f
|
2565 2566 2567 2568 2569 |
root.mnt = ns->root; root.dentry = ns->root->mnt_root; set_fs_pwd(current->fs, &root); set_fs_root(current->fs, &root); |
1da177e4c
|
2570 |
} |
74bf17cff
|
2571 |
void __init mnt_init(void) |
1da177e4c
|
2572 |
{ |
13f14b4d8
|
2573 |
unsigned u; |
15a67dd8c
|
2574 |
int err; |
1da177e4c
|
2575 |
|
390c68436
|
2576 |
init_rwsem(&namespace_sem); |
1da177e4c
|
2577 |
mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct vfsmount), |
20c2df83d
|
2578 |
0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); |
1da177e4c
|
2579 |
|
b58fed8b1
|
2580 |
mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC); |
1da177e4c
|
2581 2582 2583 2584 |
if (!mount_hashtable) panic("Failed to allocate mount hash table "); |
80cdc6dae
|
2585 2586 |
printk(KERN_INFO "Mount-cache hash table entries: %lu ", HASH_SIZE); |
13f14b4d8
|
2587 2588 2589 |
for (u = 0; u < HASH_SIZE; u++) INIT_LIST_HEAD(&mount_hashtable[u]); |
1da177e4c
|
2590 |
|
99b7db7b8
|
2591 |
br_lock_init(vfsmount_lock); |
15a67dd8c
|
2592 2593 2594 2595 |
err = sysfs_init(); if (err) printk(KERN_WARNING "%s: sysfs_init error: %d ", |
8e24eea72
|
2596 |
__func__, err); |
00d266662
|
2597 2598 |
fs_kobj = kobject_create_and_add("fs", NULL); if (!fs_kobj) |
8e24eea72
|
2599 2600 |
printk(KERN_WARNING "%s: kobj create error ", __func__); |
1da177e4c
|
2601 2602 2603 |
init_rootfs(); init_mount_tree(); } |
616511d03
|
2604 |
void put_mnt_ns(struct mnt_namespace *ns) |
1da177e4c
|
2605 |
{ |
70fbcdf4d
|
2606 |
LIST_HEAD(umount_list); |
616511d03
|
2607 |
|
d498b25a4
|
2608 |
if (!atomic_dec_and_test(&ns->count)) |
616511d03
|
2609 |
return; |
390c68436
|
2610 |
down_write(&namespace_sem); |
99b7db7b8
|
2611 |
br_write_lock(vfsmount_lock); |
d498b25a4
|
2612 |
umount_tree(ns->root, 0, &umount_list); |
99b7db7b8
|
2613 |
br_write_unlock(vfsmount_lock); |
390c68436
|
2614 |
up_write(&namespace_sem); |
70fbcdf4d
|
2615 |
release_mounts(&umount_list); |
6b3286ed1
|
2616 |
kfree(ns); |
1da177e4c
|
2617 |
} |
cf8d2c11c
|
2618 |
EXPORT_SYMBOL(put_mnt_ns); |
9d412a43c
|
2619 2620 2621 |
struct vfsmount *kern_mount_data(struct file_system_type *type, void *data) { |
423e0ab08
|
2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 |
struct vfsmount *mnt; mnt = vfs_kern_mount(type, MS_KERNMOUNT, type->name, data); if (!IS_ERR(mnt)) { /* * it is a longterm mount, don't release mnt until * we unmount before file sys is unregistered */ mnt_make_longterm(mnt); } return mnt; |
9d412a43c
|
2632 2633 |
} EXPORT_SYMBOL_GPL(kern_mount_data); |
423e0ab08
|
2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 |
void kern_unmount(struct vfsmount *mnt) { /* release long term mount so mount point can be released */ if (!IS_ERR_OR_NULL(mnt)) { mnt_make_shortterm(mnt); mntput(mnt); } } EXPORT_SYMBOL(kern_unmount); |