Blame view
fs/namespace.c
57.2 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 13 14 |
#include <linux/syscalls.h> #include <linux/slab.h> #include <linux/sched.h> #include <linux/smp_lock.h> #include <linux/init.h> |
15a67dd8c
|
15 |
#include <linux/kernel.h> |
1da177e4c
|
16 |
#include <linux/acct.h> |
16f7e0fe2
|
17 |
#include <linux/capability.h> |
3d733633a
|
18 |
#include <linux/cpumask.h> |
1da177e4c
|
19 |
#include <linux/module.h> |
f20a9ead0
|
20 |
#include <linux/sysfs.h> |
1da177e4c
|
21 |
#include <linux/seq_file.h> |
6b3286ed1
|
22 |
#include <linux/mnt_namespace.h> |
1da177e4c
|
23 |
#include <linux/namei.h> |
b43f3cbd2
|
24 |
#include <linux/nsproxy.h> |
1da177e4c
|
25 26 |
#include <linux/security.h> #include <linux/mount.h> |
07f3f05c1
|
27 |
#include <linux/ramfs.h> |
13f14b4d8
|
28 |
#include <linux/log2.h> |
73cd49ecd
|
29 |
#include <linux/idr.h> |
5ad4e53bd
|
30 |
#include <linux/fs_struct.h> |
1da177e4c
|
31 32 |
#include <asm/uaccess.h> #include <asm/unistd.h> |
07b20889e
|
33 |
#include "pnode.h" |
948730b0e
|
34 |
#include "internal.h" |
1da177e4c
|
35 |
|
13f14b4d8
|
36 37 |
#define HASH_SHIFT ilog2(PAGE_SIZE / sizeof(struct list_head)) #define HASH_SIZE (1UL << HASH_SHIFT) |
1da177e4c
|
38 |
/* spinlock for vfsmount related operations, inplace of dcache_lock */ |
5addc5dd8
|
39 40 41 |
__cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock); static int event; |
73cd49ecd
|
42 |
static DEFINE_IDA(mnt_id_ida); |
719f5d7f0
|
43 |
static DEFINE_IDA(mnt_group_ida); |
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 |
|
1da177e4c
|
55 56 |
static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry) { |
b58fed8b1
|
57 58 |
unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES); tmp += ((unsigned long)dentry / L1_CACHE_BYTES); |
13f14b4d8
|
59 60 |
tmp = tmp + (tmp >> HASH_SHIFT); return tmp & (HASH_SIZE - 1); |
1da177e4c
|
61 |
} |
3d733633a
|
62 |
#define MNT_WRITER_UNDERFLOW_LIMIT -(1<<16) |
73cd49ecd
|
63 64 65 66 67 68 69 70 |
/* allocation is serialized by namespace_sem */ static int mnt_alloc_id(struct vfsmount *mnt) { int res; retry: ida_pre_get(&mnt_id_ida, GFP_KERNEL); spin_lock(&vfsmount_lock); |
f21f62208
|
71 72 73 |
res = ida_get_new_above(&mnt_id_ida, mnt_id_start, &mnt->mnt_id); if (!res) mnt_id_start = mnt->mnt_id + 1; |
73cd49ecd
|
74 75 76 77 78 79 80 81 82 |
spin_unlock(&vfsmount_lock); if (res == -EAGAIN) goto retry; return res; } static void mnt_free_id(struct vfsmount *mnt) { |
f21f62208
|
83 |
int id = mnt->mnt_id; |
73cd49ecd
|
84 |
spin_lock(&vfsmount_lock); |
f21f62208
|
85 86 87 |
ida_remove(&mnt_id_ida, id); if (mnt_id_start > id) mnt_id_start = id; |
73cd49ecd
|
88 89 |
spin_unlock(&vfsmount_lock); } |
719f5d7f0
|
90 91 92 93 94 95 96 |
/* * Allocate a new peer group ID * * mnt_group_ida is protected by namespace_sem */ static int mnt_alloc_group_id(struct vfsmount *mnt) { |
f21f62208
|
97 |
int res; |
719f5d7f0
|
98 99 |
if (!ida_pre_get(&mnt_group_ida, GFP_KERNEL)) return -ENOMEM; |
f21f62208
|
100 101 102 103 104 105 106 |
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
|
107 108 109 110 111 112 113 |
} /* * Release a peer group ID */ void mnt_release_group_id(struct vfsmount *mnt) { |
f21f62208
|
114 115 116 117 |
int id = mnt->mnt_group_id; ida_remove(&mnt_group_ida, id); if (mnt_group_start > id) mnt_group_start = id; |
719f5d7f0
|
118 119 |
mnt->mnt_group_id = 0; } |
1da177e4c
|
120 121 |
struct vfsmount *alloc_vfsmnt(const char *name) { |
c37622296
|
122 |
struct vfsmount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL); |
1da177e4c
|
123 |
if (mnt) { |
73cd49ecd
|
124 125 126 |
int err; err = mnt_alloc_id(mnt); |
88b387824
|
127 128 129 130 131 132 133 |
if (err) goto out_free_cache; if (name) { mnt->mnt_devname = kstrdup(name, GFP_KERNEL); if (!mnt->mnt_devname) goto out_free_id; |
73cd49ecd
|
134 |
} |
b58fed8b1
|
135 |
atomic_set(&mnt->mnt_count, 1); |
1da177e4c
|
136 137 138 139 |
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
|
140 |
INIT_LIST_HEAD(&mnt->mnt_expire); |
03e06e68f
|
141 |
INIT_LIST_HEAD(&mnt->mnt_share); |
a58b0eb8e
|
142 143 |
INIT_LIST_HEAD(&mnt->mnt_slave_list); INIT_LIST_HEAD(&mnt->mnt_slave); |
d3ef3d735
|
144 145 146 147 148 149 150 |
#ifdef CONFIG_SMP mnt->mnt_writers = alloc_percpu(int); if (!mnt->mnt_writers) goto out_free_devname; #else mnt->mnt_writers = 0; #endif |
1da177e4c
|
151 152 |
} return mnt; |
88b387824
|
153 |
|
d3ef3d735
|
154 155 156 157 |
#ifdef CONFIG_SMP out_free_devname: kfree(mnt->mnt_devname); #endif |
88b387824
|
158 159 160 161 162 |
out_free_id: mnt_free_id(mnt); out_free_cache: kmem_cache_free(mnt_cache, mnt); return NULL; |
1da177e4c
|
163 |
} |
8366025eb
|
164 165 166 167 168 169 170 171 |
/* * 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
|
172 173 174 175 176 177 178 179 180 181 182 183 184 |
/* * __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
|
185 186 187 188 189 |
if (mnt->mnt_flags & MNT_READONLY) return 1; if (mnt->mnt_sb->s_flags & MS_RDONLY) return 1; return 0; |
3d733633a
|
190 191 |
} EXPORT_SYMBOL_GPL(__mnt_is_readonly); |
d3ef3d735
|
192 193 194 195 196 197 198 199 |
static inline void inc_mnt_writers(struct vfsmount *mnt) { #ifdef CONFIG_SMP (*per_cpu_ptr(mnt->mnt_writers, smp_processor_id()))++; #else mnt->mnt_writers++; #endif } |
3d733633a
|
200 |
|
d3ef3d735
|
201 |
static inline void dec_mnt_writers(struct vfsmount *mnt) |
3d733633a
|
202 |
{ |
d3ef3d735
|
203 204 205 206 207 |
#ifdef CONFIG_SMP (*per_cpu_ptr(mnt->mnt_writers, smp_processor_id()))--; #else mnt->mnt_writers--; #endif |
3d733633a
|
208 |
} |
3d733633a
|
209 |
|
d3ef3d735
|
210 |
static unsigned int count_mnt_writers(struct vfsmount *mnt) |
3d733633a
|
211 |
{ |
d3ef3d735
|
212 213 |
#ifdef CONFIG_SMP unsigned int count = 0; |
3d733633a
|
214 |
int cpu; |
3d733633a
|
215 216 |
for_each_possible_cpu(cpu) { |
d3ef3d735
|
217 |
count += *per_cpu_ptr(mnt->mnt_writers, cpu); |
3d733633a
|
218 |
} |
3d733633a
|
219 |
|
d3ef3d735
|
220 221 222 223 |
return count; #else return mnt->mnt_writers; #endif |
3d733633a
|
224 225 226 227 228 229 230 231 232 233 |
} /* * 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
|
234 235 236 237 238 239 240 241 242 243 244 245 |
/** * 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
|
246 |
int ret = 0; |
3d733633a
|
247 |
|
d3ef3d735
|
248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 |
preempt_disable(); inc_mnt_writers(mnt); /* * The store to inc_mnt_writers must be visible before we pass * 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
|
264 |
if (__mnt_is_readonly(mnt)) { |
d3ef3d735
|
265 |
dec_mnt_writers(mnt); |
3d733633a
|
266 267 268 |
ret = -EROFS; goto out; } |
3d733633a
|
269 |
out: |
d3ef3d735
|
270 |
preempt_enable(); |
3d733633a
|
271 |
return ret; |
8366025eb
|
272 273 274 275 |
} EXPORT_SYMBOL_GPL(mnt_want_write); /** |
96029c4e0
|
276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 |
* 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(); inc_mnt_writers(mnt); 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
|
308 309 |
struct inode *inode = file->f_dentry->d_inode; if (!(file->f_mode & FMODE_WRITE) || special_file(inode->i_mode)) |
96029c4e0
|
310 311 312 313 314 315 316 |
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
|
317 318 319 320 321 322 323 324 325 |
* 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
|
326 327 328 |
preempt_disable(); dec_mnt_writers(mnt); preempt_enable(); |
8366025eb
|
329 330 |
} EXPORT_SYMBOL_GPL(mnt_drop_write); |
2e4b7fcd9
|
331 |
static int mnt_make_readonly(struct vfsmount *mnt) |
8366025eb
|
332 |
{ |
3d733633a
|
333 |
int ret = 0; |
d3ef3d735
|
334 335 |
spin_lock(&vfsmount_lock); mnt->mnt_flags |= MNT_WRITE_HOLD; |
3d733633a
|
336 |
/* |
d3ef3d735
|
337 338 |
* After storing MNT_WRITE_HOLD, we'll read the counters. This store * should be visible before we do. |
3d733633a
|
339 |
*/ |
d3ef3d735
|
340 |
smp_mb(); |
3d733633a
|
341 |
/* |
d3ef3d735
|
342 343 344 345 346 347 348 349 350 351 352 353 354 355 |
* 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
|
356 |
*/ |
d3ef3d735
|
357 358 359 |
if (count_mnt_writers(mnt) > 0) ret = -EBUSY; else |
2e4b7fcd9
|
360 |
mnt->mnt_flags |= MNT_READONLY; |
d3ef3d735
|
361 362 363 364 365 366 |
/* * 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; |
2e4b7fcd9
|
367 |
spin_unlock(&vfsmount_lock); |
3d733633a
|
368 |
return ret; |
8366025eb
|
369 |
} |
8366025eb
|
370 |
|
2e4b7fcd9
|
371 372 373 374 375 376 |
static void __mnt_unmake_readonly(struct vfsmount *mnt) { spin_lock(&vfsmount_lock); mnt->mnt_flags &= ~MNT_READONLY; spin_unlock(&vfsmount_lock); } |
a3ec947c8
|
377 |
void simple_set_mnt(struct vfsmount *mnt, struct super_block *sb) |
454e2398b
|
378 379 380 |
{ mnt->mnt_sb = sb; mnt->mnt_root = dget(sb->s_root); |
454e2398b
|
381 382 383 |
} EXPORT_SYMBOL(simple_set_mnt); |
1da177e4c
|
384 385 386 |
void free_vfsmnt(struct vfsmount *mnt) { kfree(mnt->mnt_devname); |
73cd49ecd
|
387 |
mnt_free_id(mnt); |
d3ef3d735
|
388 389 390 |
#ifdef CONFIG_SMP free_percpu(mnt->mnt_writers); #endif |
1da177e4c
|
391 392 393 394 |
kmem_cache_free(mnt_cache, mnt); } /* |
a05964f39
|
395 396 |
* 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. |
1da177e4c
|
397 |
*/ |
a05964f39
|
398 399 |
struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry, int dir) |
1da177e4c
|
400 |
{ |
b58fed8b1
|
401 402 |
struct list_head *head = mount_hashtable + hash(mnt, dentry); struct list_head *tmp = head; |
1da177e4c
|
403 |
struct vfsmount *p, *found = NULL; |
1da177e4c
|
404 |
for (;;) { |
a05964f39
|
405 |
tmp = dir ? tmp->next : tmp->prev; |
1da177e4c
|
406 407 408 409 410 |
p = NULL; if (tmp == head) break; p = list_entry(tmp, struct vfsmount, mnt_hash); if (p->mnt_parent == mnt && p->mnt_mountpoint == dentry) { |
a05964f39
|
411 |
found = p; |
1da177e4c
|
412 413 414 |
break; } } |
1da177e4c
|
415 416 |
return found; } |
a05964f39
|
417 418 419 420 |
/* * lookup_mnt increments the ref count before returning * the vfsmount struct. */ |
1c755af4d
|
421 |
struct vfsmount *lookup_mnt(struct path *path) |
a05964f39
|
422 423 424 |
{ struct vfsmount *child_mnt; spin_lock(&vfsmount_lock); |
1c755af4d
|
425 |
if ((child_mnt = __lookup_mnt(path->mnt, path->dentry, 1))) |
a05964f39
|
426 427 428 429 |
mntget(child_mnt); spin_unlock(&vfsmount_lock); return child_mnt; } |
1da177e4c
|
430 431 |
static inline int check_mnt(struct vfsmount *mnt) { |
6b3286ed1
|
432 |
return mnt->mnt_ns == current->nsproxy->mnt_ns; |
1da177e4c
|
433 |
} |
6b3286ed1
|
434 |
static void touch_mnt_namespace(struct mnt_namespace *ns) |
5addc5dd8
|
435 436 437 438 439 440 |
{ if (ns) { ns->event = ++event; wake_up_interruptible(&ns->poll); } } |
6b3286ed1
|
441 |
static void __touch_mnt_namespace(struct mnt_namespace *ns) |
5addc5dd8
|
442 443 444 445 446 447 |
{ if (ns && ns->event != event) { ns->event = event; wake_up_interruptible(&ns->poll); } } |
1a3906895
|
448 |
static void detach_mnt(struct vfsmount *mnt, struct path *old_path) |
1da177e4c
|
449 |
{ |
1a3906895
|
450 451 |
old_path->dentry = mnt->mnt_mountpoint; old_path->mnt = mnt->mnt_parent; |
1da177e4c
|
452 453 454 455 |
mnt->mnt_parent = mnt; mnt->mnt_mountpoint = mnt->mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); |
1a3906895
|
456 |
old_path->dentry->d_mounted--; |
1da177e4c
|
457 |
} |
b90fa9ae8
|
458 459 460 461 462 463 464 |
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); dentry->d_mounted++; } |
1a3906895
|
465 |
static void attach_mnt(struct vfsmount *mnt, struct path *path) |
1da177e4c
|
466 |
{ |
1a3906895
|
467 |
mnt_set_mountpoint(path->mnt, path->dentry, mnt); |
b90fa9ae8
|
468 |
list_add_tail(&mnt->mnt_hash, mount_hashtable + |
1a3906895
|
469 470 |
hash(path->mnt, path->dentry)); list_add_tail(&mnt->mnt_child, &path->mnt->mnt_mounts); |
b90fa9ae8
|
471 472 473 474 475 476 477 478 479 480 |
} /* * the caller must hold vfsmount_lock */ static void commit_tree(struct vfsmount *mnt) { struct vfsmount *parent = mnt->mnt_parent; struct vfsmount *m; LIST_HEAD(head); |
6b3286ed1
|
481 |
struct mnt_namespace *n = parent->mnt_ns; |
b90fa9ae8
|
482 483 484 485 486 |
BUG_ON(parent == mnt); list_add_tail(&head, &mnt->mnt_list); list_for_each_entry(m, &head, mnt_list) |
6b3286ed1
|
487 |
m->mnt_ns = n; |
b90fa9ae8
|
488 489 490 491 492 |
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
|
493 |
touch_mnt_namespace(n); |
1da177e4c
|
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 |
} 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
|
511 512 513 514 515 516 517 518 519 |
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; } |
36341f645
|
520 521 |
static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, int flag) |
1da177e4c
|
522 523 524 525 526 |
{ struct super_block *sb = old->mnt_sb; struct vfsmount *mnt = alloc_vfsmnt(old->mnt_devname); if (mnt) { |
719f5d7f0
|
527 528 529 530 531 532 533 534 535 536 |
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; } |
1da177e4c
|
537 538 539 540 541 542 |
mnt->mnt_flags = old->mnt_flags; 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
|
543 |
|
5afe00221
|
544 545 546 547 |
if (flag & CL_SLAVE) { list_add(&mnt->mnt_slave, &old->mnt_slave_list); mnt->mnt_master = old; CLEAR_MNT_SHARED(mnt); |
8aec08094
|
548 |
} else if (!(flag & CL_PRIVATE)) { |
5afe00221
|
549 550 551 552 553 554 |
if ((flag & CL_PROPAGATION) || IS_MNT_SHARED(old)) 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
|
555 556 |
if (flag & CL_MAKE_SHARED) set_mnt_shared(mnt); |
1da177e4c
|
557 558 559 |
/* stick the duplicate mount on the same expiry list * as the original if that was on one */ |
36341f645
|
560 |
if (flag & CL_EXPIRE) { |
36341f645
|
561 562 |
if (!list_empty(&old->mnt_expire)) list_add(&mnt->mnt_expire, &old->mnt_expire); |
36341f645
|
563 |
} |
1da177e4c
|
564 565 |
} return mnt; |
719f5d7f0
|
566 567 568 569 |
out_free: free_vfsmnt(mnt); return NULL; |
1da177e4c
|
570 |
} |
7b7b1ace2
|
571 |
static inline void __mntput(struct vfsmount *mnt) |
1da177e4c
|
572 573 |
{ struct super_block *sb = mnt->mnt_sb; |
3d733633a
|
574 |
/* |
3d733633a
|
575 576 577 578 579 |
* 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
|
580 581 582 583 584 |
/* * atomic_dec_and_lock() used to deal with ->mnt_count decrements * provides barriers, so count_mnt_writers() below is safe. AV */ WARN_ON(count_mnt_writers(mnt)); |
1da177e4c
|
585 586 587 588 |
dput(mnt->mnt_root); free_vfsmnt(mnt); deactivate_super(sb); } |
7b7b1ace2
|
589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 |
void mntput_no_expire(struct vfsmount *mnt) { repeat: if (atomic_dec_and_lock(&mnt->mnt_count, &vfsmount_lock)) { if (likely(!mnt->mnt_pinned)) { spin_unlock(&vfsmount_lock); __mntput(mnt); return; } atomic_add(mnt->mnt_pinned + 1, &mnt->mnt_count); mnt->mnt_pinned = 0; spin_unlock(&vfsmount_lock); acct_auto_close_mnt(mnt); security_sb_umount_close(mnt); goto repeat; } } EXPORT_SYMBOL(mntput_no_expire); void mnt_pin(struct vfsmount *mnt) { spin_lock(&vfsmount_lock); mnt->mnt_pinned++; spin_unlock(&vfsmount_lock); } EXPORT_SYMBOL(mnt_pin); void mnt_unpin(struct vfsmount *mnt) { spin_lock(&vfsmount_lock); if (mnt->mnt_pinned) { atomic_inc(&mnt->mnt_count); mnt->mnt_pinned--; } spin_unlock(&vfsmount_lock); } EXPORT_SYMBOL(mnt_unpin); |
1da177e4c
|
629 |
|
b3b304a23
|
630 631 632 633 634 635 636 637 638 639 640 641 642 643 |
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
|
644 645 646 647 |
const char *options; rcu_read_lock(); options = rcu_dereference(mnt->mnt_sb->s_options); |
b3b304a23
|
648 649 650 651 652 |
if (options != NULL && options[0]) { seq_putc(m, ','); mangle(m, options); } |
2a32cebd6
|
653 |
rcu_read_unlock(); |
b3b304a23
|
654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 |
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
|
674 675 |
BUG_ON(sb->s_options); rcu_assign_pointer(sb->s_options, kstrdup(options, GFP_KERNEL)); |
b3b304a23
|
676 677 |
} EXPORT_SYMBOL(save_mount_options); |
2a32cebd6
|
678 679 680 681 682 683 684 685 686 687 |
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
|
688 |
#ifdef CONFIG_PROC_FS |
1da177e4c
|
689 690 691 |
/* iterator */ static void *m_start(struct seq_file *m, loff_t *pos) { |
a1a2c409b
|
692 |
struct proc_mounts *p = m->private; |
1da177e4c
|
693 |
|
390c68436
|
694 |
down_read(&namespace_sem); |
a1a2c409b
|
695 |
return seq_list_start(&p->ns->list, *pos); |
1da177e4c
|
696 697 698 699 |
} static void *m_next(struct seq_file *m, void *v, loff_t *pos) { |
a1a2c409b
|
700 |
struct proc_mounts *p = m->private; |
b0765fb85
|
701 |
|
a1a2c409b
|
702 |
return seq_list_next(v, &p->ns->list, pos); |
1da177e4c
|
703 704 705 706 |
} static void m_stop(struct seq_file *m, void *v) { |
390c68436
|
707 |
up_read(&namespace_sem); |
1da177e4c
|
708 |
} |
2d4d4864a
|
709 710 711 712 |
struct proc_fs_info { int flag; const char *str; }; |
2069f4578
|
713 |
static int show_sb_opts(struct seq_file *m, struct super_block *sb) |
1da177e4c
|
714 |
{ |
2d4d4864a
|
715 |
static const struct proc_fs_info fs_info[] = { |
1da177e4c
|
716 717 718 |
{ MS_SYNCHRONOUS, ",sync" }, { MS_DIRSYNC, ",dirsync" }, { MS_MANDLOCK, ",mand" }, |
1da177e4c
|
719 720 |
{ 0, NULL } }; |
2d4d4864a
|
721 722 723 724 725 726 |
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
|
727 728 |
return security_sb_show_options(m, sb); |
2d4d4864a
|
729 730 731 732 733 |
} static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt) { static const struct proc_fs_info mnt_info[] = { |
1da177e4c
|
734 735 736 |
{ MNT_NOSUID, ",nosuid" }, { MNT_NODEV, ",nodev" }, { MNT_NOEXEC, ",noexec" }, |
fc33a7bb9
|
737 738 |
{ MNT_NOATIME, ",noatime" }, { MNT_NODIRATIME, ",nodiratime" }, |
47ae32d6a
|
739 |
{ MNT_RELATIME, ",relatime" }, |
d0adde574
|
740 |
{ MNT_STRICTATIME, ",strictatime" }, |
1da177e4c
|
741 742 |
{ 0, NULL } }; |
2d4d4864a
|
743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 |
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
|
764 |
struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; |
1da177e4c
|
765 766 767 |
mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); seq_putc(m, ' '); |
c32c2f63a
|
768 769 |
seq_path(m, &mnt_path, " \t \\"); |
1da177e4c
|
770 |
seq_putc(m, ' '); |
2d4d4864a
|
771 |
show_type(m, mnt->mnt_sb); |
2e4b7fcd9
|
772 |
seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw"); |
2069f4578
|
773 774 775 |
err = show_sb_opts(m, mnt->mnt_sb); if (err) goto out; |
2d4d4864a
|
776 |
show_mnt_opts(m, mnt); |
1da177e4c
|
777 778 779 780 |
if (mnt->mnt_sb->s_op->show_options) err = mnt->mnt_sb->s_op->show_options(m, mnt); seq_puts(m, " 0 0 "); |
2069f4578
|
781 |
out: |
1da177e4c
|
782 783 |
return err; } |
a1a2c409b
|
784 |
const struct seq_operations mounts_op = { |
1da177e4c
|
785 786 787 788 789 |
.start = m_start, .next = m_next, .stop = m_stop, .show = show_vfsmnt }; |
2d4d4864a
|
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 |
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)); seq_dentry(m, mnt->mnt_root, " \t \\"); 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
|
820 821 822 823 824 825 826 |
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
|
827 828 829 830 831 832 833 834 835 |
if (IS_MNT_UNBINDABLE(mnt)) seq_puts(m, " unbindable"); /* Filesystem specific data */ seq_puts(m, " - "); show_type(m, sb); seq_putc(m, ' '); mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw"); |
2069f4578
|
836 837 838 |
err = show_sb_opts(m, sb); if (err) goto out; |
2d4d4864a
|
839 840 841 842 |
if (sb->s_op->show_options) err = sb->s_op->show_options(m, mnt); seq_putc(m, ' '); |
2069f4578
|
843 |
out: |
2d4d4864a
|
844 845 846 847 848 849 850 851 852 |
return err; } const struct seq_operations mountinfo_op = { .start = m_start, .next = m_next, .stop = m_stop, .show = show_mountinfo, }; |
b4629fe2f
|
853 854 |
static int show_vfsstat(struct seq_file *m, void *v) { |
b0765fb85
|
855 |
struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); |
c32c2f63a
|
856 |
struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; |
b4629fe2f
|
857 858 859 860 861 862 863 864 865 866 867 |
int err = 0; /* device */ if (mnt->mnt_devname) { seq_puts(m, "device "); mangle(m, mnt->mnt_devname); } else seq_puts(m, "no device"); /* mount point */ seq_puts(m, " mounted on "); |
c32c2f63a
|
868 869 |
seq_path(m, &mnt_path, " \t \\"); |
b4629fe2f
|
870 871 872 873 |
seq_putc(m, ' '); /* file system type */ seq_puts(m, "with fstype "); |
2d4d4864a
|
874 |
show_type(m, mnt->mnt_sb); |
b4629fe2f
|
875 876 877 878 879 880 881 882 883 884 885 |
/* optional statistics */ if (mnt->mnt_sb->s_op->show_stats) { seq_putc(m, ' '); err = mnt->mnt_sb->s_op->show_stats(m, mnt); } seq_putc(m, ' '); return err; } |
a1a2c409b
|
886 |
const struct seq_operations mountstats_op = { |
b4629fe2f
|
887 888 889 890 891 |
.start = m_start, .next = m_next, .stop = m_stop, .show = show_vfsstat, }; |
a1a2c409b
|
892 |
#endif /* CONFIG_PROC_FS */ |
b4629fe2f
|
893 |
|
1da177e4c
|
894 895 896 897 898 899 900 901 902 903 |
/** * 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
|
904 905 906 |
int actual_refs = 0; int minimum_refs = 0; struct vfsmount *p; |
1da177e4c
|
907 908 |
spin_lock(&vfsmount_lock); |
36341f645
|
909 |
for (p = mnt; p; p = next_mnt(p, mnt)) { |
1da177e4c
|
910 911 |
actual_refs += atomic_read(&p->mnt_count); minimum_refs += 2; |
1da177e4c
|
912 913 914 915 |
} spin_unlock(&vfsmount_lock); if (actual_refs > minimum_refs) |
e3474a8eb
|
916 |
return 0; |
1da177e4c
|
917 |
|
e3474a8eb
|
918 |
return 1; |
1da177e4c
|
919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 |
} 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
|
938 |
int ret = 1; |
a05964f39
|
939 940 |
spin_lock(&vfsmount_lock); if (propagate_mount_busy(mnt, 2)) |
e3474a8eb
|
941 |
ret = 0; |
a05964f39
|
942 943 |
spin_unlock(&vfsmount_lock); return ret; |
1da177e4c
|
944 945 946 |
} EXPORT_SYMBOL(may_umount); |
b90fa9ae8
|
947 |
void release_mounts(struct list_head *head) |
70fbcdf4d
|
948 949 |
{ struct vfsmount *mnt; |
bf066c7db
|
950 |
while (!list_empty(head)) { |
b5e618181
|
951 |
mnt = list_first_entry(head, struct vfsmount, mnt_hash); |
70fbcdf4d
|
952 953 954 955 956 957 958 959 960 |
list_del_init(&mnt->mnt_hash); if (mnt->mnt_parent != mnt) { struct dentry *dentry; struct vfsmount *m; spin_lock(&vfsmount_lock); dentry = mnt->mnt_mountpoint; m = mnt->mnt_parent; mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; |
7c4b93d82
|
961 |
m->mnt_ghosts--; |
70fbcdf4d
|
962 963 964 965 966 967 968 |
spin_unlock(&vfsmount_lock); dput(dentry); mntput(m); } mntput(mnt); } } |
a05964f39
|
969 |
void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) |
1da177e4c
|
970 971 |
{ struct vfsmount *p; |
1da177e4c
|
972 |
|
1bfba4e8e
|
973 974 |
for (p = mnt; p; p = next_mnt(p, mnt)) list_move(&p->mnt_hash, kill); |
1da177e4c
|
975 |
|
a05964f39
|
976 977 |
if (propagate) propagate_umount(kill); |
70fbcdf4d
|
978 979 980 |
list_for_each_entry(p, kill, mnt_hash) { list_del_init(&p->mnt_expire); list_del_init(&p->mnt_list); |
6b3286ed1
|
981 982 |
__touch_mnt_namespace(p->mnt_ns); p->mnt_ns = NULL; |
70fbcdf4d
|
983 |
list_del_init(&p->mnt_child); |
7c4b93d82
|
984 985 |
if (p->mnt_parent != p) { p->mnt_parent->mnt_ghosts++; |
f30ac319f
|
986 |
p->mnt_mountpoint->d_mounted--; |
7c4b93d82
|
987 |
} |
a05964f39
|
988 |
change_mnt_propagation(p, MS_PRIVATE); |
1da177e4c
|
989 990 |
} } |
c35038bec
|
991 |
static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts); |
1da177e4c
|
992 993 |
static int do_umount(struct vfsmount *mnt, int flags) { |
b58fed8b1
|
994 |
struct super_block *sb = mnt->mnt_sb; |
1da177e4c
|
995 |
int retval; |
70fbcdf4d
|
996 |
LIST_HEAD(umount_list); |
1da177e4c
|
997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 |
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
|
1009 |
if (mnt == current->fs->root.mnt || |
1da177e4c
|
1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 |
flags & (MNT_FORCE | MNT_DETACH)) return -EINVAL; if (atomic_read(&mnt->mnt_count) != 2) return -EBUSY; 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
|
1029 |
if (flags & MNT_FORCE && sb->s_op->umount_begin) { |
42faad996
|
1030 |
sb->s_op->umount_begin(sb); |
42faad996
|
1031 |
} |
1da177e4c
|
1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 |
/* * 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
|
1042 |
if (mnt == current->fs->root.mnt && !(flags & MNT_DETACH)) { |
1da177e4c
|
1043 1044 1045 1046 1047 |
/* * Special case for "unmounting" root ... * we just try to remount it readonly. */ down_write(&sb->s_umount); |
4aa98cf76
|
1048 |
if (!(sb->s_flags & MS_RDONLY)) |
1da177e4c
|
1049 |
retval = do_remount_sb(sb, MS_RDONLY, NULL, 0); |
1da177e4c
|
1050 1051 1052 |
up_write(&sb->s_umount); return retval; } |
390c68436
|
1053 |
down_write(&namespace_sem); |
1da177e4c
|
1054 |
spin_lock(&vfsmount_lock); |
5addc5dd8
|
1055 |
event++; |
1da177e4c
|
1056 |
|
c35038bec
|
1057 1058 |
if (!(flags & MNT_DETACH)) shrink_submounts(mnt, &umount_list); |
1da177e4c
|
1059 |
retval = -EBUSY; |
a05964f39
|
1060 |
if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) { |
1da177e4c
|
1061 |
if (!list_empty(&mnt->mnt_list)) |
a05964f39
|
1062 |
umount_tree(mnt, 1, &umount_list); |
1da177e4c
|
1063 1064 1065 1066 1067 |
retval = 0; } spin_unlock(&vfsmount_lock); if (retval) security_sb_umount_busy(mnt); |
390c68436
|
1068 |
up_write(&namespace_sem); |
70fbcdf4d
|
1069 |
release_mounts(&umount_list); |
1da177e4c
|
1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 |
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
|
1080 |
SYSCALL_DEFINE2(umount, char __user *, name, int, flags) |
1da177e4c
|
1081 |
{ |
2d8f30380
|
1082 |
struct path path; |
1da177e4c
|
1083 |
int retval; |
2d8f30380
|
1084 |
retval = user_path(name, &path); |
1da177e4c
|
1085 1086 1087 |
if (retval) goto out; retval = -EINVAL; |
2d8f30380
|
1088 |
if (path.dentry != path.mnt->mnt_root) |
1da177e4c
|
1089 |
goto dput_and_out; |
2d8f30380
|
1090 |
if (!check_mnt(path.mnt)) |
1da177e4c
|
1091 1092 1093 1094 1095 |
goto dput_and_out; retval = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto dput_and_out; |
2d8f30380
|
1096 |
retval = do_umount(path.mnt, flags); |
1da177e4c
|
1097 |
dput_and_out: |
429731b15
|
1098 |
/* we mustn't call path_put() as that would clear mnt_expiry_mark */ |
2d8f30380
|
1099 1100 |
dput(path.dentry); mntput_no_expire(path.mnt); |
1da177e4c
|
1101 1102 1103 1104 1105 1106 1107 |
out: return retval; } #ifdef __ARCH_WANT_SYS_OLDUMOUNT /* |
b58fed8b1
|
1108 |
* The 2.0 compatible umount. No flags. |
1da177e4c
|
1109 |
*/ |
bdc480e3b
|
1110 |
SYSCALL_DEFINE1(oldumount, char __user *, name) |
1da177e4c
|
1111 |
{ |
b58fed8b1
|
1112 |
return sys_umount(name, 0); |
1da177e4c
|
1113 1114 1115 |
} #endif |
2d92ab3c6
|
1116 |
static int mount_is_safe(struct path *path) |
1da177e4c
|
1117 1118 1119 1120 1121 |
{ if (capable(CAP_SYS_ADMIN)) return 0; return -EPERM; #ifdef notyet |
2d92ab3c6
|
1122 |
if (S_ISLNK(path->dentry->d_inode->i_mode)) |
1da177e4c
|
1123 |
return -EPERM; |
2d92ab3c6
|
1124 |
if (path->dentry->d_inode->i_mode & S_ISVTX) { |
da9592ede
|
1125 |
if (current_uid() != path->dentry->d_inode->i_uid) |
1da177e4c
|
1126 1127 |
return -EPERM; } |
2d92ab3c6
|
1128 |
if (inode_permission(path->dentry->d_inode, MAY_WRITE)) |
1da177e4c
|
1129 1130 1131 1132 |
return -EPERM; return 0; #endif } |
b90fa9ae8
|
1133 |
struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, |
36341f645
|
1134 |
int flag) |
1da177e4c
|
1135 1136 |
{ struct vfsmount *res, *p, *q, *r, *s; |
1a3906895
|
1137 |
struct path path; |
1da177e4c
|
1138 |
|
9676f0c63
|
1139 1140 |
if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) return NULL; |
36341f645
|
1141 |
res = q = clone_mnt(mnt, dentry, flag); |
1da177e4c
|
1142 1143 1144 1145 1146 |
if (!q) goto Enomem; q->mnt_mountpoint = mnt->mnt_mountpoint; p = mnt; |
fdadd65fb
|
1147 |
list_for_each_entry(r, &mnt->mnt_mounts, mnt_child) { |
7ec02ef15
|
1148 |
if (!is_subdir(r->mnt_mountpoint, dentry)) |
1da177e4c
|
1149 1150 1151 |
continue; for (s = r; s; s = next_mnt(s, r)) { |
9676f0c63
|
1152 1153 1154 1155 |
if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(s)) { s = skip_mnt_tree(s); continue; } |
1da177e4c
|
1156 1157 1158 1159 1160 |
while (p != s->mnt_parent) { p = p->mnt_parent; q = q->mnt_parent; } p = s; |
1a3906895
|
1161 1162 |
path.mnt = q; path.dentry = p->mnt_mountpoint; |
36341f645
|
1163 |
q = clone_mnt(p, p->mnt_root, flag); |
1da177e4c
|
1164 1165 1166 1167 |
if (!q) goto Enomem; spin_lock(&vfsmount_lock); list_add_tail(&q->mnt_list, &res->mnt_list); |
1a3906895
|
1168 |
attach_mnt(q, &path); |
1da177e4c
|
1169 1170 1171 1172 |
spin_unlock(&vfsmount_lock); } } return res; |
b58fed8b1
|
1173 |
Enomem: |
1da177e4c
|
1174 |
if (res) { |
70fbcdf4d
|
1175 |
LIST_HEAD(umount_list); |
1da177e4c
|
1176 |
spin_lock(&vfsmount_lock); |
a05964f39
|
1177 |
umount_tree(res, 0, &umount_list); |
1da177e4c
|
1178 |
spin_unlock(&vfsmount_lock); |
70fbcdf4d
|
1179 |
release_mounts(&umount_list); |
1da177e4c
|
1180 1181 1182 |
} return NULL; } |
589ff870e
|
1183 |
struct vfsmount *collect_mounts(struct path *path) |
8aec08094
|
1184 1185 |
{ struct vfsmount *tree; |
1a60a2807
|
1186 |
down_write(&namespace_sem); |
589ff870e
|
1187 |
tree = copy_tree(path->mnt, path->dentry, CL_COPY_ALL | CL_PRIVATE); |
1a60a2807
|
1188 |
up_write(&namespace_sem); |
8aec08094
|
1189 1190 1191 1192 1193 1194 |
return tree; } void drop_collected_mounts(struct vfsmount *mnt) { LIST_HEAD(umount_list); |
1a60a2807
|
1195 |
down_write(&namespace_sem); |
8aec08094
|
1196 1197 1198 |
spin_lock(&vfsmount_lock); umount_tree(mnt, 0, &umount_list); spin_unlock(&vfsmount_lock); |
1a60a2807
|
1199 |
up_write(&namespace_sem); |
8aec08094
|
1200 1201 |
release_mounts(&umount_list); } |
719f5d7f0
|
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 |
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
|
1228 1229 |
/* * @source_mnt : mount tree to be attached |
214444032
|
1230 1231 1232 1233 |
* @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
|
1234 1235 1236 |
* * 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
|
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 |
* --------------------------------------------------------------------------- * | BIND MOUNT OPERATION | * |************************************************************************** * | source-->| shared | private | slave | unbindable | * | dest | | | | | * | | | | | | | * | v | | | | | * |************************************************************************** * | shared | shared (++) | shared (+) | shared(+++)| invalid | * | | | | | | * |non-shared| shared (+) | private | slave (*) | invalid | * *************************************************************************** |
b90fa9ae8
|
1249 1250 1251 1252 1253 1254 1255 1256 1257 |
* 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
|
1258 1259 1260 1261 1262 1263 1264 |
* (+++) 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
|
1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 |
* --------------------------------------------------------------------------- * | MOVE MOUNT OPERATION | * |************************************************************************** * | source-->| shared | private | slave | unbindable | * | dest | | | | | * | | | | | | | * | v | | | | | * |************************************************************************** * | shared | shared (+) | shared (+) | shared(+++) | invalid | * | | | | | | * |non-shared| shared (+*) | private | slave (*) | unbindable | * *************************************************************************** |
5afe00221
|
1277 1278 1279 |
* * (+) the mount is moved to the destination. And is then propagated to * all the mounts in the propagation tree of the destination mount. |
214444032
|
1280 |
* (+*) the mount is moved to the destination. |
5afe00221
|
1281 1282 1283 1284 |
* (+++) 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
|
1285 1286 1287 1288 1289 1290 1291 |
* * 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
|
1292 |
struct path *path, struct path *parent_path) |
b90fa9ae8
|
1293 1294 |
{ LIST_HEAD(tree_list); |
1a3906895
|
1295 1296 |
struct vfsmount *dest_mnt = path->mnt; struct dentry *dest_dentry = path->dentry; |
b90fa9ae8
|
1297 |
struct vfsmount *child, *p; |
719f5d7f0
|
1298 |
int err; |
b90fa9ae8
|
1299 |
|
719f5d7f0
|
1300 1301 1302 1303 1304 1305 1306 1307 |
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
|
1308 1309 1310 1311 1312 1313 1314 |
if (IS_MNT_SHARED(dest_mnt)) { for (p = source_mnt; p; p = next_mnt(p, source_mnt)) set_mnt_shared(p); } spin_lock(&vfsmount_lock); |
1a3906895
|
1315 1316 1317 |
if (parent_path) { detach_mnt(source_mnt, parent_path); attach_mnt(source_mnt, path); |
e5d67f071
|
1318 |
touch_mnt_namespace(parent_path->mnt->mnt_ns); |
214444032
|
1319 1320 1321 1322 |
} else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); commit_tree(source_mnt); } |
b90fa9ae8
|
1323 1324 1325 1326 1327 1328 1329 |
list_for_each_entry_safe(child, p, &tree_list, mnt_hash) { list_del_init(&child->mnt_hash); commit_tree(child); } spin_unlock(&vfsmount_lock); return 0; |
719f5d7f0
|
1330 1331 1332 1333 1334 1335 |
out_cleanup_ids: if (IS_MNT_SHARED(dest_mnt)) cleanup_group_ids(source_mnt, NULL); out: return err; |
b90fa9ae8
|
1336 |
} |
8c3ee42e8
|
1337 |
static int graft_tree(struct vfsmount *mnt, struct path *path) |
1da177e4c
|
1338 1339 1340 1341 |
{ int err; if (mnt->mnt_sb->s_flags & MS_NOUSER) return -EINVAL; |
8c3ee42e8
|
1342 |
if (S_ISDIR(path->dentry->d_inode->i_mode) != |
1da177e4c
|
1343 1344 1345 1346 |
S_ISDIR(mnt->mnt_root->d_inode->i_mode)) return -ENOTDIR; err = -ENOENT; |
8c3ee42e8
|
1347 1348 |
mutex_lock(&path->dentry->d_inode->i_mutex); if (IS_DEADDIR(path->dentry->d_inode)) |
1da177e4c
|
1349 |
goto out_unlock; |
8c3ee42e8
|
1350 |
err = security_sb_check_sb(mnt, path); |
1da177e4c
|
1351 1352 1353 1354 |
if (err) goto out_unlock; err = -ENOENT; |
f3da392e9
|
1355 |
if (!d_unlinked(path->dentry)) |
8c3ee42e8
|
1356 |
err = attach_recursive_mnt(mnt, path, NULL); |
1da177e4c
|
1357 |
out_unlock: |
8c3ee42e8
|
1358 |
mutex_unlock(&path->dentry->d_inode->i_mutex); |
1da177e4c
|
1359 |
if (!err) |
8c3ee42e8
|
1360 |
security_sb_post_addmount(mnt, path); |
1da177e4c
|
1361 1362 1363 1364 |
return err; } /* |
07b20889e
|
1365 1366 |
* recursively change the type of the mountpoint. */ |
0a0d8a467
|
1367 |
static int do_change_type(struct path *path, int flag) |
07b20889e
|
1368 |
{ |
2d92ab3c6
|
1369 |
struct vfsmount *m, *mnt = path->mnt; |
07b20889e
|
1370 1371 |
int recurse = flag & MS_REC; int type = flag & ~MS_REC; |
719f5d7f0
|
1372 |
int err = 0; |
07b20889e
|
1373 |
|
ee6f95829
|
1374 1375 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
2d92ab3c6
|
1376 |
if (path->dentry != path->mnt->mnt_root) |
07b20889e
|
1377 1378 1379 |
return -EINVAL; down_write(&namespace_sem); |
719f5d7f0
|
1380 1381 1382 1383 1384 |
if (type == MS_SHARED) { err = invent_group_ids(mnt, recurse); if (err) goto out_unlock; } |
07b20889e
|
1385 1386 1387 1388 |
spin_lock(&vfsmount_lock); for (m = mnt; m; m = (recurse ? next_mnt(m, mnt) : NULL)) change_mnt_propagation(m, type); spin_unlock(&vfsmount_lock); |
719f5d7f0
|
1389 1390 |
out_unlock: |
07b20889e
|
1391 |
up_write(&namespace_sem); |
719f5d7f0
|
1392 |
return err; |
07b20889e
|
1393 1394 1395 |
} /* |
1da177e4c
|
1396 1397 |
* do loopback mount. */ |
0a0d8a467
|
1398 |
static int do_loopback(struct path *path, char *old_name, |
2dafe1c4d
|
1399 |
int recurse) |
1da177e4c
|
1400 |
{ |
2d92ab3c6
|
1401 |
struct path old_path; |
1da177e4c
|
1402 |
struct vfsmount *mnt = NULL; |
2d92ab3c6
|
1403 |
int err = mount_is_safe(path); |
1da177e4c
|
1404 1405 1406 1407 |
if (err) return err; if (!old_name || !*old_name) return -EINVAL; |
2d92ab3c6
|
1408 |
err = kern_path(old_name, LOOKUP_FOLLOW, &old_path); |
1da177e4c
|
1409 1410 |
if (err) return err; |
390c68436
|
1411 |
down_write(&namespace_sem); |
1da177e4c
|
1412 |
err = -EINVAL; |
2d92ab3c6
|
1413 |
if (IS_MNT_UNBINDABLE(old_path.mnt)) |
4ac913785
|
1414 |
goto out; |
9676f0c63
|
1415 |
|
2d92ab3c6
|
1416 |
if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) |
ccd48bc7f
|
1417 |
goto out; |
1da177e4c
|
1418 |
|
ccd48bc7f
|
1419 1420 |
err = -ENOMEM; if (recurse) |
2d92ab3c6
|
1421 |
mnt = copy_tree(old_path.mnt, old_path.dentry, 0); |
ccd48bc7f
|
1422 |
else |
2d92ab3c6
|
1423 |
mnt = clone_mnt(old_path.mnt, old_path.dentry, 0); |
ccd48bc7f
|
1424 1425 1426 |
if (!mnt) goto out; |
2d92ab3c6
|
1427 |
err = graft_tree(mnt, path); |
ccd48bc7f
|
1428 |
if (err) { |
70fbcdf4d
|
1429 |
LIST_HEAD(umount_list); |
1da177e4c
|
1430 |
spin_lock(&vfsmount_lock); |
a05964f39
|
1431 |
umount_tree(mnt, 0, &umount_list); |
1da177e4c
|
1432 |
spin_unlock(&vfsmount_lock); |
70fbcdf4d
|
1433 |
release_mounts(&umount_list); |
5b83d2c5c
|
1434 |
} |
1da177e4c
|
1435 |
|
ccd48bc7f
|
1436 |
out: |
390c68436
|
1437 |
up_write(&namespace_sem); |
2d92ab3c6
|
1438 |
path_put(&old_path); |
1da177e4c
|
1439 1440 |
return err; } |
2e4b7fcd9
|
1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 |
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
|
1457 1458 1459 1460 1461 |
/* * 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
|
1462 |
static int do_remount(struct path *path, int flags, int mnt_flags, |
1da177e4c
|
1463 1464 1465 |
void *data) { int err; |
2d92ab3c6
|
1466 |
struct super_block *sb = path->mnt->mnt_sb; |
1da177e4c
|
1467 1468 1469 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
2d92ab3c6
|
1470 |
if (!check_mnt(path->mnt)) |
1da177e4c
|
1471 |
return -EINVAL; |
2d92ab3c6
|
1472 |
if (path->dentry != path->mnt->mnt_root) |
1da177e4c
|
1473 1474 1475 |
return -EINVAL; down_write(&sb->s_umount); |
2e4b7fcd9
|
1476 |
if (flags & MS_BIND) |
2d92ab3c6
|
1477 |
err = change_mount_flags(path->mnt, flags); |
4aa98cf76
|
1478 |
else |
2e4b7fcd9
|
1479 |
err = do_remount_sb(sb, flags, data, 0); |
1da177e4c
|
1480 |
if (!err) |
2d92ab3c6
|
1481 |
path->mnt->mnt_flags = mnt_flags; |
1da177e4c
|
1482 |
up_write(&sb->s_umount); |
0e55a7cca
|
1483 |
if (!err) { |
2d92ab3c6
|
1484 |
security_sb_post_remount(path->mnt, flags, data); |
0e55a7cca
|
1485 1486 1487 1488 1489 |
spin_lock(&vfsmount_lock); touch_mnt_namespace(path->mnt->mnt_ns); spin_unlock(&vfsmount_lock); } |
1da177e4c
|
1490 1491 |
return err; } |
9676f0c63
|
1492 1493 1494 1495 1496 1497 1498 1499 1500 |
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
|
1501 |
static int do_move_mount(struct path *path, char *old_name) |
1da177e4c
|
1502 |
{ |
2d92ab3c6
|
1503 |
struct path old_path, parent_path; |
1da177e4c
|
1504 1505 1506 1507 1508 1509 |
struct vfsmount *p; int err = 0; if (!capable(CAP_SYS_ADMIN)) return -EPERM; if (!old_name || !*old_name) return -EINVAL; |
2d92ab3c6
|
1510 |
err = kern_path(old_name, LOOKUP_FOLLOW, &old_path); |
1da177e4c
|
1511 1512 |
if (err) return err; |
390c68436
|
1513 |
down_write(&namespace_sem); |
2d92ab3c6
|
1514 |
while (d_mountpoint(path->dentry) && |
9393bd07c
|
1515 |
follow_down(path)) |
1da177e4c
|
1516 1517 |
; err = -EINVAL; |
2d92ab3c6
|
1518 |
if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) |
1da177e4c
|
1519 1520 1521 |
goto out; err = -ENOENT; |
2d92ab3c6
|
1522 1523 |
mutex_lock(&path->dentry->d_inode->i_mutex); if (IS_DEADDIR(path->dentry->d_inode)) |
1da177e4c
|
1524 |
goto out1; |
f3da392e9
|
1525 |
if (d_unlinked(path->dentry)) |
214444032
|
1526 |
goto out1; |
1da177e4c
|
1527 1528 |
err = -EINVAL; |
2d92ab3c6
|
1529 |
if (old_path.dentry != old_path.mnt->mnt_root) |
214444032
|
1530 |
goto out1; |
1da177e4c
|
1531 |
|
2d92ab3c6
|
1532 |
if (old_path.mnt == old_path.mnt->mnt_parent) |
214444032
|
1533 |
goto out1; |
1da177e4c
|
1534 |
|
2d92ab3c6
|
1535 1536 |
if (S_ISDIR(path->dentry->d_inode->i_mode) != S_ISDIR(old_path.dentry->d_inode->i_mode)) |
214444032
|
1537 1538 1539 1540 |
goto out1; /* * Don't move a mount residing in a shared parent. */ |
2d92ab3c6
|
1541 1542 |
if (old_path.mnt->mnt_parent && IS_MNT_SHARED(old_path.mnt->mnt_parent)) |
214444032
|
1543 |
goto out1; |
9676f0c63
|
1544 1545 1546 1547 |
/* * Don't move a mount tree containing unbindable mounts to a destination * mount which is shared. */ |
2d92ab3c6
|
1548 1549 |
if (IS_MNT_SHARED(path->mnt) && tree_contains_unbindable(old_path.mnt)) |
9676f0c63
|
1550 |
goto out1; |
1da177e4c
|
1551 |
err = -ELOOP; |
2d92ab3c6
|
1552 1553 |
for (p = path->mnt; p->mnt_parent != p; p = p->mnt_parent) if (p == old_path.mnt) |
214444032
|
1554 |
goto out1; |
1da177e4c
|
1555 |
|
2d92ab3c6
|
1556 |
err = attach_recursive_mnt(old_path.mnt, path, &parent_path); |
4ac913785
|
1557 |
if (err) |
214444032
|
1558 |
goto out1; |
1da177e4c
|
1559 1560 1561 |
/* if the mount is moved, it should no longer be expire * automatically */ |
2d92ab3c6
|
1562 |
list_del_init(&old_path.mnt->mnt_expire); |
1da177e4c
|
1563 |
out1: |
2d92ab3c6
|
1564 |
mutex_unlock(&path->dentry->d_inode->i_mutex); |
1da177e4c
|
1565 |
out: |
390c68436
|
1566 |
up_write(&namespace_sem); |
1da177e4c
|
1567 |
if (!err) |
1a3906895
|
1568 |
path_put(&parent_path); |
2d92ab3c6
|
1569 |
path_put(&old_path); |
1da177e4c
|
1570 1571 1572 1573 1574 1575 1576 |
return err; } /* * create a new mount for userspace and request it to be added into the * namespace's tree */ |
0a0d8a467
|
1577 |
static int do_new_mount(struct path *path, char *type, int flags, |
1da177e4c
|
1578 1579 1580 |
int mnt_flags, char *name, void *data) { struct vfsmount *mnt; |
eca6f534e
|
1581 |
if (!type) |
1da177e4c
|
1582 1583 1584 1585 1586 |
return -EINVAL; /* we need capabilities... */ if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
7f78d4cd4
|
1587 |
lock_kernel(); |
1da177e4c
|
1588 |
mnt = do_kern_mount(type, flags, name, data); |
7f78d4cd4
|
1589 |
unlock_kernel(); |
1da177e4c
|
1590 1591 |
if (IS_ERR(mnt)) return PTR_ERR(mnt); |
2d92ab3c6
|
1592 |
return do_add_mount(mnt, path, mnt_flags, NULL); |
1da177e4c
|
1593 1594 1595 1596 1597 1598 |
} /* * add a mount into a namespace's mount tree * - provide the option of adding the new mount to an expiration list */ |
8d66bf548
|
1599 |
int do_add_mount(struct vfsmount *newmnt, struct path *path, |
1da177e4c
|
1600 1601 1602 |
int mnt_flags, struct list_head *fslist) { int err; |
390c68436
|
1603 |
down_write(&namespace_sem); |
1da177e4c
|
1604 |
/* Something was mounted here while we slept */ |
8d66bf548
|
1605 |
while (d_mountpoint(path->dentry) && |
9393bd07c
|
1606 |
follow_down(path)) |
1da177e4c
|
1607 1608 |
; err = -EINVAL; |
dd5cae6e9
|
1609 |
if (!(mnt_flags & MNT_SHRINKABLE) && !check_mnt(path->mnt)) |
1da177e4c
|
1610 1611 1612 1613 |
goto unlock; /* Refuse the same filesystem on the same mount point */ err = -EBUSY; |
8d66bf548
|
1614 1615 |
if (path->mnt->mnt_sb == newmnt->mnt_sb && path->mnt->mnt_root == path->dentry) |
1da177e4c
|
1616 1617 1618 1619 1620 1621 1622 |
goto unlock; err = -EINVAL; if (S_ISLNK(newmnt->mnt_root->d_inode->i_mode)) goto unlock; newmnt->mnt_flags = mnt_flags; |
8d66bf548
|
1623 |
if ((err = graft_tree(newmnt, path))) |
5b83d2c5c
|
1624 |
goto unlock; |
1da177e4c
|
1625 |
|
6758f953d
|
1626 |
if (fslist) /* add to the specified expiration list */ |
55e700b92
|
1627 |
list_add_tail(&newmnt->mnt_expire, fslist); |
6758f953d
|
1628 |
|
390c68436
|
1629 |
up_write(&namespace_sem); |
5b83d2c5c
|
1630 |
return 0; |
1da177e4c
|
1631 1632 |
unlock: |
390c68436
|
1633 |
up_write(&namespace_sem); |
1da177e4c
|
1634 1635 1636 1637 1638 |
mntput(newmnt); return err; } EXPORT_SYMBOL_GPL(do_add_mount); |
5528f911b
|
1639 |
/* |
1da177e4c
|
1640 1641 1642 1643 1644 1645 |
* 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
|
1646 1647 |
struct vfsmount *mnt, *next; LIST_HEAD(graveyard); |
bcc5c7d2b
|
1648 |
LIST_HEAD(umounts); |
1da177e4c
|
1649 1650 1651 |
if (list_empty(mounts)) return; |
bcc5c7d2b
|
1652 |
down_write(&namespace_sem); |
1da177e4c
|
1653 1654 1655 1656 1657 1658 1659 1660 |
spin_lock(&vfsmount_lock); /* 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
|
1661 |
list_for_each_entry_safe(mnt, next, mounts, mnt_expire) { |
1da177e4c
|
1662 |
if (!xchg(&mnt->mnt_expiry_mark, 1) || |
bcc5c7d2b
|
1663 |
propagate_mount_busy(mnt, 1)) |
1da177e4c
|
1664 |
continue; |
55e700b92
|
1665 |
list_move(&mnt->mnt_expire, &graveyard); |
1da177e4c
|
1666 |
} |
bcc5c7d2b
|
1667 1668 1669 1670 1671 |
while (!list_empty(&graveyard)) { mnt = list_first_entry(&graveyard, struct vfsmount, mnt_expire); touch_mnt_namespace(mnt->mnt_ns); umount_tree(mnt, 1, &umounts); } |
5528f911b
|
1672 |
spin_unlock(&vfsmount_lock); |
bcc5c7d2b
|
1673 1674 1675 |
up_write(&namespace_sem); release_mounts(&umounts); |
5528f911b
|
1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 |
} 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
|
1701 |
continue; |
5528f911b
|
1702 1703 1704 1705 1706 1707 1708 |
/* * Descend a level if the d_mounts list is non-empty. */ if (!list_empty(&mnt->mnt_mounts)) { this_parent = mnt; goto repeat; } |
1da177e4c
|
1709 |
|
5528f911b
|
1710 |
if (!propagate_mount_busy(mnt, 1)) { |
5528f911b
|
1711 1712 1713 |
list_move_tail(&mnt->mnt_expire, graveyard); found++; } |
1da177e4c
|
1714 |
} |
5528f911b
|
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 |
/* * 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 */ |
c35038bec
|
1730 |
static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts) |
5528f911b
|
1731 1732 |
{ LIST_HEAD(graveyard); |
c35038bec
|
1733 |
struct vfsmount *m; |
5528f911b
|
1734 |
|
5528f911b
|
1735 |
/* extract submounts of 'mountpoint' from the expiration list */ |
c35038bec
|
1736 |
while (select_submounts(mnt, &graveyard)) { |
bcc5c7d2b
|
1737 |
while (!list_empty(&graveyard)) { |
c35038bec
|
1738 |
m = list_first_entry(&graveyard, struct vfsmount, |
bcc5c7d2b
|
1739 |
mnt_expire); |
afef80b3d
|
1740 1741 |
touch_mnt_namespace(m->mnt_ns); umount_tree(m, 1, umounts); |
bcc5c7d2b
|
1742 1743 |
} } |
1da177e4c
|
1744 |
} |
1da177e4c
|
1745 1746 1747 1748 1749 1750 |
/* * 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
|
1751 1752 |
static long exact_copy_from_user(void *to, const void __user * from, unsigned long n) |
1da177e4c
|
1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 |
{ 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
|
1772 |
int copy_mount_options(const void __user * data, unsigned long *where) |
1da177e4c
|
1773 1774 1775 1776 |
{ int i; unsigned long page; unsigned long size; |
b58fed8b1
|
1777 |
|
1da177e4c
|
1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 |
*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
|
1796 |
free_page(page); |
1da177e4c
|
1797 1798 1799 1800 1801 1802 1803 |
return -EFAULT; } if (i != PAGE_SIZE) memset((char *)page + i, 0, PAGE_SIZE - i); *where = page; return 0; } |
eca6f534e
|
1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 |
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
|
1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 |
/* * 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
|
1834 |
long do_mount(char *dev_name, char *dir_name, char *type_page, |
1da177e4c
|
1835 1836 |
unsigned long flags, void *data_page) { |
2d92ab3c6
|
1837 |
struct path path; |
1da177e4c
|
1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 |
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
|
1849 1850 1851 |
if (data_page) ((char *)data_page)[PAGE_SIZE - 1] = 0; |
613cbe3d4
|
1852 1853 1854 |
/* Default to relatime unless overriden */ if (!(flags & MS_NOATIME)) mnt_flags |= MNT_RELATIME; |
0a1c01c94
|
1855 |
|
1da177e4c
|
1856 1857 1858 1859 1860 1861 1862 |
/* 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
|
1863 1864 1865 1866 |
if (flags & MS_NOATIME) mnt_flags |= MNT_NOATIME; if (flags & MS_NODIRATIME) mnt_flags |= MNT_NODIRATIME; |
d0adde574
|
1867 1868 |
if (flags & MS_STRICTATIME) mnt_flags &= ~(MNT_RELATIME | MNT_NOATIME); |
2e4b7fcd9
|
1869 1870 |
if (flags & MS_RDONLY) mnt_flags |= MNT_READONLY; |
fc33a7bb9
|
1871 1872 |
flags &= ~(MS_NOSUID | MS_NOEXEC | MS_NODEV | MS_ACTIVE | |
d0adde574
|
1873 1874 |
MS_NOATIME | MS_NODIRATIME | MS_RELATIME| MS_KERNMOUNT | MS_STRICTATIME); |
1da177e4c
|
1875 1876 |
/* ... and get the mountpoint */ |
2d92ab3c6
|
1877 |
retval = kern_path(dir_name, LOOKUP_FOLLOW, &path); |
1da177e4c
|
1878 1879 |
if (retval) return retval; |
2d92ab3c6
|
1880 |
retval = security_sb_mount(dev_name, &path, |
b5266eb4c
|
1881 |
type_page, flags, data_page); |
1da177e4c
|
1882 1883 1884 1885 |
if (retval) goto dput_out; if (flags & MS_REMOUNT) |
2d92ab3c6
|
1886 |
retval = do_remount(&path, flags & ~MS_REMOUNT, mnt_flags, |
1da177e4c
|
1887 1888 |
data_page); else if (flags & MS_BIND) |
2d92ab3c6
|
1889 |
retval = do_loopback(&path, dev_name, flags & MS_REC); |
9676f0c63
|
1890 |
else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) |
2d92ab3c6
|
1891 |
retval = do_change_type(&path, flags); |
1da177e4c
|
1892 |
else if (flags & MS_MOVE) |
2d92ab3c6
|
1893 |
retval = do_move_mount(&path, dev_name); |
1da177e4c
|
1894 |
else |
2d92ab3c6
|
1895 |
retval = do_new_mount(&path, type_page, flags, mnt_flags, |
1da177e4c
|
1896 1897 |
dev_name, data_page); dput_out: |
2d92ab3c6
|
1898 |
path_put(&path); |
1da177e4c
|
1899 1900 |
return retval; } |
cf8d2c11c
|
1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 |
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; } |
741a29513
|
1915 1916 1917 1918 |
/* * Allocate a new namespace structure and populate it with contents * copied from the namespace of the passed in task structure. */ |
e3222c4ec
|
1919 |
static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, |
6b3286ed1
|
1920 |
struct fs_struct *fs) |
1da177e4c
|
1921 |
{ |
6b3286ed1
|
1922 |
struct mnt_namespace *new_ns; |
7f2da1e7d
|
1923 |
struct vfsmount *rootmnt = NULL, *pwdmnt = NULL; |
1da177e4c
|
1924 |
struct vfsmount *p, *q; |
cf8d2c11c
|
1925 1926 1927 |
new_ns = alloc_mnt_ns(); if (IS_ERR(new_ns)) return new_ns; |
1da177e4c
|
1928 |
|
390c68436
|
1929 |
down_write(&namespace_sem); |
1da177e4c
|
1930 |
/* First pass: copy the tree topology */ |
6b3286ed1
|
1931 |
new_ns->root = copy_tree(mnt_ns->root, mnt_ns->root->mnt_root, |
9676f0c63
|
1932 |
CL_COPY_ALL | CL_EXPIRE); |
1da177e4c
|
1933 |
if (!new_ns->root) { |
390c68436
|
1934 |
up_write(&namespace_sem); |
1da177e4c
|
1935 |
kfree(new_ns); |
5cc4a0341
|
1936 |
return ERR_PTR(-ENOMEM); |
1da177e4c
|
1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 |
} spin_lock(&vfsmount_lock); list_add_tail(&new_ns->list, &new_ns->root->mnt_list); spin_unlock(&vfsmount_lock); /* * 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
|
1947 |
p = mnt_ns->root; |
1da177e4c
|
1948 1949 |
q = new_ns->root; while (p) { |
6b3286ed1
|
1950 |
q->mnt_ns = new_ns; |
1da177e4c
|
1951 |
if (fs) { |
6ac08c39a
|
1952 |
if (p == fs->root.mnt) { |
1da177e4c
|
1953 |
rootmnt = p; |
6ac08c39a
|
1954 |
fs->root.mnt = mntget(q); |
1da177e4c
|
1955 |
} |
6ac08c39a
|
1956 |
if (p == fs->pwd.mnt) { |
1da177e4c
|
1957 |
pwdmnt = p; |
6ac08c39a
|
1958 |
fs->pwd.mnt = mntget(q); |
1da177e4c
|
1959 |
} |
1da177e4c
|
1960 |
} |
6b3286ed1
|
1961 |
p = next_mnt(p, mnt_ns->root); |
1da177e4c
|
1962 1963 |
q = next_mnt(q, new_ns->root); } |
390c68436
|
1964 |
up_write(&namespace_sem); |
1da177e4c
|
1965 |
|
1da177e4c
|
1966 1967 1968 1969 |
if (rootmnt) mntput(rootmnt); if (pwdmnt) mntput(pwdmnt); |
1da177e4c
|
1970 |
|
741a29513
|
1971 1972 |
return new_ns; } |
213dd266d
|
1973 |
struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns, |
e3222c4ec
|
1974 |
struct fs_struct *new_fs) |
741a29513
|
1975 |
{ |
6b3286ed1
|
1976 |
struct mnt_namespace *new_ns; |
741a29513
|
1977 |
|
e3222c4ec
|
1978 |
BUG_ON(!ns); |
6b3286ed1
|
1979 |
get_mnt_ns(ns); |
741a29513
|
1980 1981 |
if (!(flags & CLONE_NEWNS)) |
e3222c4ec
|
1982 |
return ns; |
741a29513
|
1983 |
|
e3222c4ec
|
1984 |
new_ns = dup_mnt_ns(ns, new_fs); |
741a29513
|
1985 |
|
6b3286ed1
|
1986 |
put_mnt_ns(ns); |
e3222c4ec
|
1987 |
return new_ns; |
1da177e4c
|
1988 |
} |
cf8d2c11c
|
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 |
/** * create_mnt_ns - creates a private namespace and adds a root filesystem * @mnt: pointer to the new root filesystem mountpoint */ struct mnt_namespace *create_mnt_ns(struct vfsmount *mnt) { struct mnt_namespace *new_ns; new_ns = alloc_mnt_ns(); if (!IS_ERR(new_ns)) { mnt->mnt_ns = new_ns; new_ns->root = mnt; list_add(&new_ns->list, &new_ns->root->mnt_list); } return new_ns; } EXPORT_SYMBOL(create_mnt_ns); |
bdc480e3b
|
2006 2007 |
SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, char __user *, type, unsigned long, flags, void __user *, data) |
1da177e4c
|
2008 |
{ |
eca6f534e
|
2009 2010 2011 2012 |
int ret; char *kernel_type; char *kernel_dir; char *kernel_dev; |
1da177e4c
|
2013 |
unsigned long data_page; |
1da177e4c
|
2014 |
|
eca6f534e
|
2015 2016 2017 |
ret = copy_mount_string(type, &kernel_type); if (ret < 0) goto out_type; |
1da177e4c
|
2018 |
|
eca6f534e
|
2019 2020 2021 2022 2023 |
kernel_dir = getname(dir_name); if (IS_ERR(kernel_dir)) { ret = PTR_ERR(kernel_dir); goto out_dir; } |
1da177e4c
|
2024 |
|
eca6f534e
|
2025 2026 2027 |
ret = copy_mount_string(dev_name, &kernel_dev); if (ret < 0) goto out_dev; |
1da177e4c
|
2028 |
|
eca6f534e
|
2029 2030 2031 |
ret = copy_mount_options(data, &data_page); if (ret < 0) goto out_data; |
1da177e4c
|
2032 |
|
eca6f534e
|
2033 2034 |
ret = do_mount(kernel_dev, kernel_dir, kernel_type, flags, (void *) data_page); |
1da177e4c
|
2035 |
|
eca6f534e
|
2036 2037 2038 2039 2040 2041 2042 2043 2044 |
free_page(data_page); out_data: kfree(kernel_dev); out_dev: putname(kernel_dir); out_dir: kfree(kernel_type); out_type: return ret; |
1da177e4c
|
2045 2046 2047 |
} /* |
1da177e4c
|
2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 |
* 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
|
2060 2061 2062 2063 |
* 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
|
2064 2065 2066 2067 2068 2069 2070 2071 |
* 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
|
2072 2073 |
SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, const char __user *, put_old) |
1da177e4c
|
2074 2075 |
{ struct vfsmount *tmp; |
2d8f30380
|
2076 |
struct path new, old, parent_path, root_parent, root; |
1da177e4c
|
2077 2078 2079 2080 |
int error; if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
2d8f30380
|
2081 |
error = user_path_dir(new_root, &new); |
1da177e4c
|
2082 2083 2084 |
if (error) goto out0; error = -EINVAL; |
2d8f30380
|
2085 |
if (!check_mnt(new.mnt)) |
1da177e4c
|
2086 |
goto out1; |
2d8f30380
|
2087 |
error = user_path_dir(put_old, &old); |
1da177e4c
|
2088 2089 |
if (error) goto out1; |
2d8f30380
|
2090 |
error = security_sb_pivotroot(&old, &new); |
1da177e4c
|
2091 |
if (error) { |
2d8f30380
|
2092 |
path_put(&old); |
1da177e4c
|
2093 2094 2095 2096 |
goto out1; } read_lock(¤t->fs->lock); |
8c3ee42e8
|
2097 |
root = current->fs->root; |
6ac08c39a
|
2098 |
path_get(¤t->fs->root); |
1da177e4c
|
2099 |
read_unlock(¤t->fs->lock); |
390c68436
|
2100 |
down_write(&namespace_sem); |
2d8f30380
|
2101 |
mutex_lock(&old.dentry->d_inode->i_mutex); |
1da177e4c
|
2102 |
error = -EINVAL; |
2d8f30380
|
2103 2104 |
if (IS_MNT_SHARED(old.mnt) || IS_MNT_SHARED(new.mnt->mnt_parent) || |
8c3ee42e8
|
2105 |
IS_MNT_SHARED(root.mnt->mnt_parent)) |
214444032
|
2106 |
goto out2; |
8c3ee42e8
|
2107 |
if (!check_mnt(root.mnt)) |
1da177e4c
|
2108 2109 |
goto out2; error = -ENOENT; |
2d8f30380
|
2110 |
if (IS_DEADDIR(new.dentry->d_inode)) |
1da177e4c
|
2111 |
goto out2; |
f3da392e9
|
2112 |
if (d_unlinked(new.dentry)) |
1da177e4c
|
2113 |
goto out2; |
f3da392e9
|
2114 |
if (d_unlinked(old.dentry)) |
1da177e4c
|
2115 2116 |
goto out2; error = -EBUSY; |
2d8f30380
|
2117 2118 |
if (new.mnt == root.mnt || old.mnt == root.mnt) |
1da177e4c
|
2119 2120 |
goto out2; /* loop, on the same file system */ error = -EINVAL; |
8c3ee42e8
|
2121 |
if (root.mnt->mnt_root != root.dentry) |
1da177e4c
|
2122 |
goto out2; /* not a mountpoint */ |
8c3ee42e8
|
2123 |
if (root.mnt->mnt_parent == root.mnt) |
0bb6fcc13
|
2124 |
goto out2; /* not attached */ |
2d8f30380
|
2125 |
if (new.mnt->mnt_root != new.dentry) |
1da177e4c
|
2126 |
goto out2; /* not a mountpoint */ |
2d8f30380
|
2127 |
if (new.mnt->mnt_parent == new.mnt) |
0bb6fcc13
|
2128 |
goto out2; /* not attached */ |
4ac913785
|
2129 |
/* make sure we can reach put_old from new_root */ |
2d8f30380
|
2130 |
tmp = old.mnt; |
1da177e4c
|
2131 |
spin_lock(&vfsmount_lock); |
2d8f30380
|
2132 |
if (tmp != new.mnt) { |
1da177e4c
|
2133 2134 2135 |
for (;;) { if (tmp->mnt_parent == tmp) goto out3; /* already mounted on put_old */ |
2d8f30380
|
2136 |
if (tmp->mnt_parent == new.mnt) |
1da177e4c
|
2137 2138 2139 |
break; tmp = tmp->mnt_parent; } |
2d8f30380
|
2140 |
if (!is_subdir(tmp->mnt_mountpoint, new.dentry)) |
1da177e4c
|
2141 |
goto out3; |
2d8f30380
|
2142 |
} else if (!is_subdir(old.dentry, new.dentry)) |
1da177e4c
|
2143 |
goto out3; |
2d8f30380
|
2144 |
detach_mnt(new.mnt, &parent_path); |
8c3ee42e8
|
2145 |
detach_mnt(root.mnt, &root_parent); |
4ac913785
|
2146 |
/* mount old root on put_old */ |
2d8f30380
|
2147 |
attach_mnt(root.mnt, &old); |
4ac913785
|
2148 |
/* mount new_root on / */ |
2d8f30380
|
2149 |
attach_mnt(new.mnt, &root_parent); |
6b3286ed1
|
2150 |
touch_mnt_namespace(current->nsproxy->mnt_ns); |
1da177e4c
|
2151 |
spin_unlock(&vfsmount_lock); |
2d8f30380
|
2152 2153 |
chroot_fs_refs(&root, &new); security_sb_post_pivotroot(&root, &new); |
1da177e4c
|
2154 |
error = 0; |
1a3906895
|
2155 2156 |
path_put(&root_parent); path_put(&parent_path); |
1da177e4c
|
2157 |
out2: |
2d8f30380
|
2158 |
mutex_unlock(&old.dentry->d_inode->i_mutex); |
390c68436
|
2159 |
up_write(&namespace_sem); |
8c3ee42e8
|
2160 |
path_put(&root); |
2d8f30380
|
2161 |
path_put(&old); |
1da177e4c
|
2162 |
out1: |
2d8f30380
|
2163 |
path_put(&new); |
1da177e4c
|
2164 |
out0: |
1da177e4c
|
2165 2166 2167 2168 2169 2170 2171 2172 2173 |
return error; out3: spin_unlock(&vfsmount_lock); goto out2; } static void __init init_mount_tree(void) { struct vfsmount *mnt; |
6b3286ed1
|
2174 |
struct mnt_namespace *ns; |
ac748a09f
|
2175 |
struct path root; |
1da177e4c
|
2176 2177 2178 2179 |
mnt = do_kern_mount("rootfs", 0, "rootfs", NULL); if (IS_ERR(mnt)) panic("Can't create rootfs"); |
3b22edc57
|
2180 2181 |
ns = create_mnt_ns(mnt); if (IS_ERR(ns)) |
1da177e4c
|
2182 |
panic("Can't allocate initial namespace"); |
6b3286ed1
|
2183 2184 2185 |
init_task.nsproxy->mnt_ns = ns; get_mnt_ns(ns); |
ac748a09f
|
2186 2187 2188 2189 2190 |
root.mnt = ns->root; root.dentry = ns->root->mnt_root; set_fs_pwd(current->fs, &root); set_fs_root(current->fs, &root); |
1da177e4c
|
2191 |
} |
74bf17cff
|
2192 |
void __init mnt_init(void) |
1da177e4c
|
2193 |
{ |
13f14b4d8
|
2194 |
unsigned u; |
15a67dd8c
|
2195 |
int err; |
1da177e4c
|
2196 |
|
390c68436
|
2197 |
init_rwsem(&namespace_sem); |
1da177e4c
|
2198 |
mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct vfsmount), |
20c2df83d
|
2199 |
0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); |
1da177e4c
|
2200 |
|
b58fed8b1
|
2201 |
mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC); |
1da177e4c
|
2202 2203 2204 2205 |
if (!mount_hashtable) panic("Failed to allocate mount hash table "); |
13f14b4d8
|
2206 2207 2208 2209 2210 |
printk("Mount-cache hash table entries: %lu ", HASH_SIZE); for (u = 0; u < HASH_SIZE; u++) INIT_LIST_HEAD(&mount_hashtable[u]); |
1da177e4c
|
2211 |
|
15a67dd8c
|
2212 2213 2214 2215 |
err = sysfs_init(); if (err) printk(KERN_WARNING "%s: sysfs_init error: %d ", |
8e24eea72
|
2216 |
__func__, err); |
00d266662
|
2217 2218 |
fs_kobj = kobject_create_and_add("fs", NULL); if (!fs_kobj) |
8e24eea72
|
2219 2220 |
printk(KERN_WARNING "%s: kobj create error ", __func__); |
1da177e4c
|
2221 2222 2223 |
init_rootfs(); init_mount_tree(); } |
616511d03
|
2224 |
void put_mnt_ns(struct mnt_namespace *ns) |
1da177e4c
|
2225 |
{ |
616511d03
|
2226 |
struct vfsmount *root; |
70fbcdf4d
|
2227 |
LIST_HEAD(umount_list); |
616511d03
|
2228 2229 2230 2231 |
if (!atomic_dec_and_lock(&ns->count, &vfsmount_lock)) return; root = ns->root; |
6b3286ed1
|
2232 |
ns->root = NULL; |
1ce88cf46
|
2233 |
spin_unlock(&vfsmount_lock); |
390c68436
|
2234 |
down_write(&namespace_sem); |
1da177e4c
|
2235 |
spin_lock(&vfsmount_lock); |
a05964f39
|
2236 |
umount_tree(root, 0, &umount_list); |
1da177e4c
|
2237 |
spin_unlock(&vfsmount_lock); |
390c68436
|
2238 |
up_write(&namespace_sem); |
70fbcdf4d
|
2239 |
release_mounts(&umount_list); |
6b3286ed1
|
2240 |
kfree(ns); |
1da177e4c
|
2241 |
} |
cf8d2c11c
|
2242 |
EXPORT_SYMBOL(put_mnt_ns); |