Blame view
fs/inode.c
49.9 KB
1da177e4c
|
1 |
/* |
1da177e4c
|
2 |
* (C) 1997 Linus Torvalds |
4b4563dc8
|
3 |
* (C) 1999 Andrea Arcangeli <andrea@suse.de> (dynamic inode allocation) |
1da177e4c
|
4 |
*/ |
e59cc473c
|
5 |
#include <linux/export.h> |
1da177e4c
|
6 7 |
#include <linux/fs.h> #include <linux/mm.h> |
1da177e4c
|
8 |
#include <linux/backing-dev.h> |
1da177e4c
|
9 10 11 |
#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> |
1da177e4c
|
12 13 |
#include <linux/cdev.h> #include <linux/bootmem.h> |
3be25f49b
|
14 |
#include <linux/fsnotify.h> |
fc33a7bb9
|
15 |
#include <linux/mount.h> |
f19d4a8fa
|
16 |
#include <linux/posix_acl.h> |
9ce6e0be0
|
17 |
#include <linux/prefetch.h> |
4b4563dc8
|
18 |
#include <linux/buffer_head.h> /* for inode_has_buffers */ |
7ada4db88
|
19 |
#include <linux/ratelimit.h> |
bc3b14cb2
|
20 |
#include <linux/list_lru.h> |
a66979aba
|
21 |
#include "internal.h" |
1da177e4c
|
22 23 |
/* |
4b4563dc8
|
24 |
* Inode locking rules: |
250df6ed2
|
25 26 27 |
* * inode->i_lock protects: * inode->i_state, inode->i_hash, __iget() |
bc3b14cb2
|
28 |
* Inode LRU list locks protect: |
98b745c64
|
29 |
* inode->i_sb->s_inode_lru, inode->i_lru |
55fa6091d
|
30 31 |
* inode_sb_list_lock protects: * sb->s_inodes, inode->i_sb_list |
f758eeabe
|
32 |
* bdi->wb.list_lock protects: |
a66979aba
|
33 |
* bdi->wb.b_{dirty,io,more_io}, inode->i_wb_list |
67a23c494
|
34 35 |
* inode_hash_lock protects: * inode_hashtable, inode->i_hash |
250df6ed2
|
36 37 |
* * Lock ordering: |
55fa6091d
|
38 39 40 |
* * inode_sb_list_lock * inode->i_lock |
bc3b14cb2
|
41 |
* Inode LRU list locks |
a66979aba
|
42 |
* |
f758eeabe
|
43 |
* bdi->wb.list_lock |
a66979aba
|
44 |
* inode->i_lock |
67a23c494
|
45 46 47 48 49 50 51 |
* * inode_hash_lock * inode_sb_list_lock * inode->i_lock * * iunique_lock * inode_hash_lock |
250df6ed2
|
52 |
*/ |
fa3536cc1
|
53 54 |
static unsigned int i_hash_mask __read_mostly; static unsigned int i_hash_shift __read_mostly; |
67a23c494
|
55 56 |
static struct hlist_head *inode_hashtable __read_mostly; static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock); |
1da177e4c
|
57 |
|
55fa6091d
|
58 |
__cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_sb_list_lock); |
1da177e4c
|
59 |
/* |
7dcda1c96
|
60 61 62 63 64 65 66 67 |
* Empty aops. Can be used for the cases where the user does not * define any of the address_space operations. */ const struct address_space_operations empty_aops = { }; EXPORT_SYMBOL(empty_aops); /* |
1da177e4c
|
68 69 70 |
* Statistics gathering.. */ struct inodes_stat_t inodes_stat; |
3942c07cc
|
71 72 |
static DEFINE_PER_CPU(unsigned long, nr_inodes); static DEFINE_PER_CPU(unsigned long, nr_unused); |
cffbc8aa3
|
73 |
|
6b3304b53
|
74 |
static struct kmem_cache *inode_cachep __read_mostly; |
1da177e4c
|
75 |
|
3942c07cc
|
76 |
static long get_nr_inodes(void) |
cffbc8aa3
|
77 |
{ |
3e880fb5e
|
78 |
int i; |
3942c07cc
|
79 |
long sum = 0; |
3e880fb5e
|
80 81 82 |
for_each_possible_cpu(i) sum += per_cpu(nr_inodes, i); return sum < 0 ? 0 : sum; |
cffbc8aa3
|
83 |
} |
3942c07cc
|
84 |
static inline long get_nr_inodes_unused(void) |
cffbc8aa3
|
85 |
{ |
fcb94f72d
|
86 |
int i; |
3942c07cc
|
87 |
long sum = 0; |
fcb94f72d
|
88 89 90 |
for_each_possible_cpu(i) sum += per_cpu(nr_unused, i); return sum < 0 ? 0 : sum; |
cffbc8aa3
|
91 |
} |
3942c07cc
|
92 |
long get_nr_dirty_inodes(void) |
cffbc8aa3
|
93 |
{ |
3e880fb5e
|
94 |
/* not actually dirty inodes, but a wild approximation */ |
3942c07cc
|
95 |
long nr_dirty = get_nr_inodes() - get_nr_inodes_unused(); |
cffbc8aa3
|
96 |
return nr_dirty > 0 ? nr_dirty : 0; |
cffbc8aa3
|
97 98 99 100 101 102 103 104 105 106 |
} /* * Handle nr_inode sysctl */ #ifdef CONFIG_SYSCTL int proc_nr_inodes(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { inodes_stat.nr_inodes = get_nr_inodes(); |
fcb94f72d
|
107 |
inodes_stat.nr_unused = get_nr_inodes_unused(); |
3942c07cc
|
108 |
return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); |
cffbc8aa3
|
109 110 |
} #endif |
2cb1599f9
|
111 112 |
/** * inode_init_always - perform inode structure intialisation |
0bc02f3fa
|
113 114 |
* @sb: superblock inode belongs to * @inode: inode to initialise |
2cb1599f9
|
115 116 117 118 |
* * These are initializations that need to be done on every inode * allocation as the fields are not initialised by slab allocation. */ |
54e346215
|
119 |
int inode_init_always(struct super_block *sb, struct inode *inode) |
1da177e4c
|
120 |
{ |
6e1d5dcc2
|
121 |
static const struct inode_operations empty_iops; |
99ac48f54
|
122 |
static const struct file_operations empty_fops; |
6b3304b53
|
123 |
struct address_space *const mapping = &inode->i_data; |
2cb1599f9
|
124 125 126 127 128 129 130 |
inode->i_sb = sb; inode->i_blkbits = sb->s_blocksize_bits; inode->i_flags = 0; atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; inode->i_fop = &empty_fops; |
a78ef704a
|
131 |
inode->__i_nlink = 1; |
3ddcd0569
|
132 |
inode->i_opflags = 0; |
92361636e
|
133 134 |
i_uid_write(inode, 0); i_gid_write(inode, 0); |
2cb1599f9
|
135 136 137 138 139 |
atomic_set(&inode->i_writecount, 0); inode->i_size = 0; inode->i_blocks = 0; inode->i_bytes = 0; inode->i_generation = 0; |
1da177e4c
|
140 |
#ifdef CONFIG_QUOTA |
2cb1599f9
|
141 |
memset(&inode->i_dquot, 0, sizeof(inode->i_dquot)); |
1da177e4c
|
142 |
#endif |
2cb1599f9
|
143 144 145 146 147 |
inode->i_pipe = NULL; inode->i_bdev = NULL; inode->i_cdev = NULL; inode->i_rdev = 0; inode->dirtied_when = 0; |
6146f0d5e
|
148 149 |
if (security_inode_alloc(inode)) |
54e346215
|
150 |
goto out; |
2cb1599f9
|
151 152 153 154 155 |
spin_lock_init(&inode->i_lock); lockdep_set_class(&inode->i_lock, &sb->s_type->i_lock_key); mutex_init(&inode->i_mutex); lockdep_set_class(&inode->i_mutex, &sb->s_type->i_mutex_key); |
bd5fe6c5e
|
156 |
atomic_set(&inode->i_dio_count, 0); |
2cb1599f9
|
157 158 159 160 |
mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags = 0; |
3c1d43787
|
161 |
mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE); |
252aa6f5b
|
162 |
mapping->private_data = NULL; |
2cb1599f9
|
163 164 165 166 167 168 169 170 171 172 |
mapping->backing_dev_info = &default_backing_dev_info; mapping->writeback_index = 0; /* * If the block_device provides a backing_dev_info for client * inodes then use that. Otherwise the inode share the bdev's * backing_dev_info. */ if (sb->s_bdev) { struct backing_dev_info *bdi; |
2c96ce9f2
|
173 |
bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info; |
2cb1599f9
|
174 175 176 177 |
mapping->backing_dev_info = bdi; } inode->i_private = NULL; inode->i_mapping = mapping; |
b3d9b7a3c
|
178 |
INIT_HLIST_HEAD(&inode->i_dentry); /* buggered by rcu freeing */ |
f19d4a8fa
|
179 180 181 |
#ifdef CONFIG_FS_POSIX_ACL inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED; #endif |
2cb1599f9
|
182 |
|
3be25f49b
|
183 184 185 |
#ifdef CONFIG_FSNOTIFY inode->i_fsnotify_mask = 0; #endif |
3e880fb5e
|
186 |
this_cpu_inc(nr_inodes); |
cffbc8aa3
|
187 |
|
54e346215
|
188 |
return 0; |
54e346215
|
189 190 |
out: return -ENOMEM; |
1da177e4c
|
191 |
} |
2cb1599f9
|
192 193 194 195 196 197 198 199 200 201 |
EXPORT_SYMBOL(inode_init_always); static struct inode *alloc_inode(struct super_block *sb) { struct inode *inode; if (sb->s_op->alloc_inode) inode = sb->s_op->alloc_inode(sb); else inode = kmem_cache_alloc(inode_cachep, GFP_KERNEL); |
54e346215
|
202 203 204 205 206 207 208 209 210 211 212 213 |
if (!inode) return NULL; if (unlikely(inode_init_always(sb, inode))) { if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode(inode); else kmem_cache_free(inode_cachep, inode); return NULL; } return inode; |
2cb1599f9
|
214 |
} |
1da177e4c
|
215 |
|
ff0c7d15f
|
216 217 218 219 220 |
void free_inode_nonrcu(struct inode *inode) { kmem_cache_free(inode_cachep, inode); } EXPORT_SYMBOL(free_inode_nonrcu); |
2e00c97e2
|
221 |
void __destroy_inode(struct inode *inode) |
1da177e4c
|
222 |
{ |
b7542f8c7
|
223 |
BUG_ON(inode_has_buffers(inode)); |
1da177e4c
|
224 |
security_inode_free(inode); |
3be25f49b
|
225 |
fsnotify_inode_delete(inode); |
7ada4db88
|
226 227 228 229 |
if (!inode->i_nlink) { WARN_ON(atomic_long_read(&inode->i_sb->s_remove_count) == 0); atomic_long_dec(&inode->i_sb->s_remove_count); } |
f19d4a8fa
|
230 231 232 233 234 235 |
#ifdef CONFIG_FS_POSIX_ACL if (inode->i_acl && inode->i_acl != ACL_NOT_CACHED) posix_acl_release(inode->i_acl); if (inode->i_default_acl && inode->i_default_acl != ACL_NOT_CACHED) posix_acl_release(inode->i_default_acl); #endif |
3e880fb5e
|
236 |
this_cpu_dec(nr_inodes); |
2e00c97e2
|
237 238 |
} EXPORT_SYMBOL(__destroy_inode); |
fa0d7e3de
|
239 240 241 |
static void i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); |
fa0d7e3de
|
242 243 |
kmem_cache_free(inode_cachep, inode); } |
56b0dacfa
|
244 |
static void destroy_inode(struct inode *inode) |
2e00c97e2
|
245 |
{ |
7ccf19a80
|
246 |
BUG_ON(!list_empty(&inode->i_lru)); |
2e00c97e2
|
247 |
__destroy_inode(inode); |
1da177e4c
|
248 249 250 |
if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode(inode); else |
fa0d7e3de
|
251 |
call_rcu(&inode->i_rcu, i_callback); |
1da177e4c
|
252 |
} |
1da177e4c
|
253 |
|
7ada4db88
|
254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 |
/** * drop_nlink - directly drop an inode's link count * @inode: inode * * This is a low-level filesystem helper to replace any * direct filesystem manipulation of i_nlink. In cases * where we are attempting to track writes to the * filesystem, a decrement to zero means an imminent * write when the file is truncated and actually unlinked * on the filesystem. */ void drop_nlink(struct inode *inode) { WARN_ON(inode->i_nlink == 0); inode->__i_nlink--; if (!inode->i_nlink) atomic_long_inc(&inode->i_sb->s_remove_count); } EXPORT_SYMBOL(drop_nlink); /** * clear_nlink - directly zero an inode's link count * @inode: inode * * This is a low-level filesystem helper to replace any * direct filesystem manipulation of i_nlink. See * drop_nlink() for why we care about i_nlink hitting zero. */ void clear_nlink(struct inode *inode) { if (inode->i_nlink) { inode->__i_nlink = 0; atomic_long_inc(&inode->i_sb->s_remove_count); } } EXPORT_SYMBOL(clear_nlink); /** * set_nlink - directly set an inode's link count * @inode: inode * @nlink: new nlink (should be non-zero) * * This is a low-level filesystem helper to replace any * direct filesystem manipulation of i_nlink. */ void set_nlink(struct inode *inode, unsigned int nlink) { if (!nlink) { |
7ada4db88
|
302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 |
clear_nlink(inode); } else { /* Yes, some filesystems do change nlink from zero to one */ if (inode->i_nlink == 0) atomic_long_dec(&inode->i_sb->s_remove_count); inode->__i_nlink = nlink; } } EXPORT_SYMBOL(set_nlink); /** * inc_nlink - directly increment an inode's link count * @inode: inode * * This is a low-level filesystem helper to replace any * direct filesystem manipulation of i_nlink. Currently, * it is only here for parity with dec_nlink(). */ void inc_nlink(struct inode *inode) { |
f4e0c30c1
|
323 324 |
if (unlikely(inode->i_nlink == 0)) { WARN_ON(!(inode->i_state & I_LINKABLE)); |
7ada4db88
|
325 |
atomic_long_dec(&inode->i_sb->s_remove_count); |
f4e0c30c1
|
326 |
} |
7ada4db88
|
327 328 329 330 |
inode->__i_nlink++; } EXPORT_SYMBOL(inc_nlink); |
2aa15890f
|
331 332 333 334 335 |
void address_space_init_once(struct address_space *mapping) { memset(mapping, 0, sizeof(*mapping)); INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC); spin_lock_init(&mapping->tree_lock); |
3d48ae45e
|
336 |
mutex_init(&mapping->i_mmap_mutex); |
2aa15890f
|
337 338 |
INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); |
6b2dbba8b
|
339 |
mapping->i_mmap = RB_ROOT; |
2aa15890f
|
340 |
INIT_LIST_HEAD(&mapping->i_mmap_nonlinear); |
2aa15890f
|
341 342 |
} EXPORT_SYMBOL(address_space_init_once); |
1da177e4c
|
343 344 345 346 347 348 349 350 351 |
/* * These are initializations that only need to be done * once, because the fields are idempotent across use * of the inode, so let the slab aware of that. */ void inode_init_once(struct inode *inode) { memset(inode, 0, sizeof(*inode)); INIT_HLIST_NODE(&inode->i_hash); |
1da177e4c
|
352 |
INIT_LIST_HEAD(&inode->i_devices); |
7ccf19a80
|
353 354 |
INIT_LIST_HEAD(&inode->i_wb_list); INIT_LIST_HEAD(&inode->i_lru); |
2aa15890f
|
355 |
address_space_init_once(&inode->i_data); |
1da177e4c
|
356 |
i_size_ordered_init(inode); |
3be25f49b
|
357 |
#ifdef CONFIG_FSNOTIFY |
e61ce8673
|
358 |
INIT_HLIST_HEAD(&inode->i_fsnotify_marks); |
3be25f49b
|
359 |
#endif |
1da177e4c
|
360 |
} |
1da177e4c
|
361 |
EXPORT_SYMBOL(inode_init_once); |
51cc50685
|
362 |
static void init_once(void *foo) |
1da177e4c
|
363 |
{ |
6b3304b53
|
364 |
struct inode *inode = (struct inode *) foo; |
1da177e4c
|
365 |
|
a35afb830
|
366 |
inode_init_once(inode); |
1da177e4c
|
367 368 369 |
} /* |
250df6ed2
|
370 |
* inode->i_lock must be held |
1da177e4c
|
371 |
*/ |
6b3304b53
|
372 |
void __iget(struct inode *inode) |
1da177e4c
|
373 |
{ |
9e38d86ff
|
374 375 |
atomic_inc(&inode->i_count); } |
2e147f1ef
|
376 |
|
7de9c6ee3
|
377 378 379 380 381 382 383 384 |
/* * get additional reference to inode; caller must already hold one. */ void ihold(struct inode *inode) { WARN_ON(atomic_inc_return(&inode->i_count) < 2); } EXPORT_SYMBOL(ihold); |
9e38d86ff
|
385 386 |
static void inode_lru_list_add(struct inode *inode) { |
bc3b14cb2
|
387 |
if (list_lru_add(&inode->i_sb->s_inode_lru, &inode->i_lru)) |
fcb94f72d
|
388 |
this_cpu_inc(nr_unused); |
9e38d86ff
|
389 |
} |
2e147f1ef
|
390 |
|
4eff96dd5
|
391 392 393 394 395 396 397 398 399 400 401 |
/* * Add inode to LRU if needed (inode is unused and clean). * * Needs inode->i_lock held. */ void inode_add_lru(struct inode *inode) { if (!(inode->i_state & (I_DIRTY | I_SYNC | I_FREEING | I_WILL_FREE)) && !atomic_read(&inode->i_count) && inode->i_sb->s_flags & MS_ACTIVE) inode_lru_list_add(inode); } |
9e38d86ff
|
402 403 |
static void inode_lru_list_del(struct inode *inode) { |
bc3b14cb2
|
404 405 |
if (list_lru_del(&inode->i_sb->s_inode_lru, &inode->i_lru)) |
fcb94f72d
|
406 |
this_cpu_dec(nr_unused); |
1da177e4c
|
407 |
} |
646ec4615
|
408 409 410 411 412 413 |
/** * inode_sb_list_add - add inode to the superblock list of inodes * @inode: inode to add */ void inode_sb_list_add(struct inode *inode) { |
55fa6091d
|
414 415 416 |
spin_lock(&inode_sb_list_lock); list_add(&inode->i_sb_list, &inode->i_sb->s_inodes); spin_unlock(&inode_sb_list_lock); |
646ec4615
|
417 418 |
} EXPORT_SYMBOL_GPL(inode_sb_list_add); |
55fa6091d
|
419 |
static inline void inode_sb_list_del(struct inode *inode) |
646ec4615
|
420 |
{ |
a209dfc7b
|
421 422 423 424 425 |
if (!list_empty(&inode->i_sb_list)) { spin_lock(&inode_sb_list_lock); list_del_init(&inode->i_sb_list); spin_unlock(&inode_sb_list_lock); } |
646ec4615
|
426 |
} |
4c51acbc6
|
427 428 429 430 431 432 |
static unsigned long hash(struct super_block *sb, unsigned long hashval) { unsigned long tmp; tmp = (hashval * (unsigned long)sb) ^ (GOLDEN_RATIO_PRIME + hashval) / L1_CACHE_BYTES; |
4b4563dc8
|
433 434 |
tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> i_hash_shift); return tmp & i_hash_mask; |
4c51acbc6
|
435 436 437 438 439 440 441 442 443 444 445 446 |
} /** * __insert_inode_hash - hash an inode * @inode: unhashed inode * @hashval: unsigned long value used to locate this object in the * inode_hashtable. * * Add an inode to the inode hash for this superblock. */ void __insert_inode_hash(struct inode *inode, unsigned long hashval) { |
646ec4615
|
447 |
struct hlist_head *b = inode_hashtable + hash(inode->i_sb, hashval); |
67a23c494
|
448 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
449 |
spin_lock(&inode->i_lock); |
646ec4615
|
450 |
hlist_add_head(&inode->i_hash, b); |
250df6ed2
|
451 |
spin_unlock(&inode->i_lock); |
67a23c494
|
452 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
453 454 455 456 |
} EXPORT_SYMBOL(__insert_inode_hash); /** |
f2ee7abf4
|
457 |
* __remove_inode_hash - remove an inode from the hash |
4c51acbc6
|
458 459 460 461 |
* @inode: inode to unhash * * Remove an inode from the superblock. */ |
f2ee7abf4
|
462 |
void __remove_inode_hash(struct inode *inode) |
4c51acbc6
|
463 |
{ |
67a23c494
|
464 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
465 |
spin_lock(&inode->i_lock); |
4c51acbc6
|
466 |
hlist_del_init(&inode->i_hash); |
250df6ed2
|
467 |
spin_unlock(&inode->i_lock); |
67a23c494
|
468 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
469 |
} |
f2ee7abf4
|
470 |
EXPORT_SYMBOL(__remove_inode_hash); |
4c51acbc6
|
471 |
|
dbd5768f8
|
472 |
void clear_inode(struct inode *inode) |
b0683aa63
|
473 474 |
{ might_sleep(); |
08142579b
|
475 476 477 478 479 480 |
/* * We have to cycle tree_lock here because reclaim can be still in the * process of removing the last page (in __delete_from_page_cache()) * and we must not free mapping under it. */ spin_lock_irq(&inode->i_data.tree_lock); |
b0683aa63
|
481 |
BUG_ON(inode->i_data.nrpages); |
91b0abe36
|
482 |
BUG_ON(inode->i_data.nrshadows); |
08142579b
|
483 |
spin_unlock_irq(&inode->i_data.tree_lock); |
b0683aa63
|
484 485 486 |
BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); |
fa0d7e3de
|
487 |
/* don't need i_lock here, no concurrent mods to i_state */ |
b0683aa63
|
488 489 |
inode->i_state = I_FREEING | I_CLEAR; } |
dbd5768f8
|
490 |
EXPORT_SYMBOL(clear_inode); |
b0683aa63
|
491 |
|
b2b2af8e6
|
492 493 494 495 496 497 498 499 500 501 502 503 504 |
/* * Free the inode passed in, removing it from the lists it is still connected * to. We remove any pages still attached to the inode and wait for any IO that * is still in progress before finally destroying the inode. * * An inode must already be marked I_FREEING so that we avoid the inode being * moved back onto lists if we race with other code that manipulates the lists * (e.g. writeback_single_inode). The caller is responsible for setting this. * * An inode must already be removed from the LRU list before being evicted from * the cache. This should occur atomically with setting the I_FREEING state * flag, so no inodes here should ever be on the LRU when being evicted. */ |
644da5960
|
505 |
static void evict(struct inode *inode) |
b4272d4c8
|
506 507 |
{ const struct super_operations *op = inode->i_sb->s_op; |
b2b2af8e6
|
508 509 |
BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(!list_empty(&inode->i_lru)); |
b12362bdb
|
510 511 |
if (!list_empty(&inode->i_wb_list)) inode_wb_list_del(inode); |
55fa6091d
|
512 |
inode_sb_list_del(inode); |
169ebd901
|
513 514 515 516 517 518 519 |
/* * Wait for flusher thread to be done with the inode so that filesystem * does not start destroying it while writeback is still running. Since * the inode has I_FREEING set, flusher thread won't start new work on * the inode. We just have to wait for running writeback to finish. */ inode_wait_for_writeback(inode); |
7994e6f72
|
520 |
|
be7ce4161
|
521 522 |
if (op->evict_inode) { op->evict_inode(inode); |
b4272d4c8
|
523 |
} else { |
91b0abe36
|
524 |
truncate_inode_pages_final(&inode->i_data); |
dbd5768f8
|
525 |
clear_inode(inode); |
b4272d4c8
|
526 |
} |
661074e91
|
527 528 529 530 |
if (S_ISBLK(inode->i_mode) && inode->i_bdev) bd_forget(inode); if (S_ISCHR(inode->i_mode) && inode->i_cdev) cd_forget(inode); |
b2b2af8e6
|
531 532 533 534 535 536 537 538 539 |
remove_inode_hash(inode); spin_lock(&inode->i_lock); wake_up_bit(&inode->i_state, __I_NEW); BUG_ON(inode->i_state != (I_FREEING | I_CLEAR)); spin_unlock(&inode->i_lock); destroy_inode(inode); |
b4272d4c8
|
540 |
} |
1da177e4c
|
541 542 543 544 545 546 547 548 549 |
/* * dispose_list - dispose of the contents of a local list * @head: the head of the list to free * * Dispose-list gets a local list with local inodes in it, so it doesn't * need to worry about list corruption and SMP locks. */ static void dispose_list(struct list_head *head) { |
1da177e4c
|
550 551 |
while (!list_empty(head)) { struct inode *inode; |
7ccf19a80
|
552 553 |
inode = list_first_entry(head, struct inode, i_lru); list_del_init(&inode->i_lru); |
1da177e4c
|
554 |
|
644da5960
|
555 |
evict(inode); |
1da177e4c
|
556 |
} |
1da177e4c
|
557 |
} |
a03187867
|
558 |
/** |
63997e98a
|
559 560 561 562 563 564 565 |
* evict_inodes - evict all evictable inodes for a superblock * @sb: superblock to operate on * * Make sure that no inodes with zero refcount are retained. This is * called by superblock shutdown after having MS_ACTIVE flag removed, * so any inode reaching zero refcount during or after that call will * be immediately evicted. |
1da177e4c
|
566 |
*/ |
63997e98a
|
567 |
void evict_inodes(struct super_block *sb) |
1da177e4c
|
568 |
{ |
63997e98a
|
569 570 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
571 |
|
55fa6091d
|
572 |
spin_lock(&inode_sb_list_lock); |
63997e98a
|
573 574 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { if (atomic_read(&inode->i_count)) |
aabb8fdb4
|
575 |
continue; |
250df6ed2
|
576 577 578 579 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
1da177e4c
|
580 |
continue; |
250df6ed2
|
581 |
} |
63997e98a
|
582 583 |
inode->i_state |= I_FREEING; |
02afc410f
|
584 |
inode_lru_list_del(inode); |
250df6ed2
|
585 |
spin_unlock(&inode->i_lock); |
02afc410f
|
586 |
list_add(&inode->i_lru, &dispose); |
1da177e4c
|
587 |
} |
55fa6091d
|
588 |
spin_unlock(&inode_sb_list_lock); |
63997e98a
|
589 590 |
dispose_list(&dispose); |
1da177e4c
|
591 |
} |
1da177e4c
|
592 |
/** |
a03187867
|
593 594 |
* invalidate_inodes - attempt to free all inodes on a superblock * @sb: superblock to operate on |
93b270f76
|
595 |
* @kill_dirty: flag to guide handling of dirty inodes |
1da177e4c
|
596 |
* |
a03187867
|
597 598 |
* Attempts to free all inodes for a given superblock. If there were any * busy inodes return a non-zero value, else zero. |
93b270f76
|
599 600 |
* If @kill_dirty is set, discard dirty inodes too, otherwise treat * them as busy. |
1da177e4c
|
601 |
*/ |
93b270f76
|
602 |
int invalidate_inodes(struct super_block *sb, bool kill_dirty) |
1da177e4c
|
603 |
{ |
cffbc8aa3
|
604 |
int busy = 0; |
a03187867
|
605 606 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
607 |
|
55fa6091d
|
608 |
spin_lock(&inode_sb_list_lock); |
a03187867
|
609 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { |
250df6ed2
|
610 611 612 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
aabb8fdb4
|
613 |
continue; |
250df6ed2
|
614 |
} |
93b270f76
|
615 |
if (inode->i_state & I_DIRTY && !kill_dirty) { |
250df6ed2
|
616 |
spin_unlock(&inode->i_lock); |
93b270f76
|
617 618 619 |
busy = 1; continue; } |
99a389192
|
620 |
if (atomic_read(&inode->i_count)) { |
250df6ed2
|
621 |
spin_unlock(&inode->i_lock); |
99a389192
|
622 |
busy = 1; |
1da177e4c
|
623 624 |
continue; } |
99a389192
|
625 |
|
99a389192
|
626 |
inode->i_state |= I_FREEING; |
02afc410f
|
627 |
inode_lru_list_del(inode); |
250df6ed2
|
628 |
spin_unlock(&inode->i_lock); |
02afc410f
|
629 |
list_add(&inode->i_lru, &dispose); |
1da177e4c
|
630 |
} |
55fa6091d
|
631 |
spin_unlock(&inode_sb_list_lock); |
1da177e4c
|
632 |
|
a03187867
|
633 |
dispose_list(&dispose); |
1da177e4c
|
634 635 636 |
return busy; } |
1da177e4c
|
637 |
|
1da177e4c
|
638 |
/* |
bc3b14cb2
|
639 |
* Isolate the inode from the LRU in preparation for freeing it. |
1da177e4c
|
640 641 |
* * Any inodes which are pinned purely because of attached pagecache have their |
9e38d86ff
|
642 643 |
* pagecache removed. If the inode has metadata buffers attached to * mapping->private_list then try to remove them. |
1da177e4c
|
644 |
* |
9e38d86ff
|
645 646 647 648 649 650 651 |
* If the inode has the I_REFERENCED flag set, then it means that it has been * used recently - the flag is set in iput_final(). When we encounter such an * inode, clear the flag and move it to the back of the LRU so it gets another * pass through the LRU before it gets reclaimed. This is necessary because of * the fact we are doing lazy LRU updates to minimise lock contention so the * LRU does not have strict ordering. Hence we don't want to reclaim inodes * with this flag set because they are the inodes that are out of order. |
1da177e4c
|
652 |
*/ |
bc3b14cb2
|
653 654 |
static enum lru_status inode_lru_isolate(struct list_head *item, spinlock_t *lru_lock, void *arg) |
1da177e4c
|
655 |
{ |
bc3b14cb2
|
656 657 |
struct list_head *freeable = arg; struct inode *inode = container_of(item, struct inode, i_lru); |
1da177e4c
|
658 |
|
bc3b14cb2
|
659 660 661 662 663 664 |
/* * we are inverting the lru lock/inode->i_lock here, so use a trylock. * If we fail to get the lock, just skip it. */ if (!spin_trylock(&inode->i_lock)) return LRU_SKIP; |
1da177e4c
|
665 |
|
bc3b14cb2
|
666 667 668 669 670 671 672 673 674 675 676 |
/* * Referenced or dirty inodes are still in use. Give them another pass * through the LRU as we canot reclaim them now. */ if (atomic_read(&inode->i_count) || (inode->i_state & ~I_REFERENCED)) { list_del_init(&inode->i_lru); spin_unlock(&inode->i_lock); this_cpu_dec(nr_unused); return LRU_REMOVED; } |
1da177e4c
|
677 |
|
bc3b14cb2
|
678 679 680 681 682 683 |
/* recently referenced inodes get one more pass */ if (inode->i_state & I_REFERENCED) { inode->i_state &= ~I_REFERENCED; spin_unlock(&inode->i_lock); return LRU_ROTATE; } |
1da177e4c
|
684 |
|
bc3b14cb2
|
685 686 687 688 689 690 691 692 693 694 695 696 697 |
if (inode_has_buffers(inode) || inode->i_data.nrpages) { __iget(inode); spin_unlock(&inode->i_lock); spin_unlock(lru_lock); if (remove_inode_buffers(inode)) { unsigned long reap; reap = invalidate_mapping_pages(&inode->i_data, 0, -1); if (current_is_kswapd()) __count_vm_events(KSWAPD_INODESTEAL, reap); else __count_vm_events(PGINODESTEAL, reap); if (current->reclaim_state) current->reclaim_state->reclaimed_slab += reap; |
02afc410f
|
698 |
} |
bc3b14cb2
|
699 700 701 702 |
iput(inode); spin_lock(lru_lock); return LRU_RETRY; } |
02afc410f
|
703 |
|
bc3b14cb2
|
704 705 |
WARN_ON(inode->i_state & I_NEW); inode->i_state |= I_FREEING; |
d38fa6986
|
706 |
list_move(&inode->i_lru, freeable); |
bc3b14cb2
|
707 |
spin_unlock(&inode->i_lock); |
9e38d86ff
|
708 |
|
bc3b14cb2
|
709 710 711 |
this_cpu_dec(nr_unused); return LRU_REMOVED; } |
7ccf19a80
|
712 |
|
bc3b14cb2
|
713 714 715 716 717 718 |
/* * Walk the superblock inode LRU for freeable inodes and attempt to free them. * This is called from the superblock shrinker function with a number of inodes * to trim from the LRU. Inodes to be freed are moved to a temporary list and * then are freed outside inode_lock by dispose_list(). */ |
9b17c6238
|
719 720 |
long prune_icache_sb(struct super_block *sb, unsigned long nr_to_scan, int nid) |
bc3b14cb2
|
721 722 723 |
{ LIST_HEAD(freeable); long freed; |
1da177e4c
|
724 |
|
9b17c6238
|
725 726 |
freed = list_lru_walk_node(&sb->s_inode_lru, nid, inode_lru_isolate, &freeable, &nr_to_scan); |
1da177e4c
|
727 |
dispose_list(&freeable); |
0a234c6dc
|
728 |
return freed; |
1da177e4c
|
729 |
} |
1da177e4c
|
730 731 732 |
static void __wait_on_freeing_inode(struct inode *inode); /* * Called with the inode lock held. |
1da177e4c
|
733 |
*/ |
6b3304b53
|
734 735 736 737 |
static struct inode *find_inode(struct super_block *sb, struct hlist_head *head, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
738 |
{ |
6b3304b53
|
739 |
struct inode *inode = NULL; |
1da177e4c
|
740 741 |
repeat: |
b67bfe0d4
|
742 |
hlist_for_each_entry(inode, head, i_hash) { |
5a3cd9928
|
743 |
if (inode->i_sb != sb) |
1da177e4c
|
744 |
continue; |
5a3cd9928
|
745 |
if (!test(inode, data)) |
1da177e4c
|
746 |
continue; |
5a3cd9928
|
747 |
spin_lock(&inode->i_lock); |
a4ffdde6e
|
748 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
749 750 751 |
__wait_on_freeing_inode(inode); goto repeat; } |
f7899bd54
|
752 |
__iget(inode); |
250df6ed2
|
753 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
754 |
return inode; |
1da177e4c
|
755 |
} |
f7899bd54
|
756 |
return NULL; |
1da177e4c
|
757 758 759 760 761 762 |
} /* * find_inode_fast is the fast path version of find_inode, see the comment at * iget_locked for details. */ |
6b3304b53
|
763 764 |
static struct inode *find_inode_fast(struct super_block *sb, struct hlist_head *head, unsigned long ino) |
1da177e4c
|
765 |
{ |
6b3304b53
|
766 |
struct inode *inode = NULL; |
1da177e4c
|
767 768 |
repeat: |
b67bfe0d4
|
769 |
hlist_for_each_entry(inode, head, i_hash) { |
5a3cd9928
|
770 |
if (inode->i_ino != ino) |
1da177e4c
|
771 |
continue; |
5a3cd9928
|
772 |
if (inode->i_sb != sb) |
1da177e4c
|
773 |
continue; |
5a3cd9928
|
774 |
spin_lock(&inode->i_lock); |
a4ffdde6e
|
775 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
776 777 778 |
__wait_on_freeing_inode(inode); goto repeat; } |
f7899bd54
|
779 |
__iget(inode); |
250df6ed2
|
780 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
781 |
return inode; |
1da177e4c
|
782 |
} |
f7899bd54
|
783 |
return NULL; |
8290c35f8
|
784 |
} |
f991bd2e1
|
785 786 787 788 |
/* * Each cpu owns a range of LAST_INO_BATCH numbers. * 'shared_last_ino' is dirtied only once out of LAST_INO_BATCH allocations, * to renew the exhausted range. |
8290c35f8
|
789 |
* |
f991bd2e1
|
790 791 792 793 794 795 796 797 798 |
* This does not significantly increase overflow rate because every CPU can * consume at most LAST_INO_BATCH-1 unused inode numbers. So there is * NR_CPUS*(LAST_INO_BATCH-1) wastage. At 4096 and 1024, this is ~0.1% of the * 2^32 range, and is a worst-case. Even a 50% wastage would only increase * overflow rate by 2x, which does not seem too significant. * * On a 32bit, non LFS stat() call, glibc will generate an EOVERFLOW * error if st_ino won't fit in target struct field. Use 32bit counter * here to attempt to avoid that. |
8290c35f8
|
799 |
*/ |
f991bd2e1
|
800 801 |
#define LAST_INO_BATCH 1024 static DEFINE_PER_CPU(unsigned int, last_ino); |
85fe4025c
|
802 |
unsigned int get_next_ino(void) |
8290c35f8
|
803 |
{ |
f991bd2e1
|
804 805 |
unsigned int *p = &get_cpu_var(last_ino); unsigned int res = *p; |
8290c35f8
|
806 |
|
f991bd2e1
|
807 808 809 810 811 812 813 814 815 816 817 818 |
#ifdef CONFIG_SMP if (unlikely((res & (LAST_INO_BATCH-1)) == 0)) { static atomic_t shared_last_ino; int next = atomic_add_return(LAST_INO_BATCH, &shared_last_ino); res = next - LAST_INO_BATCH; } #endif *p = ++res; put_cpu_var(last_ino); return res; |
8290c35f8
|
819 |
} |
85fe4025c
|
820 |
EXPORT_SYMBOL(get_next_ino); |
8290c35f8
|
821 |
|
1da177e4c
|
822 |
/** |
a209dfc7b
|
823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 |
* new_inode_pseudo - obtain an inode * @sb: superblock * * Allocates a new inode for given superblock. * Inode wont be chained in superblock s_inodes list * This means : * - fs can't be unmount * - quotas, fsnotify, writeback can't work */ struct inode *new_inode_pseudo(struct super_block *sb) { struct inode *inode = alloc_inode(sb); if (inode) { spin_lock(&inode->i_lock); inode->i_state = 0; spin_unlock(&inode->i_lock); INIT_LIST_HEAD(&inode->i_sb_list); } return inode; } /** |
1da177e4c
|
846 847 848 |
* new_inode - obtain an inode * @sb: superblock * |
769848c03
|
849 |
* Allocates a new inode for given superblock. The default gfp_mask |
3c1d43787
|
850 |
* for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE. |
769848c03
|
851 852 853 854 855 |
* If HIGHMEM pages are unsuitable or it is known that pages allocated * for the page cache are not reclaimable or migratable, * mapping_set_gfp_mask() must be called with suitable flags on the * newly created inode's mapping * |
1da177e4c
|
856 857 858 |
*/ struct inode *new_inode(struct super_block *sb) { |
6b3304b53
|
859 |
struct inode *inode; |
1da177e4c
|
860 |
|
55fa6091d
|
861 |
spin_lock_prefetch(&inode_sb_list_lock); |
6b3304b53
|
862 |
|
a209dfc7b
|
863 864 |
inode = new_inode_pseudo(sb); if (inode) |
55fa6091d
|
865 |
inode_sb_list_add(inode); |
1da177e4c
|
866 867 |
return inode; } |
1da177e4c
|
868 |
EXPORT_SYMBOL(new_inode); |
14358e6dd
|
869 |
#ifdef CONFIG_DEBUG_LOCK_ALLOC |
e096d0c7e
|
870 871 |
void lockdep_annotate_inode_mutex_key(struct inode *inode) { |
a3314a0ed
|
872 |
if (S_ISDIR(inode->i_mode)) { |
1e89a5e15
|
873 |
struct file_system_type *type = inode->i_sb->s_type; |
9a7aa12f3
|
874 |
/* Set new key only if filesystem hasn't already changed it */ |
978d6d8c4
|
875 |
if (lockdep_match_class(&inode->i_mutex, &type->i_mutex_key)) { |
9a7aa12f3
|
876 877 878 879 880 881 882 883 |
/* * ensure nobody is actually holding i_mutex */ mutex_destroy(&inode->i_mutex); mutex_init(&inode->i_mutex); lockdep_set_class(&inode->i_mutex, &type->i_mutex_dir_key); } |
1e89a5e15
|
884 |
} |
e096d0c7e
|
885 886 |
} EXPORT_SYMBOL(lockdep_annotate_inode_mutex_key); |
14358e6dd
|
887 |
#endif |
e096d0c7e
|
888 889 890 891 892 893 894 895 896 897 898 |
/** * unlock_new_inode - clear the I_NEW state and wake up any waiters * @inode: new inode to unlock * * Called when the inode is fully initialised to clear the new state of the * inode and wake up anyone waiting for the inode to finish initialisation. */ void unlock_new_inode(struct inode *inode) { lockdep_annotate_inode_mutex_key(inode); |
250df6ed2
|
899 |
spin_lock(&inode->i_lock); |
eaff8079d
|
900 901 |
WARN_ON(!(inode->i_state & I_NEW)); inode->i_state &= ~I_NEW; |
310fa7a36
|
902 |
smp_mb(); |
250df6ed2
|
903 904 |
wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); |
1da177e4c
|
905 |
} |
1da177e4c
|
906 |
EXPORT_SYMBOL(unlock_new_inode); |
0b2d0724e
|
907 |
/** |
375e289ea
|
908 |
* lock_two_nondirectories - take two i_mutexes on non-directory objects |
4fd699ae3
|
909 910 911 912 |
* * Lock any non-NULL argument that is not a directory. * Zero, one or two objects may be locked by this function. * |
375e289ea
|
913 914 915 916 917 |
* @inode1: first inode to lock * @inode2: second inode to lock */ void lock_two_nondirectories(struct inode *inode1, struct inode *inode2) { |
4fd699ae3
|
918 919 920 921 |
if (inode1 > inode2) swap(inode1, inode2); if (inode1 && !S_ISDIR(inode1->i_mode)) |
275555163
|
922 |
mutex_lock(&inode1->i_mutex); |
4fd699ae3
|
923 |
if (inode2 && !S_ISDIR(inode2->i_mode) && inode2 != inode1) |
40bd22c9f
|
924 |
mutex_lock_nested(&inode2->i_mutex, I_MUTEX_NONDIR2); |
375e289ea
|
925 926 927 928 929 930 931 932 933 934 |
} EXPORT_SYMBOL(lock_two_nondirectories); /** * unlock_two_nondirectories - release locks from lock_two_nondirectories() * @inode1: first inode to unlock * @inode2: second inode to unlock */ void unlock_two_nondirectories(struct inode *inode1, struct inode *inode2) { |
4fd699ae3
|
935 936 937 |
if (inode1 && !S_ISDIR(inode1->i_mode)) mutex_unlock(&inode1->i_mutex); if (inode2 && !S_ISDIR(inode2->i_mode) && inode2 != inode1) |
375e289ea
|
938 939 940 941 942 |
mutex_unlock(&inode2->i_mutex); } EXPORT_SYMBOL(unlock_two_nondirectories); /** |
0b2d0724e
|
943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 |
* iget5_locked - obtain an inode from a mounted file system * @sb: super block of file system * @hashval: hash value (usually inode number) to get * @test: callback used for comparisons between inodes * @set: callback used to initialize a new struct inode * @data: opaque data pointer to pass to @test and @set * * Search for the inode specified by @hashval and @data in the inode cache, * and if present it is return it with an increased reference count. This is * a generalized version of iget_locked() for file systems where the inode * number is not sufficient for unique identification of an inode. * * If the inode is not in cache, allocate a new inode and return it locked, * hashed, and with the I_NEW flag set. The file system gets to fill it in * before unlocking it via unlock_new_inode(). |
1da177e4c
|
958 |
* |
0b2d0724e
|
959 960 |
* Note both @test and @set are called with the inode_hash_lock held, so can't * sleep. |
1da177e4c
|
961 |
*/ |
0b2d0724e
|
962 963 964 |
struct inode *iget5_locked(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data) |
1da177e4c
|
965 |
{ |
0b2d0724e
|
966 |
struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
6b3304b53
|
967 |
struct inode *inode; |
1da177e4c
|
968 |
|
0b2d0724e
|
969 970 971 972 973 974 975 976 |
spin_lock(&inode_hash_lock); inode = find_inode(sb, head, test, data); spin_unlock(&inode_hash_lock); if (inode) { wait_on_inode(inode); return inode; } |
1da177e4c
|
977 978 |
inode = alloc_inode(sb); if (inode) { |
6b3304b53
|
979 |
struct inode *old; |
1da177e4c
|
980 |
|
67a23c494
|
981 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
982 983 984 985 986 |
/* We released the lock, so.. */ old = find_inode(sb, head, test, data); if (!old) { if (set(inode, data)) goto set_failed; |
250df6ed2
|
987 988 |
spin_lock(&inode->i_lock); inode->i_state = I_NEW; |
646ec4615
|
989 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
990 |
spin_unlock(&inode->i_lock); |
55fa6091d
|
991 |
inode_sb_list_add(inode); |
67a23c494
|
992 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 |
/* Return the locked inode with I_NEW set, the * caller is responsible for filling in the contents */ return inode; } /* * Uhhuh, somebody else created the same inode under * us. Use the old inode instead of the one we just * allocated. */ |
67a23c494
|
1005 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1006 1007 1008 1009 1010 1011 1012 |
destroy_inode(inode); inode = old; wait_on_inode(inode); } return inode; set_failed: |
67a23c494
|
1013 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1014 1015 1016 |
destroy_inode(inode); return NULL; } |
0b2d0724e
|
1017 |
EXPORT_SYMBOL(iget5_locked); |
1da177e4c
|
1018 |
|
0b2d0724e
|
1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 |
/** * iget_locked - obtain an inode from a mounted file system * @sb: super block of file system * @ino: inode number to get * * Search for the inode specified by @ino in the inode cache and if present * return it with an increased reference count. This is for file systems * where the inode number is sufficient for unique identification of an inode. * * If the inode is not in cache, allocate a new inode and return it locked, * hashed, and with the I_NEW flag set. The file system gets to fill it in * before unlocking it via unlock_new_inode(). |
1da177e4c
|
1031 |
*/ |
0b2d0724e
|
1032 |
struct inode *iget_locked(struct super_block *sb, unsigned long ino) |
1da177e4c
|
1033 |
{ |
0b2d0724e
|
1034 |
struct hlist_head *head = inode_hashtable + hash(sb, ino); |
6b3304b53
|
1035 |
struct inode *inode; |
1da177e4c
|
1036 |
|
0b2d0724e
|
1037 1038 1039 1040 1041 1042 1043 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); if (inode) { wait_on_inode(inode); return inode; } |
1da177e4c
|
1044 1045 |
inode = alloc_inode(sb); if (inode) { |
6b3304b53
|
1046 |
struct inode *old; |
1da177e4c
|
1047 |
|
67a23c494
|
1048 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1049 1050 1051 1052 |
/* We released the lock, so.. */ old = find_inode_fast(sb, head, ino); if (!old) { inode->i_ino = ino; |
250df6ed2
|
1053 1054 |
spin_lock(&inode->i_lock); inode->i_state = I_NEW; |
646ec4615
|
1055 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1056 |
spin_unlock(&inode->i_lock); |
55fa6091d
|
1057 |
inode_sb_list_add(inode); |
67a23c494
|
1058 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 |
/* Return the locked inode with I_NEW set, the * caller is responsible for filling in the contents */ return inode; } /* * Uhhuh, somebody else created the same inode under * us. Use the old inode instead of the one we just * allocated. */ |
67a23c494
|
1071 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1072 1073 1074 1075 1076 1077 |
destroy_inode(inode); inode = old; wait_on_inode(inode); } return inode; } |
0b2d0724e
|
1078 |
EXPORT_SYMBOL(iget_locked); |
1da177e4c
|
1079 |
|
ad5e195ac
|
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 |
/* * search the inode cache for a matching inode number. * If we find one, then the inode number we are trying to * allocate is not unique and so we should not use it. * * Returns 1 if the inode number is unique, 0 if it is not. */ static int test_inode_iunique(struct super_block *sb, unsigned long ino) { struct hlist_head *b = inode_hashtable + hash(sb, ino); |
ad5e195ac
|
1090 |
struct inode *inode; |
67a23c494
|
1091 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1092 |
hlist_for_each_entry(inode, b, i_hash) { |
67a23c494
|
1093 1094 |
if (inode->i_ino == ino && inode->i_sb == sb) { spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1095 |
return 0; |
67a23c494
|
1096 |
} |
ad5e195ac
|
1097 |
} |
67a23c494
|
1098 |
spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1099 1100 1101 |
return 1; } |
1da177e4c
|
1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 |
/** * iunique - get a unique inode number * @sb: superblock * @max_reserved: highest reserved inode number * * Obtain an inode number that is unique on the system for a given * superblock. This is used by file systems that have no natural * permanent inode numbering system. An inode number is returned that * is higher than the reserved limit but unique. * * BUGS: * With a large number of inodes live on the file system this function * currently becomes quite slow. */ ino_t iunique(struct super_block *sb, ino_t max_reserved) { |
866b04fcc
|
1118 1119 1120 1121 1122 |
/* * On a 32bit, non LFS stat() call, glibc will generate an EOVERFLOW * error if st_ino won't fit in target struct field. Use 32bit counter * here to attempt to avoid that. */ |
ad5e195ac
|
1123 |
static DEFINE_SPINLOCK(iunique_lock); |
866b04fcc
|
1124 |
static unsigned int counter; |
1da177e4c
|
1125 |
ino_t res; |
3361c7beb
|
1126 |
|
ad5e195ac
|
1127 |
spin_lock(&iunique_lock); |
3361c7beb
|
1128 1129 1130 |
do { if (counter <= max_reserved) counter = max_reserved + 1; |
1da177e4c
|
1131 |
res = counter++; |
ad5e195ac
|
1132 1133 |
} while (!test_inode_iunique(sb, res)); spin_unlock(&iunique_lock); |
1da177e4c
|
1134 |
|
3361c7beb
|
1135 1136 |
return res; } |
1da177e4c
|
1137 1138 1139 1140 |
EXPORT_SYMBOL(iunique); struct inode *igrab(struct inode *inode) { |
250df6ed2
|
1141 1142 |
spin_lock(&inode->i_lock); if (!(inode->i_state & (I_FREEING|I_WILL_FREE))) { |
1da177e4c
|
1143 |
__iget(inode); |
250df6ed2
|
1144 1145 1146 |
spin_unlock(&inode->i_lock); } else { spin_unlock(&inode->i_lock); |
1da177e4c
|
1147 1148 1149 1150 1151 1152 |
/* * Handle the case where s_op->clear_inode is not been * called yet, and somebody is calling igrab * while the inode is getting freed. */ inode = NULL; |
250df6ed2
|
1153 |
} |
1da177e4c
|
1154 1155 |
return inode; } |
1da177e4c
|
1156 1157 1158 |
EXPORT_SYMBOL(igrab); /** |
0b2d0724e
|
1159 |
* ilookup5_nowait - search for an inode in the inode cache |
1da177e4c
|
1160 |
* @sb: super block of file system to search |
0b2d0724e
|
1161 |
* @hashval: hash value (usually inode number) to search for |
1da177e4c
|
1162 1163 |
* @test: callback used for comparisons between inodes * @data: opaque data pointer to pass to @test |
1da177e4c
|
1164 |
* |
0b2d0724e
|
1165 |
* Search for the inode specified by @hashval and @data in the inode cache. |
1da177e4c
|
1166 1167 1168 |
* If the inode is in the cache, the inode is returned with an incremented * reference count. * |
0b2d0724e
|
1169 1170 |
* Note: I_NEW is not waited upon so you have to be very careful what you do * with the returned inode. You probably should be using ilookup5() instead. |
1da177e4c
|
1171 |
* |
b6d0ad686
|
1172 |
* Note2: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1173 |
*/ |
0b2d0724e
|
1174 1175 |
struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
1176 |
{ |
0b2d0724e
|
1177 |
struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
1da177e4c
|
1178 |
struct inode *inode; |
67a23c494
|
1179 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1180 |
inode = find_inode(sb, head, test, data); |
67a23c494
|
1181 |
spin_unlock(&inode_hash_lock); |
88bd5121d
|
1182 |
|
0b2d0724e
|
1183 |
return inode; |
88bd5121d
|
1184 |
} |
88bd5121d
|
1185 1186 1187 1188 1189 1190 1191 1192 1193 |
EXPORT_SYMBOL(ilookup5_nowait); /** * ilookup5 - search for an inode in the inode cache * @sb: super block of file system to search * @hashval: hash value (usually inode number) to search for * @test: callback used for comparisons between inodes * @data: opaque data pointer to pass to @test * |
0b2d0724e
|
1194 1195 1196 |
* Search for the inode specified by @hashval and @data in the inode cache, * and if the inode is in the cache, return the inode with an incremented * reference count. Waits on I_NEW before returning the inode. |
88bd5121d
|
1197 |
* returned with an incremented reference count. |
1da177e4c
|
1198 |
* |
0b2d0724e
|
1199 1200 |
* This is a generalized version of ilookup() for file systems where the * inode number is not sufficient for unique identification of an inode. |
1da177e4c
|
1201 |
* |
0b2d0724e
|
1202 |
* Note: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1203 1204 1205 1206 |
*/ struct inode *ilookup5(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { |
0b2d0724e
|
1207 |
struct inode *inode = ilookup5_nowait(sb, hashval, test, data); |
1da177e4c
|
1208 |
|
0b2d0724e
|
1209 1210 1211 |
if (inode) wait_on_inode(inode); return inode; |
1da177e4c
|
1212 |
} |
1da177e4c
|
1213 1214 1215 1216 1217 1218 1219 |
EXPORT_SYMBOL(ilookup5); /** * ilookup - search for an inode in the inode cache * @sb: super block of file system to search * @ino: inode number to search for * |
0b2d0724e
|
1220 1221 |
* Search for the inode @ino in the inode cache, and if the inode is in the * cache, the inode is returned with an incremented reference count. |
1da177e4c
|
1222 1223 1224 1225 |
*/ struct inode *ilookup(struct super_block *sb, unsigned long ino) { struct hlist_head *head = inode_hashtable + hash(sb, ino); |
1da177e4c
|
1226 |
struct inode *inode; |
0b2d0724e
|
1227 1228 1229 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); |
1da177e4c
|
1230 |
|
1da177e4c
|
1231 |
if (inode) |
0b2d0724e
|
1232 1233 |
wait_on_inode(inode); return inode; |
1da177e4c
|
1234 |
} |
0b2d0724e
|
1235 |
EXPORT_SYMBOL(ilookup); |
1da177e4c
|
1236 |
|
261bca86e
|
1237 1238 1239 1240 1241 |
int insert_inode_locked(struct inode *inode) { struct super_block *sb = inode->i_sb; ino_t ino = inode->i_ino; struct hlist_head *head = inode_hashtable + hash(sb, ino); |
261bca86e
|
1242 |
|
261bca86e
|
1243 |
while (1) { |
72a43d63c
|
1244 |
struct inode *old = NULL; |
67a23c494
|
1245 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1246 |
hlist_for_each_entry(old, head, i_hash) { |
72a43d63c
|
1247 1248 1249 1250 |
if (old->i_ino != ino) continue; if (old->i_sb != sb) continue; |
250df6ed2
|
1251 1252 1253 |
spin_lock(&old->i_lock); if (old->i_state & (I_FREEING|I_WILL_FREE)) { spin_unlock(&old->i_lock); |
72a43d63c
|
1254 |
continue; |
250df6ed2
|
1255 |
} |
72a43d63c
|
1256 1257 |
break; } |
b67bfe0d4
|
1258 |
if (likely(!old)) { |
250df6ed2
|
1259 1260 |
spin_lock(&inode->i_lock); inode->i_state |= I_NEW; |
261bca86e
|
1261 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1262 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1263 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1264 1265 1266 |
return 0; } __iget(old); |
250df6ed2
|
1267 |
spin_unlock(&old->i_lock); |
67a23c494
|
1268 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1269 |
wait_on_inode(old); |
1d3382cbf
|
1270 |
if (unlikely(!inode_unhashed(old))) { |
261bca86e
|
1271 1272 1273 1274 1275 1276 |
iput(old); return -EBUSY; } iput(old); } } |
261bca86e
|
1277 1278 1279 1280 1281 1282 1283 |
EXPORT_SYMBOL(insert_inode_locked); int insert_inode_locked4(struct inode *inode, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { struct super_block *sb = inode->i_sb; struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
261bca86e
|
1284 |
|
261bca86e
|
1285 |
while (1) { |
72a43d63c
|
1286 |
struct inode *old = NULL; |
67a23c494
|
1287 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1288 |
hlist_for_each_entry(old, head, i_hash) { |
72a43d63c
|
1289 1290 1291 1292 |
if (old->i_sb != sb) continue; if (!test(old, data)) continue; |
250df6ed2
|
1293 1294 1295 |
spin_lock(&old->i_lock); if (old->i_state & (I_FREEING|I_WILL_FREE)) { spin_unlock(&old->i_lock); |
72a43d63c
|
1296 |
continue; |
250df6ed2
|
1297 |
} |
72a43d63c
|
1298 1299 |
break; } |
b67bfe0d4
|
1300 |
if (likely(!old)) { |
250df6ed2
|
1301 1302 |
spin_lock(&inode->i_lock); inode->i_state |= I_NEW; |
261bca86e
|
1303 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1304 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1305 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1306 1307 1308 |
return 0; } __iget(old); |
250df6ed2
|
1309 |
spin_unlock(&old->i_lock); |
67a23c494
|
1310 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1311 |
wait_on_inode(old); |
1d3382cbf
|
1312 |
if (unlikely(!inode_unhashed(old))) { |
261bca86e
|
1313 1314 1315 1316 1317 1318 |
iput(old); return -EBUSY; } iput(old); } } |
261bca86e
|
1319 |
EXPORT_SYMBOL(insert_inode_locked4); |
1da177e4c
|
1320 |
|
45321ac54
|
1321 1322 1323 1324 1325 |
int generic_delete_inode(struct inode *inode) { return 1; } EXPORT_SYMBOL(generic_delete_inode); |
1da177e4c
|
1326 |
/* |
45321ac54
|
1327 1328 |
* Called when we're dropping the last reference * to an inode. |
22fe40421
|
1329 |
* |
45321ac54
|
1330 1331 1332 1333 1334 |
* Call the FS "drop_inode()" function, defaulting to * the legacy UNIX filesystem behaviour. If it tells * us to evict inode, do so. Otherwise, retain inode * in cache if fs is alive, sync and evict if fs is * shutting down. |
22fe40421
|
1335 |
*/ |
45321ac54
|
1336 |
static void iput_final(struct inode *inode) |
1da177e4c
|
1337 1338 |
{ struct super_block *sb = inode->i_sb; |
45321ac54
|
1339 1340 |
const struct super_operations *op = inode->i_sb->s_op; int drop; |
250df6ed2
|
1341 |
WARN_ON(inode->i_state & I_NEW); |
e7f590970
|
1342 |
if (op->drop_inode) |
45321ac54
|
1343 1344 1345 |
drop = op->drop_inode(inode); else drop = generic_drop_inode(inode); |
1da177e4c
|
1346 |
|
b2b2af8e6
|
1347 1348 |
if (!drop && (sb->s_flags & MS_ACTIVE)) { inode->i_state |= I_REFERENCED; |
4eff96dd5
|
1349 |
inode_add_lru(inode); |
b2b2af8e6
|
1350 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1351 1352 |
return; } |
45321ac54
|
1353 |
if (!drop) { |
991114c6f
|
1354 |
inode->i_state |= I_WILL_FREE; |
250df6ed2
|
1355 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1356 |
write_inode_now(inode, 1); |
250df6ed2
|
1357 |
spin_lock(&inode->i_lock); |
7ef0d7377
|
1358 |
WARN_ON(inode->i_state & I_NEW); |
991114c6f
|
1359 |
inode->i_state &= ~I_WILL_FREE; |
1da177e4c
|
1360 |
} |
7ccf19a80
|
1361 |
|
991114c6f
|
1362 |
inode->i_state |= I_FREEING; |
c4ae0c654
|
1363 1364 |
if (!list_empty(&inode->i_lru)) inode_lru_list_del(inode); |
b2b2af8e6
|
1365 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1366 |
|
644da5960
|
1367 |
evict(inode); |
1da177e4c
|
1368 |
} |
1da177e4c
|
1369 |
/** |
6b3304b53
|
1370 |
* iput - put an inode |
1da177e4c
|
1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 |
* @inode: inode to put * * Puts an inode, dropping its usage count. If the inode use count hits * zero, the inode is then freed and may also be destroyed. * * Consequently, iput() can sleep. */ void iput(struct inode *inode) { if (inode) { |
a4ffdde6e
|
1381 |
BUG_ON(inode->i_state & I_CLEAR); |
1da177e4c
|
1382 |
|
f283c86af
|
1383 |
if (atomic_dec_and_lock(&inode->i_count, &inode->i_lock)) |
1da177e4c
|
1384 1385 1386 |
iput_final(inode); } } |
1da177e4c
|
1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 |
EXPORT_SYMBOL(iput); /** * bmap - find a block number in a file * @inode: inode of file * @block: block to find * * Returns the block number on the device holding the inode that * is the disk block number for the block of the file requested. * That is, asked for block 4 of inode 1 the function will return the |
6b3304b53
|
1397 |
* disk block relative to the disk start that holds that block of the |
1da177e4c
|
1398 1399 |
* file. */ |
6b3304b53
|
1400 |
sector_t bmap(struct inode *inode, sector_t block) |
1da177e4c
|
1401 1402 1403 1404 1405 1406 |
{ sector_t res = 0; if (inode->i_mapping->a_ops->bmap) res = inode->i_mapping->a_ops->bmap(inode->i_mapping, block); return res; } |
1da177e4c
|
1407 |
EXPORT_SYMBOL(bmap); |
11ff6f05f
|
1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 |
/* * With relative atime, only update atime if the previous atime is * earlier than either the ctime or mtime or if at least a day has * passed since the last atime update. */ static int relatime_need_update(struct vfsmount *mnt, struct inode *inode, struct timespec now) { if (!(mnt->mnt_flags & MNT_RELATIME)) return 1; /* * Is mtime younger than atime? If yes, update atime: */ if (timespec_compare(&inode->i_mtime, &inode->i_atime) >= 0) return 1; /* * Is ctime younger than atime? If yes, update atime: */ if (timespec_compare(&inode->i_ctime, &inode->i_atime) >= 0) return 1; /* * Is the previous atime value older than a day? If yes, * update atime: */ if ((long)(now.tv_sec - inode->i_atime.tv_sec) >= 24*60*60) return 1; /* * Good, we can skip the atime update: */ return 0; } |
c3b2da314
|
1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 |
/* * This does the actual work of updating an inodes time or version. Must have * had called mnt_want_write() before calling this. */ static int update_time(struct inode *inode, struct timespec *time, int flags) { if (inode->i_op->update_time) return inode->i_op->update_time(inode, time, flags); if (flags & S_ATIME) inode->i_atime = *time; if (flags & S_VERSION) inode_inc_iversion(inode); if (flags & S_CTIME) inode->i_ctime = *time; if (flags & S_MTIME) inode->i_mtime = *time; mark_inode_dirty_sync(inode); return 0; } |
1da177e4c
|
1461 |
/** |
869243a0f
|
1462 |
* touch_atime - update the access time |
185553b22
|
1463 |
* @path: the &struct path to update |
1da177e4c
|
1464 1465 1466 1467 1468 |
* * Update the accessed time on an inode and mark it for writeback. * This function automatically handles read only file systems and media, * as well as the "noatime" flag and inode specific "noatime" markers. */ |
badcf2b7b
|
1469 |
void touch_atime(const struct path *path) |
1da177e4c
|
1470 |
{ |
68ac1234f
|
1471 1472 |
struct vfsmount *mnt = path->mnt; struct inode *inode = path->dentry->d_inode; |
1da177e4c
|
1473 |
struct timespec now; |
cdb70f3f7
|
1474 |
if (inode->i_flags & S_NOATIME) |
b12536c27
|
1475 |
return; |
37756ced1
|
1476 |
if (IS_NOATIME(inode)) |
b12536c27
|
1477 |
return; |
b22761384
|
1478 |
if ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode)) |
b12536c27
|
1479 |
return; |
47ae32d6a
|
1480 |
|
cdb70f3f7
|
1481 |
if (mnt->mnt_flags & MNT_NOATIME) |
b12536c27
|
1482 |
return; |
cdb70f3f7
|
1483 |
if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)) |
b12536c27
|
1484 |
return; |
1da177e4c
|
1485 1486 |
now = current_fs_time(inode->i_sb); |
11ff6f05f
|
1487 1488 |
if (!relatime_need_update(mnt, inode, now)) |
b12536c27
|
1489 |
return; |
11ff6f05f
|
1490 |
|
47ae32d6a
|
1491 |
if (timespec_equal(&inode->i_atime, &now)) |
b12536c27
|
1492 |
return; |
5d37e9e6d
|
1493 |
if (!sb_start_write_trylock(inode->i_sb)) |
b12536c27
|
1494 |
return; |
47ae32d6a
|
1495 |
|
5d37e9e6d
|
1496 1497 |
if (__mnt_want_write(mnt)) goto skip_update; |
c3b2da314
|
1498 1499 1500 1501 1502 1503 |
/* * File systems can error out when updating inodes if they need to * allocate new space to modify an inode (such is the case for * Btrfs), but since we touch atime while walking down the path we * really don't care if we failed to update the atime of the file, * so just ignore the return value. |
2bc556528
|
1504 1505 |
* We may also fail on filesystems that have the ability to make parts * of the fs read only, e.g. subvolumes in Btrfs. |
c3b2da314
|
1506 1507 |
*/ update_time(inode, &now, S_ATIME); |
5d37e9e6d
|
1508 1509 1510 |
__mnt_drop_write(mnt); skip_update: sb_end_write(inode->i_sb); |
1da177e4c
|
1511 |
} |
869243a0f
|
1512 |
EXPORT_SYMBOL(touch_atime); |
1da177e4c
|
1513 |
|
3ed37648e
|
1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 |
/* * The logic we want is * * if suid or (sgid and xgrp) * remove privs */ int should_remove_suid(struct dentry *dentry) { umode_t mode = dentry->d_inode->i_mode; int kill = 0; /* suid always must be killed */ if (unlikely(mode & S_ISUID)) kill = ATTR_KILL_SUID; /* * sgid without any exec bits is just a mandatory locking mark; leave * it alone. If some exec bits are set, it's a real sgid; kill it. */ if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) kill |= ATTR_KILL_SGID; if (unlikely(kill && !capable(CAP_FSETID) && S_ISREG(mode))) return kill; return 0; } EXPORT_SYMBOL(should_remove_suid); static int __remove_suid(struct dentry *dentry, int kill) { struct iattr newattrs; newattrs.ia_valid = ATTR_FORCE | kill; |
27ac0ffea
|
1548 1549 1550 1551 1552 |
/* * Note we call this on write, so notify_change will not * encounter any conflicting delegations: */ return notify_change(dentry, &newattrs, NULL); |
3ed37648e
|
1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 |
} int file_remove_suid(struct file *file) { struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; int killsuid; int killpriv; int error = 0; /* Fast path for nothing security related */ if (IS_NOSEC(inode)) return 0; killsuid = should_remove_suid(dentry); killpriv = security_inode_need_killpriv(dentry); if (killpriv < 0) return killpriv; if (killpriv) error = security_inode_killpriv(dentry); if (!error && killsuid) error = __remove_suid(dentry, killsuid); if (!error && (inode->i_sb->s_flags & MS_NOSEC)) inode->i_flags |= S_NOSEC; return error; } EXPORT_SYMBOL(file_remove_suid); |
1da177e4c
|
1582 |
/** |
870f48179
|
1583 1584 |
* file_update_time - update mtime and ctime time * @file: file accessed |
1da177e4c
|
1585 |
* |
870f48179
|
1586 1587 1588 1589 |
* Update the mtime and ctime members of an inode and mark the inode * for writeback. Note that this function is meant exclusively for * usage in the file write path of filesystems, and filesystems may * choose to explicitly ignore update via this function with the |
2eadfc0ed
|
1590 |
* S_NOCMTIME inode flag, e.g. for network filesystem where these |
c3b2da314
|
1591 1592 |
* timestamps are handled by the server. This can return an error for * file systems who need to allocate space in order to update an inode. |
1da177e4c
|
1593 |
*/ |
c3b2da314
|
1594 |
int file_update_time(struct file *file) |
1da177e4c
|
1595 |
{ |
496ad9aa8
|
1596 |
struct inode *inode = file_inode(file); |
1da177e4c
|
1597 |
struct timespec now; |
c3b2da314
|
1598 1599 |
int sync_it = 0; int ret; |
1da177e4c
|
1600 |
|
ce06e0b21
|
1601 |
/* First try to exhaust all avenues to not sync */ |
1da177e4c
|
1602 |
if (IS_NOCMTIME(inode)) |
c3b2da314
|
1603 |
return 0; |
20ddee2c7
|
1604 |
|
1da177e4c
|
1605 |
now = current_fs_time(inode->i_sb); |
ce06e0b21
|
1606 1607 |
if (!timespec_equal(&inode->i_mtime, &now)) sync_it = S_MTIME; |
1da177e4c
|
1608 |
|
ce06e0b21
|
1609 1610 |
if (!timespec_equal(&inode->i_ctime, &now)) sync_it |= S_CTIME; |
870f48179
|
1611 |
|
ce06e0b21
|
1612 1613 |
if (IS_I_VERSION(inode)) sync_it |= S_VERSION; |
7a224228e
|
1614 |
|
ce06e0b21
|
1615 |
if (!sync_it) |
c3b2da314
|
1616 |
return 0; |
ce06e0b21
|
1617 1618 |
/* Finally allowed to write? Takes lock. */ |
eb04c2828
|
1619 |
if (__mnt_want_write_file(file)) |
c3b2da314
|
1620 |
return 0; |
ce06e0b21
|
1621 |
|
c3b2da314
|
1622 |
ret = update_time(inode, &now, sync_it); |
eb04c2828
|
1623 |
__mnt_drop_write_file(file); |
c3b2da314
|
1624 1625 |
return ret; |
1da177e4c
|
1626 |
} |
870f48179
|
1627 |
EXPORT_SYMBOL(file_update_time); |
1da177e4c
|
1628 1629 1630 1631 1632 1633 1634 1635 1636 |
int inode_needs_sync(struct inode *inode) { if (IS_SYNC(inode)) return 1; if (S_ISDIR(inode->i_mode) && IS_DIRSYNC(inode)) return 1; return 0; } |
1da177e4c
|
1637 |
EXPORT_SYMBOL(inode_needs_sync); |
1da177e4c
|
1638 1639 1640 1641 1642 |
int inode_wait(void *word) { schedule(); return 0; } |
d44dab8d1
|
1643 |
EXPORT_SYMBOL(inode_wait); |
1da177e4c
|
1644 1645 |
/* |
168a9fd6a
|
1646 1647 1648 1649 1650 1651 |
* If we try to find an inode in the inode hash while it is being * deleted, we have to wait until the filesystem completes its * deletion before reporting that it isn't found. This function waits * until the deletion _might_ have completed. Callers are responsible * to recheck inode state. * |
eaff8079d
|
1652 |
* It doesn't matter if I_NEW is not set initially, a call to |
250df6ed2
|
1653 1654 |
* wake_up_bit(&inode->i_state, __I_NEW) after removing from the hash list * will DTRT. |
1da177e4c
|
1655 1656 1657 1658 |
*/ static void __wait_on_freeing_inode(struct inode *inode) { wait_queue_head_t *wq; |
eaff8079d
|
1659 1660 |
DEFINE_WAIT_BIT(wait, &inode->i_state, __I_NEW); wq = bit_waitqueue(&inode->i_state, __I_NEW); |
1da177e4c
|
1661 |
prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE); |
250df6ed2
|
1662 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1663 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1664 1665 |
schedule(); finish_wait(wq, &wait.wait); |
67a23c494
|
1666 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1667 |
} |
1da177e4c
|
1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 |
static __initdata unsigned long ihash_entries; static int __init set_ihash_entries(char *str) { if (!str) return 0; ihash_entries = simple_strtoul(str, &str, 0); return 1; } __setup("ihash_entries=", set_ihash_entries); /* * Initialize the waitqueues and inode hash table. */ void __init inode_init_early(void) { |
074b85175
|
1683 |
unsigned int loop; |
1da177e4c
|
1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 |
/* If hashes are distributed across NUMA nodes, defer * hash allocation until vmalloc space is available. */ if (hashdist) return; inode_hashtable = alloc_large_system_hash("Inode-cache", sizeof(struct hlist_head), ihash_entries, 14, HASH_EARLY, &i_hash_shift, &i_hash_mask, |
31fe62b95
|
1699 |
0, |
1da177e4c
|
1700 |
0); |
074b85175
|
1701 |
for (loop = 0; loop < (1U << i_hash_shift); loop++) |
1da177e4c
|
1702 1703 |
INIT_HLIST_HEAD(&inode_hashtable[loop]); } |
74bf17cff
|
1704 |
void __init inode_init(void) |
1da177e4c
|
1705 |
{ |
074b85175
|
1706 |
unsigned int loop; |
1da177e4c
|
1707 1708 |
/* inode slab cache */ |
b0196009d
|
1709 1710 1711 1712 1713 |
inode_cachep = kmem_cache_create("inode_cache", sizeof(struct inode), 0, (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC| SLAB_MEM_SPREAD), |
20c2df83d
|
1714 |
init_once); |
1da177e4c
|
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 |
/* Hash may have been set up in inode_init_early */ if (!hashdist) return; inode_hashtable = alloc_large_system_hash("Inode-cache", sizeof(struct hlist_head), ihash_entries, 14, 0, &i_hash_shift, &i_hash_mask, |
31fe62b95
|
1728 |
0, |
1da177e4c
|
1729 |
0); |
074b85175
|
1730 |
for (loop = 0; loop < (1U << i_hash_shift); loop++) |
1da177e4c
|
1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 |
INIT_HLIST_HEAD(&inode_hashtable[loop]); } void init_special_inode(struct inode *inode, umode_t mode, dev_t rdev) { inode->i_mode = mode; if (S_ISCHR(mode)) { inode->i_fop = &def_chr_fops; inode->i_rdev = rdev; } else if (S_ISBLK(mode)) { inode->i_fop = &def_blk_fops; inode->i_rdev = rdev; } else if (S_ISFIFO(mode)) |
599a0ac14
|
1744 |
inode->i_fop = &pipefifo_fops; |
1da177e4c
|
1745 1746 1747 |
else if (S_ISSOCK(mode)) inode->i_fop = &bad_sock_fops; else |
af0d9ae81
|
1748 1749 1750 1751 |
printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o) for" " inode %s:%lu ", mode, inode->i_sb->s_id, inode->i_ino); |
1da177e4c
|
1752 1753 |
} EXPORT_SYMBOL(init_special_inode); |
a1bd120d1
|
1754 1755 |
/** |
eaae668d0
|
1756 |
* inode_init_owner - Init uid,gid,mode for new inode according to posix standards |
a1bd120d1
|
1757 1758 1759 1760 1761 |
* @inode: New inode * @dir: Directory inode * @mode: mode of the new inode */ void inode_init_owner(struct inode *inode, const struct inode *dir, |
62bb10917
|
1762 |
umode_t mode) |
a1bd120d1
|
1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 |
{ inode->i_uid = current_fsuid(); if (dir && dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; if (S_ISDIR(mode)) mode |= S_ISGID; } else inode->i_gid = current_fsgid(); inode->i_mode = mode; } EXPORT_SYMBOL(inode_init_owner); |
e795b7179
|
1774 |
|
2e1496707
|
1775 1776 1777 1778 1779 1780 |
/** * inode_owner_or_capable - check current task permissions to inode * @inode: inode being checked * * Return true if current either has CAP_FOWNER to the inode, or * owns the file. |
e795b7179
|
1781 |
*/ |
2e1496707
|
1782 |
bool inode_owner_or_capable(const struct inode *inode) |
e795b7179
|
1783 |
{ |
92361636e
|
1784 |
if (uid_eq(current_fsuid(), inode->i_uid)) |
e795b7179
|
1785 |
return true; |
1a48e2ac0
|
1786 |
if (inode_capable(inode, CAP_FOWNER)) |
e795b7179
|
1787 1788 1789 |
return true; return false; } |
2e1496707
|
1790 |
EXPORT_SYMBOL(inode_owner_or_capable); |
1d59d61f6
|
1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 |
/* * Direct i/o helper functions */ static void __inode_dio_wait(struct inode *inode) { wait_queue_head_t *wq = bit_waitqueue(&inode->i_state, __I_DIO_WAKEUP); DEFINE_WAIT_BIT(q, &inode->i_state, __I_DIO_WAKEUP); do { prepare_to_wait(wq, &q.wait, TASK_UNINTERRUPTIBLE); if (atomic_read(&inode->i_dio_count)) schedule(); } while (atomic_read(&inode->i_dio_count)); finish_wait(wq, &q.wait); } /** * inode_dio_wait - wait for outstanding DIO requests to finish * @inode: inode to wait for * * Waits for all pending direct I/O requests to finish so that we can * proceed with a truncate or equivalent operation. * * Must be called under a lock that serializes taking new references * to i_dio_count, usually by inode->i_mutex. */ void inode_dio_wait(struct inode *inode) { if (atomic_read(&inode->i_dio_count)) __inode_dio_wait(inode); } EXPORT_SYMBOL(inode_dio_wait); /* * inode_dio_done - signal finish of a direct I/O requests * @inode: inode the direct I/O happens on * * This is called once we've finished processing a direct I/O request, * and is used to wake up callers waiting for direct I/O to be quiesced. */ void inode_dio_done(struct inode *inode) { if (atomic_dec_and_test(&inode->i_dio_count)) wake_up_bit(&inode->i_state, __I_DIO_WAKEUP); } EXPORT_SYMBOL(inode_dio_done); |
5f16f3225
|
1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 |
/* * inode_set_flags - atomically set some inode flags * * Note: the caller should be holding i_mutex, or else be sure that * they have exclusive access to the inode structure (i.e., while the * inode is being instantiated). The reason for the cmpxchg() loop * --- which wouldn't be necessary if all code paths which modify * i_flags actually followed this rule, is that there is at least one * code path which doesn't today --- for example, * __generic_file_aio_write() calls file_remove_suid() without holding * i_mutex --- so we use cmpxchg() out of an abundance of caution. * * In the long run, i_mutex is overkill, and we should probably look * at using the i_lock spinlock to protect i_flags, and then make sure * it is so documented in include/linux/fs.h and that all code follows * the locking convention!! */ void inode_set_flags(struct inode *inode, unsigned int flags, unsigned int mask) { unsigned int old_flags, new_flags; WARN_ON_ONCE(flags & ~mask); do { old_flags = ACCESS_ONCE(inode->i_flags); new_flags = (old_flags & ~mask) | flags; } while (unlikely(cmpxchg(&inode->i_flags, old_flags, new_flags) != old_flags)); } EXPORT_SYMBOL(inode_set_flags); |