Blame view
fs/inode.c
60.8 KB
457c89965
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
1da177e4c
|
2 |
/* |
1da177e4c
|
3 |
* (C) 1997 Linus Torvalds |
4b4563dc8
|
4 |
* (C) 1999 Andrea Arcangeli <andrea@suse.de> (dynamic inode allocation) |
1da177e4c
|
5 |
*/ |
e59cc473c
|
6 |
#include <linux/export.h> |
1da177e4c
|
7 8 |
#include <linux/fs.h> #include <linux/mm.h> |
1da177e4c
|
9 |
#include <linux/backing-dev.h> |
1da177e4c
|
10 11 12 |
#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> |
1da177e4c
|
13 |
#include <linux/cdev.h> |
57c8a661d
|
14 |
#include <linux/memblock.h> |
3be25f49b
|
15 |
#include <linux/fsnotify.h> |
fc33a7bb9
|
16 |
#include <linux/mount.h> |
f19d4a8fa
|
17 |
#include <linux/posix_acl.h> |
9ce6e0be0
|
18 |
#include <linux/prefetch.h> |
4b4563dc8
|
19 |
#include <linux/buffer_head.h> /* for inode_has_buffers */ |
7ada4db88
|
20 |
#include <linux/ratelimit.h> |
bc3b14cb2
|
21 |
#include <linux/list_lru.h> |
ae5e165d8
|
22 |
#include <linux/iversion.h> |
0ae45f63d
|
23 |
#include <trace/events/writeback.h> |
a66979aba
|
24 |
#include "internal.h" |
1da177e4c
|
25 26 |
/* |
4b4563dc8
|
27 |
* Inode locking rules: |
250df6ed2
|
28 29 30 |
* * inode->i_lock protects: * inode->i_state, inode->i_hash, __iget() |
bc3b14cb2
|
31 |
* Inode LRU list locks protect: |
98b745c64
|
32 |
* inode->i_sb->s_inode_lru, inode->i_lru |
74278da9f
|
33 34 |
* inode->i_sb->s_inode_list_lock protects: * inode->i_sb->s_inodes, inode->i_sb_list |
f758eeabe
|
35 |
* bdi->wb.list_lock protects: |
c7f540849
|
36 |
* bdi->wb.b_{dirty,io,more_io,dirty_time}, inode->i_io_list |
67a23c494
|
37 38 |
* inode_hash_lock protects: * inode_hashtable, inode->i_hash |
250df6ed2
|
39 40 |
* * Lock ordering: |
55fa6091d
|
41 |
* |
74278da9f
|
42 |
* inode->i_sb->s_inode_list_lock |
55fa6091d
|
43 |
* inode->i_lock |
bc3b14cb2
|
44 |
* Inode LRU list locks |
a66979aba
|
45 |
* |
f758eeabe
|
46 |
* bdi->wb.list_lock |
a66979aba
|
47 |
* inode->i_lock |
67a23c494
|
48 49 |
* * inode_hash_lock |
74278da9f
|
50 |
* inode->i_sb->s_inode_list_lock |
67a23c494
|
51 52 53 54 |
* inode->i_lock * * iunique_lock * inode_hash_lock |
250df6ed2
|
55 |
*/ |
fa3536cc1
|
56 57 |
static unsigned int i_hash_mask __read_mostly; static unsigned int i_hash_shift __read_mostly; |
67a23c494
|
58 59 |
static struct hlist_head *inode_hashtable __read_mostly; static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock); |
1da177e4c
|
60 |
|
1da177e4c
|
61 |
/* |
7dcda1c96
|
62 63 64 65 66 67 68 69 |
* 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
|
70 71 72 |
* Statistics gathering.. */ struct inodes_stat_t inodes_stat; |
3942c07cc
|
73 74 |
static DEFINE_PER_CPU(unsigned long, nr_inodes); static DEFINE_PER_CPU(unsigned long, nr_unused); |
cffbc8aa3
|
75 |
|
6b3304b53
|
76 |
static struct kmem_cache *inode_cachep __read_mostly; |
1da177e4c
|
77 |
|
3942c07cc
|
78 |
static long get_nr_inodes(void) |
cffbc8aa3
|
79 |
{ |
3e880fb5e
|
80 |
int i; |
3942c07cc
|
81 |
long sum = 0; |
3e880fb5e
|
82 83 84 |
for_each_possible_cpu(i) sum += per_cpu(nr_inodes, i); return sum < 0 ? 0 : sum; |
cffbc8aa3
|
85 |
} |
3942c07cc
|
86 |
static inline long get_nr_inodes_unused(void) |
cffbc8aa3
|
87 |
{ |
fcb94f72d
|
88 |
int i; |
3942c07cc
|
89 |
long sum = 0; |
fcb94f72d
|
90 91 92 |
for_each_possible_cpu(i) sum += per_cpu(nr_unused, i); return sum < 0 ? 0 : sum; |
cffbc8aa3
|
93 |
} |
3942c07cc
|
94 |
long get_nr_dirty_inodes(void) |
cffbc8aa3
|
95 |
{ |
3e880fb5e
|
96 |
/* not actually dirty inodes, but a wild approximation */ |
3942c07cc
|
97 |
long nr_dirty = get_nr_inodes() - get_nr_inodes_unused(); |
cffbc8aa3
|
98 |
return nr_dirty > 0 ? nr_dirty : 0; |
cffbc8aa3
|
99 100 101 102 103 104 |
} /* * Handle nr_inode sysctl */ #ifdef CONFIG_SYSCTL |
1f7e0616c
|
105 |
int proc_nr_inodes(struct ctl_table *table, int write, |
cffbc8aa3
|
106 107 108 |
void __user *buffer, size_t *lenp, loff_t *ppos) { inodes_stat.nr_inodes = get_nr_inodes(); |
fcb94f72d
|
109 |
inodes_stat.nr_unused = get_nr_inodes_unused(); |
3942c07cc
|
110 |
return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); |
cffbc8aa3
|
111 112 |
} #endif |
bd9b51e79
|
113 114 115 116 |
static int no_open(struct inode *inode, struct file *file) { return -ENXIO; } |
2cb1599f9
|
117 |
/** |
6e7c2b4dd
|
118 |
* inode_init_always - perform inode structure initialisation |
0bc02f3fa
|
119 120 |
* @sb: superblock inode belongs to * @inode: inode to initialise |
2cb1599f9
|
121 122 123 124 |
* * These are initializations that need to be done on every inode * allocation as the fields are not initialised by slab allocation. */ |
54e346215
|
125 |
int inode_init_always(struct super_block *sb, struct inode *inode) |
1da177e4c
|
126 |
{ |
6e1d5dcc2
|
127 |
static const struct inode_operations empty_iops; |
bd9b51e79
|
128 |
static const struct file_operations no_open_fops = {.open = no_open}; |
6b3304b53
|
129 |
struct address_space *const mapping = &inode->i_data; |
2cb1599f9
|
130 131 132 133 |
inode->i_sb = sb; inode->i_blkbits = sb->s_blocksize_bits; inode->i_flags = 0; |
553d46b07
|
134 |
atomic64_set(&inode->i_sequence, 0); |
2cb1599f9
|
135 136 |
atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; |
bd9b51e79
|
137 |
inode->i_fop = &no_open_fops; |
a78ef704a
|
138 |
inode->__i_nlink = 1; |
3ddcd0569
|
139 |
inode->i_opflags = 0; |
d0a5b995a
|
140 141 |
if (sb->s_xattr) inode->i_opflags |= IOP_XATTR; |
92361636e
|
142 143 |
i_uid_write(inode, 0); i_gid_write(inode, 0); |
2cb1599f9
|
144 145 |
atomic_set(&inode->i_writecount, 0); inode->i_size = 0; |
c75b1d942
|
146 |
inode->i_write_hint = WRITE_LIFE_NOT_SET; |
2cb1599f9
|
147 148 149 |
inode->i_blocks = 0; inode->i_bytes = 0; inode->i_generation = 0; |
2cb1599f9
|
150 151 152 |
inode->i_pipe = NULL; inode->i_bdev = NULL; inode->i_cdev = NULL; |
61ba64fc0
|
153 |
inode->i_link = NULL; |
84e710da2
|
154 |
inode->i_dir_seq = 0; |
2cb1599f9
|
155 156 |
inode->i_rdev = 0; inode->dirtied_when = 0; |
6146f0d5e
|
157 |
|
3d65ae463
|
158 159 160 161 162 |
#ifdef CONFIG_CGROUP_WRITEBACK inode->i_wb_frn_winner = 0; inode->i_wb_frn_avg_time = 0; inode->i_wb_frn_history = 0; #endif |
6146f0d5e
|
163 |
if (security_inode_alloc(inode)) |
54e346215
|
164 |
goto out; |
2cb1599f9
|
165 166 |
spin_lock_init(&inode->i_lock); lockdep_set_class(&inode->i_lock, &sb->s_type->i_lock_key); |
9902af79c
|
167 168 |
init_rwsem(&inode->i_rwsem); lockdep_set_class(&inode->i_rwsem, &sb->s_type->i_mutex_key); |
2cb1599f9
|
169 |
|
bd5fe6c5e
|
170 |
atomic_set(&inode->i_dio_count, 0); |
2cb1599f9
|
171 172 173 174 |
mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags = 0; |
829bc787c
|
175 |
mapping->wb_err = 0; |
4bb5f5d93
|
176 |
atomic_set(&mapping->i_mmap_writable, 0); |
09d91cda0
|
177 178 179 |
#ifdef CONFIG_READ_ONLY_THP_FOR_FS atomic_set(&mapping->nr_thps, 0); #endif |
3c1d43787
|
180 |
mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE); |
252aa6f5b
|
181 |
mapping->private_data = NULL; |
2cb1599f9
|
182 |
mapping->writeback_index = 0; |
2cb1599f9
|
183 184 |
inode->i_private = NULL; inode->i_mapping = mapping; |
b3d9b7a3c
|
185 |
INIT_HLIST_HEAD(&inode->i_dentry); /* buggered by rcu freeing */ |
f19d4a8fa
|
186 187 188 |
#ifdef CONFIG_FS_POSIX_ACL inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED; #endif |
2cb1599f9
|
189 |
|
3be25f49b
|
190 191 192 |
#ifdef CONFIG_FSNOTIFY inode->i_fsnotify_mask = 0; #endif |
4a075e39c
|
193 |
inode->i_flctx = NULL; |
3e880fb5e
|
194 |
this_cpu_inc(nr_inodes); |
cffbc8aa3
|
195 |
|
54e346215
|
196 |
return 0; |
54e346215
|
197 198 |
out: return -ENOMEM; |
1da177e4c
|
199 |
} |
2cb1599f9
|
200 |
EXPORT_SYMBOL(inode_init_always); |
fdb0da89f
|
201 202 203 204 205 206 207 208 209 210 211 212 213 214 |
void free_inode_nonrcu(struct inode *inode) { kmem_cache_free(inode_cachep, inode); } EXPORT_SYMBOL(free_inode_nonrcu); static void i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); if (inode->free_inode) inode->free_inode(inode); else free_inode_nonrcu(inode); } |
2cb1599f9
|
215 216 |
static struct inode *alloc_inode(struct super_block *sb) { |
fdb0da89f
|
217 |
const struct super_operations *ops = sb->s_op; |
2cb1599f9
|
218 |
struct inode *inode; |
fdb0da89f
|
219 220 |
if (ops->alloc_inode) inode = ops->alloc_inode(sb); |
2cb1599f9
|
221 222 |
else inode = kmem_cache_alloc(inode_cachep, GFP_KERNEL); |
54e346215
|
223 224 225 226 |
if (!inode) return NULL; if (unlikely(inode_init_always(sb, inode))) { |
fdb0da89f
|
227 228 229 230 231 232 233 |
if (ops->destroy_inode) { ops->destroy_inode(inode); if (!ops->free_inode) return NULL; } inode->free_inode = ops->free_inode; i_callback(&inode->i_rcu); |
54e346215
|
234 235 236 237 |
return NULL; } return inode; |
2cb1599f9
|
238 |
} |
1da177e4c
|
239 |
|
2e00c97e2
|
240 |
void __destroy_inode(struct inode *inode) |
1da177e4c
|
241 |
{ |
b7542f8c7
|
242 |
BUG_ON(inode_has_buffers(inode)); |
52ebea749
|
243 |
inode_detach_wb(inode); |
1da177e4c
|
244 |
security_inode_free(inode); |
3be25f49b
|
245 |
fsnotify_inode_delete(inode); |
f27a0fe08
|
246 |
locks_free_lock_context(inode); |
7ada4db88
|
247 248 249 250 |
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
|
251 |
#ifdef CONFIG_FS_POSIX_ACL |
b8a7a3a66
|
252 |
if (inode->i_acl && !is_uncached_acl(inode->i_acl)) |
f19d4a8fa
|
253 |
posix_acl_release(inode->i_acl); |
b8a7a3a66
|
254 |
if (inode->i_default_acl && !is_uncached_acl(inode->i_default_acl)) |
f19d4a8fa
|
255 256 |
posix_acl_release(inode->i_default_acl); #endif |
3e880fb5e
|
257 |
this_cpu_dec(nr_inodes); |
2e00c97e2
|
258 259 |
} EXPORT_SYMBOL(__destroy_inode); |
56b0dacfa
|
260 |
static void destroy_inode(struct inode *inode) |
2e00c97e2
|
261 |
{ |
fdb0da89f
|
262 |
const struct super_operations *ops = inode->i_sb->s_op; |
7ccf19a80
|
263 |
BUG_ON(!list_empty(&inode->i_lru)); |
2e00c97e2
|
264 |
__destroy_inode(inode); |
fdb0da89f
|
265 266 267 268 269 270 271 |
if (ops->destroy_inode) { ops->destroy_inode(inode); if (!ops->free_inode) return; } inode->free_inode = ops->free_inode; call_rcu(&inode->i_rcu, i_callback); |
1da177e4c
|
272 |
} |
1da177e4c
|
273 |
|
7ada4db88
|
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 |
/** * 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
|
322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 |
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
|
343 344 |
if (unlikely(inode->i_nlink == 0)) { WARN_ON(!(inode->i_state & I_LINKABLE)); |
7ada4db88
|
345 |
atomic_long_dec(&inode->i_sb->s_remove_count); |
f4e0c30c1
|
346 |
} |
7ada4db88
|
347 348 349 350 |
inode->__i_nlink++; } EXPORT_SYMBOL(inc_nlink); |
ae23395d8
|
351 |
static void __address_space_init_once(struct address_space *mapping) |
2aa15890f
|
352 |
{ |
7b785645e
|
353 |
xa_init_flags(&mapping->i_pages, XA_FLAGS_LOCK_IRQ | XA_FLAGS_ACCOUNT); |
c8c06efa8
|
354 |
init_rwsem(&mapping->i_mmap_rwsem); |
2aa15890f
|
355 356 |
INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); |
f808c13fd
|
357 |
mapping->i_mmap = RB_ROOT_CACHED; |
2aa15890f
|
358 |
} |
ae23395d8
|
359 360 361 362 363 364 |
void address_space_init_once(struct address_space *mapping) { memset(mapping, 0, sizeof(*mapping)); __address_space_init_once(mapping); } |
2aa15890f
|
365 |
EXPORT_SYMBOL(address_space_init_once); |
1da177e4c
|
366 367 368 369 370 371 372 373 374 |
/* * 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
|
375 |
INIT_LIST_HEAD(&inode->i_devices); |
c7f540849
|
376 |
INIT_LIST_HEAD(&inode->i_io_list); |
6c60d2b57
|
377 |
INIT_LIST_HEAD(&inode->i_wb_list); |
7ccf19a80
|
378 |
INIT_LIST_HEAD(&inode->i_lru); |
ae23395d8
|
379 |
__address_space_init_once(&inode->i_data); |
1da177e4c
|
380 381 |
i_size_ordered_init(inode); } |
1da177e4c
|
382 |
EXPORT_SYMBOL(inode_init_once); |
51cc50685
|
383 |
static void init_once(void *foo) |
1da177e4c
|
384 |
{ |
6b3304b53
|
385 |
struct inode *inode = (struct inode *) foo; |
1da177e4c
|
386 |
|
a35afb830
|
387 |
inode_init_once(inode); |
1da177e4c
|
388 389 390 |
} /* |
250df6ed2
|
391 |
* inode->i_lock must be held |
1da177e4c
|
392 |
*/ |
6b3304b53
|
393 |
void __iget(struct inode *inode) |
1da177e4c
|
394 |
{ |
9e38d86ff
|
395 396 |
atomic_inc(&inode->i_count); } |
2e147f1ef
|
397 |
|
7de9c6ee3
|
398 399 400 401 402 403 404 405 |
/* * 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
|
406 407 |
static void inode_lru_list_add(struct inode *inode) { |
bc3b14cb2
|
408 |
if (list_lru_add(&inode->i_sb->s_inode_lru, &inode->i_lru)) |
fcb94f72d
|
409 |
this_cpu_inc(nr_unused); |
563f40019
|
410 411 |
else inode->i_state |= I_REFERENCED; |
9e38d86ff
|
412 |
} |
2e147f1ef
|
413 |
|
4eff96dd5
|
414 415 416 417 418 419 420 |
/* * Add inode to LRU if needed (inode is unused and clean). * * Needs inode->i_lock held. */ void inode_add_lru(struct inode *inode) { |
0ae45f63d
|
421 422 |
if (!(inode->i_state & (I_DIRTY_ALL | I_SYNC | I_FREEING | I_WILL_FREE)) && |
1751e8a6c
|
423 |
!atomic_read(&inode->i_count) && inode->i_sb->s_flags & SB_ACTIVE) |
4eff96dd5
|
424 425 |
inode_lru_list_add(inode); } |
9e38d86ff
|
426 427 |
static void inode_lru_list_del(struct inode *inode) { |
bc3b14cb2
|
428 429 |
if (list_lru_del(&inode->i_sb->s_inode_lru, &inode->i_lru)) |
fcb94f72d
|
430 |
this_cpu_dec(nr_unused); |
1da177e4c
|
431 |
} |
646ec4615
|
432 433 434 435 436 437 |
/** * inode_sb_list_add - add inode to the superblock list of inodes * @inode: inode to add */ void inode_sb_list_add(struct inode *inode) { |
74278da9f
|
438 |
spin_lock(&inode->i_sb->s_inode_list_lock); |
55fa6091d
|
439 |
list_add(&inode->i_sb_list, &inode->i_sb->s_inodes); |
74278da9f
|
440 |
spin_unlock(&inode->i_sb->s_inode_list_lock); |
646ec4615
|
441 442 |
} EXPORT_SYMBOL_GPL(inode_sb_list_add); |
55fa6091d
|
443 |
static inline void inode_sb_list_del(struct inode *inode) |
646ec4615
|
444 |
{ |
a209dfc7b
|
445 |
if (!list_empty(&inode->i_sb_list)) { |
74278da9f
|
446 |
spin_lock(&inode->i_sb->s_inode_list_lock); |
a209dfc7b
|
447 |
list_del_init(&inode->i_sb_list); |
74278da9f
|
448 |
spin_unlock(&inode->i_sb->s_inode_list_lock); |
a209dfc7b
|
449 |
} |
646ec4615
|
450 |
} |
4c51acbc6
|
451 452 453 454 455 456 |
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
|
457 458 |
tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> i_hash_shift); return tmp & i_hash_mask; |
4c51acbc6
|
459 460 461 462 463 464 465 466 467 468 469 470 |
} /** * __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
|
471 |
struct hlist_head *b = inode_hashtable + hash(inode->i_sb, hashval); |
67a23c494
|
472 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
473 |
spin_lock(&inode->i_lock); |
646ec4615
|
474 |
hlist_add_head(&inode->i_hash, b); |
250df6ed2
|
475 |
spin_unlock(&inode->i_lock); |
67a23c494
|
476 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
477 478 479 480 |
} EXPORT_SYMBOL(__insert_inode_hash); /** |
f2ee7abf4
|
481 |
* __remove_inode_hash - remove an inode from the hash |
4c51acbc6
|
482 483 484 485 |
* @inode: inode to unhash * * Remove an inode from the superblock. */ |
f2ee7abf4
|
486 |
void __remove_inode_hash(struct inode *inode) |
4c51acbc6
|
487 |
{ |
67a23c494
|
488 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
489 |
spin_lock(&inode->i_lock); |
4c51acbc6
|
490 |
hlist_del_init(&inode->i_hash); |
250df6ed2
|
491 |
spin_unlock(&inode->i_lock); |
67a23c494
|
492 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
493 |
} |
f2ee7abf4
|
494 |
EXPORT_SYMBOL(__remove_inode_hash); |
4c51acbc6
|
495 |
|
dbd5768f8
|
496 |
void clear_inode(struct inode *inode) |
b0683aa63
|
497 |
{ |
08142579b
|
498 |
/* |
b93b01631
|
499 |
* We have to cycle the i_pages lock here because reclaim can be in the |
08142579b
|
500 |
* process of removing the last page (in __delete_from_page_cache()) |
b93b01631
|
501 |
* and we must not free the mapping under it. |
08142579b
|
502 |
*/ |
b93b01631
|
503 |
xa_lock_irq(&inode->i_data.i_pages); |
b0683aa63
|
504 |
BUG_ON(inode->i_data.nrpages); |
f9fe48bec
|
505 |
BUG_ON(inode->i_data.nrexceptional); |
b93b01631
|
506 |
xa_unlock_irq(&inode->i_data.i_pages); |
b0683aa63
|
507 508 509 |
BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); |
6c60d2b57
|
510 |
BUG_ON(!list_empty(&inode->i_wb_list)); |
fa0d7e3de
|
511 |
/* don't need i_lock here, no concurrent mods to i_state */ |
b0683aa63
|
512 513 |
inode->i_state = I_FREEING | I_CLEAR; } |
dbd5768f8
|
514 |
EXPORT_SYMBOL(clear_inode); |
b0683aa63
|
515 |
|
b2b2af8e6
|
516 517 518 519 520 521 522 523 524 525 526 527 528 |
/* * 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
|
529 |
static void evict(struct inode *inode) |
b4272d4c8
|
530 531 |
{ const struct super_operations *op = inode->i_sb->s_op; |
b2b2af8e6
|
532 533 |
BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(!list_empty(&inode->i_lru)); |
c7f540849
|
534 535 |
if (!list_empty(&inode->i_io_list)) inode_io_list_del(inode); |
b12362bdb
|
536 |
|
55fa6091d
|
537 |
inode_sb_list_del(inode); |
169ebd901
|
538 539 540 541 542 543 544 |
/* * 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
|
545 |
|
be7ce4161
|
546 547 |
if (op->evict_inode) { op->evict_inode(inode); |
b4272d4c8
|
548 |
} else { |
91b0abe36
|
549 |
truncate_inode_pages_final(&inode->i_data); |
dbd5768f8
|
550 |
clear_inode(inode); |
b4272d4c8
|
551 |
} |
661074e91
|
552 553 554 555 |
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
|
556 557 558 559 560 561 562 563 564 |
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
|
565 |
} |
1da177e4c
|
566 567 568 569 570 571 572 573 574 |
/* * 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
|
575 576 |
while (!list_empty(head)) { struct inode *inode; |
7ccf19a80
|
577 578 |
inode = list_first_entry(head, struct inode, i_lru); list_del_init(&inode->i_lru); |
1da177e4c
|
579 |
|
644da5960
|
580 |
evict(inode); |
ac05fbb40
|
581 |
cond_resched(); |
1da177e4c
|
582 |
} |
1da177e4c
|
583 |
} |
a03187867
|
584 |
/** |
63997e98a
|
585 586 587 588 |
* 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 |
1751e8a6c
|
589 |
* called by superblock shutdown after having SB_ACTIVE flag removed, |
63997e98a
|
590 591 |
* so any inode reaching zero refcount during or after that call will * be immediately evicted. |
1da177e4c
|
592 |
*/ |
63997e98a
|
593 |
void evict_inodes(struct super_block *sb) |
1da177e4c
|
594 |
{ |
63997e98a
|
595 596 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
597 |
|
ac05fbb40
|
598 |
again: |
74278da9f
|
599 |
spin_lock(&sb->s_inode_list_lock); |
63997e98a
|
600 601 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { if (atomic_read(&inode->i_count)) |
aabb8fdb4
|
602 |
continue; |
250df6ed2
|
603 604 605 606 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
1da177e4c
|
607 |
continue; |
250df6ed2
|
608 |
} |
63997e98a
|
609 610 |
inode->i_state |= I_FREEING; |
02afc410f
|
611 |
inode_lru_list_del(inode); |
250df6ed2
|
612 |
spin_unlock(&inode->i_lock); |
02afc410f
|
613 |
list_add(&inode->i_lru, &dispose); |
ac05fbb40
|
614 615 616 617 618 619 620 621 622 623 624 625 |
/* * We can have a ton of inodes to evict at unmount time given * enough memory, check to see if we need to go to sleep for a * bit so we don't livelock. */ if (need_resched()) { spin_unlock(&sb->s_inode_list_lock); cond_resched(); dispose_list(&dispose); goto again; } |
1da177e4c
|
626 |
} |
74278da9f
|
627 |
spin_unlock(&sb->s_inode_list_lock); |
63997e98a
|
628 629 |
dispose_list(&dispose); |
1da177e4c
|
630 |
} |
799ea9e9c
|
631 |
EXPORT_SYMBOL_GPL(evict_inodes); |
1da177e4c
|
632 |
|
1da177e4c
|
633 |
/** |
a03187867
|
634 635 |
* invalidate_inodes - attempt to free all inodes on a superblock * @sb: superblock to operate on |
93b270f76
|
636 |
* @kill_dirty: flag to guide handling of dirty inodes |
1da177e4c
|
637 |
* |
a03187867
|
638 639 |
* Attempts to free all inodes for a given superblock. If there were any * busy inodes return a non-zero value, else zero. |
93b270f76
|
640 641 |
* If @kill_dirty is set, discard dirty inodes too, otherwise treat * them as busy. |
1da177e4c
|
642 |
*/ |
93b270f76
|
643 |
int invalidate_inodes(struct super_block *sb, bool kill_dirty) |
1da177e4c
|
644 |
{ |
cffbc8aa3
|
645 |
int busy = 0; |
a03187867
|
646 647 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
648 |
|
16b730a45
|
649 |
again: |
74278da9f
|
650 |
spin_lock(&sb->s_inode_list_lock); |
a03187867
|
651 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { |
250df6ed2
|
652 653 654 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
aabb8fdb4
|
655 |
continue; |
250df6ed2
|
656 |
} |
0ae45f63d
|
657 |
if (inode->i_state & I_DIRTY_ALL && !kill_dirty) { |
250df6ed2
|
658 |
spin_unlock(&inode->i_lock); |
93b270f76
|
659 660 661 |
busy = 1; continue; } |
99a389192
|
662 |
if (atomic_read(&inode->i_count)) { |
250df6ed2
|
663 |
spin_unlock(&inode->i_lock); |
99a389192
|
664 |
busy = 1; |
1da177e4c
|
665 666 |
continue; } |
99a389192
|
667 |
|
99a389192
|
668 |
inode->i_state |= I_FREEING; |
02afc410f
|
669 |
inode_lru_list_del(inode); |
250df6ed2
|
670 |
spin_unlock(&inode->i_lock); |
02afc410f
|
671 |
list_add(&inode->i_lru, &dispose); |
16b730a45
|
672 673 674 675 676 677 |
if (need_resched()) { spin_unlock(&sb->s_inode_list_lock); cond_resched(); dispose_list(&dispose); goto again; } |
1da177e4c
|
678 |
} |
74278da9f
|
679 |
spin_unlock(&sb->s_inode_list_lock); |
1da177e4c
|
680 |
|
a03187867
|
681 |
dispose_list(&dispose); |
1da177e4c
|
682 683 684 |
return busy; } |
1da177e4c
|
685 |
|
1da177e4c
|
686 |
/* |
bc3b14cb2
|
687 |
* Isolate the inode from the LRU in preparation for freeing it. |
1da177e4c
|
688 689 |
* * Any inodes which are pinned purely because of attached pagecache have their |
9e38d86ff
|
690 691 |
* pagecache removed. If the inode has metadata buffers attached to * mapping->private_list then try to remove them. |
1da177e4c
|
692 |
* |
9e38d86ff
|
693 694 695 696 697 698 699 |
* 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
|
700 |
*/ |
3f97b1632
|
701 702 |
static enum lru_status inode_lru_isolate(struct list_head *item, struct list_lru_one *lru, spinlock_t *lru_lock, void *arg) |
1da177e4c
|
703 |
{ |
bc3b14cb2
|
704 705 |
struct list_head *freeable = arg; struct inode *inode = container_of(item, struct inode, i_lru); |
1da177e4c
|
706 |
|
bc3b14cb2
|
707 708 709 710 711 712 |
/* * 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
|
713 |
|
bc3b14cb2
|
714 715 716 717 718 719 |
/* * 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)) { |
3f97b1632
|
720 |
list_lru_isolate(lru, &inode->i_lru); |
bc3b14cb2
|
721 722 723 724 |
spin_unlock(&inode->i_lock); this_cpu_dec(nr_unused); return LRU_REMOVED; } |
1da177e4c
|
725 |
|
69056ee6a
|
726 727 |
/* recently referenced inodes get one more pass */ if (inode->i_state & I_REFERENCED) { |
bc3b14cb2
|
728 729 730 731 |
inode->i_state &= ~I_REFERENCED; spin_unlock(&inode->i_lock); return LRU_ROTATE; } |
1da177e4c
|
732 |
|
bc3b14cb2
|
733 734 735 736 737 738 739 740 741 742 743 744 745 |
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
|
746 |
} |
bc3b14cb2
|
747 748 749 750 |
iput(inode); spin_lock(lru_lock); return LRU_RETRY; } |
02afc410f
|
751 |
|
bc3b14cb2
|
752 753 |
WARN_ON(inode->i_state & I_NEW); inode->i_state |= I_FREEING; |
3f97b1632
|
754 |
list_lru_isolate_move(lru, &inode->i_lru, freeable); |
bc3b14cb2
|
755 |
spin_unlock(&inode->i_lock); |
9e38d86ff
|
756 |
|
bc3b14cb2
|
757 758 759 |
this_cpu_dec(nr_unused); return LRU_REMOVED; } |
7ccf19a80
|
760 |
|
bc3b14cb2
|
761 762 763 764 765 766 |
/* * 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(). */ |
503c358cf
|
767 |
long prune_icache_sb(struct super_block *sb, struct shrink_control *sc) |
bc3b14cb2
|
768 769 770 |
{ LIST_HEAD(freeable); long freed; |
1da177e4c
|
771 |
|
503c358cf
|
772 773 |
freed = list_lru_shrink_walk(&sb->s_inode_lru, sc, inode_lru_isolate, &freeable); |
1da177e4c
|
774 |
dispose_list(&freeable); |
0a234c6dc
|
775 |
return freed; |
1da177e4c
|
776 |
} |
1da177e4c
|
777 778 779 |
static void __wait_on_freeing_inode(struct inode *inode); /* * Called with the inode lock held. |
1da177e4c
|
780 |
*/ |
6b3304b53
|
781 782 783 784 |
static struct inode *find_inode(struct super_block *sb, struct hlist_head *head, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
785 |
{ |
6b3304b53
|
786 |
struct inode *inode = NULL; |
1da177e4c
|
787 788 |
repeat: |
b67bfe0d4
|
789 |
hlist_for_each_entry(inode, head, i_hash) { |
5a3cd9928
|
790 |
if (inode->i_sb != sb) |
1da177e4c
|
791 |
continue; |
5a3cd9928
|
792 |
if (!test(inode, data)) |
1da177e4c
|
793 |
continue; |
5a3cd9928
|
794 |
spin_lock(&inode->i_lock); |
a4ffdde6e
|
795 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
796 797 798 |
__wait_on_freeing_inode(inode); goto repeat; } |
c2b6d621c
|
799 800 801 802 |
if (unlikely(inode->i_state & I_CREATING)) { spin_unlock(&inode->i_lock); return ERR_PTR(-ESTALE); } |
f7899bd54
|
803 |
__iget(inode); |
250df6ed2
|
804 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
805 |
return inode; |
1da177e4c
|
806 |
} |
f7899bd54
|
807 |
return NULL; |
1da177e4c
|
808 809 810 811 812 813 |
} /* * find_inode_fast is the fast path version of find_inode, see the comment at * iget_locked for details. */ |
6b3304b53
|
814 815 |
static struct inode *find_inode_fast(struct super_block *sb, struct hlist_head *head, unsigned long ino) |
1da177e4c
|
816 |
{ |
6b3304b53
|
817 |
struct inode *inode = NULL; |
1da177e4c
|
818 819 |
repeat: |
b67bfe0d4
|
820 |
hlist_for_each_entry(inode, head, i_hash) { |
5a3cd9928
|
821 |
if (inode->i_ino != ino) |
1da177e4c
|
822 |
continue; |
5a3cd9928
|
823 |
if (inode->i_sb != sb) |
1da177e4c
|
824 |
continue; |
5a3cd9928
|
825 |
spin_lock(&inode->i_lock); |
a4ffdde6e
|
826 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
827 828 829 |
__wait_on_freeing_inode(inode); goto repeat; } |
c2b6d621c
|
830 831 832 833 |
if (unlikely(inode->i_state & I_CREATING)) { spin_unlock(&inode->i_lock); return ERR_PTR(-ESTALE); } |
f7899bd54
|
834 |
__iget(inode); |
250df6ed2
|
835 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
836 |
return inode; |
1da177e4c
|
837 |
} |
f7899bd54
|
838 |
return NULL; |
8290c35f8
|
839 |
} |
f991bd2e1
|
840 841 842 843 |
/* * 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
|
844 |
* |
f991bd2e1
|
845 846 847 848 849 850 851 852 853 |
* 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
|
854 |
*/ |
f991bd2e1
|
855 856 |
#define LAST_INO_BATCH 1024 static DEFINE_PER_CPU(unsigned int, last_ino); |
85fe4025c
|
857 |
unsigned int get_next_ino(void) |
8290c35f8
|
858 |
{ |
f991bd2e1
|
859 860 |
unsigned int *p = &get_cpu_var(last_ino); unsigned int res = *p; |
8290c35f8
|
861 |
|
f991bd2e1
|
862 863 864 865 866 867 868 869 |
#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 |
2adc376c5
|
870 871 872 873 874 |
res++; /* get_next_ino should not provide a 0 inode number */ if (unlikely(!res)) res++; *p = res; |
f991bd2e1
|
875 876 |
put_cpu_var(last_ino); return res; |
8290c35f8
|
877 |
} |
85fe4025c
|
878 |
EXPORT_SYMBOL(get_next_ino); |
8290c35f8
|
879 |
|
1da177e4c
|
880 |
/** |
a209dfc7b
|
881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 |
* 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
|
904 905 906 |
* new_inode - obtain an inode * @sb: superblock * |
769848c03
|
907 |
* Allocates a new inode for given superblock. The default gfp_mask |
3c1d43787
|
908 |
* for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE. |
769848c03
|
909 910 911 912 913 |
* 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
|
914 915 916 |
*/ struct inode *new_inode(struct super_block *sb) { |
6b3304b53
|
917 |
struct inode *inode; |
1da177e4c
|
918 |
|
74278da9f
|
919 |
spin_lock_prefetch(&sb->s_inode_list_lock); |
6b3304b53
|
920 |
|
a209dfc7b
|
921 922 |
inode = new_inode_pseudo(sb); if (inode) |
55fa6091d
|
923 |
inode_sb_list_add(inode); |
1da177e4c
|
924 925 |
return inode; } |
1da177e4c
|
926 |
EXPORT_SYMBOL(new_inode); |
14358e6dd
|
927 |
#ifdef CONFIG_DEBUG_LOCK_ALLOC |
e096d0c7e
|
928 929 |
void lockdep_annotate_inode_mutex_key(struct inode *inode) { |
a3314a0ed
|
930 |
if (S_ISDIR(inode->i_mode)) { |
1e89a5e15
|
931 |
struct file_system_type *type = inode->i_sb->s_type; |
9a7aa12f3
|
932 |
/* Set new key only if filesystem hasn't already changed it */ |
9902af79c
|
933 |
if (lockdep_match_class(&inode->i_rwsem, &type->i_mutex_key)) { |
9a7aa12f3
|
934 935 936 |
/* * ensure nobody is actually holding i_mutex */ |
9902af79c
|
937 938 939 |
// mutex_destroy(&inode->i_mutex); init_rwsem(&inode->i_rwsem); lockdep_set_class(&inode->i_rwsem, |
9a7aa12f3
|
940 941 |
&type->i_mutex_dir_key); } |
1e89a5e15
|
942 |
} |
e096d0c7e
|
943 944 |
} EXPORT_SYMBOL(lockdep_annotate_inode_mutex_key); |
14358e6dd
|
945 |
#endif |
e096d0c7e
|
946 947 948 949 950 951 952 953 954 955 956 |
/** * 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
|
957 |
spin_lock(&inode->i_lock); |
eaff8079d
|
958 |
WARN_ON(!(inode->i_state & I_NEW)); |
c2b6d621c
|
959 |
inode->i_state &= ~I_NEW & ~I_CREATING; |
310fa7a36
|
960 |
smp_mb(); |
250df6ed2
|
961 962 |
wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); |
1da177e4c
|
963 |
} |
1da177e4c
|
964 |
EXPORT_SYMBOL(unlock_new_inode); |
c2b6d621c
|
965 966 967 968 969 970 971 972 973 974 975 976 |
void discard_new_inode(struct inode *inode) { lockdep_annotate_inode_mutex_key(inode); spin_lock(&inode->i_lock); WARN_ON(!(inode->i_state & I_NEW)); inode->i_state &= ~I_NEW; smp_mb(); wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); iput(inode); } EXPORT_SYMBOL(discard_new_inode); |
0b2d0724e
|
977 |
/** |
375e289ea
|
978 |
* lock_two_nondirectories - take two i_mutexes on non-directory objects |
4fd699ae3
|
979 980 981 982 |
* * Lock any non-NULL argument that is not a directory. * Zero, one or two objects may be locked by this function. * |
375e289ea
|
983 984 985 986 987 |
* @inode1: first inode to lock * @inode2: second inode to lock */ void lock_two_nondirectories(struct inode *inode1, struct inode *inode2) { |
4fd699ae3
|
988 989 990 991 |
if (inode1 > inode2) swap(inode1, inode2); if (inode1 && !S_ISDIR(inode1->i_mode)) |
5955102c9
|
992 |
inode_lock(inode1); |
4fd699ae3
|
993 |
if (inode2 && !S_ISDIR(inode2->i_mode) && inode2 != inode1) |
5955102c9
|
994 |
inode_lock_nested(inode2, I_MUTEX_NONDIR2); |
375e289ea
|
995 996 997 998 999 1000 1001 1002 1003 1004 |
} 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
|
1005 |
if (inode1 && !S_ISDIR(inode1->i_mode)) |
5955102c9
|
1006 |
inode_unlock(inode1); |
4fd699ae3
|
1007 |
if (inode2 && !S_ISDIR(inode2->i_mode) && inode2 != inode1) |
5955102c9
|
1008 |
inode_unlock(inode2); |
375e289ea
|
1009 1010 1011 1012 |
} EXPORT_SYMBOL(unlock_two_nondirectories); /** |
80ea09a00
|
1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 |
* inode_insert5 - obtain an inode from a mounted file system * @inode: pre-allocated inode to use for insert to cache * @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 variant of iget5_locked() for callers that don't want to fail on memory * allocation of inode. * * If the inode is not in cache, insert the pre-allocated inode to cache 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(). * * Note both @test and @set are called with the inode_hash_lock held, so can't * sleep. */ struct inode *inode_insert5(struct inode *inode, unsigned long hashval, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data) { struct hlist_head *head = inode_hashtable + hash(inode->i_sb, hashval); struct inode *old; |
e950564b9
|
1038 |
bool creating = inode->i_state & I_CREATING; |
80ea09a00
|
1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 |
again: spin_lock(&inode_hash_lock); old = find_inode(inode->i_sb, head, test, data); if (unlikely(old)) { /* * Uhhuh, somebody else created the same inode under us. * Use the old inode instead of the preallocated one. */ spin_unlock(&inode_hash_lock); |
c2b6d621c
|
1049 1050 |
if (IS_ERR(old)) return NULL; |
80ea09a00
|
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 |
wait_on_inode(old); if (unlikely(inode_unhashed(old))) { iput(old); goto again; } return old; } if (set && unlikely(set(inode, data))) { inode = NULL; goto unlock; } /* * Return the locked inode with I_NEW set, the * caller is responsible for filling in the contents */ spin_lock(&inode->i_lock); inode->i_state |= I_NEW; hlist_add_head(&inode->i_hash, head); spin_unlock(&inode->i_lock); |
e950564b9
|
1072 1073 |
if (!creating) inode_sb_list_add(inode); |
80ea09a00
|
1074 1075 1076 1077 1078 1079 1080 1081 |
unlock: spin_unlock(&inode_hash_lock); return inode; } EXPORT_SYMBOL(inode_insert5); /** |
0b2d0724e
|
1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 |
* 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
|
1097 |
* |
0b2d0724e
|
1098 1099 |
* Note both @test and @set are called with the inode_hash_lock held, so can't * sleep. |
1da177e4c
|
1100 |
*/ |
0b2d0724e
|
1101 1102 1103 |
struct inode *iget5_locked(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data) |
1da177e4c
|
1104 |
{ |
80ea09a00
|
1105 |
struct inode *inode = ilookup5(sb, hashval, test, data); |
0b2d0724e
|
1106 |
|
80ea09a00
|
1107 |
if (!inode) { |
e950564b9
|
1108 |
struct inode *new = alloc_inode(sb); |
0b2d0724e
|
1109 |
|
80ea09a00
|
1110 |
if (new) { |
e950564b9
|
1111 |
new->i_state = 0; |
80ea09a00
|
1112 1113 |
inode = inode_insert5(new, hashval, test, set, data); if (unlikely(inode != new)) |
e950564b9
|
1114 |
destroy_inode(new); |
2864f3014
|
1115 |
} |
1da177e4c
|
1116 1117 |
} return inode; |
1da177e4c
|
1118 |
} |
0b2d0724e
|
1119 |
EXPORT_SYMBOL(iget5_locked); |
1da177e4c
|
1120 |
|
0b2d0724e
|
1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 |
/** * 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
|
1133 |
*/ |
0b2d0724e
|
1134 |
struct inode *iget_locked(struct super_block *sb, unsigned long ino) |
1da177e4c
|
1135 |
{ |
0b2d0724e
|
1136 |
struct hlist_head *head = inode_hashtable + hash(sb, ino); |
6b3304b53
|
1137 |
struct inode *inode; |
2864f3014
|
1138 |
again: |
0b2d0724e
|
1139 1140 1141 1142 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); if (inode) { |
c2b6d621c
|
1143 1144 |
if (IS_ERR(inode)) return NULL; |
0b2d0724e
|
1145 |
wait_on_inode(inode); |
2864f3014
|
1146 1147 1148 1149 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } |
0b2d0724e
|
1150 1151 |
return inode; } |
1da177e4c
|
1152 1153 |
inode = alloc_inode(sb); if (inode) { |
6b3304b53
|
1154 |
struct inode *old; |
1da177e4c
|
1155 |
|
67a23c494
|
1156 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1157 1158 1159 1160 |
/* We released the lock, so.. */ old = find_inode_fast(sb, head, ino); if (!old) { inode->i_ino = ino; |
250df6ed2
|
1161 1162 |
spin_lock(&inode->i_lock); inode->i_state = I_NEW; |
646ec4615
|
1163 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1164 |
spin_unlock(&inode->i_lock); |
55fa6091d
|
1165 |
inode_sb_list_add(inode); |
67a23c494
|
1166 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 |
/* 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
|
1179 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1180 |
destroy_inode(inode); |
c2b6d621c
|
1181 1182 |
if (IS_ERR(old)) return NULL; |
1da177e4c
|
1183 1184 |
inode = old; wait_on_inode(inode); |
2864f3014
|
1185 1186 1187 1188 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } |
1da177e4c
|
1189 1190 1191 |
} return inode; } |
0b2d0724e
|
1192 |
EXPORT_SYMBOL(iget_locked); |
1da177e4c
|
1193 |
|
ad5e195ac
|
1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 |
/* * 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
|
1204 |
struct inode *inode; |
67a23c494
|
1205 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1206 |
hlist_for_each_entry(inode, b, i_hash) { |
67a23c494
|
1207 1208 |
if (inode->i_ino == ino && inode->i_sb == sb) { spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1209 |
return 0; |
67a23c494
|
1210 |
} |
ad5e195ac
|
1211 |
} |
67a23c494
|
1212 |
spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1213 1214 1215 |
return 1; } |
1da177e4c
|
1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 |
/** * 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
|
1232 1233 1234 1235 1236 |
/* * 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
|
1237 |
static DEFINE_SPINLOCK(iunique_lock); |
866b04fcc
|
1238 |
static unsigned int counter; |
1da177e4c
|
1239 |
ino_t res; |
3361c7beb
|
1240 |
|
ad5e195ac
|
1241 |
spin_lock(&iunique_lock); |
3361c7beb
|
1242 1243 1244 |
do { if (counter <= max_reserved) counter = max_reserved + 1; |
1da177e4c
|
1245 |
res = counter++; |
ad5e195ac
|
1246 1247 |
} while (!test_inode_iunique(sb, res)); spin_unlock(&iunique_lock); |
1da177e4c
|
1248 |
|
3361c7beb
|
1249 1250 |
return res; } |
1da177e4c
|
1251 1252 1253 1254 |
EXPORT_SYMBOL(iunique); struct inode *igrab(struct inode *inode) { |
250df6ed2
|
1255 1256 |
spin_lock(&inode->i_lock); if (!(inode->i_state & (I_FREEING|I_WILL_FREE))) { |
1da177e4c
|
1257 |
__iget(inode); |
250df6ed2
|
1258 1259 1260 |
spin_unlock(&inode->i_lock); } else { spin_unlock(&inode->i_lock); |
1da177e4c
|
1261 1262 1263 1264 1265 1266 |
/* * 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
|
1267 |
} |
1da177e4c
|
1268 1269 |
return inode; } |
1da177e4c
|
1270 1271 1272 |
EXPORT_SYMBOL(igrab); /** |
0b2d0724e
|
1273 |
* ilookup5_nowait - search for an inode in the inode cache |
1da177e4c
|
1274 |
* @sb: super block of file system to search |
0b2d0724e
|
1275 |
* @hashval: hash value (usually inode number) to search for |
1da177e4c
|
1276 1277 |
* @test: callback used for comparisons between inodes * @data: opaque data pointer to pass to @test |
1da177e4c
|
1278 |
* |
0b2d0724e
|
1279 |
* Search for the inode specified by @hashval and @data in the inode cache. |
1da177e4c
|
1280 1281 1282 |
* If the inode is in the cache, the inode is returned with an incremented * reference count. * |
0b2d0724e
|
1283 1284 |
* 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
|
1285 |
* |
b6d0ad686
|
1286 |
* Note2: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1287 |
*/ |
0b2d0724e
|
1288 1289 |
struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
1290 |
{ |
0b2d0724e
|
1291 |
struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
1da177e4c
|
1292 |
struct inode *inode; |
67a23c494
|
1293 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1294 |
inode = find_inode(sb, head, test, data); |
67a23c494
|
1295 |
spin_unlock(&inode_hash_lock); |
88bd5121d
|
1296 |
|
c2b6d621c
|
1297 |
return IS_ERR(inode) ? NULL : inode; |
88bd5121d
|
1298 |
} |
88bd5121d
|
1299 1300 1301 1302 1303 1304 1305 1306 1307 |
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
|
1308 1309 1310 |
* 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
|
1311 |
* returned with an incremented reference count. |
1da177e4c
|
1312 |
* |
0b2d0724e
|
1313 1314 |
* This is a generalized version of ilookup() for file systems where the * inode number is not sufficient for unique identification of an inode. |
1da177e4c
|
1315 |
* |
0b2d0724e
|
1316 |
* Note: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1317 1318 1319 1320 |
*/ struct inode *ilookup5(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { |
2864f3014
|
1321 1322 1323 1324 |
struct inode *inode; again: inode = ilookup5_nowait(sb, hashval, test, data); if (inode) { |
0b2d0724e
|
1325 |
wait_on_inode(inode); |
2864f3014
|
1326 1327 1328 1329 1330 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } } |
0b2d0724e
|
1331 |
return inode; |
1da177e4c
|
1332 |
} |
1da177e4c
|
1333 1334 1335 1336 1337 1338 1339 |
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
|
1340 1341 |
* 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
|
1342 1343 1344 1345 |
*/ struct inode *ilookup(struct super_block *sb, unsigned long ino) { struct hlist_head *head = inode_hashtable + hash(sb, ino); |
1da177e4c
|
1346 |
struct inode *inode; |
2864f3014
|
1347 |
again: |
0b2d0724e
|
1348 1349 1350 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); |
1da177e4c
|
1351 |
|
2864f3014
|
1352 |
if (inode) { |
c2b6d621c
|
1353 1354 |
if (IS_ERR(inode)) return NULL; |
0b2d0724e
|
1355 |
wait_on_inode(inode); |
2864f3014
|
1356 1357 1358 1359 1360 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } } |
0b2d0724e
|
1361 |
return inode; |
1da177e4c
|
1362 |
} |
0b2d0724e
|
1363 |
EXPORT_SYMBOL(ilookup); |
1da177e4c
|
1364 |
|
fe032c422
|
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 |
/** * find_inode_nowait - find an inode in the inode cache * @sb: super block of file system to search * @hashval: hash value (usually inode number) to search for * @match: callback used for comparisons between inodes * @data: opaque data pointer to pass to @match * * Search for the inode specified by @hashval and @data in the inode * cache, where the helper function @match will return 0 if the inode * does not match, 1 if the inode does match, and -1 if the search * should be stopped. The @match function must be responsible for * taking the i_lock spin_lock and checking i_state for an inode being * freed or being initialized, and incrementing the reference count * before returning 1. It also must not sleep, since it is called with * the inode_hash_lock spinlock held. * * This is a even more generalized version of ilookup5() when the * function must never block --- find_inode() can block in * __wait_on_freeing_inode() --- or when the caller can not increment * the reference count because the resulting iput() might cause an * inode eviction. The tradeoff is that the @match funtion must be * very carefully implemented. */ struct inode *find_inode_nowait(struct super_block *sb, unsigned long hashval, int (*match)(struct inode *, unsigned long, void *), void *data) { struct hlist_head *head = inode_hashtable + hash(sb, hashval); struct inode *inode, *ret_inode = NULL; int mval; spin_lock(&inode_hash_lock); hlist_for_each_entry(inode, head, i_hash) { if (inode->i_sb != sb) continue; mval = match(inode, hashval, data); if (mval == 0) continue; if (mval == 1) ret_inode = inode; goto out; } out: spin_unlock(&inode_hash_lock); return ret_inode; } EXPORT_SYMBOL(find_inode_nowait); |
261bca86e
|
1414 1415 1416 1417 1418 |
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
|
1419 |
|
261bca86e
|
1420 |
while (1) { |
72a43d63c
|
1421 |
struct inode *old = NULL; |
67a23c494
|
1422 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1423 |
hlist_for_each_entry(old, head, i_hash) { |
72a43d63c
|
1424 1425 1426 1427 |
if (old->i_ino != ino) continue; if (old->i_sb != sb) continue; |
250df6ed2
|
1428 1429 1430 |
spin_lock(&old->i_lock); if (old->i_state & (I_FREEING|I_WILL_FREE)) { spin_unlock(&old->i_lock); |
72a43d63c
|
1431 |
continue; |
250df6ed2
|
1432 |
} |
72a43d63c
|
1433 1434 |
break; } |
b67bfe0d4
|
1435 |
if (likely(!old)) { |
250df6ed2
|
1436 |
spin_lock(&inode->i_lock); |
c2b6d621c
|
1437 |
inode->i_state |= I_NEW | I_CREATING; |
261bca86e
|
1438 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1439 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1440 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1441 1442 |
return 0; } |
c2b6d621c
|
1443 1444 1445 1446 1447 |
if (unlikely(old->i_state & I_CREATING)) { spin_unlock(&old->i_lock); spin_unlock(&inode_hash_lock); return -EBUSY; } |
261bca86e
|
1448 |
__iget(old); |
250df6ed2
|
1449 |
spin_unlock(&old->i_lock); |
67a23c494
|
1450 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1451 |
wait_on_inode(old); |
1d3382cbf
|
1452 |
if (unlikely(!inode_unhashed(old))) { |
261bca86e
|
1453 1454 1455 1456 1457 1458 |
iput(old); return -EBUSY; } iput(old); } } |
261bca86e
|
1459 1460 1461 1462 1463 |
EXPORT_SYMBOL(insert_inode_locked); int insert_inode_locked4(struct inode *inode, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { |
c2b6d621c
|
1464 1465 1466 1467 |
struct inode *old; inode->i_state |= I_CREATING; old = inode_insert5(inode, hashval, test, NULL, data); |
261bca86e
|
1468 |
|
80ea09a00
|
1469 |
if (old != inode) { |
261bca86e
|
1470 |
iput(old); |
80ea09a00
|
1471 |
return -EBUSY; |
261bca86e
|
1472 |
} |
80ea09a00
|
1473 |
return 0; |
261bca86e
|
1474 |
} |
261bca86e
|
1475 |
EXPORT_SYMBOL(insert_inode_locked4); |
1da177e4c
|
1476 |
|
45321ac54
|
1477 1478 1479 1480 1481 |
int generic_delete_inode(struct inode *inode) { return 1; } EXPORT_SYMBOL(generic_delete_inode); |
1da177e4c
|
1482 |
/* |
45321ac54
|
1483 1484 |
* Called when we're dropping the last reference * to an inode. |
22fe40421
|
1485 |
* |
45321ac54
|
1486 1487 1488 1489 1490 |
* 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
|
1491 |
*/ |
45321ac54
|
1492 |
static void iput_final(struct inode *inode) |
1da177e4c
|
1493 1494 |
{ struct super_block *sb = inode->i_sb; |
45321ac54
|
1495 1496 |
const struct super_operations *op = inode->i_sb->s_op; int drop; |
250df6ed2
|
1497 |
WARN_ON(inode->i_state & I_NEW); |
e7f590970
|
1498 |
if (op->drop_inode) |
45321ac54
|
1499 1500 1501 |
drop = op->drop_inode(inode); else drop = generic_drop_inode(inode); |
1da177e4c
|
1502 |
|
1751e8a6c
|
1503 |
if (!drop && (sb->s_flags & SB_ACTIVE)) { |
4eff96dd5
|
1504 |
inode_add_lru(inode); |
b2b2af8e6
|
1505 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1506 1507 |
return; } |
45321ac54
|
1508 |
if (!drop) { |
991114c6f
|
1509 |
inode->i_state |= I_WILL_FREE; |
250df6ed2
|
1510 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1511 |
write_inode_now(inode, 1); |
250df6ed2
|
1512 |
spin_lock(&inode->i_lock); |
7ef0d7377
|
1513 |
WARN_ON(inode->i_state & I_NEW); |
991114c6f
|
1514 |
inode->i_state &= ~I_WILL_FREE; |
1da177e4c
|
1515 |
} |
7ccf19a80
|
1516 |
|
991114c6f
|
1517 |
inode->i_state |= I_FREEING; |
c4ae0c654
|
1518 1519 |
if (!list_empty(&inode->i_lru)) inode_lru_list_del(inode); |
b2b2af8e6
|
1520 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1521 |
|
644da5960
|
1522 |
evict(inode); |
1da177e4c
|
1523 |
} |
1da177e4c
|
1524 |
/** |
6b3304b53
|
1525 |
* iput - put an inode |
1da177e4c
|
1526 1527 1528 1529 1530 1531 1532 1533 1534 |
* @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) { |
0ae45f63d
|
1535 1536 1537 1538 1539 1540 1541 |
if (!inode) return; BUG_ON(inode->i_state & I_CLEAR); retry: if (atomic_dec_and_lock(&inode->i_count, &inode->i_lock)) { if (inode->i_nlink && (inode->i_state & I_DIRTY_TIME)) { atomic_inc(&inode->i_count); |
0ae45f63d
|
1542 1543 1544 1545 1546 1547 |
spin_unlock(&inode->i_lock); trace_writeback_lazytime_iput(inode); mark_inode_dirty_sync(inode); goto retry; } iput_final(inode); |
1da177e4c
|
1548 1549 |
} } |
1da177e4c
|
1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 |
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
|
1560 |
* disk block relative to the disk start that holds that block of the |
1da177e4c
|
1561 1562 |
* file. */ |
6b3304b53
|
1563 |
sector_t bmap(struct inode *inode, sector_t block) |
1da177e4c
|
1564 1565 1566 1567 1568 1569 |
{ 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
|
1570 |
EXPORT_SYMBOL(bmap); |
11ff6f05f
|
1571 1572 1573 1574 1575 |
/* * 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. */ |
c67185434
|
1576 |
static int relatime_need_update(struct vfsmount *mnt, struct inode *inode, |
6f22b6649
|
1577 |
struct timespec64 now) |
11ff6f05f
|
1578 |
{ |
c67185434
|
1579 |
if (!(mnt->mnt_flags & MNT_RELATIME)) |
11ff6f05f
|
1580 1581 1582 1583 |
return 1; /* * Is mtime younger than atime? If yes, update atime: */ |
95582b008
|
1584 |
if (timespec64_compare(&inode->i_mtime, &inode->i_atime) >= 0) |
11ff6f05f
|
1585 1586 1587 1588 |
return 1; /* * Is ctime younger than atime? If yes, update atime: */ |
95582b008
|
1589 |
if (timespec64_compare(&inode->i_ctime, &inode->i_atime) >= 0) |
11ff6f05f
|
1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 |
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; } |
95582b008
|
1603 |
int generic_update_time(struct inode *inode, struct timespec64 *time, int flags) |
c3b2da314
|
1604 |
{ |
0ae45f63d
|
1605 |
int iflags = I_DIRTY_TIME; |
e38cf302b
|
1606 |
bool dirty = false; |
c3b2da314
|
1607 1608 1609 1610 |
if (flags & S_ATIME) inode->i_atime = *time; if (flags & S_VERSION) |
e38cf302b
|
1611 |
dirty = inode_maybe_inc_iversion(inode, false); |
c3b2da314
|
1612 1613 1614 1615 |
if (flags & S_CTIME) inode->i_ctime = *time; if (flags & S_MTIME) inode->i_mtime = *time; |
e38cf302b
|
1616 1617 1618 |
if ((flags & (S_ATIME | S_CTIME | S_MTIME)) && !(inode->i_sb->s_flags & SB_LAZYTIME)) dirty = true; |
0ae45f63d
|
1619 |
|
e38cf302b
|
1620 |
if (dirty) |
0ae45f63d
|
1621 1622 |
iflags |= I_DIRTY_SYNC; __mark_inode_dirty(inode, iflags); |
c3b2da314
|
1623 1624 |
return 0; } |
0ae45f63d
|
1625 1626 1627 1628 1629 1630 |
EXPORT_SYMBOL(generic_update_time); /* * This does the actual work of updating an inodes time or version. Must have * had called mnt_want_write() before calling this. */ |
95582b008
|
1631 |
static int update_time(struct inode *inode, struct timespec64 *time, int flags) |
0ae45f63d
|
1632 |
{ |
95582b008
|
1633 |
int (*update_time)(struct inode *, struct timespec64 *, int); |
0ae45f63d
|
1634 1635 1636 1637 1638 1639 |
update_time = inode->i_op->update_time ? inode->i_op->update_time : generic_update_time; return update_time(inode, time, flags); } |
c3b2da314
|
1640 |
|
1da177e4c
|
1641 |
/** |
869243a0f
|
1642 |
* touch_atime - update the access time |
185553b22
|
1643 |
* @path: the &struct path to update |
30fdc8ee0
|
1644 |
* @inode: inode to update |
1da177e4c
|
1645 1646 1647 1648 1649 |
* * 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. */ |
c67185434
|
1650 |
bool atime_needs_update(const struct path *path, struct inode *inode) |
1da177e4c
|
1651 |
{ |
68ac1234f
|
1652 |
struct vfsmount *mnt = path->mnt; |
95582b008
|
1653 |
struct timespec64 now; |
1da177e4c
|
1654 |
|
cdb70f3f7
|
1655 |
if (inode->i_flags & S_NOATIME) |
8fa9dd246
|
1656 |
return false; |
0bd23d09b
|
1657 1658 1659 1660 1661 1662 |
/* Atime updates will likely cause i_uid and i_gid to be written * back improprely if their true value is unknown to the vfs. */ if (HAS_UNMAPPED_ID(inode)) return false; |
37756ced1
|
1663 |
if (IS_NOATIME(inode)) |
8fa9dd246
|
1664 |
return false; |
1751e8a6c
|
1665 |
if ((inode->i_sb->s_flags & SB_NODIRATIME) && S_ISDIR(inode->i_mode)) |
8fa9dd246
|
1666 |
return false; |
47ae32d6a
|
1667 |
|
cdb70f3f7
|
1668 |
if (mnt->mnt_flags & MNT_NOATIME) |
8fa9dd246
|
1669 |
return false; |
cdb70f3f7
|
1670 |
if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)) |
8fa9dd246
|
1671 |
return false; |
1da177e4c
|
1672 |
|
c2050a454
|
1673 |
now = current_time(inode); |
11ff6f05f
|
1674 |
|
6f22b6649
|
1675 |
if (!relatime_need_update(mnt, inode, now)) |
8fa9dd246
|
1676 |
return false; |
11ff6f05f
|
1677 |
|
95582b008
|
1678 |
if (timespec64_equal(&inode->i_atime, &now)) |
8fa9dd246
|
1679 1680 1681 1682 1683 1684 1685 1686 1687 |
return false; return true; } void touch_atime(const struct path *path) { struct vfsmount *mnt = path->mnt; struct inode *inode = d_inode(path->dentry); |
95582b008
|
1688 |
struct timespec64 now; |
8fa9dd246
|
1689 |
|
c67185434
|
1690 |
if (!atime_needs_update(path, inode)) |
b12536c27
|
1691 |
return; |
5d37e9e6d
|
1692 |
if (!sb_start_write_trylock(inode->i_sb)) |
b12536c27
|
1693 |
return; |
47ae32d6a
|
1694 |
|
8fa9dd246
|
1695 |
if (__mnt_want_write(mnt) != 0) |
5d37e9e6d
|
1696 |
goto skip_update; |
c3b2da314
|
1697 1698 1699 1700 1701 1702 |
/* * 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
|
1703 1704 |
* We may also fail on filesystems that have the ability to make parts * of the fs read only, e.g. subvolumes in Btrfs. |
c3b2da314
|
1705 |
*/ |
c2050a454
|
1706 |
now = current_time(inode); |
c3b2da314
|
1707 |
update_time(inode, &now, S_ATIME); |
5d37e9e6d
|
1708 1709 1710 |
__mnt_drop_write(mnt); skip_update: sb_end_write(inode->i_sb); |
1da177e4c
|
1711 |
} |
869243a0f
|
1712 |
EXPORT_SYMBOL(touch_atime); |
1da177e4c
|
1713 |
|
3ed37648e
|
1714 1715 1716 1717 1718 1719 1720 1721 |
/* * The logic we want is * * if suid or (sgid and xgrp) * remove privs */ int should_remove_suid(struct dentry *dentry) { |
df2b1afde
|
1722 |
umode_t mode = d_inode(dentry)->i_mode; |
3ed37648e
|
1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 |
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); |
dbfae0cdc
|
1742 1743 1744 1745 1746 |
/* * Return mask of changes for notify_change() that need to be done as a * response to write or truncate. Return 0 if nothing has to be changed. * Negative value on error (change should be denied). */ |
45f147a1b
|
1747 |
int dentry_needs_remove_privs(struct dentry *dentry) |
dbfae0cdc
|
1748 |
{ |
dbfae0cdc
|
1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 |
struct inode *inode = d_inode(dentry); int mask = 0; int ret; if (IS_NOSEC(inode)) return 0; mask = should_remove_suid(dentry); ret = security_inode_need_killpriv(dentry); if (ret < 0) return ret; if (ret) mask |= ATTR_KILL_PRIV; return mask; } |
dbfae0cdc
|
1764 1765 |
static int __remove_privs(struct dentry *dentry, int kill) |
3ed37648e
|
1766 1767 1768 1769 |
{ struct iattr newattrs; newattrs.ia_valid = ATTR_FORCE | kill; |
27ac0ffea
|
1770 1771 1772 1773 1774 |
/* * Note we call this on write, so notify_change will not * encounter any conflicting delegations: */ return notify_change(dentry, &newattrs, NULL); |
3ed37648e
|
1775 |
} |
5fa8e0a1c
|
1776 1777 1778 1779 1780 |
/* * Remove special file priviledges (suid, capabilities) when file is written * to or truncated. */ int file_remove_privs(struct file *file) |
3ed37648e
|
1781 |
{ |
c1892c377
|
1782 1783 |
struct dentry *dentry = file_dentry(file); struct inode *inode = file_inode(file); |
dbfae0cdc
|
1784 |
int kill; |
3ed37648e
|
1785 |
int error = 0; |
f69e749a4
|
1786 1787 1788 1789 1790 1791 1792 |
/* * Fast path for nothing security related. * As well for non-regular files, e.g. blkdev inodes. * For example, blkdev_write_iter() might get here * trying to remove privs which it is not allowed to. */ if (IS_NOSEC(inode) || !S_ISREG(inode->i_mode)) |
3ed37648e
|
1793 |
return 0; |
c1892c377
|
1794 |
kill = dentry_needs_remove_privs(dentry); |
dbfae0cdc
|
1795 1796 1797 1798 |
if (kill < 0) return kill; if (kill) error = __remove_privs(dentry, kill); |
2426f3910
|
1799 1800 |
if (!error) inode_has_no_xattr(inode); |
3ed37648e
|
1801 1802 1803 |
return error; } |
5fa8e0a1c
|
1804 |
EXPORT_SYMBOL(file_remove_privs); |
3ed37648e
|
1805 |
|
1da177e4c
|
1806 |
/** |
870f48179
|
1807 1808 |
* file_update_time - update mtime and ctime time * @file: file accessed |
1da177e4c
|
1809 |
* |
870f48179
|
1810 1811 1812 1813 |
* 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
|
1814 |
* S_NOCMTIME inode flag, e.g. for network filesystem where these |
c3b2da314
|
1815 1816 |
* 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
|
1817 |
*/ |
c3b2da314
|
1818 |
int file_update_time(struct file *file) |
1da177e4c
|
1819 |
{ |
496ad9aa8
|
1820 |
struct inode *inode = file_inode(file); |
95582b008
|
1821 |
struct timespec64 now; |
c3b2da314
|
1822 1823 |
int sync_it = 0; int ret; |
1da177e4c
|
1824 |
|
ce06e0b21
|
1825 |
/* First try to exhaust all avenues to not sync */ |
1da177e4c
|
1826 |
if (IS_NOCMTIME(inode)) |
c3b2da314
|
1827 |
return 0; |
20ddee2c7
|
1828 |
|
c2050a454
|
1829 |
now = current_time(inode); |
95582b008
|
1830 |
if (!timespec64_equal(&inode->i_mtime, &now)) |
ce06e0b21
|
1831 |
sync_it = S_MTIME; |
1da177e4c
|
1832 |
|
95582b008
|
1833 |
if (!timespec64_equal(&inode->i_ctime, &now)) |
ce06e0b21
|
1834 |
sync_it |= S_CTIME; |
870f48179
|
1835 |
|
e38cf302b
|
1836 |
if (IS_I_VERSION(inode) && inode_iversion_need_inc(inode)) |
ce06e0b21
|
1837 |
sync_it |= S_VERSION; |
7a224228e
|
1838 |
|
ce06e0b21
|
1839 |
if (!sync_it) |
c3b2da314
|
1840 |
return 0; |
ce06e0b21
|
1841 1842 |
/* Finally allowed to write? Takes lock. */ |
eb04c2828
|
1843 |
if (__mnt_want_write_file(file)) |
c3b2da314
|
1844 |
return 0; |
ce06e0b21
|
1845 |
|
c3b2da314
|
1846 |
ret = update_time(inode, &now, sync_it); |
eb04c2828
|
1847 |
__mnt_drop_write_file(file); |
c3b2da314
|
1848 1849 |
return ret; |
1da177e4c
|
1850 |
} |
870f48179
|
1851 |
EXPORT_SYMBOL(file_update_time); |
1da177e4c
|
1852 |
|
e38f7f53c
|
1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 |
/* Caller must hold the file's inode lock */ int file_modified(struct file *file) { int err; /* * Clear the security bits if the process is not being run by root. * This keeps people from modifying setuid and setgid binaries. */ err = file_remove_privs(file); if (err) return err; if (unlikely(file->f_mode & FMODE_NOCMTIME)) return 0; return file_update_time(file); } EXPORT_SYMBOL(file_modified); |
1da177e4c
|
1872 1873 1874 1875 1876 1877 1878 1879 |
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
|
1880 |
EXPORT_SYMBOL(inode_needs_sync); |
1da177e4c
|
1881 |
/* |
168a9fd6a
|
1882 1883 1884 1885 1886 1887 |
* 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
|
1888 |
* It doesn't matter if I_NEW is not set initially, a call to |
250df6ed2
|
1889 1890 |
* wake_up_bit(&inode->i_state, __I_NEW) after removing from the hash list * will DTRT. |
1da177e4c
|
1891 1892 1893 1894 |
*/ static void __wait_on_freeing_inode(struct inode *inode) { wait_queue_head_t *wq; |
eaff8079d
|
1895 1896 |
DEFINE_WAIT_BIT(wait, &inode->i_state, __I_NEW); wq = bit_waitqueue(&inode->i_state, __I_NEW); |
214171361
|
1897 |
prepare_to_wait(wq, &wait.wq_entry, TASK_UNINTERRUPTIBLE); |
250df6ed2
|
1898 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1899 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1900 |
schedule(); |
214171361
|
1901 |
finish_wait(wq, &wait.wq_entry); |
67a23c494
|
1902 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1903 |
} |
1da177e4c
|
1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 |
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) { |
1da177e4c
|
1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 |
/* 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, |
3d375d785
|
1930 |
HASH_EARLY | HASH_ZERO, |
1da177e4c
|
1931 1932 |
&i_hash_shift, &i_hash_mask, |
31fe62b95
|
1933 |
0, |
1da177e4c
|
1934 |
0); |
1da177e4c
|
1935 |
} |
74bf17cff
|
1936 |
void __init inode_init(void) |
1da177e4c
|
1937 |
{ |
1da177e4c
|
1938 |
/* inode slab cache */ |
b0196009d
|
1939 1940 1941 1942 |
inode_cachep = kmem_cache_create("inode_cache", sizeof(struct inode), 0, (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC| |
5d097056c
|
1943 |
SLAB_MEM_SPREAD|SLAB_ACCOUNT), |
20c2df83d
|
1944 |
init_once); |
1da177e4c
|
1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 |
/* 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, |
3d375d785
|
1955 |
HASH_ZERO, |
1da177e4c
|
1956 1957 |
&i_hash_shift, &i_hash_mask, |
31fe62b95
|
1958 |
0, |
1da177e4c
|
1959 |
0); |
1da177e4c
|
1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 |
} 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
|
1972 |
inode->i_fop = &pipefifo_fops; |
1da177e4c
|
1973 |
else if (S_ISSOCK(mode)) |
bd9b51e79
|
1974 |
; /* leave it no_open_fops */ |
1da177e4c
|
1975 |
else |
af0d9ae81
|
1976 1977 1978 1979 |
printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o) for" " inode %s:%lu ", mode, inode->i_sb->s_id, inode->i_ino); |
1da177e4c
|
1980 1981 |
} EXPORT_SYMBOL(init_special_inode); |
a1bd120d1
|
1982 1983 |
/** |
eaae668d0
|
1984 |
* inode_init_owner - Init uid,gid,mode for new inode according to posix standards |
a1bd120d1
|
1985 1986 1987 1988 1989 |
* @inode: New inode * @dir: Directory inode * @mode: mode of the new inode */ void inode_init_owner(struct inode *inode, const struct inode *dir, |
62bb10917
|
1990 |
umode_t mode) |
a1bd120d1
|
1991 1992 1993 1994 |
{ inode->i_uid = current_fsuid(); if (dir && dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; |
0fa3ecd87
|
1995 1996 |
/* Directories are special, and always inherit S_ISGID */ |
a1bd120d1
|
1997 1998 |
if (S_ISDIR(mode)) mode |= S_ISGID; |
0fa3ecd87
|
1999 2000 2001 2002 |
else if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP) && !in_group_p(inode->i_gid) && !capable_wrt_inode_uidgid(dir, CAP_FSETID)) mode &= ~S_ISGID; |
a1bd120d1
|
2003 2004 2005 2006 2007 |
} else inode->i_gid = current_fsgid(); inode->i_mode = mode; } EXPORT_SYMBOL(inode_init_owner); |
e795b7179
|
2008 |
|
2e1496707
|
2009 2010 2011 2012 |
/** * inode_owner_or_capable - check current task permissions to inode * @inode: inode being checked * |
23adbe12e
|
2013 2014 |
* Return true if current either has CAP_FOWNER in a namespace with the * inode owner uid mapped, or owns the file. |
e795b7179
|
2015 |
*/ |
2e1496707
|
2016 |
bool inode_owner_or_capable(const struct inode *inode) |
e795b7179
|
2017 |
{ |
23adbe12e
|
2018 |
struct user_namespace *ns; |
92361636e
|
2019 |
if (uid_eq(current_fsuid(), inode->i_uid)) |
e795b7179
|
2020 |
return true; |
23adbe12e
|
2021 2022 |
ns = current_user_ns(); |
cc658db47
|
2023 |
if (kuid_has_mapping(ns, inode->i_uid) && ns_capable(ns, CAP_FOWNER)) |
e795b7179
|
2024 2025 2026 |
return true; return false; } |
2e1496707
|
2027 |
EXPORT_SYMBOL(inode_owner_or_capable); |
1d59d61f6
|
2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 |
/* * 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 { |
214171361
|
2038 |
prepare_to_wait(wq, &q.wq_entry, TASK_UNINTERRUPTIBLE); |
1d59d61f6
|
2039 2040 2041 |
if (atomic_read(&inode->i_dio_count)) schedule(); } while (atomic_read(&inode->i_dio_count)); |
214171361
|
2042 |
finish_wait(wq, &q.wq_entry); |
1d59d61f6
|
2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 |
} /** * 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); /* |
5f16f3225
|
2063 2064 2065 2066 2067 2068 2069 |
* 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 |
5fa8e0a1c
|
2070 2071 |
* code path which doesn't today so we use cmpxchg() out of an abundance * of caution. |
5f16f3225
|
2072 2073 2074 2075 2076 2077 2078 2079 2080 |
* * 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) { |
5f16f3225
|
2081 |
WARN_ON_ONCE(flags & ~mask); |
a905737fd
|
2082 |
set_mask_bits(&inode->i_flags, mask, flags); |
5f16f3225
|
2083 2084 |
} EXPORT_SYMBOL(inode_set_flags); |
21fc61c73
|
2085 2086 2087 2088 2089 2090 |
void inode_nohighmem(struct inode *inode) { mapping_set_gfp_mask(inode->i_mapping, GFP_USER); } EXPORT_SYMBOL(inode_nohighmem); |
3cd886666
|
2091 2092 |
/** |
8efd6894f
|
2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 |
* timespec64_trunc - Truncate timespec64 to a granularity * @t: Timespec64 * @gran: Granularity in ns. * * Truncate a timespec64 to a granularity. Always rounds down. gran must * not be 0 nor greater than a second (NSEC_PER_SEC, or 10^9 ns). */ struct timespec64 timespec64_trunc(struct timespec64 t, unsigned gran) { /* Avoid division in the common cases 1 ns and 1 s. */ if (gran == 1) { /* nothing */ } else if (gran == NSEC_PER_SEC) { t.tv_nsec = 0; } else if (gran > 1 && gran < NSEC_PER_SEC) { t.tv_nsec -= t.tv_nsec % gran; } else { WARN(1, "illegal file time granularity: %u", gran); } return t; } EXPORT_SYMBOL(timespec64_trunc); /** |
50e17c000
|
2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 |
* timestamp_truncate - Truncate timespec to a granularity * @t: Timespec * @inode: inode being updated * * Truncate a timespec to the granularity supported by the fs * containing the inode. Always rounds down. gran must * not be 0 nor greater than a second (NSEC_PER_SEC, or 10^9 ns). */ struct timespec64 timestamp_truncate(struct timespec64 t, struct inode *inode) { struct super_block *sb = inode->i_sb; unsigned int gran = sb->s_time_gran; t.tv_sec = clamp(t.tv_sec, sb->s_time_min, sb->s_time_max); if (unlikely(t.tv_sec == sb->s_time_max || t.tv_sec == sb->s_time_min)) t.tv_nsec = 0; /* Avoid division in the common cases 1 ns and 1 s. */ if (gran == 1) ; /* nothing */ else if (gran == NSEC_PER_SEC) t.tv_nsec = 0; else if (gran > 1 && gran < NSEC_PER_SEC) t.tv_nsec -= t.tv_nsec % gran; else WARN(1, "invalid file time granularity: %u", gran); return t; } EXPORT_SYMBOL(timestamp_truncate); /** |
3cd886666
|
2148 2149 2150 2151 2152 2153 2154 2155 2156 |
* current_time - Return FS time * @inode: inode. * * Return the current time truncated to the time granularity supported by * the fs. * * Note that inode and inode->sb cannot be NULL. * Otherwise, the function warns and returns time without truncation. */ |
95582b008
|
2157 |
struct timespec64 current_time(struct inode *inode) |
3cd886666
|
2158 |
{ |
d651d1607
|
2159 2160 2161 |
struct timespec64 now; ktime_get_coarse_real_ts64(&now); |
3cd886666
|
2162 2163 2164 2165 2166 |
if (unlikely(!inode->i_sb)) { WARN(1, "current_time() called with uninitialized super_block in the inode"); return now; } |
50e17c000
|
2167 |
return timestamp_truncate(now, inode); |
3cd886666
|
2168 2169 |
} EXPORT_SYMBOL(current_time); |
5aca28421
|
2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 |
/* * Generic function to check FS_IOC_SETFLAGS values and reject any invalid * configurations. * * Note: the caller should be holding i_mutex, or else be sure that they have * exclusive access to the inode structure. */ int vfs_ioc_setflags_prepare(struct inode *inode, unsigned int oldflags, unsigned int flags) { /* * The IMMUTABLE and APPEND_ONLY flags can only be changed by * the relevant capability. * * This test looks nicer. Thanks to Pauline Middelink */ if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL) && !capable(CAP_LINUX_IMMUTABLE)) return -EPERM; return 0; } EXPORT_SYMBOL(vfs_ioc_setflags_prepare); |
7b0e492e6
|
2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 |
/* * Generic function to check FS_IOC_FSSETXATTR values and reject any invalid * configurations. * * Note: the caller should be holding i_mutex, or else be sure that they have * exclusive access to the inode structure. */ int vfs_ioc_fssetxattr_check(struct inode *inode, const struct fsxattr *old_fa, struct fsxattr *fa) { /* * Can't modify an immutable/append-only file unless we have * appropriate permission. */ if ((old_fa->fsx_xflags ^ fa->fsx_xflags) & (FS_XFLAG_IMMUTABLE | FS_XFLAG_APPEND) && !capable(CAP_LINUX_IMMUTABLE)) return -EPERM; |
f991492ed
|
2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 |
/* * Project Quota ID state is only allowed to change from within the init * namespace. Enforce that restriction only if we are trying to change * the quota ID state. Everything else is allowed in user namespaces. */ if (current_user_ns() != &init_user_ns) { if (old_fa->fsx_projid != fa->fsx_projid) return -EINVAL; if ((old_fa->fsx_xflags ^ fa->fsx_xflags) & FS_XFLAG_PROJINHERIT) return -EINVAL; } |
ca29be753
|
2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 |
/* Check extent size hints. */ if ((fa->fsx_xflags & FS_XFLAG_EXTSIZE) && !S_ISREG(inode->i_mode)) return -EINVAL; if ((fa->fsx_xflags & FS_XFLAG_EXTSZINHERIT) && !S_ISDIR(inode->i_mode)) return -EINVAL; if ((fa->fsx_xflags & FS_XFLAG_COWEXTSIZE) && !S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode)) return -EINVAL; |
dbc77f31e
|
2236 2237 2238 2239 2240 2241 2242 |
/* * It is only valid to set the DAX flag on regular files and * directories on filesystems. */ if ((fa->fsx_xflags & FS_XFLAG_DAX) && !(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) return -EINVAL; |
ca29be753
|
2243 2244 2245 2246 2247 |
/* Extent size hints of zero turn off the flags. */ if (fa->fsx_extsize == 0) fa->fsx_xflags &= ~(FS_XFLAG_EXTSIZE | FS_XFLAG_EXTSZINHERIT); if (fa->fsx_cowextsize == 0) fa->fsx_xflags &= ~FS_XFLAG_COWEXTSIZE; |
7b0e492e6
|
2248 2249 2250 |
return 0; } EXPORT_SYMBOL(vfs_ioc_fssetxattr_check); |