Blame view
fs/inode.c
59.7 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
|
130 131 132 133 134 135 |
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; |
bd9b51e79
|
136 |
inode->i_fop = &no_open_fops; |
a78ef704a
|
137 |
inode->__i_nlink = 1; |
3ddcd0569
|
138 |
inode->i_opflags = 0; |
d0a5b995a
|
139 140 |
if (sb->s_xattr) inode->i_opflags |= IOP_XATTR; |
92361636e
|
141 142 |
i_uid_write(inode, 0); i_gid_write(inode, 0); |
2cb1599f9
|
143 144 |
atomic_set(&inode->i_writecount, 0); inode->i_size = 0; |
c75b1d942
|
145 |
inode->i_write_hint = WRITE_LIFE_NOT_SET; |
2cb1599f9
|
146 147 148 |
inode->i_blocks = 0; inode->i_bytes = 0; inode->i_generation = 0; |
2cb1599f9
|
149 150 151 |
inode->i_pipe = NULL; inode->i_bdev = NULL; inode->i_cdev = NULL; |
61ba64fc0
|
152 |
inode->i_link = NULL; |
84e710da2
|
153 |
inode->i_dir_seq = 0; |
2cb1599f9
|
154 155 |
inode->i_rdev = 0; inode->dirtied_when = 0; |
6146f0d5e
|
156 |
|
3d65ae463
|
157 158 159 160 161 |
#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
|
162 |
if (security_inode_alloc(inode)) |
54e346215
|
163 |
goto out; |
2cb1599f9
|
164 165 |
spin_lock_init(&inode->i_lock); lockdep_set_class(&inode->i_lock, &sb->s_type->i_lock_key); |
9902af79c
|
166 167 |
init_rwsem(&inode->i_rwsem); lockdep_set_class(&inode->i_rwsem, &sb->s_type->i_mutex_key); |
2cb1599f9
|
168 |
|
bd5fe6c5e
|
169 |
atomic_set(&inode->i_dio_count, 0); |
2cb1599f9
|
170 171 172 173 |
mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags = 0; |
829bc787c
|
174 |
mapping->wb_err = 0; |
4bb5f5d93
|
175 |
atomic_set(&mapping->i_mmap_writable, 0); |
3c1d43787
|
176 |
mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE); |
252aa6f5b
|
177 |
mapping->private_data = NULL; |
2cb1599f9
|
178 |
mapping->writeback_index = 0; |
2cb1599f9
|
179 180 |
inode->i_private = NULL; inode->i_mapping = mapping; |
b3d9b7a3c
|
181 |
INIT_HLIST_HEAD(&inode->i_dentry); /* buggered by rcu freeing */ |
f19d4a8fa
|
182 183 184 |
#ifdef CONFIG_FS_POSIX_ACL inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED; #endif |
2cb1599f9
|
185 |
|
3be25f49b
|
186 187 188 |
#ifdef CONFIG_FSNOTIFY inode->i_fsnotify_mask = 0; #endif |
4a075e39c
|
189 |
inode->i_flctx = NULL; |
3e880fb5e
|
190 |
this_cpu_inc(nr_inodes); |
cffbc8aa3
|
191 |
|
54e346215
|
192 |
return 0; |
54e346215
|
193 194 |
out: return -ENOMEM; |
1da177e4c
|
195 |
} |
2cb1599f9
|
196 |
EXPORT_SYMBOL(inode_init_always); |
fdb0da89f
|
197 198 199 200 201 202 203 204 205 206 207 208 209 210 |
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
|
211 212 |
static struct inode *alloc_inode(struct super_block *sb) { |
fdb0da89f
|
213 |
const struct super_operations *ops = sb->s_op; |
2cb1599f9
|
214 |
struct inode *inode; |
fdb0da89f
|
215 216 |
if (ops->alloc_inode) inode = ops->alloc_inode(sb); |
2cb1599f9
|
217 218 |
else inode = kmem_cache_alloc(inode_cachep, GFP_KERNEL); |
54e346215
|
219 220 221 222 |
if (!inode) return NULL; if (unlikely(inode_init_always(sb, inode))) { |
fdb0da89f
|
223 224 225 226 227 228 229 |
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
|
230 231 232 233 |
return NULL; } return inode; |
2cb1599f9
|
234 |
} |
1da177e4c
|
235 |
|
2e00c97e2
|
236 |
void __destroy_inode(struct inode *inode) |
1da177e4c
|
237 |
{ |
b7542f8c7
|
238 |
BUG_ON(inode_has_buffers(inode)); |
52ebea749
|
239 |
inode_detach_wb(inode); |
1da177e4c
|
240 |
security_inode_free(inode); |
3be25f49b
|
241 |
fsnotify_inode_delete(inode); |
f27a0fe08
|
242 |
locks_free_lock_context(inode); |
7ada4db88
|
243 244 245 246 |
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
|
247 |
#ifdef CONFIG_FS_POSIX_ACL |
b8a7a3a66
|
248 |
if (inode->i_acl && !is_uncached_acl(inode->i_acl)) |
f19d4a8fa
|
249 |
posix_acl_release(inode->i_acl); |
b8a7a3a66
|
250 |
if (inode->i_default_acl && !is_uncached_acl(inode->i_default_acl)) |
f19d4a8fa
|
251 252 |
posix_acl_release(inode->i_default_acl); #endif |
3e880fb5e
|
253 |
this_cpu_dec(nr_inodes); |
2e00c97e2
|
254 255 |
} EXPORT_SYMBOL(__destroy_inode); |
56b0dacfa
|
256 |
static void destroy_inode(struct inode *inode) |
2e00c97e2
|
257 |
{ |
fdb0da89f
|
258 |
const struct super_operations *ops = inode->i_sb->s_op; |
7ccf19a80
|
259 |
BUG_ON(!list_empty(&inode->i_lru)); |
2e00c97e2
|
260 |
__destroy_inode(inode); |
fdb0da89f
|
261 262 263 264 265 266 267 |
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
|
268 |
} |
1da177e4c
|
269 |
|
7ada4db88
|
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 |
/** * 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
|
318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 |
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
|
339 340 |
if (unlikely(inode->i_nlink == 0)) { WARN_ON(!(inode->i_state & I_LINKABLE)); |
7ada4db88
|
341 |
atomic_long_dec(&inode->i_sb->s_remove_count); |
f4e0c30c1
|
342 |
} |
7ada4db88
|
343 344 345 346 |
inode->__i_nlink++; } EXPORT_SYMBOL(inc_nlink); |
ae23395d8
|
347 |
static void __address_space_init_once(struct address_space *mapping) |
2aa15890f
|
348 |
{ |
7b785645e
|
349 |
xa_init_flags(&mapping->i_pages, XA_FLAGS_LOCK_IRQ | XA_FLAGS_ACCOUNT); |
c8c06efa8
|
350 |
init_rwsem(&mapping->i_mmap_rwsem); |
2aa15890f
|
351 352 |
INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); |
f808c13fd
|
353 |
mapping->i_mmap = RB_ROOT_CACHED; |
2aa15890f
|
354 |
} |
ae23395d8
|
355 356 357 358 359 360 |
void address_space_init_once(struct address_space *mapping) { memset(mapping, 0, sizeof(*mapping)); __address_space_init_once(mapping); } |
2aa15890f
|
361 |
EXPORT_SYMBOL(address_space_init_once); |
1da177e4c
|
362 363 364 365 366 367 368 369 370 |
/* * 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
|
371 |
INIT_LIST_HEAD(&inode->i_devices); |
c7f540849
|
372 |
INIT_LIST_HEAD(&inode->i_io_list); |
6c60d2b57
|
373 |
INIT_LIST_HEAD(&inode->i_wb_list); |
7ccf19a80
|
374 |
INIT_LIST_HEAD(&inode->i_lru); |
ae23395d8
|
375 |
__address_space_init_once(&inode->i_data); |
1da177e4c
|
376 377 |
i_size_ordered_init(inode); } |
1da177e4c
|
378 |
EXPORT_SYMBOL(inode_init_once); |
51cc50685
|
379 |
static void init_once(void *foo) |
1da177e4c
|
380 |
{ |
6b3304b53
|
381 |
struct inode *inode = (struct inode *) foo; |
1da177e4c
|
382 |
|
a35afb830
|
383 |
inode_init_once(inode); |
1da177e4c
|
384 385 386 |
} /* |
250df6ed2
|
387 |
* inode->i_lock must be held |
1da177e4c
|
388 |
*/ |
6b3304b53
|
389 |
void __iget(struct inode *inode) |
1da177e4c
|
390 |
{ |
9e38d86ff
|
391 392 |
atomic_inc(&inode->i_count); } |
2e147f1ef
|
393 |
|
7de9c6ee3
|
394 395 396 397 398 399 400 401 |
/* * 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
|
402 403 |
static void inode_lru_list_add(struct inode *inode) { |
bc3b14cb2
|
404 |
if (list_lru_add(&inode->i_sb->s_inode_lru, &inode->i_lru)) |
fcb94f72d
|
405 |
this_cpu_inc(nr_unused); |
563f40019
|
406 407 |
else inode->i_state |= I_REFERENCED; |
9e38d86ff
|
408 |
} |
2e147f1ef
|
409 |
|
4eff96dd5
|
410 411 412 413 414 415 416 |
/* * Add inode to LRU if needed (inode is unused and clean). * * Needs inode->i_lock held. */ void inode_add_lru(struct inode *inode) { |
0ae45f63d
|
417 418 |
if (!(inode->i_state & (I_DIRTY_ALL | I_SYNC | I_FREEING | I_WILL_FREE)) && |
1751e8a6c
|
419 |
!atomic_read(&inode->i_count) && inode->i_sb->s_flags & SB_ACTIVE) |
4eff96dd5
|
420 421 |
inode_lru_list_add(inode); } |
9e38d86ff
|
422 423 |
static void inode_lru_list_del(struct inode *inode) { |
bc3b14cb2
|
424 425 |
if (list_lru_del(&inode->i_sb->s_inode_lru, &inode->i_lru)) |
fcb94f72d
|
426 |
this_cpu_dec(nr_unused); |
1da177e4c
|
427 |
} |
646ec4615
|
428 429 430 431 432 433 |
/** * 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
|
434 |
spin_lock(&inode->i_sb->s_inode_list_lock); |
55fa6091d
|
435 |
list_add(&inode->i_sb_list, &inode->i_sb->s_inodes); |
74278da9f
|
436 |
spin_unlock(&inode->i_sb->s_inode_list_lock); |
646ec4615
|
437 438 |
} EXPORT_SYMBOL_GPL(inode_sb_list_add); |
55fa6091d
|
439 |
static inline void inode_sb_list_del(struct inode *inode) |
646ec4615
|
440 |
{ |
a209dfc7b
|
441 |
if (!list_empty(&inode->i_sb_list)) { |
74278da9f
|
442 |
spin_lock(&inode->i_sb->s_inode_list_lock); |
a209dfc7b
|
443 |
list_del_init(&inode->i_sb_list); |
74278da9f
|
444 |
spin_unlock(&inode->i_sb->s_inode_list_lock); |
a209dfc7b
|
445 |
} |
646ec4615
|
446 |
} |
4c51acbc6
|
447 448 449 450 451 452 |
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
|
453 454 |
tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> i_hash_shift); return tmp & i_hash_mask; |
4c51acbc6
|
455 456 457 458 459 460 461 462 463 464 465 466 |
} /** * __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
|
467 |
struct hlist_head *b = inode_hashtable + hash(inode->i_sb, hashval); |
67a23c494
|
468 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
469 |
spin_lock(&inode->i_lock); |
646ec4615
|
470 |
hlist_add_head(&inode->i_hash, b); |
250df6ed2
|
471 |
spin_unlock(&inode->i_lock); |
67a23c494
|
472 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
473 474 475 476 |
} EXPORT_SYMBOL(__insert_inode_hash); /** |
f2ee7abf4
|
477 |
* __remove_inode_hash - remove an inode from the hash |
4c51acbc6
|
478 479 480 481 |
* @inode: inode to unhash * * Remove an inode from the superblock. */ |
f2ee7abf4
|
482 |
void __remove_inode_hash(struct inode *inode) |
4c51acbc6
|
483 |
{ |
67a23c494
|
484 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
485 |
spin_lock(&inode->i_lock); |
4c51acbc6
|
486 |
hlist_del_init(&inode->i_hash); |
250df6ed2
|
487 |
spin_unlock(&inode->i_lock); |
67a23c494
|
488 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
489 |
} |
f2ee7abf4
|
490 |
EXPORT_SYMBOL(__remove_inode_hash); |
4c51acbc6
|
491 |
|
dbd5768f8
|
492 |
void clear_inode(struct inode *inode) |
b0683aa63
|
493 |
{ |
08142579b
|
494 |
/* |
b93b01631
|
495 |
* We have to cycle the i_pages lock here because reclaim can be in the |
08142579b
|
496 |
* process of removing the last page (in __delete_from_page_cache()) |
b93b01631
|
497 |
* and we must not free the mapping under it. |
08142579b
|
498 |
*/ |
b93b01631
|
499 |
xa_lock_irq(&inode->i_data.i_pages); |
b0683aa63
|
500 |
BUG_ON(inode->i_data.nrpages); |
f9fe48bec
|
501 |
BUG_ON(inode->i_data.nrexceptional); |
b93b01631
|
502 |
xa_unlock_irq(&inode->i_data.i_pages); |
b0683aa63
|
503 504 505 |
BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); |
6c60d2b57
|
506 |
BUG_ON(!list_empty(&inode->i_wb_list)); |
fa0d7e3de
|
507 |
/* don't need i_lock here, no concurrent mods to i_state */ |
b0683aa63
|
508 509 |
inode->i_state = I_FREEING | I_CLEAR; } |
dbd5768f8
|
510 |
EXPORT_SYMBOL(clear_inode); |
b0683aa63
|
511 |
|
b2b2af8e6
|
512 513 514 515 516 517 518 519 520 521 522 523 524 |
/* * 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
|
525 |
static void evict(struct inode *inode) |
b4272d4c8
|
526 527 |
{ const struct super_operations *op = inode->i_sb->s_op; |
b2b2af8e6
|
528 529 |
BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(!list_empty(&inode->i_lru)); |
c7f540849
|
530 531 |
if (!list_empty(&inode->i_io_list)) inode_io_list_del(inode); |
b12362bdb
|
532 |
|
55fa6091d
|
533 |
inode_sb_list_del(inode); |
169ebd901
|
534 535 536 537 538 539 540 |
/* * 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
|
541 |
|
be7ce4161
|
542 543 |
if (op->evict_inode) { op->evict_inode(inode); |
b4272d4c8
|
544 |
} else { |
91b0abe36
|
545 |
truncate_inode_pages_final(&inode->i_data); |
dbd5768f8
|
546 |
clear_inode(inode); |
b4272d4c8
|
547 |
} |
661074e91
|
548 549 550 551 |
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
|
552 553 554 555 556 557 558 559 560 |
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
|
561 |
} |
1da177e4c
|
562 563 564 565 566 567 568 569 570 |
/* * 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
|
571 572 |
while (!list_empty(head)) { struct inode *inode; |
7ccf19a80
|
573 574 |
inode = list_first_entry(head, struct inode, i_lru); list_del_init(&inode->i_lru); |
1da177e4c
|
575 |
|
644da5960
|
576 |
evict(inode); |
ac05fbb40
|
577 |
cond_resched(); |
1da177e4c
|
578 |
} |
1da177e4c
|
579 |
} |
a03187867
|
580 |
/** |
63997e98a
|
581 582 583 584 |
* 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
|
585 |
* called by superblock shutdown after having SB_ACTIVE flag removed, |
63997e98a
|
586 587 |
* so any inode reaching zero refcount during or after that call will * be immediately evicted. |
1da177e4c
|
588 |
*/ |
63997e98a
|
589 |
void evict_inodes(struct super_block *sb) |
1da177e4c
|
590 |
{ |
63997e98a
|
591 592 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
593 |
|
ac05fbb40
|
594 |
again: |
74278da9f
|
595 |
spin_lock(&sb->s_inode_list_lock); |
63997e98a
|
596 597 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { if (atomic_read(&inode->i_count)) |
aabb8fdb4
|
598 |
continue; |
250df6ed2
|
599 600 601 602 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
1da177e4c
|
603 |
continue; |
250df6ed2
|
604 |
} |
63997e98a
|
605 606 |
inode->i_state |= I_FREEING; |
02afc410f
|
607 |
inode_lru_list_del(inode); |
250df6ed2
|
608 |
spin_unlock(&inode->i_lock); |
02afc410f
|
609 |
list_add(&inode->i_lru, &dispose); |
ac05fbb40
|
610 611 612 613 614 615 616 617 618 619 620 621 |
/* * 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
|
622 |
} |
74278da9f
|
623 |
spin_unlock(&sb->s_inode_list_lock); |
63997e98a
|
624 625 |
dispose_list(&dispose); |
1da177e4c
|
626 |
} |
799ea9e9c
|
627 |
EXPORT_SYMBOL_GPL(evict_inodes); |
1da177e4c
|
628 |
|
1da177e4c
|
629 |
/** |
a03187867
|
630 631 |
* invalidate_inodes - attempt to free all inodes on a superblock * @sb: superblock to operate on |
93b270f76
|
632 |
* @kill_dirty: flag to guide handling of dirty inodes |
1da177e4c
|
633 |
* |
a03187867
|
634 635 |
* Attempts to free all inodes for a given superblock. If there were any * busy inodes return a non-zero value, else zero. |
93b270f76
|
636 637 |
* If @kill_dirty is set, discard dirty inodes too, otherwise treat * them as busy. |
1da177e4c
|
638 |
*/ |
93b270f76
|
639 |
int invalidate_inodes(struct super_block *sb, bool kill_dirty) |
1da177e4c
|
640 |
{ |
cffbc8aa3
|
641 |
int busy = 0; |
a03187867
|
642 643 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
644 |
|
74278da9f
|
645 |
spin_lock(&sb->s_inode_list_lock); |
a03187867
|
646 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { |
250df6ed2
|
647 648 649 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
aabb8fdb4
|
650 |
continue; |
250df6ed2
|
651 |
} |
0ae45f63d
|
652 |
if (inode->i_state & I_DIRTY_ALL && !kill_dirty) { |
250df6ed2
|
653 |
spin_unlock(&inode->i_lock); |
93b270f76
|
654 655 656 |
busy = 1; continue; } |
99a389192
|
657 |
if (atomic_read(&inode->i_count)) { |
250df6ed2
|
658 |
spin_unlock(&inode->i_lock); |
99a389192
|
659 |
busy = 1; |
1da177e4c
|
660 661 |
continue; } |
99a389192
|
662 |
|
99a389192
|
663 |
inode->i_state |= I_FREEING; |
02afc410f
|
664 |
inode_lru_list_del(inode); |
250df6ed2
|
665 |
spin_unlock(&inode->i_lock); |
02afc410f
|
666 |
list_add(&inode->i_lru, &dispose); |
1da177e4c
|
667 |
} |
74278da9f
|
668 |
spin_unlock(&sb->s_inode_list_lock); |
1da177e4c
|
669 |
|
a03187867
|
670 |
dispose_list(&dispose); |
1da177e4c
|
671 672 673 |
return busy; } |
1da177e4c
|
674 |
|
1da177e4c
|
675 |
/* |
bc3b14cb2
|
676 |
* Isolate the inode from the LRU in preparation for freeing it. |
1da177e4c
|
677 678 |
* * Any inodes which are pinned purely because of attached pagecache have their |
9e38d86ff
|
679 680 |
* pagecache removed. If the inode has metadata buffers attached to * mapping->private_list then try to remove them. |
1da177e4c
|
681 |
* |
9e38d86ff
|
682 683 684 685 686 687 688 |
* 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
|
689 |
*/ |
3f97b1632
|
690 691 |
static enum lru_status inode_lru_isolate(struct list_head *item, struct list_lru_one *lru, spinlock_t *lru_lock, void *arg) |
1da177e4c
|
692 |
{ |
bc3b14cb2
|
693 694 |
struct list_head *freeable = arg; struct inode *inode = container_of(item, struct inode, i_lru); |
1da177e4c
|
695 |
|
bc3b14cb2
|
696 697 698 699 700 701 |
/* * 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
|
702 |
|
bc3b14cb2
|
703 704 705 706 707 708 |
/* * 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
|
709 |
list_lru_isolate(lru, &inode->i_lru); |
bc3b14cb2
|
710 711 712 713 |
spin_unlock(&inode->i_lock); this_cpu_dec(nr_unused); return LRU_REMOVED; } |
1da177e4c
|
714 |
|
69056ee6a
|
715 716 |
/* recently referenced inodes get one more pass */ if (inode->i_state & I_REFERENCED) { |
bc3b14cb2
|
717 718 719 720 |
inode->i_state &= ~I_REFERENCED; spin_unlock(&inode->i_lock); return LRU_ROTATE; } |
1da177e4c
|
721 |
|
bc3b14cb2
|
722 723 724 725 726 727 728 729 730 731 732 733 734 |
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
|
735 |
} |
bc3b14cb2
|
736 737 738 739 |
iput(inode); spin_lock(lru_lock); return LRU_RETRY; } |
02afc410f
|
740 |
|
bc3b14cb2
|
741 742 |
WARN_ON(inode->i_state & I_NEW); inode->i_state |= I_FREEING; |
3f97b1632
|
743 |
list_lru_isolate_move(lru, &inode->i_lru, freeable); |
bc3b14cb2
|
744 |
spin_unlock(&inode->i_lock); |
9e38d86ff
|
745 |
|
bc3b14cb2
|
746 747 748 |
this_cpu_dec(nr_unused); return LRU_REMOVED; } |
7ccf19a80
|
749 |
|
bc3b14cb2
|
750 751 752 753 754 755 |
/* * 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
|
756 |
long prune_icache_sb(struct super_block *sb, struct shrink_control *sc) |
bc3b14cb2
|
757 758 759 |
{ LIST_HEAD(freeable); long freed; |
1da177e4c
|
760 |
|
503c358cf
|
761 762 |
freed = list_lru_shrink_walk(&sb->s_inode_lru, sc, inode_lru_isolate, &freeable); |
1da177e4c
|
763 |
dispose_list(&freeable); |
0a234c6dc
|
764 |
return freed; |
1da177e4c
|
765 |
} |
1da177e4c
|
766 767 768 |
static void __wait_on_freeing_inode(struct inode *inode); /* * Called with the inode lock held. |
1da177e4c
|
769 |
*/ |
6b3304b53
|
770 771 772 773 |
static struct inode *find_inode(struct super_block *sb, struct hlist_head *head, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
774 |
{ |
6b3304b53
|
775 |
struct inode *inode = NULL; |
1da177e4c
|
776 777 |
repeat: |
b67bfe0d4
|
778 |
hlist_for_each_entry(inode, head, i_hash) { |
5a3cd9928
|
779 |
if (inode->i_sb != sb) |
1da177e4c
|
780 |
continue; |
5a3cd9928
|
781 |
if (!test(inode, data)) |
1da177e4c
|
782 |
continue; |
5a3cd9928
|
783 |
spin_lock(&inode->i_lock); |
a4ffdde6e
|
784 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
785 786 787 |
__wait_on_freeing_inode(inode); goto repeat; } |
c2b6d621c
|
788 789 790 791 |
if (unlikely(inode->i_state & I_CREATING)) { spin_unlock(&inode->i_lock); return ERR_PTR(-ESTALE); } |
f7899bd54
|
792 |
__iget(inode); |
250df6ed2
|
793 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
794 |
return inode; |
1da177e4c
|
795 |
} |
f7899bd54
|
796 |
return NULL; |
1da177e4c
|
797 798 799 800 801 802 |
} /* * find_inode_fast is the fast path version of find_inode, see the comment at * iget_locked for details. */ |
6b3304b53
|
803 804 |
static struct inode *find_inode_fast(struct super_block *sb, struct hlist_head *head, unsigned long ino) |
1da177e4c
|
805 |
{ |
6b3304b53
|
806 |
struct inode *inode = NULL; |
1da177e4c
|
807 808 |
repeat: |
b67bfe0d4
|
809 |
hlist_for_each_entry(inode, head, i_hash) { |
5a3cd9928
|
810 |
if (inode->i_ino != ino) |
1da177e4c
|
811 |
continue; |
5a3cd9928
|
812 |
if (inode->i_sb != sb) |
1da177e4c
|
813 |
continue; |
5a3cd9928
|
814 |
spin_lock(&inode->i_lock); |
a4ffdde6e
|
815 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
816 817 818 |
__wait_on_freeing_inode(inode); goto repeat; } |
c2b6d621c
|
819 820 821 822 |
if (unlikely(inode->i_state & I_CREATING)) { spin_unlock(&inode->i_lock); return ERR_PTR(-ESTALE); } |
f7899bd54
|
823 |
__iget(inode); |
250df6ed2
|
824 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
825 |
return inode; |
1da177e4c
|
826 |
} |
f7899bd54
|
827 |
return NULL; |
8290c35f8
|
828 |
} |
f991bd2e1
|
829 830 831 832 |
/* * 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
|
833 |
* |
f991bd2e1
|
834 835 836 837 838 839 840 841 842 |
* 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
|
843 |
*/ |
f991bd2e1
|
844 845 |
#define LAST_INO_BATCH 1024 static DEFINE_PER_CPU(unsigned int, last_ino); |
85fe4025c
|
846 |
unsigned int get_next_ino(void) |
8290c35f8
|
847 |
{ |
f991bd2e1
|
848 849 |
unsigned int *p = &get_cpu_var(last_ino); unsigned int res = *p; |
8290c35f8
|
850 |
|
f991bd2e1
|
851 852 853 854 855 856 857 858 |
#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
|
859 860 861 862 863 |
res++; /* get_next_ino should not provide a 0 inode number */ if (unlikely(!res)) res++; *p = res; |
f991bd2e1
|
864 865 |
put_cpu_var(last_ino); return res; |
8290c35f8
|
866 |
} |
85fe4025c
|
867 |
EXPORT_SYMBOL(get_next_ino); |
8290c35f8
|
868 |
|
1da177e4c
|
869 |
/** |
a209dfc7b
|
870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 |
* 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
|
893 894 895 |
* new_inode - obtain an inode * @sb: superblock * |
769848c03
|
896 |
* Allocates a new inode for given superblock. The default gfp_mask |
3c1d43787
|
897 |
* for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE. |
769848c03
|
898 899 900 901 902 |
* 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
|
903 904 905 |
*/ struct inode *new_inode(struct super_block *sb) { |
6b3304b53
|
906 |
struct inode *inode; |
1da177e4c
|
907 |
|
74278da9f
|
908 |
spin_lock_prefetch(&sb->s_inode_list_lock); |
6b3304b53
|
909 |
|
a209dfc7b
|
910 911 |
inode = new_inode_pseudo(sb); if (inode) |
55fa6091d
|
912 |
inode_sb_list_add(inode); |
1da177e4c
|
913 914 |
return inode; } |
1da177e4c
|
915 |
EXPORT_SYMBOL(new_inode); |
14358e6dd
|
916 |
#ifdef CONFIG_DEBUG_LOCK_ALLOC |
e096d0c7e
|
917 918 |
void lockdep_annotate_inode_mutex_key(struct inode *inode) { |
a3314a0ed
|
919 |
if (S_ISDIR(inode->i_mode)) { |
1e89a5e15
|
920 |
struct file_system_type *type = inode->i_sb->s_type; |
9a7aa12f3
|
921 |
/* Set new key only if filesystem hasn't already changed it */ |
9902af79c
|
922 |
if (lockdep_match_class(&inode->i_rwsem, &type->i_mutex_key)) { |
9a7aa12f3
|
923 924 925 |
/* * ensure nobody is actually holding i_mutex */ |
9902af79c
|
926 927 928 |
// mutex_destroy(&inode->i_mutex); init_rwsem(&inode->i_rwsem); lockdep_set_class(&inode->i_rwsem, |
9a7aa12f3
|
929 930 |
&type->i_mutex_dir_key); } |
1e89a5e15
|
931 |
} |
e096d0c7e
|
932 933 |
} EXPORT_SYMBOL(lockdep_annotate_inode_mutex_key); |
14358e6dd
|
934 |
#endif |
e096d0c7e
|
935 936 937 938 939 940 941 942 943 944 945 |
/** * 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
|
946 |
spin_lock(&inode->i_lock); |
eaff8079d
|
947 |
WARN_ON(!(inode->i_state & I_NEW)); |
c2b6d621c
|
948 |
inode->i_state &= ~I_NEW & ~I_CREATING; |
310fa7a36
|
949 |
smp_mb(); |
250df6ed2
|
950 951 |
wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); |
1da177e4c
|
952 |
} |
1da177e4c
|
953 |
EXPORT_SYMBOL(unlock_new_inode); |
c2b6d621c
|
954 955 956 957 958 959 960 961 962 963 964 965 |
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
|
966 |
/** |
375e289ea
|
967 |
* lock_two_nondirectories - take two i_mutexes on non-directory objects |
4fd699ae3
|
968 969 970 971 |
* * Lock any non-NULL argument that is not a directory. * Zero, one or two objects may be locked by this function. * |
375e289ea
|
972 973 974 975 976 |
* @inode1: first inode to lock * @inode2: second inode to lock */ void lock_two_nondirectories(struct inode *inode1, struct inode *inode2) { |
4fd699ae3
|
977 978 979 980 |
if (inode1 > inode2) swap(inode1, inode2); if (inode1 && !S_ISDIR(inode1->i_mode)) |
5955102c9
|
981 |
inode_lock(inode1); |
4fd699ae3
|
982 |
if (inode2 && !S_ISDIR(inode2->i_mode) && inode2 != inode1) |
5955102c9
|
983 |
inode_lock_nested(inode2, I_MUTEX_NONDIR2); |
375e289ea
|
984 985 986 987 988 989 990 991 992 993 |
} 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
|
994 |
if (inode1 && !S_ISDIR(inode1->i_mode)) |
5955102c9
|
995 |
inode_unlock(inode1); |
4fd699ae3
|
996 |
if (inode2 && !S_ISDIR(inode2->i_mode) && inode2 != inode1) |
5955102c9
|
997 |
inode_unlock(inode2); |
375e289ea
|
998 999 1000 1001 |
} EXPORT_SYMBOL(unlock_two_nondirectories); /** |
80ea09a00
|
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 |
* 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
|
1027 |
bool creating = inode->i_state & I_CREATING; |
80ea09a00
|
1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 |
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
|
1038 1039 |
if (IS_ERR(old)) return NULL; |
80ea09a00
|
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 |
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
|
1061 1062 |
if (!creating) inode_sb_list_add(inode); |
80ea09a00
|
1063 1064 1065 1066 1067 1068 1069 1070 |
unlock: spin_unlock(&inode_hash_lock); return inode; } EXPORT_SYMBOL(inode_insert5); /** |
0b2d0724e
|
1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 |
* 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
|
1086 |
* |
0b2d0724e
|
1087 1088 |
* Note both @test and @set are called with the inode_hash_lock held, so can't * sleep. |
1da177e4c
|
1089 |
*/ |
0b2d0724e
|
1090 1091 1092 |
struct inode *iget5_locked(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data) |
1da177e4c
|
1093 |
{ |
80ea09a00
|
1094 |
struct inode *inode = ilookup5(sb, hashval, test, data); |
0b2d0724e
|
1095 |
|
80ea09a00
|
1096 |
if (!inode) { |
e950564b9
|
1097 |
struct inode *new = alloc_inode(sb); |
0b2d0724e
|
1098 |
|
80ea09a00
|
1099 |
if (new) { |
e950564b9
|
1100 |
new->i_state = 0; |
80ea09a00
|
1101 1102 |
inode = inode_insert5(new, hashval, test, set, data); if (unlikely(inode != new)) |
e950564b9
|
1103 |
destroy_inode(new); |
2864f3014
|
1104 |
} |
1da177e4c
|
1105 1106 |
} return inode; |
1da177e4c
|
1107 |
} |
0b2d0724e
|
1108 |
EXPORT_SYMBOL(iget5_locked); |
1da177e4c
|
1109 |
|
0b2d0724e
|
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 |
/** * 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
|
1122 |
*/ |
0b2d0724e
|
1123 |
struct inode *iget_locked(struct super_block *sb, unsigned long ino) |
1da177e4c
|
1124 |
{ |
0b2d0724e
|
1125 |
struct hlist_head *head = inode_hashtable + hash(sb, ino); |
6b3304b53
|
1126 |
struct inode *inode; |
2864f3014
|
1127 |
again: |
0b2d0724e
|
1128 1129 1130 1131 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); if (inode) { |
c2b6d621c
|
1132 1133 |
if (IS_ERR(inode)) return NULL; |
0b2d0724e
|
1134 |
wait_on_inode(inode); |
2864f3014
|
1135 1136 1137 1138 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } |
0b2d0724e
|
1139 1140 |
return inode; } |
1da177e4c
|
1141 1142 |
inode = alloc_inode(sb); if (inode) { |
6b3304b53
|
1143 |
struct inode *old; |
1da177e4c
|
1144 |
|
67a23c494
|
1145 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1146 1147 1148 1149 |
/* We released the lock, so.. */ old = find_inode_fast(sb, head, ino); if (!old) { inode->i_ino = ino; |
250df6ed2
|
1150 1151 |
spin_lock(&inode->i_lock); inode->i_state = I_NEW; |
646ec4615
|
1152 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1153 |
spin_unlock(&inode->i_lock); |
55fa6091d
|
1154 |
inode_sb_list_add(inode); |
67a23c494
|
1155 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 |
/* 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
|
1168 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1169 |
destroy_inode(inode); |
c2b6d621c
|
1170 1171 |
if (IS_ERR(old)) return NULL; |
1da177e4c
|
1172 1173 |
inode = old; wait_on_inode(inode); |
2864f3014
|
1174 1175 1176 1177 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } |
1da177e4c
|
1178 1179 1180 |
} return inode; } |
0b2d0724e
|
1181 |
EXPORT_SYMBOL(iget_locked); |
1da177e4c
|
1182 |
|
ad5e195ac
|
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 |
/* * 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
|
1193 |
struct inode *inode; |
67a23c494
|
1194 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1195 |
hlist_for_each_entry(inode, b, i_hash) { |
67a23c494
|
1196 1197 |
if (inode->i_ino == ino && inode->i_sb == sb) { spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1198 |
return 0; |
67a23c494
|
1199 |
} |
ad5e195ac
|
1200 |
} |
67a23c494
|
1201 |
spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1202 1203 1204 |
return 1; } |
1da177e4c
|
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 |
/** * 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
|
1221 1222 1223 1224 1225 |
/* * 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
|
1226 |
static DEFINE_SPINLOCK(iunique_lock); |
866b04fcc
|
1227 |
static unsigned int counter; |
1da177e4c
|
1228 |
ino_t res; |
3361c7beb
|
1229 |
|
ad5e195ac
|
1230 |
spin_lock(&iunique_lock); |
3361c7beb
|
1231 1232 1233 |
do { if (counter <= max_reserved) counter = max_reserved + 1; |
1da177e4c
|
1234 |
res = counter++; |
ad5e195ac
|
1235 1236 |
} while (!test_inode_iunique(sb, res)); spin_unlock(&iunique_lock); |
1da177e4c
|
1237 |
|
3361c7beb
|
1238 1239 |
return res; } |
1da177e4c
|
1240 1241 1242 1243 |
EXPORT_SYMBOL(iunique); struct inode *igrab(struct inode *inode) { |
250df6ed2
|
1244 1245 |
spin_lock(&inode->i_lock); if (!(inode->i_state & (I_FREEING|I_WILL_FREE))) { |
1da177e4c
|
1246 |
__iget(inode); |
250df6ed2
|
1247 1248 1249 |
spin_unlock(&inode->i_lock); } else { spin_unlock(&inode->i_lock); |
1da177e4c
|
1250 1251 1252 1253 1254 1255 |
/* * 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
|
1256 |
} |
1da177e4c
|
1257 1258 |
return inode; } |
1da177e4c
|
1259 1260 1261 |
EXPORT_SYMBOL(igrab); /** |
0b2d0724e
|
1262 |
* ilookup5_nowait - search for an inode in the inode cache |
1da177e4c
|
1263 |
* @sb: super block of file system to search |
0b2d0724e
|
1264 |
* @hashval: hash value (usually inode number) to search for |
1da177e4c
|
1265 1266 |
* @test: callback used for comparisons between inodes * @data: opaque data pointer to pass to @test |
1da177e4c
|
1267 |
* |
0b2d0724e
|
1268 |
* Search for the inode specified by @hashval and @data in the inode cache. |
1da177e4c
|
1269 1270 1271 |
* If the inode is in the cache, the inode is returned with an incremented * reference count. * |
0b2d0724e
|
1272 1273 |
* 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
|
1274 |
* |
b6d0ad686
|
1275 |
* Note2: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1276 |
*/ |
0b2d0724e
|
1277 1278 |
struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
1279 |
{ |
0b2d0724e
|
1280 |
struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
1da177e4c
|
1281 |
struct inode *inode; |
67a23c494
|
1282 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1283 |
inode = find_inode(sb, head, test, data); |
67a23c494
|
1284 |
spin_unlock(&inode_hash_lock); |
88bd5121d
|
1285 |
|
c2b6d621c
|
1286 |
return IS_ERR(inode) ? NULL : inode; |
88bd5121d
|
1287 |
} |
88bd5121d
|
1288 1289 1290 1291 1292 1293 1294 1295 1296 |
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
|
1297 1298 1299 |
* 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
|
1300 |
* returned with an incremented reference count. |
1da177e4c
|
1301 |
* |
0b2d0724e
|
1302 1303 |
* This is a generalized version of ilookup() for file systems where the * inode number is not sufficient for unique identification of an inode. |
1da177e4c
|
1304 |
* |
0b2d0724e
|
1305 |
* Note: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1306 1307 1308 1309 |
*/ struct inode *ilookup5(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { |
2864f3014
|
1310 1311 1312 1313 |
struct inode *inode; again: inode = ilookup5_nowait(sb, hashval, test, data); if (inode) { |
0b2d0724e
|
1314 |
wait_on_inode(inode); |
2864f3014
|
1315 1316 1317 1318 1319 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } } |
0b2d0724e
|
1320 |
return inode; |
1da177e4c
|
1321 |
} |
1da177e4c
|
1322 1323 1324 1325 1326 1327 1328 |
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
|
1329 1330 |
* 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
|
1331 1332 1333 1334 |
*/ struct inode *ilookup(struct super_block *sb, unsigned long ino) { struct hlist_head *head = inode_hashtable + hash(sb, ino); |
1da177e4c
|
1335 |
struct inode *inode; |
2864f3014
|
1336 |
again: |
0b2d0724e
|
1337 1338 1339 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); |
1da177e4c
|
1340 |
|
2864f3014
|
1341 |
if (inode) { |
c2b6d621c
|
1342 1343 |
if (IS_ERR(inode)) return NULL; |
0b2d0724e
|
1344 |
wait_on_inode(inode); |
2864f3014
|
1345 1346 1347 1348 1349 |
if (unlikely(inode_unhashed(inode))) { iput(inode); goto again; } } |
0b2d0724e
|
1350 |
return inode; |
1da177e4c
|
1351 |
} |
0b2d0724e
|
1352 |
EXPORT_SYMBOL(ilookup); |
1da177e4c
|
1353 |
|
fe032c422
|
1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 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 |
/** * 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
|
1403 1404 1405 1406 1407 |
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
|
1408 |
|
261bca86e
|
1409 |
while (1) { |
72a43d63c
|
1410 |
struct inode *old = NULL; |
67a23c494
|
1411 |
spin_lock(&inode_hash_lock); |
b67bfe0d4
|
1412 |
hlist_for_each_entry(old, head, i_hash) { |
72a43d63c
|
1413 1414 1415 1416 |
if (old->i_ino != ino) continue; if (old->i_sb != sb) continue; |
250df6ed2
|
1417 1418 1419 |
spin_lock(&old->i_lock); if (old->i_state & (I_FREEING|I_WILL_FREE)) { spin_unlock(&old->i_lock); |
72a43d63c
|
1420 |
continue; |
250df6ed2
|
1421 |
} |
72a43d63c
|
1422 1423 |
break; } |
b67bfe0d4
|
1424 |
if (likely(!old)) { |
250df6ed2
|
1425 |
spin_lock(&inode->i_lock); |
c2b6d621c
|
1426 |
inode->i_state |= I_NEW | I_CREATING; |
261bca86e
|
1427 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1428 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1429 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1430 1431 |
return 0; } |
c2b6d621c
|
1432 1433 1434 1435 1436 |
if (unlikely(old->i_state & I_CREATING)) { spin_unlock(&old->i_lock); spin_unlock(&inode_hash_lock); return -EBUSY; } |
261bca86e
|
1437 |
__iget(old); |
250df6ed2
|
1438 |
spin_unlock(&old->i_lock); |
67a23c494
|
1439 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1440 |
wait_on_inode(old); |
1d3382cbf
|
1441 |
if (unlikely(!inode_unhashed(old))) { |
261bca86e
|
1442 1443 1444 1445 1446 1447 |
iput(old); return -EBUSY; } iput(old); } } |
261bca86e
|
1448 1449 1450 1451 1452 |
EXPORT_SYMBOL(insert_inode_locked); int insert_inode_locked4(struct inode *inode, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { |
c2b6d621c
|
1453 1454 1455 1456 |
struct inode *old; inode->i_state |= I_CREATING; old = inode_insert5(inode, hashval, test, NULL, data); |
261bca86e
|
1457 |
|
80ea09a00
|
1458 |
if (old != inode) { |
261bca86e
|
1459 |
iput(old); |
80ea09a00
|
1460 |
return -EBUSY; |
261bca86e
|
1461 |
} |
80ea09a00
|
1462 |
return 0; |
261bca86e
|
1463 |
} |
261bca86e
|
1464 |
EXPORT_SYMBOL(insert_inode_locked4); |
1da177e4c
|
1465 |
|
45321ac54
|
1466 1467 1468 1469 1470 |
int generic_delete_inode(struct inode *inode) { return 1; } EXPORT_SYMBOL(generic_delete_inode); |
1da177e4c
|
1471 |
/* |
45321ac54
|
1472 1473 |
* Called when we're dropping the last reference * to an inode. |
22fe40421
|
1474 |
* |
45321ac54
|
1475 1476 1477 1478 1479 |
* 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
|
1480 |
*/ |
45321ac54
|
1481 |
static void iput_final(struct inode *inode) |
1da177e4c
|
1482 1483 |
{ struct super_block *sb = inode->i_sb; |
45321ac54
|
1484 1485 |
const struct super_operations *op = inode->i_sb->s_op; int drop; |
250df6ed2
|
1486 |
WARN_ON(inode->i_state & I_NEW); |
e7f590970
|
1487 |
if (op->drop_inode) |
45321ac54
|
1488 1489 1490 |
drop = op->drop_inode(inode); else drop = generic_drop_inode(inode); |
1da177e4c
|
1491 |
|
1751e8a6c
|
1492 |
if (!drop && (sb->s_flags & SB_ACTIVE)) { |
4eff96dd5
|
1493 |
inode_add_lru(inode); |
b2b2af8e6
|
1494 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1495 1496 |
return; } |
45321ac54
|
1497 |
if (!drop) { |
991114c6f
|
1498 |
inode->i_state |= I_WILL_FREE; |
250df6ed2
|
1499 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1500 |
write_inode_now(inode, 1); |
250df6ed2
|
1501 |
spin_lock(&inode->i_lock); |
7ef0d7377
|
1502 |
WARN_ON(inode->i_state & I_NEW); |
991114c6f
|
1503 |
inode->i_state &= ~I_WILL_FREE; |
1da177e4c
|
1504 |
} |
7ccf19a80
|
1505 |
|
991114c6f
|
1506 |
inode->i_state |= I_FREEING; |
c4ae0c654
|
1507 1508 |
if (!list_empty(&inode->i_lru)) inode_lru_list_del(inode); |
b2b2af8e6
|
1509 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1510 |
|
644da5960
|
1511 |
evict(inode); |
1da177e4c
|
1512 |
} |
1da177e4c
|
1513 |
/** |
6b3304b53
|
1514 |
* iput - put an inode |
1da177e4c
|
1515 1516 1517 1518 1519 1520 1521 1522 1523 |
* @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
|
1524 1525 1526 1527 1528 1529 1530 |
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
|
1531 1532 1533 1534 1535 1536 |
spin_unlock(&inode->i_lock); trace_writeback_lazytime_iput(inode); mark_inode_dirty_sync(inode); goto retry; } iput_final(inode); |
1da177e4c
|
1537 1538 |
} } |
1da177e4c
|
1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 |
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
|
1549 |
* disk block relative to the disk start that holds that block of the |
1da177e4c
|
1550 1551 |
* file. */ |
6b3304b53
|
1552 |
sector_t bmap(struct inode *inode, sector_t block) |
1da177e4c
|
1553 1554 1555 1556 1557 1558 |
{ 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
|
1559 |
EXPORT_SYMBOL(bmap); |
11ff6f05f
|
1560 1561 1562 1563 1564 |
/* * 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
|
1565 |
static int relatime_need_update(struct vfsmount *mnt, struct inode *inode, |
6f22b6649
|
1566 |
struct timespec64 now) |
11ff6f05f
|
1567 |
{ |
c67185434
|
1568 |
if (!(mnt->mnt_flags & MNT_RELATIME)) |
11ff6f05f
|
1569 1570 1571 1572 |
return 1; /* * Is mtime younger than atime? If yes, update atime: */ |
95582b008
|
1573 |
if (timespec64_compare(&inode->i_mtime, &inode->i_atime) >= 0) |
11ff6f05f
|
1574 1575 1576 1577 |
return 1; /* * Is ctime younger than atime? If yes, update atime: */ |
95582b008
|
1578 |
if (timespec64_compare(&inode->i_ctime, &inode->i_atime) >= 0) |
11ff6f05f
|
1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 |
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
|
1592 |
int generic_update_time(struct inode *inode, struct timespec64 *time, int flags) |
c3b2da314
|
1593 |
{ |
0ae45f63d
|
1594 |
int iflags = I_DIRTY_TIME; |
e38cf302b
|
1595 |
bool dirty = false; |
c3b2da314
|
1596 1597 1598 1599 |
if (flags & S_ATIME) inode->i_atime = *time; if (flags & S_VERSION) |
e38cf302b
|
1600 |
dirty = inode_maybe_inc_iversion(inode, false); |
c3b2da314
|
1601 1602 1603 1604 |
if (flags & S_CTIME) inode->i_ctime = *time; if (flags & S_MTIME) inode->i_mtime = *time; |
e38cf302b
|
1605 1606 1607 |
if ((flags & (S_ATIME | S_CTIME | S_MTIME)) && !(inode->i_sb->s_flags & SB_LAZYTIME)) dirty = true; |
0ae45f63d
|
1608 |
|
e38cf302b
|
1609 |
if (dirty) |
0ae45f63d
|
1610 1611 |
iflags |= I_DIRTY_SYNC; __mark_inode_dirty(inode, iflags); |
c3b2da314
|
1612 1613 |
return 0; } |
0ae45f63d
|
1614 1615 1616 1617 1618 1619 |
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
|
1620 |
static int update_time(struct inode *inode, struct timespec64 *time, int flags) |
0ae45f63d
|
1621 |
{ |
95582b008
|
1622 |
int (*update_time)(struct inode *, struct timespec64 *, int); |
0ae45f63d
|
1623 1624 1625 1626 1627 1628 |
update_time = inode->i_op->update_time ? inode->i_op->update_time : generic_update_time; return update_time(inode, time, flags); } |
c3b2da314
|
1629 |
|
1da177e4c
|
1630 |
/** |
869243a0f
|
1631 |
* touch_atime - update the access time |
185553b22
|
1632 |
* @path: the &struct path to update |
30fdc8ee0
|
1633 |
* @inode: inode to update |
1da177e4c
|
1634 1635 1636 1637 1638 |
* * 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
|
1639 |
bool atime_needs_update(const struct path *path, struct inode *inode) |
1da177e4c
|
1640 |
{ |
68ac1234f
|
1641 |
struct vfsmount *mnt = path->mnt; |
95582b008
|
1642 |
struct timespec64 now; |
1da177e4c
|
1643 |
|
cdb70f3f7
|
1644 |
if (inode->i_flags & S_NOATIME) |
8fa9dd246
|
1645 |
return false; |
0bd23d09b
|
1646 1647 1648 1649 1650 1651 |
/* 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
|
1652 |
if (IS_NOATIME(inode)) |
8fa9dd246
|
1653 |
return false; |
1751e8a6c
|
1654 |
if ((inode->i_sb->s_flags & SB_NODIRATIME) && S_ISDIR(inode->i_mode)) |
8fa9dd246
|
1655 |
return false; |
47ae32d6a
|
1656 |
|
cdb70f3f7
|
1657 |
if (mnt->mnt_flags & MNT_NOATIME) |
8fa9dd246
|
1658 |
return false; |
cdb70f3f7
|
1659 |
if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)) |
8fa9dd246
|
1660 |
return false; |
1da177e4c
|
1661 |
|
c2050a454
|
1662 |
now = current_time(inode); |
11ff6f05f
|
1663 |
|
6f22b6649
|
1664 |
if (!relatime_need_update(mnt, inode, now)) |
8fa9dd246
|
1665 |
return false; |
11ff6f05f
|
1666 |
|
95582b008
|
1667 |
if (timespec64_equal(&inode->i_atime, &now)) |
8fa9dd246
|
1668 1669 1670 1671 1672 1673 1674 1675 1676 |
return false; return true; } void touch_atime(const struct path *path) { struct vfsmount *mnt = path->mnt; struct inode *inode = d_inode(path->dentry); |
95582b008
|
1677 |
struct timespec64 now; |
8fa9dd246
|
1678 |
|
c67185434
|
1679 |
if (!atime_needs_update(path, inode)) |
b12536c27
|
1680 |
return; |
5d37e9e6d
|
1681 |
if (!sb_start_write_trylock(inode->i_sb)) |
b12536c27
|
1682 |
return; |
47ae32d6a
|
1683 |
|
8fa9dd246
|
1684 |
if (__mnt_want_write(mnt) != 0) |
5d37e9e6d
|
1685 |
goto skip_update; |
c3b2da314
|
1686 1687 1688 1689 1690 1691 |
/* * 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
|
1692 1693 |
* We may also fail on filesystems that have the ability to make parts * of the fs read only, e.g. subvolumes in Btrfs. |
c3b2da314
|
1694 |
*/ |
c2050a454
|
1695 |
now = current_time(inode); |
c3b2da314
|
1696 |
update_time(inode, &now, S_ATIME); |
5d37e9e6d
|
1697 1698 1699 |
__mnt_drop_write(mnt); skip_update: sb_end_write(inode->i_sb); |
1da177e4c
|
1700 |
} |
869243a0f
|
1701 |
EXPORT_SYMBOL(touch_atime); |
1da177e4c
|
1702 |
|
3ed37648e
|
1703 1704 1705 1706 1707 1708 1709 1710 |
/* * The logic we want is * * if suid or (sgid and xgrp) * remove privs */ int should_remove_suid(struct dentry *dentry) { |
df2b1afde
|
1711 |
umode_t mode = d_inode(dentry)->i_mode; |
3ed37648e
|
1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 |
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
|
1731 1732 1733 1734 1735 |
/* * 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
|
1736 |
int dentry_needs_remove_privs(struct dentry *dentry) |
dbfae0cdc
|
1737 |
{ |
dbfae0cdc
|
1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 |
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
|
1753 1754 |
static int __remove_privs(struct dentry *dentry, int kill) |
3ed37648e
|
1755 1756 1757 1758 |
{ struct iattr newattrs; newattrs.ia_valid = ATTR_FORCE | kill; |
27ac0ffea
|
1759 1760 1761 1762 1763 |
/* * Note we call this on write, so notify_change will not * encounter any conflicting delegations: */ return notify_change(dentry, &newattrs, NULL); |
3ed37648e
|
1764 |
} |
5fa8e0a1c
|
1765 1766 1767 1768 1769 |
/* * Remove special file priviledges (suid, capabilities) when file is written * to or truncated. */ int file_remove_privs(struct file *file) |
3ed37648e
|
1770 |
{ |
c1892c377
|
1771 1772 |
struct dentry *dentry = file_dentry(file); struct inode *inode = file_inode(file); |
dbfae0cdc
|
1773 |
int kill; |
3ed37648e
|
1774 |
int error = 0; |
f69e749a4
|
1775 1776 1777 1778 1779 1780 1781 |
/* * 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
|
1782 |
return 0; |
c1892c377
|
1783 |
kill = dentry_needs_remove_privs(dentry); |
dbfae0cdc
|
1784 1785 1786 1787 |
if (kill < 0) return kill; if (kill) error = __remove_privs(dentry, kill); |
2426f3910
|
1788 1789 |
if (!error) inode_has_no_xattr(inode); |
3ed37648e
|
1790 1791 1792 |
return error; } |
5fa8e0a1c
|
1793 |
EXPORT_SYMBOL(file_remove_privs); |
3ed37648e
|
1794 |
|
1da177e4c
|
1795 |
/** |
870f48179
|
1796 1797 |
* file_update_time - update mtime and ctime time * @file: file accessed |
1da177e4c
|
1798 |
* |
870f48179
|
1799 1800 1801 1802 |
* 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
|
1803 |
* S_NOCMTIME inode flag, e.g. for network filesystem where these |
c3b2da314
|
1804 1805 |
* 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
|
1806 |
*/ |
c3b2da314
|
1807 |
int file_update_time(struct file *file) |
1da177e4c
|
1808 |
{ |
496ad9aa8
|
1809 |
struct inode *inode = file_inode(file); |
95582b008
|
1810 |
struct timespec64 now; |
c3b2da314
|
1811 1812 |
int sync_it = 0; int ret; |
1da177e4c
|
1813 |
|
ce06e0b21
|
1814 |
/* First try to exhaust all avenues to not sync */ |
1da177e4c
|
1815 |
if (IS_NOCMTIME(inode)) |
c3b2da314
|
1816 |
return 0; |
20ddee2c7
|
1817 |
|
c2050a454
|
1818 |
now = current_time(inode); |
95582b008
|
1819 |
if (!timespec64_equal(&inode->i_mtime, &now)) |
ce06e0b21
|
1820 |
sync_it = S_MTIME; |
1da177e4c
|
1821 |
|
95582b008
|
1822 |
if (!timespec64_equal(&inode->i_ctime, &now)) |
ce06e0b21
|
1823 |
sync_it |= S_CTIME; |
870f48179
|
1824 |
|
e38cf302b
|
1825 |
if (IS_I_VERSION(inode) && inode_iversion_need_inc(inode)) |
ce06e0b21
|
1826 |
sync_it |= S_VERSION; |
7a224228e
|
1827 |
|
ce06e0b21
|
1828 |
if (!sync_it) |
c3b2da314
|
1829 |
return 0; |
ce06e0b21
|
1830 1831 |
/* Finally allowed to write? Takes lock. */ |
eb04c2828
|
1832 |
if (__mnt_want_write_file(file)) |
c3b2da314
|
1833 |
return 0; |
ce06e0b21
|
1834 |
|
c3b2da314
|
1835 |
ret = update_time(inode, &now, sync_it); |
eb04c2828
|
1836 |
__mnt_drop_write_file(file); |
c3b2da314
|
1837 1838 |
return ret; |
1da177e4c
|
1839 |
} |
870f48179
|
1840 |
EXPORT_SYMBOL(file_update_time); |
1da177e4c
|
1841 |
|
e38f7f53c
|
1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 |
/* 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
|
1861 1862 1863 1864 1865 1866 1867 1868 |
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
|
1869 |
EXPORT_SYMBOL(inode_needs_sync); |
1da177e4c
|
1870 |
/* |
168a9fd6a
|
1871 1872 1873 1874 1875 1876 |
* 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
|
1877 |
* It doesn't matter if I_NEW is not set initially, a call to |
250df6ed2
|
1878 1879 |
* wake_up_bit(&inode->i_state, __I_NEW) after removing from the hash list * will DTRT. |
1da177e4c
|
1880 1881 1882 1883 |
*/ static void __wait_on_freeing_inode(struct inode *inode) { wait_queue_head_t *wq; |
eaff8079d
|
1884 1885 |
DEFINE_WAIT_BIT(wait, &inode->i_state, __I_NEW); wq = bit_waitqueue(&inode->i_state, __I_NEW); |
214171361
|
1886 |
prepare_to_wait(wq, &wait.wq_entry, TASK_UNINTERRUPTIBLE); |
250df6ed2
|
1887 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1888 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1889 |
schedule(); |
214171361
|
1890 |
finish_wait(wq, &wait.wq_entry); |
67a23c494
|
1891 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1892 |
} |
1da177e4c
|
1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 |
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
|
1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 |
/* 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
|
1919 |
HASH_EARLY | HASH_ZERO, |
1da177e4c
|
1920 1921 |
&i_hash_shift, &i_hash_mask, |
31fe62b95
|
1922 |
0, |
1da177e4c
|
1923 |
0); |
1da177e4c
|
1924 |
} |
74bf17cff
|
1925 |
void __init inode_init(void) |
1da177e4c
|
1926 |
{ |
1da177e4c
|
1927 |
/* inode slab cache */ |
b0196009d
|
1928 1929 1930 1931 |
inode_cachep = kmem_cache_create("inode_cache", sizeof(struct inode), 0, (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC| |
5d097056c
|
1932 |
SLAB_MEM_SPREAD|SLAB_ACCOUNT), |
20c2df83d
|
1933 |
init_once); |
1da177e4c
|
1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 |
/* 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
|
1944 |
HASH_ZERO, |
1da177e4c
|
1945 1946 |
&i_hash_shift, &i_hash_mask, |
31fe62b95
|
1947 |
0, |
1da177e4c
|
1948 |
0); |
1da177e4c
|
1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 |
} 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
|
1961 |
inode->i_fop = &pipefifo_fops; |
1da177e4c
|
1962 |
else if (S_ISSOCK(mode)) |
bd9b51e79
|
1963 |
; /* leave it no_open_fops */ |
1da177e4c
|
1964 |
else |
af0d9ae81
|
1965 1966 1967 1968 |
printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o) for" " inode %s:%lu ", mode, inode->i_sb->s_id, inode->i_ino); |
1da177e4c
|
1969 1970 |
} EXPORT_SYMBOL(init_special_inode); |
a1bd120d1
|
1971 1972 |
/** |
eaae668d0
|
1973 |
* inode_init_owner - Init uid,gid,mode for new inode according to posix standards |
a1bd120d1
|
1974 1975 1976 1977 1978 |
* @inode: New inode * @dir: Directory inode * @mode: mode of the new inode */ void inode_init_owner(struct inode *inode, const struct inode *dir, |
62bb10917
|
1979 |
umode_t mode) |
a1bd120d1
|
1980 1981 1982 1983 |
{ inode->i_uid = current_fsuid(); if (dir && dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; |
0fa3ecd87
|
1984 1985 |
/* Directories are special, and always inherit S_ISGID */ |
a1bd120d1
|
1986 1987 |
if (S_ISDIR(mode)) mode |= S_ISGID; |
0fa3ecd87
|
1988 1989 1990 1991 |
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
|
1992 1993 1994 1995 1996 |
} else inode->i_gid = current_fsgid(); inode->i_mode = mode; } EXPORT_SYMBOL(inode_init_owner); |
e795b7179
|
1997 |
|
2e1496707
|
1998 1999 2000 2001 |
/** * inode_owner_or_capable - check current task permissions to inode * @inode: inode being checked * |
23adbe12e
|
2002 2003 |
* Return true if current either has CAP_FOWNER in a namespace with the * inode owner uid mapped, or owns the file. |
e795b7179
|
2004 |
*/ |
2e1496707
|
2005 |
bool inode_owner_or_capable(const struct inode *inode) |
e795b7179
|
2006 |
{ |
23adbe12e
|
2007 |
struct user_namespace *ns; |
92361636e
|
2008 |
if (uid_eq(current_fsuid(), inode->i_uid)) |
e795b7179
|
2009 |
return true; |
23adbe12e
|
2010 2011 |
ns = current_user_ns(); |
cc658db47
|
2012 |
if (kuid_has_mapping(ns, inode->i_uid) && ns_capable(ns, CAP_FOWNER)) |
e795b7179
|
2013 2014 2015 |
return true; return false; } |
2e1496707
|
2016 |
EXPORT_SYMBOL(inode_owner_or_capable); |
1d59d61f6
|
2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 |
/* * 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
|
2027 |
prepare_to_wait(wq, &q.wq_entry, TASK_UNINTERRUPTIBLE); |
1d59d61f6
|
2028 2029 2030 |
if (atomic_read(&inode->i_dio_count)) schedule(); } while (atomic_read(&inode->i_dio_count)); |
214171361
|
2031 |
finish_wait(wq, &q.wq_entry); |
1d59d61f6
|
2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 |
} /** * 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
|
2052 2053 2054 2055 2056 2057 2058 |
* 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
|
2059 2060 |
* code path which doesn't today so we use cmpxchg() out of an abundance * of caution. |
5f16f3225
|
2061 2062 2063 2064 2065 2066 2067 2068 2069 |
* * 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
|
2070 |
WARN_ON_ONCE(flags & ~mask); |
a905737fd
|
2071 |
set_mask_bits(&inode->i_flags, mask, flags); |
5f16f3225
|
2072 2073 |
} EXPORT_SYMBOL(inode_set_flags); |
21fc61c73
|
2074 2075 2076 2077 2078 2079 |
void inode_nohighmem(struct inode *inode) { mapping_set_gfp_mask(inode->i_mapping, GFP_USER); } EXPORT_SYMBOL(inode_nohighmem); |
3cd886666
|
2080 2081 |
/** |
8efd6894f
|
2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 |
* 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); /** |
3cd886666
|
2106 2107 2108 2109 2110 2111 2112 2113 2114 |
* 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
|
2115 |
struct timespec64 current_time(struct inode *inode) |
3cd886666
|
2116 |
{ |
d651d1607
|
2117 2118 2119 |
struct timespec64 now; ktime_get_coarse_real_ts64(&now); |
3cd886666
|
2120 2121 2122 2123 2124 |
if (unlikely(!inode->i_sb)) { WARN(1, "current_time() called with uninitialized super_block in the inode"); return now; } |
95582b008
|
2125 |
return timespec64_trunc(now, inode->i_sb->s_time_gran); |
3cd886666
|
2126 2127 |
} EXPORT_SYMBOL(current_time); |
5aca28421
|
2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 |
/* * 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
|
2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 |
/* * 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
|
2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 |
/* * 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
|
2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 |
/* 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
|
2194 2195 2196 2197 2198 2199 2200 |
/* * 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
|
2201 2202 2203 2204 2205 |
/* 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
|
2206 2207 2208 |
return 0; } EXPORT_SYMBOL(vfs_ioc_fssetxattr_check); |