Blame view
fs/inode.c
42.9 KB
1da177e4c
|
1 |
/* |
1da177e4c
|
2 |
* (C) 1997 Linus Torvalds |
4b4563dc8
|
3 |
* (C) 1999 Andrea Arcangeli <andrea@suse.de> (dynamic inode allocation) |
1da177e4c
|
4 |
*/ |
1da177e4c
|
5 6 7 8 |
#include <linux/fs.h> #include <linux/mm.h> #include <linux/dcache.h> #include <linux/init.h> |
1da177e4c
|
9 10 11 12 13 |
#include <linux/slab.h> #include <linux/writeback.h> #include <linux/module.h> #include <linux/backing-dev.h> #include <linux/wait.h> |
88e0fbc45
|
14 |
#include <linux/rwsem.h> |
1da177e4c
|
15 16 17 18 19 20 |
#include <linux/hash.h> #include <linux/swap.h> #include <linux/security.h> #include <linux/pagemap.h> #include <linux/cdev.h> #include <linux/bootmem.h> |
3be25f49b
|
21 |
#include <linux/fsnotify.h> |
fc33a7bb9
|
22 |
#include <linux/mount.h> |
efaee1920
|
23 |
#include <linux/async.h> |
f19d4a8fa
|
24 |
#include <linux/posix_acl.h> |
9ce6e0be0
|
25 |
#include <linux/prefetch.h> |
a178d2027
|
26 |
#include <linux/ima.h> |
e795b7179
|
27 |
#include <linux/cred.h> |
4b4563dc8
|
28 |
#include <linux/buffer_head.h> /* for inode_has_buffers */ |
a66979aba
|
29 |
#include "internal.h" |
1da177e4c
|
30 31 |
/* |
4b4563dc8
|
32 |
* Inode locking rules: |
250df6ed2
|
33 34 35 |
* * inode->i_lock protects: * inode->i_state, inode->i_hash, __iget() |
09cc9fc7a
|
36 |
* inode->i_sb->s_inode_lru_lock protects: |
98b745c64
|
37 |
* inode->i_sb->s_inode_lru, inode->i_lru |
55fa6091d
|
38 39 |
* inode_sb_list_lock protects: * sb->s_inodes, inode->i_sb_list |
f758eeabe
|
40 |
* bdi->wb.list_lock protects: |
a66979aba
|
41 |
* bdi->wb.b_{dirty,io,more_io}, inode->i_wb_list |
67a23c494
|
42 43 |
* inode_hash_lock protects: * inode_hashtable, inode->i_hash |
250df6ed2
|
44 45 |
* * Lock ordering: |
55fa6091d
|
46 47 48 |
* * inode_sb_list_lock * inode->i_lock |
09cc9fc7a
|
49 |
* inode->i_sb->s_inode_lru_lock |
a66979aba
|
50 |
* |
f758eeabe
|
51 |
* bdi->wb.list_lock |
a66979aba
|
52 |
* inode->i_lock |
67a23c494
|
53 54 55 56 57 58 59 |
* * inode_hash_lock * inode_sb_list_lock * inode->i_lock * * iunique_lock * inode_hash_lock |
250df6ed2
|
60 |
*/ |
fa3536cc1
|
61 62 |
static unsigned int i_hash_mask __read_mostly; static unsigned int i_hash_shift __read_mostly; |
67a23c494
|
63 64 |
static struct hlist_head *inode_hashtable __read_mostly; static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock); |
1da177e4c
|
65 |
|
55fa6091d
|
66 |
__cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_sb_list_lock); |
1da177e4c
|
67 |
/* |
7dcda1c96
|
68 69 70 71 72 73 74 75 |
* 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
|
76 77 78 |
* Statistics gathering.. */ struct inodes_stat_t inodes_stat; |
3e880fb5e
|
79 |
static DEFINE_PER_CPU(unsigned int, nr_inodes); |
fcb94f72d
|
80 |
static DEFINE_PER_CPU(unsigned int, nr_unused); |
cffbc8aa3
|
81 |
|
6b3304b53
|
82 |
static struct kmem_cache *inode_cachep __read_mostly; |
1da177e4c
|
83 |
|
3e880fb5e
|
84 |
static int get_nr_inodes(void) |
cffbc8aa3
|
85 |
{ |
3e880fb5e
|
86 87 88 89 90 |
int i; int sum = 0; for_each_possible_cpu(i) sum += per_cpu(nr_inodes, i); return sum < 0 ? 0 : sum; |
cffbc8aa3
|
91 92 93 94 |
} static inline int get_nr_inodes_unused(void) { |
fcb94f72d
|
95 96 97 98 99 |
int i; int sum = 0; for_each_possible_cpu(i) sum += per_cpu(nr_unused, i); return sum < 0 ? 0 : sum; |
cffbc8aa3
|
100 101 102 103 |
} int get_nr_dirty_inodes(void) { |
3e880fb5e
|
104 |
/* not actually dirty inodes, but a wild approximation */ |
cffbc8aa3
|
105 106 |
int nr_dirty = get_nr_inodes() - get_nr_inodes_unused(); return nr_dirty > 0 ? nr_dirty : 0; |
cffbc8aa3
|
107 108 109 110 111 112 113 114 115 116 |
} /* * Handle nr_inode sysctl */ #ifdef CONFIG_SYSCTL int proc_nr_inodes(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { inodes_stat.nr_inodes = get_nr_inodes(); |
fcb94f72d
|
117 |
inodes_stat.nr_unused = get_nr_inodes_unused(); |
cffbc8aa3
|
118 119 120 |
return proc_dointvec(table, write, buffer, lenp, ppos); } #endif |
2cb1599f9
|
121 122 |
/** * inode_init_always - perform inode structure intialisation |
0bc02f3fa
|
123 124 |
* @sb: superblock inode belongs to * @inode: inode to initialise |
2cb1599f9
|
125 126 127 128 |
* * These are initializations that need to be done on every inode * allocation as the fields are not initialised by slab allocation. */ |
54e346215
|
129 |
int inode_init_always(struct super_block *sb, struct inode *inode) |
1da177e4c
|
130 |
{ |
6e1d5dcc2
|
131 |
static const struct inode_operations empty_iops; |
99ac48f54
|
132 |
static const struct file_operations empty_fops; |
6b3304b53
|
133 |
struct address_space *const mapping = &inode->i_data; |
2cb1599f9
|
134 135 136 137 138 139 140 |
inode->i_sb = sb; inode->i_blkbits = sb->s_blocksize_bits; inode->i_flags = 0; atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; inode->i_fop = &empty_fops; |
a78ef704a
|
141 |
inode->__i_nlink = 1; |
3ddcd0569
|
142 |
inode->i_opflags = 0; |
56ff5efad
|
143 144 |
inode->i_uid = 0; inode->i_gid = 0; |
2cb1599f9
|
145 146 147 148 149 |
atomic_set(&inode->i_writecount, 0); inode->i_size = 0; inode->i_blocks = 0; inode->i_bytes = 0; inode->i_generation = 0; |
1da177e4c
|
150 |
#ifdef CONFIG_QUOTA |
2cb1599f9
|
151 |
memset(&inode->i_dquot, 0, sizeof(inode->i_dquot)); |
1da177e4c
|
152 |
#endif |
2cb1599f9
|
153 154 155 156 157 |
inode->i_pipe = NULL; inode->i_bdev = NULL; inode->i_cdev = NULL; inode->i_rdev = 0; inode->dirtied_when = 0; |
6146f0d5e
|
158 159 |
if (security_inode_alloc(inode)) |
54e346215
|
160 |
goto out; |
2cb1599f9
|
161 162 163 164 165 |
spin_lock_init(&inode->i_lock); lockdep_set_class(&inode->i_lock, &sb->s_type->i_lock_key); mutex_init(&inode->i_mutex); lockdep_set_class(&inode->i_mutex, &sb->s_type->i_mutex_key); |
bd5fe6c5e
|
166 |
atomic_set(&inode->i_dio_count, 0); |
2cb1599f9
|
167 168 169 170 |
mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags = 0; |
3c1d43787
|
171 |
mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE); |
2cb1599f9
|
172 173 174 175 176 177 178 179 180 181 182 |
mapping->assoc_mapping = NULL; mapping->backing_dev_info = &default_backing_dev_info; mapping->writeback_index = 0; /* * If the block_device provides a backing_dev_info for client * inodes then use that. Otherwise the inode share the bdev's * backing_dev_info. */ if (sb->s_bdev) { struct backing_dev_info *bdi; |
2c96ce9f2
|
183 |
bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info; |
2cb1599f9
|
184 185 186 187 |
mapping->backing_dev_info = bdi; } inode->i_private = NULL; inode->i_mapping = mapping; |
f19d4a8fa
|
188 189 190 |
#ifdef CONFIG_FS_POSIX_ACL inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED; #endif |
2cb1599f9
|
191 |
|
3be25f49b
|
192 193 194 |
#ifdef CONFIG_FSNOTIFY inode->i_fsnotify_mask = 0; #endif |
3e880fb5e
|
195 |
this_cpu_inc(nr_inodes); |
cffbc8aa3
|
196 |
|
54e346215
|
197 |
return 0; |
54e346215
|
198 199 |
out: return -ENOMEM; |
1da177e4c
|
200 |
} |
2cb1599f9
|
201 202 203 204 205 206 207 208 209 210 |
EXPORT_SYMBOL(inode_init_always); static struct inode *alloc_inode(struct super_block *sb) { struct inode *inode; if (sb->s_op->alloc_inode) inode = sb->s_op->alloc_inode(sb); else inode = kmem_cache_alloc(inode_cachep, GFP_KERNEL); |
54e346215
|
211 212 213 214 215 216 217 218 219 220 221 222 |
if (!inode) return NULL; if (unlikely(inode_init_always(sb, inode))) { if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode(inode); else kmem_cache_free(inode_cachep, inode); return NULL; } return inode; |
2cb1599f9
|
223 |
} |
1da177e4c
|
224 |
|
ff0c7d15f
|
225 226 227 228 229 |
void free_inode_nonrcu(struct inode *inode) { kmem_cache_free(inode_cachep, inode); } EXPORT_SYMBOL(free_inode_nonrcu); |
2e00c97e2
|
230 |
void __destroy_inode(struct inode *inode) |
1da177e4c
|
231 |
{ |
b7542f8c7
|
232 |
BUG_ON(inode_has_buffers(inode)); |
1da177e4c
|
233 |
security_inode_free(inode); |
3be25f49b
|
234 |
fsnotify_inode_delete(inode); |
f19d4a8fa
|
235 236 237 238 239 240 |
#ifdef CONFIG_FS_POSIX_ACL if (inode->i_acl && inode->i_acl != ACL_NOT_CACHED) posix_acl_release(inode->i_acl); if (inode->i_default_acl && inode->i_default_acl != ACL_NOT_CACHED) posix_acl_release(inode->i_default_acl); #endif |
3e880fb5e
|
241 |
this_cpu_dec(nr_inodes); |
2e00c97e2
|
242 243 |
} EXPORT_SYMBOL(__destroy_inode); |
fa0d7e3de
|
244 245 246 247 248 249 |
static void i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(inode_cachep, inode); } |
56b0dacfa
|
250 |
static void destroy_inode(struct inode *inode) |
2e00c97e2
|
251 |
{ |
7ccf19a80
|
252 |
BUG_ON(!list_empty(&inode->i_lru)); |
2e00c97e2
|
253 |
__destroy_inode(inode); |
1da177e4c
|
254 255 256 |
if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode(inode); else |
fa0d7e3de
|
257 |
call_rcu(&inode->i_rcu, i_callback); |
1da177e4c
|
258 |
} |
1da177e4c
|
259 |
|
2aa15890f
|
260 261 262 263 264 |
void address_space_init_once(struct address_space *mapping) { memset(mapping, 0, sizeof(*mapping)); INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC); spin_lock_init(&mapping->tree_lock); |
3d48ae45e
|
265 |
mutex_init(&mapping->i_mmap_mutex); |
2aa15890f
|
266 267 268 269 |
INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); INIT_RAW_PRIO_TREE_ROOT(&mapping->i_mmap); INIT_LIST_HEAD(&mapping->i_mmap_nonlinear); |
2aa15890f
|
270 271 |
} EXPORT_SYMBOL(address_space_init_once); |
1da177e4c
|
272 273 274 275 276 277 278 279 280 281 282 |
/* * 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); INIT_LIST_HEAD(&inode->i_dentry); INIT_LIST_HEAD(&inode->i_devices); |
7ccf19a80
|
283 284 |
INIT_LIST_HEAD(&inode->i_wb_list); INIT_LIST_HEAD(&inode->i_lru); |
2aa15890f
|
285 |
address_space_init_once(&inode->i_data); |
1da177e4c
|
286 |
i_size_ordered_init(inode); |
3be25f49b
|
287 |
#ifdef CONFIG_FSNOTIFY |
e61ce8673
|
288 |
INIT_HLIST_HEAD(&inode->i_fsnotify_marks); |
3be25f49b
|
289 |
#endif |
1da177e4c
|
290 |
} |
1da177e4c
|
291 |
EXPORT_SYMBOL(inode_init_once); |
51cc50685
|
292 |
static void init_once(void *foo) |
1da177e4c
|
293 |
{ |
6b3304b53
|
294 |
struct inode *inode = (struct inode *) foo; |
1da177e4c
|
295 |
|
a35afb830
|
296 |
inode_init_once(inode); |
1da177e4c
|
297 298 299 |
} /* |
250df6ed2
|
300 |
* inode->i_lock must be held |
1da177e4c
|
301 |
*/ |
6b3304b53
|
302 |
void __iget(struct inode *inode) |
1da177e4c
|
303 |
{ |
9e38d86ff
|
304 305 |
atomic_inc(&inode->i_count); } |
2e147f1ef
|
306 |
|
7de9c6ee3
|
307 308 309 310 311 312 313 314 |
/* * 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
|
315 316 |
static void inode_lru_list_add(struct inode *inode) { |
09cc9fc7a
|
317 |
spin_lock(&inode->i_sb->s_inode_lru_lock); |
7ccf19a80
|
318 |
if (list_empty(&inode->i_lru)) { |
98b745c64
|
319 320 |
list_add(&inode->i_lru, &inode->i_sb->s_inode_lru); inode->i_sb->s_nr_inodes_unused++; |
fcb94f72d
|
321 |
this_cpu_inc(nr_unused); |
9e38d86ff
|
322 |
} |
09cc9fc7a
|
323 |
spin_unlock(&inode->i_sb->s_inode_lru_lock); |
9e38d86ff
|
324 |
} |
2e147f1ef
|
325 |
|
9e38d86ff
|
326 327 |
static void inode_lru_list_del(struct inode *inode) { |
09cc9fc7a
|
328 |
spin_lock(&inode->i_sb->s_inode_lru_lock); |
7ccf19a80
|
329 330 |
if (!list_empty(&inode->i_lru)) { list_del_init(&inode->i_lru); |
98b745c64
|
331 |
inode->i_sb->s_nr_inodes_unused--; |
fcb94f72d
|
332 |
this_cpu_dec(nr_unused); |
9e38d86ff
|
333 |
} |
09cc9fc7a
|
334 |
spin_unlock(&inode->i_sb->s_inode_lru_lock); |
1da177e4c
|
335 |
} |
646ec4615
|
336 337 338 339 340 341 |
/** * inode_sb_list_add - add inode to the superblock list of inodes * @inode: inode to add */ void inode_sb_list_add(struct inode *inode) { |
55fa6091d
|
342 343 344 |
spin_lock(&inode_sb_list_lock); list_add(&inode->i_sb_list, &inode->i_sb->s_inodes); spin_unlock(&inode_sb_list_lock); |
646ec4615
|
345 346 |
} EXPORT_SYMBOL_GPL(inode_sb_list_add); |
55fa6091d
|
347 |
static inline void inode_sb_list_del(struct inode *inode) |
646ec4615
|
348 |
{ |
a209dfc7b
|
349 350 351 352 353 |
if (!list_empty(&inode->i_sb_list)) { spin_lock(&inode_sb_list_lock); list_del_init(&inode->i_sb_list); spin_unlock(&inode_sb_list_lock); } |
646ec4615
|
354 |
} |
4c51acbc6
|
355 356 357 358 359 360 |
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
|
361 362 |
tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> i_hash_shift); return tmp & i_hash_mask; |
4c51acbc6
|
363 364 365 366 367 368 369 370 371 372 373 374 |
} /** * __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
|
375 |
struct hlist_head *b = inode_hashtable + hash(inode->i_sb, hashval); |
67a23c494
|
376 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
377 |
spin_lock(&inode->i_lock); |
646ec4615
|
378 |
hlist_add_head(&inode->i_hash, b); |
250df6ed2
|
379 |
spin_unlock(&inode->i_lock); |
67a23c494
|
380 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
381 382 383 384 |
} EXPORT_SYMBOL(__insert_inode_hash); /** |
f2ee7abf4
|
385 |
* __remove_inode_hash - remove an inode from the hash |
4c51acbc6
|
386 387 388 389 |
* @inode: inode to unhash * * Remove an inode from the superblock. */ |
f2ee7abf4
|
390 |
void __remove_inode_hash(struct inode *inode) |
4c51acbc6
|
391 |
{ |
67a23c494
|
392 |
spin_lock(&inode_hash_lock); |
250df6ed2
|
393 |
spin_lock(&inode->i_lock); |
4c51acbc6
|
394 |
hlist_del_init(&inode->i_hash); |
250df6ed2
|
395 |
spin_unlock(&inode->i_lock); |
67a23c494
|
396 |
spin_unlock(&inode_hash_lock); |
4c51acbc6
|
397 |
} |
f2ee7abf4
|
398 |
EXPORT_SYMBOL(__remove_inode_hash); |
4c51acbc6
|
399 |
|
b0683aa63
|
400 401 402 |
void end_writeback(struct inode *inode) { might_sleep(); |
08142579b
|
403 404 405 406 407 408 |
/* * We have to cycle tree_lock here because reclaim can be still in the * process of removing the last page (in __delete_from_page_cache()) * and we must not free mapping under it. */ spin_lock_irq(&inode->i_data.tree_lock); |
b0683aa63
|
409 |
BUG_ON(inode->i_data.nrpages); |
08142579b
|
410 |
spin_unlock_irq(&inode->i_data.tree_lock); |
b0683aa63
|
411 412 413 414 |
BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); inode_sync_wait(inode); |
fa0d7e3de
|
415 |
/* don't need i_lock here, no concurrent mods to i_state */ |
b0683aa63
|
416 417 418 |
inode->i_state = I_FREEING | I_CLEAR; } EXPORT_SYMBOL(end_writeback); |
b2b2af8e6
|
419 420 421 422 423 424 425 426 427 428 429 430 431 |
/* * 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
|
432 |
static void evict(struct inode *inode) |
b4272d4c8
|
433 434 |
{ const struct super_operations *op = inode->i_sb->s_op; |
b2b2af8e6
|
435 436 |
BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(!list_empty(&inode->i_lru)); |
b12362bdb
|
437 438 |
if (!list_empty(&inode->i_wb_list)) inode_wb_list_del(inode); |
55fa6091d
|
439 |
inode_sb_list_del(inode); |
be7ce4161
|
440 441 |
if (op->evict_inode) { op->evict_inode(inode); |
b4272d4c8
|
442 443 444 |
} else { if (inode->i_data.nrpages) truncate_inode_pages(&inode->i_data, 0); |
30140837f
|
445 |
end_writeback(inode); |
b4272d4c8
|
446 |
} |
661074e91
|
447 448 449 450 |
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
|
451 452 453 454 455 456 457 458 459 |
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
|
460 |
} |
1da177e4c
|
461 462 463 464 465 466 467 468 469 |
/* * 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
|
470 471 |
while (!list_empty(head)) { struct inode *inode; |
7ccf19a80
|
472 473 |
inode = list_first_entry(head, struct inode, i_lru); list_del_init(&inode->i_lru); |
1da177e4c
|
474 |
|
644da5960
|
475 |
evict(inode); |
1da177e4c
|
476 |
} |
1da177e4c
|
477 |
} |
a03187867
|
478 |
/** |
63997e98a
|
479 480 481 482 483 484 485 |
* evict_inodes - evict all evictable inodes for a superblock * @sb: superblock to operate on * * Make sure that no inodes with zero refcount are retained. This is * called by superblock shutdown after having MS_ACTIVE flag removed, * so any inode reaching zero refcount during or after that call will * be immediately evicted. |
1da177e4c
|
486 |
*/ |
63997e98a
|
487 |
void evict_inodes(struct super_block *sb) |
1da177e4c
|
488 |
{ |
63997e98a
|
489 490 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
491 |
|
55fa6091d
|
492 |
spin_lock(&inode_sb_list_lock); |
63997e98a
|
493 494 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { if (atomic_read(&inode->i_count)) |
aabb8fdb4
|
495 |
continue; |
250df6ed2
|
496 497 498 499 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
1da177e4c
|
500 |
continue; |
250df6ed2
|
501 |
} |
63997e98a
|
502 503 |
inode->i_state |= I_FREEING; |
02afc410f
|
504 |
inode_lru_list_del(inode); |
250df6ed2
|
505 |
spin_unlock(&inode->i_lock); |
02afc410f
|
506 |
list_add(&inode->i_lru, &dispose); |
1da177e4c
|
507 |
} |
55fa6091d
|
508 |
spin_unlock(&inode_sb_list_lock); |
63997e98a
|
509 510 |
dispose_list(&dispose); |
1da177e4c
|
511 |
} |
1da177e4c
|
512 |
/** |
a03187867
|
513 514 |
* invalidate_inodes - attempt to free all inodes on a superblock * @sb: superblock to operate on |
93b270f76
|
515 |
* @kill_dirty: flag to guide handling of dirty inodes |
1da177e4c
|
516 |
* |
a03187867
|
517 518 |
* Attempts to free all inodes for a given superblock. If there were any * busy inodes return a non-zero value, else zero. |
93b270f76
|
519 520 |
* If @kill_dirty is set, discard dirty inodes too, otherwise treat * them as busy. |
1da177e4c
|
521 |
*/ |
93b270f76
|
522 |
int invalidate_inodes(struct super_block *sb, bool kill_dirty) |
1da177e4c
|
523 |
{ |
cffbc8aa3
|
524 |
int busy = 0; |
a03187867
|
525 526 |
struct inode *inode, *next; LIST_HEAD(dispose); |
1da177e4c
|
527 |
|
55fa6091d
|
528 |
spin_lock(&inode_sb_list_lock); |
a03187867
|
529 |
list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) { |
250df6ed2
|
530 531 532 |
spin_lock(&inode->i_lock); if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) { spin_unlock(&inode->i_lock); |
aabb8fdb4
|
533 |
continue; |
250df6ed2
|
534 |
} |
93b270f76
|
535 |
if (inode->i_state & I_DIRTY && !kill_dirty) { |
250df6ed2
|
536 |
spin_unlock(&inode->i_lock); |
93b270f76
|
537 538 539 |
busy = 1; continue; } |
99a389192
|
540 |
if (atomic_read(&inode->i_count)) { |
250df6ed2
|
541 |
spin_unlock(&inode->i_lock); |
99a389192
|
542 |
busy = 1; |
1da177e4c
|
543 544 |
continue; } |
99a389192
|
545 |
|
99a389192
|
546 |
inode->i_state |= I_FREEING; |
02afc410f
|
547 |
inode_lru_list_del(inode); |
250df6ed2
|
548 |
spin_unlock(&inode->i_lock); |
02afc410f
|
549 |
list_add(&inode->i_lru, &dispose); |
1da177e4c
|
550 |
} |
55fa6091d
|
551 |
spin_unlock(&inode_sb_list_lock); |
1da177e4c
|
552 |
|
a03187867
|
553 |
dispose_list(&dispose); |
1da177e4c
|
554 555 556 |
return busy; } |
1da177e4c
|
557 558 559 |
static int can_unuse(struct inode *inode) { |
9e38d86ff
|
560 |
if (inode->i_state & ~I_REFERENCED) |
1da177e4c
|
561 562 563 564 565 566 567 568 569 570 571 |
return 0; if (inode_has_buffers(inode)) return 0; if (atomic_read(&inode->i_count)) return 0; if (inode->i_data.nrpages) return 0; return 1; } /* |
b0d40c92a
|
572 573 574 575 |
* 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(). |
1da177e4c
|
576 577 |
* * Any inodes which are pinned purely because of attached pagecache have their |
9e38d86ff
|
578 579 |
* pagecache removed. If the inode has metadata buffers attached to * mapping->private_list then try to remove them. |
1da177e4c
|
580 |
* |
9e38d86ff
|
581 582 583 584 585 586 587 |
* 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
|
588 |
*/ |
b0d40c92a
|
589 |
void prune_icache_sb(struct super_block *sb, int nr_to_scan) |
1da177e4c
|
590 591 |
{ LIST_HEAD(freeable); |
1da177e4c
|
592 593 |
int nr_scanned; unsigned long reap = 0; |
09cc9fc7a
|
594 |
spin_lock(&sb->s_inode_lru_lock); |
b0d40c92a
|
595 |
for (nr_scanned = nr_to_scan; nr_scanned >= 0; nr_scanned--) { |
1da177e4c
|
596 |
struct inode *inode; |
98b745c64
|
597 |
if (list_empty(&sb->s_inode_lru)) |
1da177e4c
|
598 |
break; |
98b745c64
|
599 |
inode = list_entry(sb->s_inode_lru.prev, struct inode, i_lru); |
1da177e4c
|
600 |
|
9e38d86ff
|
601 |
/* |
09cc9fc7a
|
602 |
* we are inverting the sb->s_inode_lru_lock/inode->i_lock here, |
02afc410f
|
603 604 605 606 |
* so use a trylock. If we fail to get the lock, just move the * inode to the back of the list so we don't spin on it. */ if (!spin_trylock(&inode->i_lock)) { |
62a3ddef6
|
607 |
list_move_tail(&inode->i_lru, &sb->s_inode_lru); |
02afc410f
|
608 609 610 611 |
continue; } /* |
9e38d86ff
|
612 613 614 615 616 |
* 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)) { |
7ccf19a80
|
617 |
list_del_init(&inode->i_lru); |
f283c86af
|
618 |
spin_unlock(&inode->i_lock); |
98b745c64
|
619 |
sb->s_nr_inodes_unused--; |
fcb94f72d
|
620 |
this_cpu_dec(nr_unused); |
9e38d86ff
|
621 622 623 624 625 |
continue; } /* recently referenced inodes get one more pass */ if (inode->i_state & I_REFERENCED) { |
9e38d86ff
|
626 |
inode->i_state &= ~I_REFERENCED; |
98b745c64
|
627 |
list_move(&inode->i_lru, &sb->s_inode_lru); |
f283c86af
|
628 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
629 630 631 632 |
continue; } if (inode_has_buffers(inode) || inode->i_data.nrpages) { __iget(inode); |
250df6ed2
|
633 |
spin_unlock(&inode->i_lock); |
09cc9fc7a
|
634 |
spin_unlock(&sb->s_inode_lru_lock); |
1da177e4c
|
635 |
if (remove_inode_buffers(inode)) |
fc0ecff69
|
636 637 |
reap += invalidate_mapping_pages(&inode->i_data, 0, -1); |
1da177e4c
|
638 |
iput(inode); |
09cc9fc7a
|
639 |
spin_lock(&sb->s_inode_lru_lock); |
1da177e4c
|
640 |
|
98b745c64
|
641 |
if (inode != list_entry(sb->s_inode_lru.next, |
7ccf19a80
|
642 |
struct inode, i_lru)) |
1da177e4c
|
643 |
continue; /* wrong inode or list_empty */ |
02afc410f
|
644 645 646 |
/* avoid lock inversions with trylock */ if (!spin_trylock(&inode->i_lock)) continue; |
250df6ed2
|
647 648 |
if (!can_unuse(inode)) { spin_unlock(&inode->i_lock); |
1da177e4c
|
649 |
continue; |
250df6ed2
|
650 |
} |
1da177e4c
|
651 |
} |
7ef0d7377
|
652 |
WARN_ON(inode->i_state & I_NEW); |
1da177e4c
|
653 |
inode->i_state |= I_FREEING; |
250df6ed2
|
654 |
spin_unlock(&inode->i_lock); |
7ccf19a80
|
655 |
|
7ccf19a80
|
656 |
list_move(&inode->i_lru, &freeable); |
98b745c64
|
657 |
sb->s_nr_inodes_unused--; |
fcb94f72d
|
658 |
this_cpu_dec(nr_unused); |
1da177e4c
|
659 |
} |
f8891e5e1
|
660 661 662 663 |
if (current_is_kswapd()) __count_vm_events(KSWAPD_INODESTEAL, reap); else __count_vm_events(PGINODESTEAL, reap); |
09cc9fc7a
|
664 |
spin_unlock(&sb->s_inode_lru_lock); |
1da177e4c
|
665 666 |
dispose_list(&freeable); |
1da177e4c
|
667 |
} |
1da177e4c
|
668 669 670 |
static void __wait_on_freeing_inode(struct inode *inode); /* * Called with the inode lock held. |
1da177e4c
|
671 |
*/ |
6b3304b53
|
672 673 674 675 |
static struct inode *find_inode(struct super_block *sb, struct hlist_head *head, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
676 677 |
{ struct hlist_node *node; |
6b3304b53
|
678 |
struct inode *inode = NULL; |
1da177e4c
|
679 680 |
repeat: |
c5c8be3ce
|
681 |
hlist_for_each_entry(inode, node, head, i_hash) { |
67a23c494
|
682 683 684 |
spin_lock(&inode->i_lock); if (inode->i_sb != sb) { spin_unlock(&inode->i_lock); |
1da177e4c
|
685 |
continue; |
67a23c494
|
686 687 688 |
} if (!test(inode, data)) { spin_unlock(&inode->i_lock); |
1da177e4c
|
689 |
continue; |
67a23c494
|
690 |
} |
a4ffdde6e
|
691 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
692 693 694 |
__wait_on_freeing_inode(inode); goto repeat; } |
f7899bd54
|
695 |
__iget(inode); |
250df6ed2
|
696 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
697 |
return inode; |
1da177e4c
|
698 |
} |
f7899bd54
|
699 |
return NULL; |
1da177e4c
|
700 701 702 703 704 705 |
} /* * find_inode_fast is the fast path version of find_inode, see the comment at * iget_locked for details. */ |
6b3304b53
|
706 707 |
static struct inode *find_inode_fast(struct super_block *sb, struct hlist_head *head, unsigned long ino) |
1da177e4c
|
708 709 |
{ struct hlist_node *node; |
6b3304b53
|
710 |
struct inode *inode = NULL; |
1da177e4c
|
711 712 |
repeat: |
c5c8be3ce
|
713 |
hlist_for_each_entry(inode, node, head, i_hash) { |
67a23c494
|
714 715 716 |
spin_lock(&inode->i_lock); if (inode->i_ino != ino) { spin_unlock(&inode->i_lock); |
1da177e4c
|
717 |
continue; |
67a23c494
|
718 719 720 |
} if (inode->i_sb != sb) { spin_unlock(&inode->i_lock); |
1da177e4c
|
721 |
continue; |
67a23c494
|
722 |
} |
a4ffdde6e
|
723 |
if (inode->i_state & (I_FREEING|I_WILL_FREE)) { |
1da177e4c
|
724 725 726 |
__wait_on_freeing_inode(inode); goto repeat; } |
f7899bd54
|
727 |
__iget(inode); |
250df6ed2
|
728 |
spin_unlock(&inode->i_lock); |
f7899bd54
|
729 |
return inode; |
1da177e4c
|
730 |
} |
f7899bd54
|
731 |
return NULL; |
8290c35f8
|
732 |
} |
f991bd2e1
|
733 734 735 736 |
/* * 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
|
737 |
* |
f991bd2e1
|
738 739 740 741 742 743 744 745 746 |
* 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
|
747 |
*/ |
f991bd2e1
|
748 749 |
#define LAST_INO_BATCH 1024 static DEFINE_PER_CPU(unsigned int, last_ino); |
85fe4025c
|
750 |
unsigned int get_next_ino(void) |
8290c35f8
|
751 |
{ |
f991bd2e1
|
752 753 |
unsigned int *p = &get_cpu_var(last_ino); unsigned int res = *p; |
8290c35f8
|
754 |
|
f991bd2e1
|
755 756 757 758 759 760 761 762 763 764 765 766 |
#ifdef CONFIG_SMP if (unlikely((res & (LAST_INO_BATCH-1)) == 0)) { static atomic_t shared_last_ino; int next = atomic_add_return(LAST_INO_BATCH, &shared_last_ino); res = next - LAST_INO_BATCH; } #endif *p = ++res; put_cpu_var(last_ino); return res; |
8290c35f8
|
767 |
} |
85fe4025c
|
768 |
EXPORT_SYMBOL(get_next_ino); |
8290c35f8
|
769 |
|
1da177e4c
|
770 |
/** |
a209dfc7b
|
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 |
* 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
|
794 795 796 |
* new_inode - obtain an inode * @sb: superblock * |
769848c03
|
797 |
* Allocates a new inode for given superblock. The default gfp_mask |
3c1d43787
|
798 |
* for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE. |
769848c03
|
799 800 801 802 803 |
* 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
|
804 805 806 |
*/ struct inode *new_inode(struct super_block *sb) { |
6b3304b53
|
807 |
struct inode *inode; |
1da177e4c
|
808 |
|
55fa6091d
|
809 |
spin_lock_prefetch(&inode_sb_list_lock); |
6b3304b53
|
810 |
|
a209dfc7b
|
811 812 |
inode = new_inode_pseudo(sb); if (inode) |
55fa6091d
|
813 |
inode_sb_list_add(inode); |
1da177e4c
|
814 815 |
return inode; } |
1da177e4c
|
816 |
EXPORT_SYMBOL(new_inode); |
14358e6dd
|
817 |
#ifdef CONFIG_DEBUG_LOCK_ALLOC |
e096d0c7e
|
818 819 |
void lockdep_annotate_inode_mutex_key(struct inode *inode) { |
a3314a0ed
|
820 |
if (S_ISDIR(inode->i_mode)) { |
1e89a5e15
|
821 |
struct file_system_type *type = inode->i_sb->s_type; |
9a7aa12f3
|
822 823 824 825 826 827 828 829 830 831 832 |
/* Set new key only if filesystem hasn't already changed it */ if (!lockdep_match_class(&inode->i_mutex, &type->i_mutex_key)) { /* * ensure nobody is actually holding i_mutex */ mutex_destroy(&inode->i_mutex); mutex_init(&inode->i_mutex); lockdep_set_class(&inode->i_mutex, &type->i_mutex_dir_key); } |
1e89a5e15
|
833 |
} |
e096d0c7e
|
834 835 |
} EXPORT_SYMBOL(lockdep_annotate_inode_mutex_key); |
14358e6dd
|
836 |
#endif |
e096d0c7e
|
837 838 839 840 841 842 843 844 845 846 847 |
/** * 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
|
848 |
spin_lock(&inode->i_lock); |
eaff8079d
|
849 850 |
WARN_ON(!(inode->i_state & I_NEW)); inode->i_state &= ~I_NEW; |
250df6ed2
|
851 852 |
wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); |
1da177e4c
|
853 |
} |
1da177e4c
|
854 |
EXPORT_SYMBOL(unlock_new_inode); |
0b2d0724e
|
855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 |
/** * 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
|
871 |
* |
0b2d0724e
|
872 873 |
* Note both @test and @set are called with the inode_hash_lock held, so can't * sleep. |
1da177e4c
|
874 |
*/ |
0b2d0724e
|
875 876 877 |
struct inode *iget5_locked(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data) |
1da177e4c
|
878 |
{ |
0b2d0724e
|
879 |
struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
6b3304b53
|
880 |
struct inode *inode; |
1da177e4c
|
881 |
|
0b2d0724e
|
882 883 884 885 886 887 888 889 |
spin_lock(&inode_hash_lock); inode = find_inode(sb, head, test, data); spin_unlock(&inode_hash_lock); if (inode) { wait_on_inode(inode); return inode; } |
1da177e4c
|
890 891 |
inode = alloc_inode(sb); if (inode) { |
6b3304b53
|
892 |
struct inode *old; |
1da177e4c
|
893 |
|
67a23c494
|
894 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
895 896 897 898 899 |
/* We released the lock, so.. */ old = find_inode(sb, head, test, data); if (!old) { if (set(inode, data)) goto set_failed; |
250df6ed2
|
900 901 |
spin_lock(&inode->i_lock); inode->i_state = I_NEW; |
646ec4615
|
902 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
903 |
spin_unlock(&inode->i_lock); |
55fa6091d
|
904 |
inode_sb_list_add(inode); |
67a23c494
|
905 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
906 907 908 909 910 911 912 913 914 915 916 917 |
/* 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
|
918 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
919 920 921 922 923 924 925 |
destroy_inode(inode); inode = old; wait_on_inode(inode); } return inode; set_failed: |
67a23c494
|
926 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
927 928 929 |
destroy_inode(inode); return NULL; } |
0b2d0724e
|
930 |
EXPORT_SYMBOL(iget5_locked); |
1da177e4c
|
931 |
|
0b2d0724e
|
932 933 934 935 936 937 938 939 940 941 942 943 |
/** * 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
|
944 |
*/ |
0b2d0724e
|
945 |
struct inode *iget_locked(struct super_block *sb, unsigned long ino) |
1da177e4c
|
946 |
{ |
0b2d0724e
|
947 |
struct hlist_head *head = inode_hashtable + hash(sb, ino); |
6b3304b53
|
948 |
struct inode *inode; |
1da177e4c
|
949 |
|
0b2d0724e
|
950 951 952 953 954 955 956 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); if (inode) { wait_on_inode(inode); return inode; } |
1da177e4c
|
957 958 |
inode = alloc_inode(sb); if (inode) { |
6b3304b53
|
959 |
struct inode *old; |
1da177e4c
|
960 |
|
67a23c494
|
961 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
962 963 964 965 |
/* We released the lock, so.. */ old = find_inode_fast(sb, head, ino); if (!old) { inode->i_ino = ino; |
250df6ed2
|
966 967 |
spin_lock(&inode->i_lock); inode->i_state = I_NEW; |
646ec4615
|
968 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
969 |
spin_unlock(&inode->i_lock); |
55fa6091d
|
970 |
inode_sb_list_add(inode); |
67a23c494
|
971 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
972 973 974 975 976 977 978 979 980 981 982 983 |
/* 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
|
984 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
985 986 987 988 989 990 |
destroy_inode(inode); inode = old; wait_on_inode(inode); } return inode; } |
0b2d0724e
|
991 |
EXPORT_SYMBOL(iget_locked); |
1da177e4c
|
992 |
|
ad5e195ac
|
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 |
/* * 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); struct hlist_node *node; struct inode *inode; |
67a23c494
|
1005 |
spin_lock(&inode_hash_lock); |
ad5e195ac
|
1006 |
hlist_for_each_entry(inode, node, b, i_hash) { |
67a23c494
|
1007 1008 |
if (inode->i_ino == ino && inode->i_sb == sb) { spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1009 |
return 0; |
67a23c494
|
1010 |
} |
ad5e195ac
|
1011 |
} |
67a23c494
|
1012 |
spin_unlock(&inode_hash_lock); |
ad5e195ac
|
1013 1014 1015 |
return 1; } |
1da177e4c
|
1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 |
/** * 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
|
1032 1033 1034 1035 1036 |
/* * 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
|
1037 |
static DEFINE_SPINLOCK(iunique_lock); |
866b04fcc
|
1038 |
static unsigned int counter; |
1da177e4c
|
1039 |
ino_t res; |
3361c7beb
|
1040 |
|
ad5e195ac
|
1041 |
spin_lock(&iunique_lock); |
3361c7beb
|
1042 1043 1044 |
do { if (counter <= max_reserved) counter = max_reserved + 1; |
1da177e4c
|
1045 |
res = counter++; |
ad5e195ac
|
1046 1047 |
} while (!test_inode_iunique(sb, res)); spin_unlock(&iunique_lock); |
1da177e4c
|
1048 |
|
3361c7beb
|
1049 1050 |
return res; } |
1da177e4c
|
1051 1052 1053 1054 |
EXPORT_SYMBOL(iunique); struct inode *igrab(struct inode *inode) { |
250df6ed2
|
1055 1056 |
spin_lock(&inode->i_lock); if (!(inode->i_state & (I_FREEING|I_WILL_FREE))) { |
1da177e4c
|
1057 |
__iget(inode); |
250df6ed2
|
1058 1059 1060 |
spin_unlock(&inode->i_lock); } else { spin_unlock(&inode->i_lock); |
1da177e4c
|
1061 1062 1063 1064 1065 1066 |
/* * 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
|
1067 |
} |
1da177e4c
|
1068 1069 |
return inode; } |
1da177e4c
|
1070 1071 1072 |
EXPORT_SYMBOL(igrab); /** |
0b2d0724e
|
1073 |
* ilookup5_nowait - search for an inode in the inode cache |
1da177e4c
|
1074 |
* @sb: super block of file system to search |
0b2d0724e
|
1075 |
* @hashval: hash value (usually inode number) to search for |
1da177e4c
|
1076 1077 |
* @test: callback used for comparisons between inodes * @data: opaque data pointer to pass to @test |
1da177e4c
|
1078 |
* |
0b2d0724e
|
1079 |
* Search for the inode specified by @hashval and @data in the inode cache. |
1da177e4c
|
1080 1081 1082 |
* If the inode is in the cache, the inode is returned with an incremented * reference count. * |
0b2d0724e
|
1083 1084 |
* 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
|
1085 |
* |
b6d0ad686
|
1086 |
* Note2: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1087 |
*/ |
0b2d0724e
|
1088 1089 |
struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) |
1da177e4c
|
1090 |
{ |
0b2d0724e
|
1091 |
struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
1da177e4c
|
1092 |
struct inode *inode; |
67a23c494
|
1093 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1094 |
inode = find_inode(sb, head, test, data); |
67a23c494
|
1095 |
spin_unlock(&inode_hash_lock); |
88bd5121d
|
1096 |
|
0b2d0724e
|
1097 |
return inode; |
88bd5121d
|
1098 |
} |
88bd5121d
|
1099 1100 1101 1102 1103 1104 1105 1106 1107 |
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
|
1108 1109 1110 |
* 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
|
1111 |
* returned with an incremented reference count. |
1da177e4c
|
1112 |
* |
0b2d0724e
|
1113 1114 |
* This is a generalized version of ilookup() for file systems where the * inode number is not sufficient for unique identification of an inode. |
1da177e4c
|
1115 |
* |
0b2d0724e
|
1116 |
* Note: @test is called with the inode_hash_lock held, so can't sleep. |
1da177e4c
|
1117 1118 1119 1120 |
*/ struct inode *ilookup5(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { |
0b2d0724e
|
1121 |
struct inode *inode = ilookup5_nowait(sb, hashval, test, data); |
1da177e4c
|
1122 |
|
0b2d0724e
|
1123 1124 1125 |
if (inode) wait_on_inode(inode); return inode; |
1da177e4c
|
1126 |
} |
1da177e4c
|
1127 1128 1129 1130 1131 1132 1133 |
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
|
1134 1135 |
* 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
|
1136 1137 1138 1139 |
*/ struct inode *ilookup(struct super_block *sb, unsigned long ino) { struct hlist_head *head = inode_hashtable + hash(sb, ino); |
1da177e4c
|
1140 |
struct inode *inode; |
0b2d0724e
|
1141 1142 1143 |
spin_lock(&inode_hash_lock); inode = find_inode_fast(sb, head, ino); spin_unlock(&inode_hash_lock); |
1da177e4c
|
1144 |
|
1da177e4c
|
1145 |
if (inode) |
0b2d0724e
|
1146 1147 |
wait_on_inode(inode); return inode; |
1da177e4c
|
1148 |
} |
0b2d0724e
|
1149 |
EXPORT_SYMBOL(ilookup); |
1da177e4c
|
1150 |
|
261bca86e
|
1151 1152 1153 1154 1155 |
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
|
1156 |
|
261bca86e
|
1157 |
while (1) { |
72a43d63c
|
1158 1159 |
struct hlist_node *node; struct inode *old = NULL; |
67a23c494
|
1160 |
spin_lock(&inode_hash_lock); |
72a43d63c
|
1161 1162 1163 1164 1165 |
hlist_for_each_entry(old, node, head, i_hash) { if (old->i_ino != ino) continue; if (old->i_sb != sb) continue; |
250df6ed2
|
1166 1167 1168 |
spin_lock(&old->i_lock); if (old->i_state & (I_FREEING|I_WILL_FREE)) { spin_unlock(&old->i_lock); |
72a43d63c
|
1169 |
continue; |
250df6ed2
|
1170 |
} |
72a43d63c
|
1171 1172 1173 |
break; } if (likely(!node)) { |
250df6ed2
|
1174 1175 |
spin_lock(&inode->i_lock); inode->i_state |= I_NEW; |
261bca86e
|
1176 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1177 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1178 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1179 1180 1181 |
return 0; } __iget(old); |
250df6ed2
|
1182 |
spin_unlock(&old->i_lock); |
67a23c494
|
1183 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1184 |
wait_on_inode(old); |
1d3382cbf
|
1185 |
if (unlikely(!inode_unhashed(old))) { |
261bca86e
|
1186 1187 1188 1189 1190 1191 |
iput(old); return -EBUSY; } iput(old); } } |
261bca86e
|
1192 1193 1194 1195 1196 1197 1198 |
EXPORT_SYMBOL(insert_inode_locked); int insert_inode_locked4(struct inode *inode, unsigned long hashval, int (*test)(struct inode *, void *), void *data) { struct super_block *sb = inode->i_sb; struct hlist_head *head = inode_hashtable + hash(sb, hashval); |
261bca86e
|
1199 |
|
261bca86e
|
1200 |
while (1) { |
72a43d63c
|
1201 1202 |
struct hlist_node *node; struct inode *old = NULL; |
67a23c494
|
1203 |
spin_lock(&inode_hash_lock); |
72a43d63c
|
1204 1205 1206 1207 1208 |
hlist_for_each_entry(old, node, head, i_hash) { if (old->i_sb != sb) continue; if (!test(old, data)) continue; |
250df6ed2
|
1209 1210 1211 |
spin_lock(&old->i_lock); if (old->i_state & (I_FREEING|I_WILL_FREE)) { spin_unlock(&old->i_lock); |
72a43d63c
|
1212 |
continue; |
250df6ed2
|
1213 |
} |
72a43d63c
|
1214 1215 1216 |
break; } if (likely(!node)) { |
250df6ed2
|
1217 1218 |
spin_lock(&inode->i_lock); inode->i_state |= I_NEW; |
261bca86e
|
1219 |
hlist_add_head(&inode->i_hash, head); |
250df6ed2
|
1220 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1221 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1222 1223 1224 |
return 0; } __iget(old); |
250df6ed2
|
1225 |
spin_unlock(&old->i_lock); |
67a23c494
|
1226 |
spin_unlock(&inode_hash_lock); |
261bca86e
|
1227 |
wait_on_inode(old); |
1d3382cbf
|
1228 |
if (unlikely(!inode_unhashed(old))) { |
261bca86e
|
1229 1230 1231 1232 1233 1234 |
iput(old); return -EBUSY; } iput(old); } } |
261bca86e
|
1235 |
EXPORT_SYMBOL(insert_inode_locked4); |
1da177e4c
|
1236 |
|
45321ac54
|
1237 1238 1239 1240 1241 |
int generic_delete_inode(struct inode *inode) { return 1; } EXPORT_SYMBOL(generic_delete_inode); |
1da177e4c
|
1242 |
/* |
45321ac54
|
1243 1244 1245 |
* Normal UNIX filesystem behaviour: delete the * inode when the usage count drops to zero, and * i_nlink is zero. |
1da177e4c
|
1246 |
*/ |
45321ac54
|
1247 |
int generic_drop_inode(struct inode *inode) |
1da177e4c
|
1248 |
{ |
1d3382cbf
|
1249 |
return !inode->i_nlink || inode_unhashed(inode); |
1da177e4c
|
1250 |
} |
45321ac54
|
1251 |
EXPORT_SYMBOL_GPL(generic_drop_inode); |
1da177e4c
|
1252 |
|
45321ac54
|
1253 1254 1255 |
/* * Called when we're dropping the last reference * to an inode. |
22fe40421
|
1256 |
* |
45321ac54
|
1257 1258 1259 1260 1261 |
* 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
|
1262 |
*/ |
45321ac54
|
1263 |
static void iput_final(struct inode *inode) |
1da177e4c
|
1264 1265 |
{ struct super_block *sb = inode->i_sb; |
45321ac54
|
1266 1267 |
const struct super_operations *op = inode->i_sb->s_op; int drop; |
250df6ed2
|
1268 |
WARN_ON(inode->i_state & I_NEW); |
e7f590970
|
1269 |
if (op->drop_inode) |
45321ac54
|
1270 1271 1272 |
drop = op->drop_inode(inode); else drop = generic_drop_inode(inode); |
1da177e4c
|
1273 |
|
b2b2af8e6
|
1274 1275 1276 1277 1278 |
if (!drop && (sb->s_flags & MS_ACTIVE)) { inode->i_state |= I_REFERENCED; if (!(inode->i_state & (I_DIRTY|I_SYNC))) inode_lru_list_add(inode); spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1279 1280 |
return; } |
45321ac54
|
1281 |
if (!drop) { |
991114c6f
|
1282 |
inode->i_state |= I_WILL_FREE; |
250df6ed2
|
1283 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1284 |
write_inode_now(inode, 1); |
250df6ed2
|
1285 |
spin_lock(&inode->i_lock); |
7ef0d7377
|
1286 |
WARN_ON(inode->i_state & I_NEW); |
991114c6f
|
1287 |
inode->i_state &= ~I_WILL_FREE; |
1da177e4c
|
1288 |
} |
7ccf19a80
|
1289 |
|
991114c6f
|
1290 |
inode->i_state |= I_FREEING; |
c4ae0c654
|
1291 1292 |
if (!list_empty(&inode->i_lru)) inode_lru_list_del(inode); |
b2b2af8e6
|
1293 |
spin_unlock(&inode->i_lock); |
b2b2af8e6
|
1294 |
|
644da5960
|
1295 |
evict(inode); |
1da177e4c
|
1296 |
} |
1da177e4c
|
1297 |
/** |
6b3304b53
|
1298 |
* iput - put an inode |
1da177e4c
|
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 |
* @inode: inode to put * * Puts an inode, dropping its usage count. If the inode use count hits * zero, the inode is then freed and may also be destroyed. * * Consequently, iput() can sleep. */ void iput(struct inode *inode) { if (inode) { |
a4ffdde6e
|
1309 |
BUG_ON(inode->i_state & I_CLEAR); |
1da177e4c
|
1310 |
|
f283c86af
|
1311 |
if (atomic_dec_and_lock(&inode->i_count, &inode->i_lock)) |
1da177e4c
|
1312 1313 1314 |
iput_final(inode); } } |
1da177e4c
|
1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 |
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
|
1325 |
* disk block relative to the disk start that holds that block of the |
1da177e4c
|
1326 1327 |
* file. */ |
6b3304b53
|
1328 |
sector_t bmap(struct inode *inode, sector_t block) |
1da177e4c
|
1329 1330 1331 1332 1333 1334 |
{ 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
|
1335 |
EXPORT_SYMBOL(bmap); |
11ff6f05f
|
1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 |
/* * With relative atime, only update atime if the previous atime is * earlier than either the ctime or mtime or if at least a day has * passed since the last atime update. */ static int relatime_need_update(struct vfsmount *mnt, struct inode *inode, struct timespec now) { if (!(mnt->mnt_flags & MNT_RELATIME)) return 1; /* * Is mtime younger than atime? If yes, update atime: */ if (timespec_compare(&inode->i_mtime, &inode->i_atime) >= 0) return 1; /* * Is ctime younger than atime? If yes, update atime: */ if (timespec_compare(&inode->i_ctime, &inode->i_atime) >= 0) return 1; /* * Is the previous atime value older than a day? If yes, * update atime: */ if ((long)(now.tv_sec - inode->i_atime.tv_sec) >= 24*60*60) return 1; /* * Good, we can skip the atime update: */ return 0; } |
1da177e4c
|
1369 |
/** |
869243a0f
|
1370 1371 |
* touch_atime - update the access time * @mnt: mount the inode is accessed on |
7045f37b1
|
1372 |
* @dentry: dentry accessed |
1da177e4c
|
1373 1374 1375 1376 1377 |
* * 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. */ |
869243a0f
|
1378 |
void touch_atime(struct vfsmount *mnt, struct dentry *dentry) |
1da177e4c
|
1379 |
{ |
869243a0f
|
1380 |
struct inode *inode = dentry->d_inode; |
1da177e4c
|
1381 |
struct timespec now; |
cdb70f3f7
|
1382 |
if (inode->i_flags & S_NOATIME) |
b12536c27
|
1383 |
return; |
37756ced1
|
1384 |
if (IS_NOATIME(inode)) |
b12536c27
|
1385 |
return; |
b22761384
|
1386 |
if ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode)) |
b12536c27
|
1387 |
return; |
47ae32d6a
|
1388 |
|
cdb70f3f7
|
1389 |
if (mnt->mnt_flags & MNT_NOATIME) |
b12536c27
|
1390 |
return; |
cdb70f3f7
|
1391 |
if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)) |
b12536c27
|
1392 |
return; |
1da177e4c
|
1393 1394 |
now = current_fs_time(inode->i_sb); |
11ff6f05f
|
1395 1396 |
if (!relatime_need_update(mnt, inode, now)) |
b12536c27
|
1397 |
return; |
11ff6f05f
|
1398 |
|
47ae32d6a
|
1399 |
if (timespec_equal(&inode->i_atime, &now)) |
b12536c27
|
1400 1401 1402 1403 |
return; if (mnt_want_write(mnt)) return; |
47ae32d6a
|
1404 1405 1406 |
inode->i_atime = now; mark_inode_dirty_sync(inode); |
cdb70f3f7
|
1407 |
mnt_drop_write(mnt); |
1da177e4c
|
1408 |
} |
869243a0f
|
1409 |
EXPORT_SYMBOL(touch_atime); |
1da177e4c
|
1410 1411 |
/** |
870f48179
|
1412 1413 |
* file_update_time - update mtime and ctime time * @file: file accessed |
1da177e4c
|
1414 |
* |
870f48179
|
1415 1416 1417 1418 |
* 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
|
1419 |
* S_NOCMTIME inode flag, e.g. for network filesystem where these |
870f48179
|
1420 |
* timestamps are handled by the server. |
1da177e4c
|
1421 |
*/ |
870f48179
|
1422 |
void file_update_time(struct file *file) |
1da177e4c
|
1423 |
{ |
0f7fc9e4d
|
1424 |
struct inode *inode = file->f_path.dentry->d_inode; |
1da177e4c
|
1425 |
struct timespec now; |
ce06e0b21
|
1426 |
enum { S_MTIME = 1, S_CTIME = 2, S_VERSION = 4 } sync_it = 0; |
1da177e4c
|
1427 |
|
ce06e0b21
|
1428 |
/* First try to exhaust all avenues to not sync */ |
1da177e4c
|
1429 1430 |
if (IS_NOCMTIME(inode)) return; |
20ddee2c7
|
1431 |
|
1da177e4c
|
1432 |
now = current_fs_time(inode->i_sb); |
ce06e0b21
|
1433 1434 |
if (!timespec_equal(&inode->i_mtime, &now)) sync_it = S_MTIME; |
1da177e4c
|
1435 |
|
ce06e0b21
|
1436 1437 |
if (!timespec_equal(&inode->i_ctime, &now)) sync_it |= S_CTIME; |
870f48179
|
1438 |
|
ce06e0b21
|
1439 1440 |
if (IS_I_VERSION(inode)) sync_it |= S_VERSION; |
7a224228e
|
1441 |
|
ce06e0b21
|
1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 |
if (!sync_it) return; /* Finally allowed to write? Takes lock. */ if (mnt_want_write_file(file)) return; /* Only change inode inside the lock region */ if (sync_it & S_VERSION) inode_inc_iversion(inode); if (sync_it & S_CTIME) inode->i_ctime = now; if (sync_it & S_MTIME) inode->i_mtime = now; mark_inode_dirty_sync(inode); |
20ddee2c7
|
1457 |
mnt_drop_write(file->f_path.mnt); |
1da177e4c
|
1458 |
} |
870f48179
|
1459 |
EXPORT_SYMBOL(file_update_time); |
1da177e4c
|
1460 1461 1462 1463 1464 1465 1466 1467 1468 |
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
|
1469 |
EXPORT_SYMBOL(inode_needs_sync); |
1da177e4c
|
1470 1471 1472 1473 1474 |
int inode_wait(void *word) { schedule(); return 0; } |
d44dab8d1
|
1475 |
EXPORT_SYMBOL(inode_wait); |
1da177e4c
|
1476 1477 |
/* |
168a9fd6a
|
1478 1479 1480 1481 1482 1483 |
* 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
|
1484 |
* It doesn't matter if I_NEW is not set initially, a call to |
250df6ed2
|
1485 1486 |
* wake_up_bit(&inode->i_state, __I_NEW) after removing from the hash list * will DTRT. |
1da177e4c
|
1487 1488 1489 1490 |
*/ static void __wait_on_freeing_inode(struct inode *inode) { wait_queue_head_t *wq; |
eaff8079d
|
1491 1492 |
DEFINE_WAIT_BIT(wait, &inode->i_state, __I_NEW); wq = bit_waitqueue(&inode->i_state, __I_NEW); |
1da177e4c
|
1493 |
prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE); |
250df6ed2
|
1494 |
spin_unlock(&inode->i_lock); |
67a23c494
|
1495 |
spin_unlock(&inode_hash_lock); |
1da177e4c
|
1496 1497 |
schedule(); finish_wait(wq, &wait.wait); |
67a23c494
|
1498 |
spin_lock(&inode_hash_lock); |
1da177e4c
|
1499 |
} |
1da177e4c
|
1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 |
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) { int loop; /* If hashes are distributed across NUMA nodes, defer * hash allocation until vmalloc space is available. */ if (hashdist) return; inode_hashtable = alloc_large_system_hash("Inode-cache", sizeof(struct hlist_head), ihash_entries, 14, HASH_EARLY, &i_hash_shift, &i_hash_mask, 0); for (loop = 0; loop < (1 << i_hash_shift); loop++) INIT_HLIST_HEAD(&inode_hashtable[loop]); } |
74bf17cff
|
1536 |
void __init inode_init(void) |
1da177e4c
|
1537 1538 1539 1540 |
{ int loop; /* inode slab cache */ |
b0196009d
|
1541 1542 1543 1544 1545 |
inode_cachep = kmem_cache_create("inode_cache", sizeof(struct inode), 0, (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC| SLAB_MEM_SPREAD), |
20c2df83d
|
1546 |
init_once); |
1da177e4c
|
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 |
/* Hash may have been set up in inode_init_early */ if (!hashdist) return; inode_hashtable = alloc_large_system_hash("Inode-cache", sizeof(struct hlist_head), ihash_entries, 14, 0, &i_hash_shift, &i_hash_mask, 0); for (loop = 0; loop < (1 << i_hash_shift); loop++) INIT_HLIST_HEAD(&inode_hashtable[loop]); } void init_special_inode(struct inode *inode, umode_t mode, dev_t rdev) { inode->i_mode = mode; if (S_ISCHR(mode)) { inode->i_fop = &def_chr_fops; inode->i_rdev = rdev; } else if (S_ISBLK(mode)) { inode->i_fop = &def_blk_fops; inode->i_rdev = rdev; } else if (S_ISFIFO(mode)) inode->i_fop = &def_fifo_fops; else if (S_ISSOCK(mode)) inode->i_fop = &bad_sock_fops; else |
af0d9ae81
|
1580 1581 1582 1583 |
printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o) for" " inode %s:%lu ", mode, inode->i_sb->s_id, inode->i_ino); |
1da177e4c
|
1584 1585 |
} EXPORT_SYMBOL(init_special_inode); |
a1bd120d1
|
1586 1587 |
/** |
eaae668d0
|
1588 |
* inode_init_owner - Init uid,gid,mode for new inode according to posix standards |
a1bd120d1
|
1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 |
* @inode: New inode * @dir: Directory inode * @mode: mode of the new inode */ void inode_init_owner(struct inode *inode, const struct inode *dir, mode_t mode) { inode->i_uid = current_fsuid(); if (dir && dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; if (S_ISDIR(mode)) mode |= S_ISGID; } else inode->i_gid = current_fsgid(); inode->i_mode = mode; } EXPORT_SYMBOL(inode_init_owner); |
e795b7179
|
1606 |
|
2e1496707
|
1607 1608 1609 1610 1611 1612 |
/** * inode_owner_or_capable - check current task permissions to inode * @inode: inode being checked * * Return true if current either has CAP_FOWNER to the inode, or * owns the file. |
e795b7179
|
1613 |
*/ |
2e1496707
|
1614 |
bool inode_owner_or_capable(const struct inode *inode) |
e795b7179
|
1615 1616 1617 1618 1619 1620 1621 1622 1623 |
{ struct user_namespace *ns = inode_userns(inode); if (current_user_ns() == ns && current_fsuid() == inode->i_uid) return true; if (ns_capable(ns, CAP_FOWNER)) return true; return false; } |
2e1496707
|
1624 |
EXPORT_SYMBOL(inode_owner_or_capable); |