Blame view
fs/dcache.c
84 KB
457c89965
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
1da177e4c
|
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 |
/* * fs/dcache.c * * Complete reimplementation * (C) 1997 Thomas Schoebel-Theuer, * with heavy changes by Linus Torvalds */ /* * Notes on the allocation strategy: * * The dcache is a master of the icache - whenever a dcache entry * exists, the inode will always exist. "iput()" is done either when * the dcache entry is deleted or garbage collected. */ |
7a5cf791a
|
17 |
#include <linux/ratelimit.h> |
1da177e4c
|
18 19 20 |
#include <linux/string.h> #include <linux/mm.h> #include <linux/fs.h> |
0bf3d5c16
|
21 |
#include <linux/fscrypt.h> |
7a91bf7f5
|
22 |
#include <linux/fsnotify.h> |
1da177e4c
|
23 24 |
#include <linux/slab.h> #include <linux/init.h> |
1da177e4c
|
25 26 |
#include <linux/hash.h> #include <linux/cache.h> |
630d9c472
|
27 |
#include <linux/export.h> |
1da177e4c
|
28 29 |
#include <linux/security.h> #include <linux/seqlock.h> |
57c8a661d
|
30 |
#include <linux/memblock.h> |
ceb5bdc2d
|
31 32 |
#include <linux/bit_spinlock.h> #include <linux/rculist_bl.h> |
f60415675
|
33 |
#include <linux/list_lru.h> |
07f3f05c1
|
34 |
#include "internal.h" |
b2dba1af3
|
35 |
#include "mount.h" |
1da177e4c
|
36 |
|
789680d1e
|
37 38 |
/* * Usage: |
873feea09
|
39 |
* dcache->d_inode->i_lock protects: |
946e51f2b
|
40 |
* - i_dentry, d_u.d_alias, d_inode of aliases |
ceb5bdc2d
|
41 42 |
* dcache_hash_bucket lock protects: * - the dcache hash table |
f1ee61621
|
43 44 |
* s_roots bl list spinlock protects: * - the s_roots list (see __d_drop) |
19156840e
|
45 |
* dentry->d_sb->s_dentry_lru_lock protects: |
230445078
|
46 47 48 49 50 |
* - the dcache lru lists and counters * d_lock protects: * - d_flags * - d_name * - d_lru |
b7ab39f63
|
51 |
* - d_count |
da5029563
|
52 |
* - d_unhashed() |
2fd6b7f50
|
53 54 |
* - d_parent and d_subdirs * - childrens' d_child and d_parent |
946e51f2b
|
55 |
* - d_u.d_alias, d_inode |
789680d1e
|
56 57 |
* * Ordering: |
873feea09
|
58 |
* dentry->d_inode->i_lock |
b5c84bf6f
|
59 |
* dentry->d_lock |
19156840e
|
60 |
* dentry->d_sb->s_dentry_lru_lock |
ceb5bdc2d
|
61 |
* dcache_hash_bucket lock |
f1ee61621
|
62 |
* s_roots lock |
789680d1e
|
63 |
* |
da5029563
|
64 65 66 67 68 69 70 |
* If there is an ancestor relationship: * dentry->d_parent->...->d_parent->d_lock * ... * dentry->d_parent->d_lock * dentry->d_lock * * If no ancestor relationship: |
076515fc9
|
71 |
* arbitrary, since it's serialized on rename_lock |
789680d1e
|
72 |
*/ |
fa3536cc1
|
73 |
int sysctl_vfs_cache_pressure __read_mostly = 100; |
1da177e4c
|
74 |
EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure); |
74c3cbe33
|
75 |
__cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); |
1da177e4c
|
76 |
|
949854d02
|
77 |
EXPORT_SYMBOL(rename_lock); |
1da177e4c
|
78 |
|
e18b890bb
|
79 |
static struct kmem_cache *dentry_cache __read_mostly; |
1da177e4c
|
80 |
|
cdf01226b
|
81 82 83 84 |
const struct qstr empty_name = QSTR_INIT("", 0); EXPORT_SYMBOL(empty_name); const struct qstr slash_name = QSTR_INIT("/", 1); EXPORT_SYMBOL(slash_name); |
1da177e4c
|
85 86 87 88 89 90 91 92 |
/* * This is the single most critical data structure when it comes * to the dcache: the hashtable for lookups. Somebody should try * to make this good - I've just made it work. * * This hash-function tries to avoid losing too many bits of hash * information, yet avoid using a prime hash-size or similar. */ |
1da177e4c
|
93 |
|
fa3536cc1
|
94 |
static unsigned int d_hash_shift __read_mostly; |
ceb5bdc2d
|
95 |
|
b07ad9967
|
96 |
static struct hlist_bl_head *dentry_hashtable __read_mostly; |
ceb5bdc2d
|
97 |
|
8387ff257
|
98 |
static inline struct hlist_bl_head *d_hash(unsigned int hash) |
ceb5bdc2d
|
99 |
{ |
854d3e634
|
100 |
return dentry_hashtable + (hash >> d_hash_shift); |
ceb5bdc2d
|
101 |
} |
94bdd655c
|
102 103 104 105 106 107 108 109 110 |
#define IN_LOOKUP_SHIFT 10 static struct hlist_bl_head in_lookup_hashtable[1 << IN_LOOKUP_SHIFT]; static inline struct hlist_bl_head *in_lookup_hash(const struct dentry *parent, unsigned int hash) { hash += (unsigned long) parent / L1_CACHE_BYTES; return in_lookup_hashtable + hash_32(hash, IN_LOOKUP_SHIFT); } |
1da177e4c
|
111 112 113 114 |
/* Statistics gathering. */ struct dentry_stat_t dentry_stat = { .age_limit = 45, }; |
3942c07cc
|
115 |
static DEFINE_PER_CPU(long, nr_dentry); |
62d36c770
|
116 |
static DEFINE_PER_CPU(long, nr_dentry_unused); |
af0c9af1b
|
117 |
static DEFINE_PER_CPU(long, nr_dentry_negative); |
312d3ca85
|
118 119 |
#if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) |
62d36c770
|
120 121 122 123 124 125 126 127 128 129 130 131 132 |
/* * Here we resort to our own counters instead of using generic per-cpu counters * for consistency with what the vfs inode code does. We are expected to harvest * better code and performance by having our own specialized counters. * * Please note that the loop is done over all possible CPUs, not over all online * CPUs. The reason for this is that we don't want to play games with CPUs going * on and off. If one of them goes off, we will just keep their counters. * * glommer: See cffbc8a for details, and if you ever intend to change this, * please update all vfs counters to match. */ |
3942c07cc
|
133 |
static long get_nr_dentry(void) |
3e880fb5e
|
134 135 |
{ int i; |
3942c07cc
|
136 |
long sum = 0; |
3e880fb5e
|
137 138 139 140 |
for_each_possible_cpu(i) sum += per_cpu(nr_dentry, i); return sum < 0 ? 0 : sum; } |
62d36c770
|
141 142 143 144 145 146 147 148 |
static long get_nr_dentry_unused(void) { int i; long sum = 0; for_each_possible_cpu(i) sum += per_cpu(nr_dentry_unused, i); return sum < 0 ? 0 : sum; } |
af0c9af1b
|
149 150 151 152 153 154 155 156 157 |
static long get_nr_dentry_negative(void) { int i; long sum = 0; for_each_possible_cpu(i) sum += per_cpu(nr_dentry_negative, i); return sum < 0 ? 0 : sum; } |
1f7e0616c
|
158 |
int proc_nr_dentry(struct ctl_table *table, int write, void __user *buffer, |
312d3ca85
|
159 160 |
size_t *lenp, loff_t *ppos) { |
3e880fb5e
|
161 |
dentry_stat.nr_dentry = get_nr_dentry(); |
62d36c770
|
162 |
dentry_stat.nr_unused = get_nr_dentry_unused(); |
af0c9af1b
|
163 |
dentry_stat.nr_negative = get_nr_dentry_negative(); |
3942c07cc
|
164 |
return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); |
312d3ca85
|
165 166 |
} #endif |
5483f18e9
|
167 168 169 170 |
/* * Compare 2 name strings, return 0 if they match, otherwise non-zero. * The strings are both count bytes long, and count is non-zero. */ |
e419b4cc5
|
171 172 173 174 175 176 177 178 179 180 181 182 |
#ifdef CONFIG_DCACHE_WORD_ACCESS #include <asm/word-at-a-time.h> /* * NOTE! 'cs' and 'scount' come from a dentry, so it has a * aligned allocation for this particular component. We don't * strictly need the load_unaligned_zeropad() safety, but it * doesn't hurt either. * * In contrast, 'ct' and 'tcount' can be from a pathname, and do * need the careful unaligned handling. */ |
94753db5e
|
183 |
static inline int dentry_string_cmp(const unsigned char *cs, const unsigned char *ct, unsigned tcount) |
5483f18e9
|
184 |
{ |
bfcfaa77b
|
185 |
unsigned long a,b,mask; |
bfcfaa77b
|
186 187 |
for (;;) { |
bfe7aa6c3
|
188 |
a = read_word_at_a_time(cs); |
e419b4cc5
|
189 |
b = load_unaligned_zeropad(ct); |
bfcfaa77b
|
190 191 192 193 194 195 196 197 198 199 |
if (tcount < sizeof(unsigned long)) break; if (unlikely(a != b)) return 1; cs += sizeof(unsigned long); ct += sizeof(unsigned long); tcount -= sizeof(unsigned long); if (!tcount) return 0; } |
a5c21dcef
|
200 |
mask = bytemask_from_count(tcount); |
bfcfaa77b
|
201 |
return unlikely(!!((a ^ b) & mask)); |
e419b4cc5
|
202 |
} |
bfcfaa77b
|
203 |
#else |
e419b4cc5
|
204 |
|
94753db5e
|
205 |
static inline int dentry_string_cmp(const unsigned char *cs, const unsigned char *ct, unsigned tcount) |
e419b4cc5
|
206 |
{ |
5483f18e9
|
207 208 209 210 211 212 213 214 215 |
do { if (*cs != *ct) return 1; cs++; ct++; tcount--; } while (tcount); return 0; } |
e419b4cc5
|
216 |
#endif |
94753db5e
|
217 218 |
static inline int dentry_cmp(const struct dentry *dentry, const unsigned char *ct, unsigned tcount) { |
94753db5e
|
219 220 |
/* * Be careful about RCU walk racing with rename: |
506458efa
|
221 |
* use 'READ_ONCE' to fetch the name pointer. |
94753db5e
|
222 223 224 225 226 227 228 229 230 231 232 233 234 |
* * NOTE! Even if a rename will mean that the length * was not loaded atomically, we don't care. The * RCU walk will check the sequence count eventually, * and catch it. And we won't overrun the buffer, * because we're reading the name pointer atomically, * and a dentry name is guaranteed to be properly * terminated with a NUL byte. * * End result: even if 'len' is wrong, we'll exit * early because the data cannot match (there can * be no NUL in the ct/tcount data) */ |
506458efa
|
235 |
const unsigned char *cs = READ_ONCE(dentry->d_name.name); |
ae0a843c7
|
236 |
|
6326c71fd
|
237 |
return dentry_string_cmp(cs, ct, tcount); |
94753db5e
|
238 |
} |
8d85b4845
|
239 240 241 242 243 244 245 246 247 248 249 250 |
struct external_name { union { atomic_t count; struct rcu_head head; } u; unsigned char name[]; }; static inline struct external_name *external_name(struct dentry *dentry) { return container_of(dentry->d_name.name, struct external_name, name[0]); } |
9c82ab9c9
|
251 |
static void __d_free(struct rcu_head *head) |
1da177e4c
|
252 |
{ |
9c82ab9c9
|
253 |
struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); |
8d85b4845
|
254 255 256 257 258 259 |
kmem_cache_free(dentry_cache, dentry); } static void __d_free_external(struct rcu_head *head) { struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); |
2e03b4bc4
|
260 |
kfree(external_name(dentry)); |
f1782c9bc
|
261 |
kmem_cache_free(dentry_cache, dentry); |
1da177e4c
|
262 |
} |
810bb1726
|
263 264 265 266 |
static inline int dname_external(const struct dentry *dentry) { return dentry->d_name.name != dentry->d_iname; } |
49d31c2f3
|
267 268 269 |
void take_dentry_name_snapshot(struct name_snapshot *name, struct dentry *dentry) { spin_lock(&dentry->d_lock); |
230c6402b
|
270 |
name->name = dentry->d_name; |
49d31c2f3
|
271 |
if (unlikely(dname_external(dentry))) { |
230c6402b
|
272 |
atomic_inc(&external_name(dentry)->u.count); |
49d31c2f3
|
273 |
} else { |
6cd00a01f
|
274 275 |
memcpy(name->inline_name, dentry->d_iname, dentry->d_name.len + 1); |
230c6402b
|
276 |
name->name.name = name->inline_name; |
49d31c2f3
|
277 |
} |
230c6402b
|
278 |
spin_unlock(&dentry->d_lock); |
49d31c2f3
|
279 280 281 282 283 |
} EXPORT_SYMBOL(take_dentry_name_snapshot); void release_dentry_name_snapshot(struct name_snapshot *name) { |
230c6402b
|
284 |
if (unlikely(name->name.name != name->inline_name)) { |
49d31c2f3
|
285 |
struct external_name *p; |
230c6402b
|
286 |
p = container_of(name->name.name, struct external_name, name[0]); |
49d31c2f3
|
287 |
if (unlikely(atomic_dec_and_test(&p->u.count))) |
2e03b4bc4
|
288 |
kfree_rcu(p, u.head); |
49d31c2f3
|
289 290 291 |
} } EXPORT_SYMBOL(release_dentry_name_snapshot); |
4bf46a272
|
292 293 294 295 296 297 298 |
static inline void __d_set_inode_and_type(struct dentry *dentry, struct inode *inode, unsigned type_flags) { unsigned flags; dentry->d_inode = inode; |
4bf46a272
|
299 300 301 |
flags = READ_ONCE(dentry->d_flags); flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); flags |= type_flags; |
2fa6b1e01
|
302 |
smp_store_release(&dentry->d_flags, flags); |
4bf46a272
|
303 |
} |
4bf46a272
|
304 305 306 307 308 309 |
static inline void __d_clear_type_and_inode(struct dentry *dentry) { unsigned flags = READ_ONCE(dentry->d_flags); flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); WRITE_ONCE(dentry->d_flags, flags); |
4bf46a272
|
310 |
dentry->d_inode = NULL; |
af0c9af1b
|
311 312 |
if (dentry->d_flags & DCACHE_LRU_LIST) this_cpu_inc(nr_dentry_negative); |
4bf46a272
|
313 |
} |
b4f0354e9
|
314 315 |
static void dentry_free(struct dentry *dentry) { |
946e51f2b
|
316 |
WARN_ON(!hlist_unhashed(&dentry->d_u.d_alias)); |
8d85b4845
|
317 318 319 320 321 322 323 |
if (unlikely(dname_external(dentry))) { struct external_name *p = external_name(dentry); if (likely(atomic_dec_and_test(&p->u.count))) { call_rcu(&dentry->d_u.d_rcu, __d_free_external); return; } } |
b4f0354e9
|
324 |
/* if dentry was never visible to RCU, immediate free is OK */ |
5467a68cb
|
325 |
if (dentry->d_flags & DCACHE_NORCU) |
b4f0354e9
|
326 327 328 329 |
__d_free(&dentry->d_u.d_rcu); else call_rcu(&dentry->d_u.d_rcu, __d_free); } |
1da177e4c
|
330 331 |
/* * Release the dentry's inode, using the filesystem |
550dce01d
|
332 |
* d_iput() operation if defined. |
31e6b01f4
|
333 334 335 |
*/ static void dentry_unlink_inode(struct dentry * dentry) __releases(dentry->d_lock) |
873feea09
|
336 |
__releases(dentry->d_inode->i_lock) |
31e6b01f4
|
337 338 |
{ struct inode *inode = dentry->d_inode; |
a528aca7f
|
339 |
|
4c0d7cd5c
|
340 |
raw_write_seqcount_begin(&dentry->d_seq); |
4bf46a272
|
341 |
__d_clear_type_and_inode(dentry); |
946e51f2b
|
342 |
hlist_del_init(&dentry->d_u.d_alias); |
4c0d7cd5c
|
343 |
raw_write_seqcount_end(&dentry->d_seq); |
31e6b01f4
|
344 |
spin_unlock(&dentry->d_lock); |
873feea09
|
345 |
spin_unlock(&inode->i_lock); |
31e6b01f4
|
346 347 348 349 350 351 352 353 354 |
if (!inode->i_nlink) fsnotify_inoderemove(inode); if (dentry->d_op && dentry->d_op->d_iput) dentry->d_op->d_iput(dentry, inode); else iput(inode); } /* |
89dc77bcd
|
355 356 357 358 359 360 361 362 363 364 |
* The DCACHE_LRU_LIST bit is set whenever the 'd_lru' entry * is in use - which includes both the "real" per-superblock * LRU list _and_ the DCACHE_SHRINK_LIST use. * * The DCACHE_SHRINK_LIST bit is set whenever the dentry is * on the shrink list (ie not on the superblock LRU list). * * The per-cpu "nr_dentry_unused" counters are updated with * the DCACHE_LRU_LIST bit. * |
af0c9af1b
|
365 366 367 368 369 |
* The per-cpu "nr_dentry_negative" counters are only updated * when deleted from or added to the per-superblock LRU list, not * from/to the shrink list. That is to avoid an unneeded dec/inc * pair when moving from LRU to shrink list in select_collect(). * |
89dc77bcd
|
370 371 372 373 374 375 376 377 378 |
* These helper functions make sure we always follow the * rules. d_lock must be held by the caller. */ #define D_FLAG_VERIFY(dentry,x) WARN_ON_ONCE(((dentry)->d_flags & (DCACHE_LRU_LIST | DCACHE_SHRINK_LIST)) != (x)) static void d_lru_add(struct dentry *dentry) { D_FLAG_VERIFY(dentry, 0); dentry->d_flags |= DCACHE_LRU_LIST; this_cpu_inc(nr_dentry_unused); |
af0c9af1b
|
379 380 |
if (d_is_negative(dentry)) this_cpu_inc(nr_dentry_negative); |
89dc77bcd
|
381 382 383 384 385 386 387 388 |
WARN_ON_ONCE(!list_lru_add(&dentry->d_sb->s_dentry_lru, &dentry->d_lru)); } static void d_lru_del(struct dentry *dentry) { D_FLAG_VERIFY(dentry, DCACHE_LRU_LIST); dentry->d_flags &= ~DCACHE_LRU_LIST; this_cpu_dec(nr_dentry_unused); |
af0c9af1b
|
389 390 |
if (d_is_negative(dentry)) this_cpu_dec(nr_dentry_negative); |
89dc77bcd
|
391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 |
WARN_ON_ONCE(!list_lru_del(&dentry->d_sb->s_dentry_lru, &dentry->d_lru)); } static void d_shrink_del(struct dentry *dentry) { D_FLAG_VERIFY(dentry, DCACHE_SHRINK_LIST | DCACHE_LRU_LIST); list_del_init(&dentry->d_lru); dentry->d_flags &= ~(DCACHE_SHRINK_LIST | DCACHE_LRU_LIST); this_cpu_dec(nr_dentry_unused); } static void d_shrink_add(struct dentry *dentry, struct list_head *list) { D_FLAG_VERIFY(dentry, 0); list_add(&dentry->d_lru, list); dentry->d_flags |= DCACHE_SHRINK_LIST | DCACHE_LRU_LIST; this_cpu_inc(nr_dentry_unused); } /* * These can only be called under the global LRU lock, ie during the * callback for freeing the LRU list. "isolate" removes it from the * LRU lists entirely, while shrink_move moves it to the indicated * private list. */ |
3f97b1632
|
416 |
static void d_lru_isolate(struct list_lru_one *lru, struct dentry *dentry) |
89dc77bcd
|
417 418 419 420 |
{ D_FLAG_VERIFY(dentry, DCACHE_LRU_LIST); dentry->d_flags &= ~DCACHE_LRU_LIST; this_cpu_dec(nr_dentry_unused); |
af0c9af1b
|
421 422 |
if (d_is_negative(dentry)) this_cpu_dec(nr_dentry_negative); |
3f97b1632
|
423 |
list_lru_isolate(lru, &dentry->d_lru); |
89dc77bcd
|
424 |
} |
3f97b1632
|
425 426 |
static void d_lru_shrink_move(struct list_lru_one *lru, struct dentry *dentry, struct list_head *list) |
89dc77bcd
|
427 428 429 |
{ D_FLAG_VERIFY(dentry, DCACHE_LRU_LIST); dentry->d_flags |= DCACHE_SHRINK_LIST; |
af0c9af1b
|
430 431 |
if (d_is_negative(dentry)) this_cpu_dec(nr_dentry_negative); |
3f97b1632
|
432 |
list_lru_isolate_move(lru, &dentry->d_lru, list); |
89dc77bcd
|
433 |
} |
d52b90864
|
434 |
/** |
789680d1e
|
435 436 437 438 439 440 441 442 443 444 445 446 |
* d_drop - drop a dentry * @dentry: dentry to drop * * d_drop() unhashes the entry from the parent dentry hashes, so that it won't * be found through a VFS lookup any more. Note that this is different from * deleting the dentry - d_delete will try to mark the dentry negative if * possible, giving a successful _negative_ lookup, while d_drop will * just make the cache lookup fail. * * d_drop() is used mainly for stuff that wants to invalidate a dentry for some * reason (NFS timeouts or autofs deletes). * |
61647823a
|
447 448 449 |
* __d_drop requires dentry->d_lock * ___d_drop doesn't mark dentry as "unhashed" * (dentry->d_hash.pprev will be LIST_POISON2, not NULL). |
789680d1e
|
450 |
*/ |
61647823a
|
451 |
static void ___d_drop(struct dentry *dentry) |
789680d1e
|
452 |
{ |
0632a9ac7
|
453 454 455 456 457 458 459 460 461 462 |
struct hlist_bl_head *b; /* * Hashed dentries are normally on the dentry hashtable, * with the exception of those newly allocated by * d_obtain_root, which are always IS_ROOT: */ if (unlikely(IS_ROOT(dentry))) b = &dentry->d_sb->s_roots; else b = d_hash(dentry->d_name.hash); |
b61625d24
|
463 |
|
0632a9ac7
|
464 465 466 |
hlist_bl_lock(b); __hlist_bl_del(&dentry->d_hash); hlist_bl_unlock(b); |
789680d1e
|
467 |
} |
61647823a
|
468 469 470 |
void __d_drop(struct dentry *dentry) { |
0632a9ac7
|
471 472 473 474 475 |
if (!d_unhashed(dentry)) { ___d_drop(dentry); dentry->d_hash.pprev = NULL; write_seqcount_invalidate(&dentry->d_seq); } |
61647823a
|
476 |
} |
789680d1e
|
477 478 479 480 |
EXPORT_SYMBOL(__d_drop); void d_drop(struct dentry *dentry) { |
789680d1e
|
481 482 483 |
spin_lock(&dentry->d_lock); __d_drop(dentry); spin_unlock(&dentry->d_lock); |
789680d1e
|
484 485 |
} EXPORT_SYMBOL(d_drop); |
ba65dc5ef
|
486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 |
static inline void dentry_unlist(struct dentry *dentry, struct dentry *parent) { struct dentry *next; /* * Inform d_walk() and shrink_dentry_list() that we are no longer * attached to the dentry tree */ dentry->d_flags |= DCACHE_DENTRY_KILLED; if (unlikely(list_empty(&dentry->d_child))) return; __list_del_entry(&dentry->d_child); /* * Cursors can move around the list of children. While we'd been * a normal list member, it didn't matter - ->d_child.next would've * been updated. However, from now on it won't be and for the * things like d_walk() it might end up with a nasty surprise. * Normally d_walk() doesn't care about cursors moving around - * ->d_lock on parent prevents that and since a cursor has no children * of its own, we get through it without ever unlocking the parent. * There is one exception, though - if we ascend from a child that * gets killed as soon as we unlock it, the next sibling is found * using the value left in its ->d_child.next. And if _that_ * pointed to a cursor, and cursor got moved (e.g. by lseek()) * before d_walk() regains parent->d_lock, we'll end up skipping * everything the cursor had been moved past. * * Solution: make sure that the pointer left behind in ->d_child.next * points to something that won't be moving around. I.e. skip the * cursors. */ while (dentry->d_child.next != &parent->d_subdirs) { next = list_entry(dentry->d_child.next, struct dentry, d_child); if (likely(!(next->d_flags & DCACHE_DENTRY_CURSOR))) break; dentry->d_child.next = next->d_child.next; } } |
e55fd0115
|
523 |
static void __dentry_kill(struct dentry *dentry) |
77812a1ef
|
524 |
{ |
41edf278f
|
525 526 |
struct dentry *parent = NULL; bool can_free = true; |
41edf278f
|
527 |
if (!IS_ROOT(dentry)) |
77812a1ef
|
528 |
parent = dentry->d_parent; |
31e6b01f4
|
529 |
|
0d98439ea
|
530 531 532 533 |
/* * The dentry is now unrecoverably dead to the world. */ lockref_mark_dead(&dentry->d_lockref); |
f0023bc61
|
534 |
/* |
f0023bc61
|
535 536 537 |
* inform the fs via d_prune that this dentry is about to be * unhashed and destroyed. */ |
292662014
|
538 |
if (dentry->d_flags & DCACHE_OP_PRUNE) |
61572bb1f
|
539 |
dentry->d_op->d_prune(dentry); |
01b603519
|
540 541 542 |
if (dentry->d_flags & DCACHE_LRU_LIST) { if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) d_lru_del(dentry); |
01b603519
|
543 |
} |
77812a1ef
|
544 545 |
/* if it was on the hash then remove it */ __d_drop(dentry); |
ba65dc5ef
|
546 |
dentry_unlist(dentry, parent); |
03b3b889e
|
547 548 |
if (parent) spin_unlock(&parent->d_lock); |
550dce01d
|
549 550 551 552 |
if (dentry->d_inode) dentry_unlink_inode(dentry); else spin_unlock(&dentry->d_lock); |
03b3b889e
|
553 554 555 |
this_cpu_dec(nr_dentry); if (dentry->d_op && dentry->d_op->d_release) dentry->d_op->d_release(dentry); |
41edf278f
|
556 557 558 559 560 561 |
spin_lock(&dentry->d_lock); if (dentry->d_flags & DCACHE_SHRINK_LIST) { dentry->d_flags |= DCACHE_MAY_FREE; can_free = false; } spin_unlock(&dentry->d_lock); |
41edf278f
|
562 563 |
if (likely(can_free)) dentry_free(dentry); |
9c5f1d301
|
564 |
cond_resched(); |
e55fd0115
|
565 |
} |
8b987a46a
|
566 |
static struct dentry *__lock_parent(struct dentry *dentry) |
046b961b4
|
567 |
{ |
8b987a46a
|
568 |
struct dentry *parent; |
046b961b4
|
569 |
rcu_read_lock(); |
c2338f2dc
|
570 |
spin_unlock(&dentry->d_lock); |
046b961b4
|
571 |
again: |
66702eb59
|
572 |
parent = READ_ONCE(dentry->d_parent); |
046b961b4
|
573 574 575 576 577 578 579 580 581 582 583 584 585 |
spin_lock(&parent->d_lock); /* * We can't blindly lock dentry until we are sure * that we won't violate the locking order. * Any changes of dentry->d_parent must have * been done with parent->d_lock held, so * spin_lock() above is enough of a barrier * for checking if it's still our child. */ if (unlikely(parent != dentry->d_parent)) { spin_unlock(&parent->d_lock); goto again; } |
65d8eb5a8
|
586 587 |
rcu_read_unlock(); if (parent != dentry) |
9f12600fe
|
588 |
spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
65d8eb5a8
|
589 |
else |
046b961b4
|
590 591 592 |
parent = NULL; return parent; } |
8b987a46a
|
593 594 595 596 597 598 599 600 601 |
static inline struct dentry *lock_parent(struct dentry *dentry) { struct dentry *parent = dentry->d_parent; if (IS_ROOT(dentry)) return NULL; if (likely(spin_trylock(&parent->d_lock))) return parent; return __lock_parent(dentry); } |
a338579f2
|
602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 |
static inline bool retain_dentry(struct dentry *dentry) { WARN_ON(d_in_lookup(dentry)); /* Unreachable? Get rid of it */ if (unlikely(d_unhashed(dentry))) return false; if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) return false; if (unlikely(dentry->d_flags & DCACHE_OP_DELETE)) { if (dentry->d_op->d_delete(dentry)) return false; } |
62d9956ce
|
617 618 619 620 621 622 |
/* retain; LRU fodder */ dentry->d_lockref.count--; if (unlikely(!(dentry->d_flags & DCACHE_LRU_LIST))) d_lru_add(dentry); else if (unlikely(!(dentry->d_flags & DCACHE_REFERENCED))) dentry->d_flags |= DCACHE_REFERENCED; |
a338579f2
|
623 624 |
return true; } |
360f54796
|
625 |
/* |
c1d0c1a2b
|
626 627 628 629 630 631 632 633 634 635 636 |
* Finish off a dentry we've decided to kill. * dentry->d_lock must be held, returns with it unlocked. * Returns dentry requiring refcount drop, or NULL if we're done. */ static struct dentry *dentry_kill(struct dentry *dentry) __releases(dentry->d_lock) { struct inode *inode = dentry->d_inode; struct dentry *parent = NULL; if (inode && unlikely(!spin_trylock(&inode->i_lock))) |
f657a666f
|
637 |
goto slow_positive; |
c1d0c1a2b
|
638 639 640 641 |
if (!IS_ROOT(dentry)) { parent = dentry->d_parent; if (unlikely(!spin_trylock(&parent->d_lock))) { |
f657a666f
|
642 643 644 645 646 647 648 649 |
parent = __lock_parent(dentry); if (likely(inode || !dentry->d_inode)) goto got_locks; /* negative that became positive */ if (parent) spin_unlock(&parent->d_lock); inode = dentry->d_inode; goto slow_positive; |
c1d0c1a2b
|
650 651 |
} } |
c1d0c1a2b
|
652 653 |
__dentry_kill(dentry); return parent; |
f657a666f
|
654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 |
slow_positive: spin_unlock(&dentry->d_lock); spin_lock(&inode->i_lock); spin_lock(&dentry->d_lock); parent = lock_parent(dentry); got_locks: if (unlikely(dentry->d_lockref.count != 1)) { dentry->d_lockref.count--; } else if (likely(!retain_dentry(dentry))) { __dentry_kill(dentry); return parent; } /* we are keeping it, after all */ if (inode) spin_unlock(&inode->i_lock); if (parent) spin_unlock(&parent->d_lock); |
c1d0c1a2b
|
671 |
spin_unlock(&dentry->d_lock); |
f657a666f
|
672 |
return NULL; |
c1d0c1a2b
|
673 674 675 |
} /* |
360f54796
|
676 677 678 679 680 681 682 683 684 685 686 687 688 689 |
* Try to do a lockless dput(), and return whether that was successful. * * If unsuccessful, we return false, having already taken the dentry lock. * * The caller needs to hold the RCU read lock, so that the dentry is * guaranteed to stay around even if the refcount goes down to zero! */ static inline bool fast_dput(struct dentry *dentry) { int ret; unsigned int d_flags; /* * If we have a d_op->d_delete() operation, we sould not |
75a6f82a0
|
690 |
* let the dentry count go to zero, so use "put_or_lock". |
360f54796
|
691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 |
*/ if (unlikely(dentry->d_flags & DCACHE_OP_DELETE)) return lockref_put_or_lock(&dentry->d_lockref); /* * .. otherwise, we can try to just decrement the * lockref optimistically. */ ret = lockref_put_return(&dentry->d_lockref); /* * If the lockref_put_return() failed due to the lock being held * by somebody else, the fast path has failed. We will need to * get the lock, and then check the count again. */ if (unlikely(ret < 0)) { spin_lock(&dentry->d_lock); if (dentry->d_lockref.count > 1) { dentry->d_lockref.count--; spin_unlock(&dentry->d_lock); |
7964410fc
|
711 |
return true; |
360f54796
|
712 |
} |
7964410fc
|
713 |
return false; |
360f54796
|
714 715 716 717 718 719 |
} /* * If we weren't the last ref, we're done. */ if (ret) |
7964410fc
|
720 |
return true; |
360f54796
|
721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 |
/* * Careful, careful. The reference count went down * to zero, but we don't hold the dentry lock, so * somebody else could get it again, and do another * dput(), and we need to not race with that. * * However, there is a very special and common case * where we don't care, because there is nothing to * do: the dentry is still hashed, it does not have * a 'delete' op, and it's referenced and already on * the LRU list. * * NOTE! Since we aren't locked, these values are * not "stable". However, it is sufficient that at * some point after we dropped the reference the * dentry was hashed and the flags had the proper * value. Other dentry users may have re-gotten * a reference to the dentry and change that, but * our work is done - we can leave the dentry * around with a zero refcount. */ smp_rmb(); |
66702eb59
|
744 |
d_flags = READ_ONCE(dentry->d_flags); |
75a6f82a0
|
745 |
d_flags &= DCACHE_REFERENCED | DCACHE_LRU_LIST | DCACHE_DISCONNECTED; |
360f54796
|
746 747 748 |
/* Nothing to do? Dropping the reference was all we needed? */ if (d_flags == (DCACHE_REFERENCED | DCACHE_LRU_LIST) && !d_unhashed(dentry)) |
7964410fc
|
749 |
return true; |
360f54796
|
750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 |
/* * Not the fast normal case? Get the lock. We've already decremented * the refcount, but we'll need to re-check the situation after * getting the lock. */ spin_lock(&dentry->d_lock); /* * Did somebody else grab a reference to it in the meantime, and * we're no longer the last user after all? Alternatively, somebody * else could have killed it and marked it dead. Either way, we * don't need to do anything else. */ if (dentry->d_lockref.count) { spin_unlock(&dentry->d_lock); |
7964410fc
|
766 |
return true; |
360f54796
|
767 768 769 770 771 772 773 774 |
} /* * Re-get the reference we optimistically dropped. We hold the * lock, and we just tested that it was zero, so we can just * set it to 1. */ dentry->d_lockref.count = 1; |
7964410fc
|
775 |
return false; |
360f54796
|
776 |
} |
1da177e4c
|
777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 |
/* * This is dput * * This is complicated by the fact that we do not want to put * dentries that are no longer on any hash chain on the unused * list: we'd much rather just get rid of them immediately. * * However, that implies that we have to traverse the dentry * tree upwards to the parents which might _also_ now be * scheduled for deletion (it may have been only waiting for * its last child to go away). * * This tail recursion is done by hand as we don't want to depend * on the compiler to always get this right (gcc generally doesn't). * Real recursion would eat up our stack space. */ /* * dput - release a dentry * @dentry: dentry to release * * Release a dentry. This will drop the usage count and if appropriate * call the dentry unlink method as well as removing it from the queues and * releasing its resources. If the parent dentries were scheduled for release * they too may now get deleted. |
1da177e4c
|
802 |
*/ |
1da177e4c
|
803 804 |
void dput(struct dentry *dentry) { |
1088a6408
|
805 806 |
while (dentry) { might_sleep(); |
1da177e4c
|
807 |
|
1088a6408
|
808 809 810 811 812 |
rcu_read_lock(); if (likely(fast_dput(dentry))) { rcu_read_unlock(); return; } |
47be61845
|
813 |
|
1088a6408
|
814 |
/* Slow case: now with the dentry lock held */ |
360f54796
|
815 |
rcu_read_unlock(); |
360f54796
|
816 |
|
1088a6408
|
817 818 819 820 |
if (likely(retain_dentry(dentry))) { spin_unlock(&dentry->d_lock); return; } |
265ac9023
|
821 |
|
1088a6408
|
822 |
dentry = dentry_kill(dentry); |
47be61845
|
823 |
} |
1da177e4c
|
824 |
} |
ec4f86059
|
825 |
EXPORT_SYMBOL(dput); |
1da177e4c
|
826 |
|
9bdebc2bd
|
827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 |
static void __dput_to_list(struct dentry *dentry, struct list_head *list) __must_hold(&dentry->d_lock) { if (dentry->d_flags & DCACHE_SHRINK_LIST) { /* let the owner of the list it's on deal with it */ --dentry->d_lockref.count; } else { if (dentry->d_flags & DCACHE_LRU_LIST) d_lru_del(dentry); if (!--dentry->d_lockref.count) d_shrink_add(dentry, list); } } void dput_to_list(struct dentry *dentry, struct list_head *list) { rcu_read_lock(); if (likely(fast_dput(dentry))) { rcu_read_unlock(); return; } rcu_read_unlock(); if (!retain_dentry(dentry)) __dput_to_list(dentry, list); spin_unlock(&dentry->d_lock); } |
1da177e4c
|
853 |
|
b5c84bf6f
|
854 |
/* This must be called with d_lock held */ |
dc0474be3
|
855 |
static inline void __dget_dlock(struct dentry *dentry) |
230445078
|
856 |
{ |
98474236f
|
857 |
dentry->d_lockref.count++; |
230445078
|
858 |
} |
dc0474be3
|
859 |
static inline void __dget(struct dentry *dentry) |
1da177e4c
|
860 |
{ |
98474236f
|
861 |
lockref_get(&dentry->d_lockref); |
1da177e4c
|
862 |
} |
b7ab39f63
|
863 864 |
struct dentry *dget_parent(struct dentry *dentry) { |
df3d0bbcd
|
865 |
int gotref; |
b7ab39f63
|
866 |
struct dentry *ret; |
e84009336
|
867 |
unsigned seq; |
b7ab39f63
|
868 |
|
df3d0bbcd
|
869 870 871 872 873 |
/* * Do optimistic parent lookup without any * locking. */ rcu_read_lock(); |
e84009336
|
874 |
seq = raw_seqcount_begin(&dentry->d_seq); |
66702eb59
|
875 |
ret = READ_ONCE(dentry->d_parent); |
df3d0bbcd
|
876 877 878 |
gotref = lockref_get_not_zero(&ret->d_lockref); rcu_read_unlock(); if (likely(gotref)) { |
e84009336
|
879 |
if (!read_seqcount_retry(&dentry->d_seq, seq)) |
df3d0bbcd
|
880 881 882 |
return ret; dput(ret); } |
b7ab39f63
|
883 |
repeat: |
a734eb458
|
884 885 886 887 888 |
/* * Don't need rcu_dereference because we re-check it was correct under * the lock. */ rcu_read_lock(); |
b7ab39f63
|
889 |
ret = dentry->d_parent; |
a734eb458
|
890 891 892 893 |
spin_lock(&ret->d_lock); if (unlikely(ret != dentry->d_parent)) { spin_unlock(&ret->d_lock); rcu_read_unlock(); |
b7ab39f63
|
894 895 |
goto repeat; } |
a734eb458
|
896 |
rcu_read_unlock(); |
98474236f
|
897 898 |
BUG_ON(!ret->d_lockref.count); ret->d_lockref.count++; |
b7ab39f63
|
899 |
spin_unlock(&ret->d_lock); |
b7ab39f63
|
900 901 902 |
return ret; } EXPORT_SYMBOL(dget_parent); |
61fec493c
|
903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 |
static struct dentry * __d_find_any_alias(struct inode *inode) { struct dentry *alias; if (hlist_empty(&inode->i_dentry)) return NULL; alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias); __dget(alias); return alias; } /** * d_find_any_alias - find any alias for a given inode * @inode: inode to find an alias for * * If any aliases exist for the given inode, take and return a * reference for one of them. If no aliases exist, return %NULL. */ struct dentry *d_find_any_alias(struct inode *inode) { struct dentry *de; spin_lock(&inode->i_lock); de = __d_find_any_alias(inode); spin_unlock(&inode->i_lock); return de; } EXPORT_SYMBOL(d_find_any_alias); |
1da177e4c
|
931 932 933 |
/** * d_find_alias - grab a hashed alias of inode * @inode: inode in question |
1da177e4c
|
934 935 936 937 938 |
* * If inode has a hashed alias, or is a directory and has any alias, * acquire the reference to alias and return it. Otherwise return NULL. * Notice that if inode is a directory there can be only one alias and * it can be unhashed only if it has no children, or if it is the root |
3ccb354d6
|
939 940 |
* of a filesystem, or if the directory was renamed and d_revalidate * was the first vfs operation to notice. |
1da177e4c
|
941 |
* |
21c0d8fdd
|
942 |
* If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer |
52ed46f0f
|
943 |
* any other hashed alias over that one. |
1da177e4c
|
944 |
*/ |
52ed46f0f
|
945 |
static struct dentry *__d_find_alias(struct inode *inode) |
1da177e4c
|
946 |
{ |
61fec493c
|
947 948 949 950 |
struct dentry *alias; if (S_ISDIR(inode->i_mode)) return __d_find_any_alias(inode); |
1da177e4c
|
951 |
|
946e51f2b
|
952 |
hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) { |
da5029563
|
953 |
spin_lock(&alias->d_lock); |
61fec493c
|
954 |
if (!d_unhashed(alias)) { |
8d80d7dab
|
955 956 957 |
__dget_dlock(alias); spin_unlock(&alias->d_lock); return alias; |
1da177e4c
|
958 |
} |
da5029563
|
959 |
spin_unlock(&alias->d_lock); |
1da177e4c
|
960 |
} |
da5029563
|
961 |
return NULL; |
1da177e4c
|
962 |
} |
da5029563
|
963 |
struct dentry *d_find_alias(struct inode *inode) |
1da177e4c
|
964 |
{ |
214fda1f6
|
965 |
struct dentry *de = NULL; |
b3d9b7a3c
|
966 |
if (!hlist_empty(&inode->i_dentry)) { |
873feea09
|
967 |
spin_lock(&inode->i_lock); |
52ed46f0f
|
968 |
de = __d_find_alias(inode); |
873feea09
|
969 |
spin_unlock(&inode->i_lock); |
214fda1f6
|
970 |
} |
1da177e4c
|
971 972 |
return de; } |
ec4f86059
|
973 |
EXPORT_SYMBOL(d_find_alias); |
1da177e4c
|
974 975 976 977 978 979 980 |
/* * Try to kill dentries associated with this inode. * WARNING: you must own a reference to inode. */ void d_prune_aliases(struct inode *inode) { |
0cdca3f98
|
981 |
struct dentry *dentry; |
1da177e4c
|
982 |
restart: |
873feea09
|
983 |
spin_lock(&inode->i_lock); |
946e51f2b
|
984 |
hlist_for_each_entry(dentry, &inode->i_dentry, d_u.d_alias) { |
1da177e4c
|
985 |
spin_lock(&dentry->d_lock); |
98474236f
|
986 |
if (!dentry->d_lockref.count) { |
29355c390
|
987 988 989 |
struct dentry *parent = lock_parent(dentry); if (likely(!dentry->d_lockref.count)) { __dentry_kill(dentry); |
4a7795d35
|
990 |
dput(parent); |
29355c390
|
991 992 993 994 |
goto restart; } if (parent) spin_unlock(&parent->d_lock); |
1da177e4c
|
995 996 997 |
} spin_unlock(&dentry->d_lock); } |
873feea09
|
998 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
999 |
} |
ec4f86059
|
1000 |
EXPORT_SYMBOL(d_prune_aliases); |
1da177e4c
|
1001 |
|
3b3f09f48
|
1002 1003 |
/* * Lock a dentry from shrink list. |
8f04da2ad
|
1004 1005 |
* Called under rcu_read_lock() and dentry->d_lock; the former * guarantees that nothing we access will be freed under us. |
3b3f09f48
|
1006 |
* Note that dentry is *not* protected from concurrent dentry_kill(), |
8f04da2ad
|
1007 1008 |
* d_delete(), etc. * |
3b3f09f48
|
1009 1010 1011 1012 1013 |
* Return false if dentry has been disrupted or grabbed, leaving * the caller to kick it off-list. Otherwise, return true and have * that dentry's inode and parent both locked. */ static bool shrink_lock_dentry(struct dentry *dentry) |
1da177e4c
|
1014 |
{ |
3b3f09f48
|
1015 1016 |
struct inode *inode; struct dentry *parent; |
da3bbdd46
|
1017 |
|
3b3f09f48
|
1018 1019 1020 1021 1022 |
if (dentry->d_lockref.count) return false; inode = dentry->d_inode; if (inode && unlikely(!spin_trylock(&inode->i_lock))) { |
3b3f09f48
|
1023 1024 |
spin_unlock(&dentry->d_lock); spin_lock(&inode->i_lock); |
ec33679d7
|
1025 |
spin_lock(&dentry->d_lock); |
3b3f09f48
|
1026 1027 1028 1029 1030 |
if (unlikely(dentry->d_lockref.count)) goto out; /* changed inode means that somebody had grabbed it */ if (unlikely(inode != dentry->d_inode)) goto out; |
3b3f09f48
|
1031 |
} |
046b961b4
|
1032 |
|
3b3f09f48
|
1033 1034 1035 |
parent = dentry->d_parent; if (IS_ROOT(dentry) || likely(spin_trylock(&parent->d_lock))) return true; |
dd1f6b2e4
|
1036 |
|
3b3f09f48
|
1037 |
spin_unlock(&dentry->d_lock); |
3b3f09f48
|
1038 1039 1040 1041 1042 1043 1044 |
spin_lock(&parent->d_lock); if (unlikely(parent != dentry->d_parent)) { spin_unlock(&parent->d_lock); spin_lock(&dentry->d_lock); goto out; } spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
8f04da2ad
|
1045 |
if (likely(!dentry->d_lockref.count)) |
3b3f09f48
|
1046 |
return true; |
3b3f09f48
|
1047 1048 1049 1050 |
spin_unlock(&parent->d_lock); out: if (inode) spin_unlock(&inode->i_lock); |
3b3f09f48
|
1051 1052 |
return false; } |
77812a1ef
|
1053 |
|
9bdebc2bd
|
1054 |
void shrink_dentry_list(struct list_head *list) |
3b3f09f48
|
1055 1056 1057 |
{ while (!list_empty(list)) { struct dentry *dentry, *parent; |
64fd72e0a
|
1058 |
|
3b3f09f48
|
1059 1060 |
dentry = list_entry(list->prev, struct dentry, d_lru); spin_lock(&dentry->d_lock); |
8f04da2ad
|
1061 |
rcu_read_lock(); |
3b3f09f48
|
1062 1063 |
if (!shrink_lock_dentry(dentry)) { bool can_free = false; |
8f04da2ad
|
1064 |
rcu_read_unlock(); |
3b3f09f48
|
1065 1066 1067 |
d_shrink_del(dentry); if (dentry->d_lockref.count < 0) can_free = dentry->d_flags & DCACHE_MAY_FREE; |
64fd72e0a
|
1068 1069 1070 1071 1072 |
spin_unlock(&dentry->d_lock); if (can_free) dentry_free(dentry); continue; } |
8f04da2ad
|
1073 |
rcu_read_unlock(); |
3b3f09f48
|
1074 1075 |
d_shrink_del(dentry); parent = dentry->d_parent; |
9bdebc2bd
|
1076 1077 |
if (parent != dentry) __dput_to_list(parent, list); |
ff2fde992
|
1078 |
__dentry_kill(dentry); |
da3bbdd46
|
1079 |
} |
3049cfe24
|
1080 |
} |
3f97b1632
|
1081 1082 |
static enum lru_status dentry_lru_isolate(struct list_head *item, struct list_lru_one *lru, spinlock_t *lru_lock, void *arg) |
f60415675
|
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 |
{ struct list_head *freeable = arg; struct dentry *dentry = container_of(item, struct dentry, d_lru); /* * we are inverting the lru lock/dentry->d_lock here, * so use a trylock. If we fail to get the lock, just skip * it */ if (!spin_trylock(&dentry->d_lock)) return LRU_SKIP; /* * Referenced dentries are still in use. If they have active * counts, just remove them from the LRU. Otherwise give them * another pass through the LRU. */ if (dentry->d_lockref.count) { |
3f97b1632
|
1102 |
d_lru_isolate(lru, dentry); |
f60415675
|
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 |
spin_unlock(&dentry->d_lock); return LRU_REMOVED; } if (dentry->d_flags & DCACHE_REFERENCED) { dentry->d_flags &= ~DCACHE_REFERENCED; spin_unlock(&dentry->d_lock); /* * The list move itself will be made by the common LRU code. At * this point, we've dropped the dentry->d_lock but keep the * lru lock. This is safe to do, since every list movement is * protected by the lru lock even if both locks are held. * * This is guaranteed by the fact that all LRU management * functions are intermediated by the LRU API calls like * list_lru_add and list_lru_del. List movement in this file * only ever occur through this functions or through callbacks * like this one, that are called from the LRU API. * * The only exceptions to this are functions like * shrink_dentry_list, and code that first checks for the * DCACHE_SHRINK_LIST flag. Those are guaranteed to be * operating only with stack provided lists after they are * properly isolated from the main list. It is thus, always a * local access. */ return LRU_ROTATE; } |
3f97b1632
|
1132 |
d_lru_shrink_move(lru, dentry, freeable); |
f60415675
|
1133 1134 1135 1136 |
spin_unlock(&dentry->d_lock); return LRU_REMOVED; } |
3049cfe24
|
1137 |
/** |
b48f03b31
|
1138 1139 |
* prune_dcache_sb - shrink the dcache * @sb: superblock |
503c358cf
|
1140 |
* @sc: shrink control, passed to list_lru_shrink_walk() |
b48f03b31
|
1141 |
* |
503c358cf
|
1142 1143 |
* Attempt to shrink the superblock dcache LRU by @sc->nr_to_scan entries. This * is done when we need more memory and called from the superblock shrinker |
b48f03b31
|
1144 |
* function. |
3049cfe24
|
1145 |
* |
b48f03b31
|
1146 1147 |
* This function may fail to free any resources if all the dentries are in * use. |
3049cfe24
|
1148 |
*/ |
503c358cf
|
1149 |
long prune_dcache_sb(struct super_block *sb, struct shrink_control *sc) |
3049cfe24
|
1150 |
{ |
f60415675
|
1151 1152 |
LIST_HEAD(dispose); long freed; |
3049cfe24
|
1153 |
|
503c358cf
|
1154 1155 |
freed = list_lru_shrink_walk(&sb->s_dentry_lru, sc, dentry_lru_isolate, &dispose); |
f60415675
|
1156 |
shrink_dentry_list(&dispose); |
0a234c6dc
|
1157 |
return freed; |
da3bbdd46
|
1158 |
} |
230445078
|
1159 |
|
4e717f5c1
|
1160 |
static enum lru_status dentry_lru_isolate_shrink(struct list_head *item, |
3f97b1632
|
1161 |
struct list_lru_one *lru, spinlock_t *lru_lock, void *arg) |
dd1f6b2e4
|
1162 |
{ |
4e717f5c1
|
1163 1164 |
struct list_head *freeable = arg; struct dentry *dentry = container_of(item, struct dentry, d_lru); |
dd1f6b2e4
|
1165 |
|
4e717f5c1
|
1166 1167 1168 1169 1170 1171 1172 |
/* * we are inverting the lru lock/dentry->d_lock here, * so use a trylock. If we fail to get the lock, just skip * it */ if (!spin_trylock(&dentry->d_lock)) return LRU_SKIP; |
3f97b1632
|
1173 |
d_lru_shrink_move(lru, dentry, freeable); |
4e717f5c1
|
1174 |
spin_unlock(&dentry->d_lock); |
ec33679d7
|
1175 |
|
4e717f5c1
|
1176 |
return LRU_REMOVED; |
da3bbdd46
|
1177 |
} |
4e717f5c1
|
1178 |
|
da3bbdd46
|
1179 |
/** |
1da177e4c
|
1180 1181 1182 |
* shrink_dcache_sb - shrink dcache for a superblock * @sb: superblock * |
3049cfe24
|
1183 1184 |
* Shrink the dcache for the specified super block. This is used to free * the dcache before unmounting a file system. |
1da177e4c
|
1185 |
*/ |
3049cfe24
|
1186 |
void shrink_dcache_sb(struct super_block *sb) |
1da177e4c
|
1187 |
{ |
4e717f5c1
|
1188 1189 |
do { LIST_HEAD(dispose); |
1dbd449c9
|
1190 |
list_lru_walk(&sb->s_dentry_lru, |
b17c070fb
|
1191 |
dentry_lru_isolate_shrink, &dispose, 1024); |
4e717f5c1
|
1192 |
shrink_dentry_list(&dispose); |
b17c070fb
|
1193 |
} while (list_lru_count(&sb->s_dentry_lru) > 0); |
1da177e4c
|
1194 |
} |
ec4f86059
|
1195 |
EXPORT_SYMBOL(shrink_dcache_sb); |
1da177e4c
|
1196 |
|
db14fc3ab
|
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 |
/** * enum d_walk_ret - action to talke during tree walk * @D_WALK_CONTINUE: contrinue walk * @D_WALK_QUIT: quit walk * @D_WALK_NORETRY: quit when retry is needed * @D_WALK_SKIP: skip this dentry and its children */ enum d_walk_ret { D_WALK_CONTINUE, D_WALK_QUIT, D_WALK_NORETRY, D_WALK_SKIP, }; |
c826cb7df
|
1210 |
|
1da177e4c
|
1211 |
/** |
db14fc3ab
|
1212 1213 1214 1215 |
* d_walk - walk the dentry tree * @parent: start of walk * @data: data passed to @enter() and @finish() * @enter: callback when first entering the dentry |
1da177e4c
|
1216 |
* |
3a8e3611e
|
1217 |
* The @enter() callbacks are called with d_lock held. |
1da177e4c
|
1218 |
*/ |
db14fc3ab
|
1219 |
static void d_walk(struct dentry *parent, void *data, |
3a8e3611e
|
1220 |
enum d_walk_ret (*enter)(void *, struct dentry *)) |
1da177e4c
|
1221 |
{ |
949854d02
|
1222 |
struct dentry *this_parent; |
1da177e4c
|
1223 |
struct list_head *next; |
48f5ec21d
|
1224 |
unsigned seq = 0; |
db14fc3ab
|
1225 1226 |
enum d_walk_ret ret; bool retry = true; |
949854d02
|
1227 |
|
58db63d08
|
1228 |
again: |
48f5ec21d
|
1229 |
read_seqbegin_or_lock(&rename_lock, &seq); |
58db63d08
|
1230 |
this_parent = parent; |
2fd6b7f50
|
1231 |
spin_lock(&this_parent->d_lock); |
db14fc3ab
|
1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 |
ret = enter(data, this_parent); switch (ret) { case D_WALK_CONTINUE: break; case D_WALK_QUIT: case D_WALK_SKIP: goto out_unlock; case D_WALK_NORETRY: retry = false; break; } |
1da177e4c
|
1244 1245 1246 1247 1248 |
repeat: next = this_parent->d_subdirs.next; resume: while (next != &this_parent->d_subdirs) { struct list_head *tmp = next; |
946e51f2b
|
1249 |
struct dentry *dentry = list_entry(tmp, struct dentry, d_child); |
1da177e4c
|
1250 |
next = tmp->next; |
2fd6b7f50
|
1251 |
|
ba65dc5ef
|
1252 1253 |
if (unlikely(dentry->d_flags & DCACHE_DENTRY_CURSOR)) continue; |
2fd6b7f50
|
1254 |
spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
db14fc3ab
|
1255 1256 1257 1258 1259 1260 |
ret = enter(data, dentry); switch (ret) { case D_WALK_CONTINUE: break; case D_WALK_QUIT: |
2fd6b7f50
|
1261 |
spin_unlock(&dentry->d_lock); |
db14fc3ab
|
1262 1263 1264 1265 1266 1267 1268 |
goto out_unlock; case D_WALK_NORETRY: retry = false; break; case D_WALK_SKIP: spin_unlock(&dentry->d_lock); continue; |
2fd6b7f50
|
1269 |
} |
db14fc3ab
|
1270 |
|
1da177e4c
|
1271 |
if (!list_empty(&dentry->d_subdirs)) { |
2fd6b7f50
|
1272 |
spin_unlock(&this_parent->d_lock); |
5facae4f3
|
1273 |
spin_release(&dentry->d_lock.dep_map, _RET_IP_); |
1da177e4c
|
1274 |
this_parent = dentry; |
2fd6b7f50
|
1275 |
spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); |
1da177e4c
|
1276 1277 |
goto repeat; } |
2fd6b7f50
|
1278 |
spin_unlock(&dentry->d_lock); |
1da177e4c
|
1279 1280 1281 1282 |
} /* * All done at this level ... ascend and resume the search. */ |
ca5358ef7
|
1283 1284 |
rcu_read_lock(); ascend: |
1da177e4c
|
1285 |
if (this_parent != parent) { |
c826cb7df
|
1286 |
struct dentry *child = this_parent; |
31dec1327
|
1287 |
this_parent = child->d_parent; |
31dec1327
|
1288 1289 |
spin_unlock(&child->d_lock); spin_lock(&this_parent->d_lock); |
ca5358ef7
|
1290 1291 |
/* might go back up the wrong parent if we have had a rename. */ if (need_seqretry(&rename_lock, seq)) |
949854d02
|
1292 |
goto rename_retry; |
2159184ea
|
1293 1294 1295 |
/* go into the first sibling still alive */ do { next = child->d_child.next; |
ca5358ef7
|
1296 1297 1298 |
if (next == &this_parent->d_subdirs) goto ascend; child = list_entry(next, struct dentry, d_child); |
2159184ea
|
1299 |
} while (unlikely(child->d_flags & DCACHE_DENTRY_KILLED)); |
31dec1327
|
1300 |
rcu_read_unlock(); |
1da177e4c
|
1301 1302 |
goto resume; } |
ca5358ef7
|
1303 |
if (need_seqretry(&rename_lock, seq)) |
949854d02
|
1304 |
goto rename_retry; |
ca5358ef7
|
1305 |
rcu_read_unlock(); |
db14fc3ab
|
1306 1307 1308 |
out_unlock: spin_unlock(&this_parent->d_lock); |
48f5ec21d
|
1309 |
done_seqretry(&rename_lock, seq); |
db14fc3ab
|
1310 |
return; |
58db63d08
|
1311 1312 |
rename_retry: |
ca5358ef7
|
1313 1314 1315 |
spin_unlock(&this_parent->d_lock); rcu_read_unlock(); BUG_ON(seq & 1); |
db14fc3ab
|
1316 1317 |
if (!retry) return; |
48f5ec21d
|
1318 |
seq = 1; |
58db63d08
|
1319 |
goto again; |
1da177e4c
|
1320 |
} |
db14fc3ab
|
1321 |
|
01619491a
|
1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 |
struct check_mount { struct vfsmount *mnt; unsigned int mounted; }; static enum d_walk_ret path_check_mount(void *data, struct dentry *dentry) { struct check_mount *info = data; struct path path = { .mnt = info->mnt, .dentry = dentry }; if (likely(!d_mountpoint(dentry))) return D_WALK_CONTINUE; if (__path_is_mountpoint(&path)) { info->mounted = 1; return D_WALK_QUIT; } return D_WALK_CONTINUE; } /** * path_has_submounts - check for mounts over a dentry in the * current namespace. * @parent: path to check. * * Return true if the parent or its subdirectories contain * a mount point in the current namespace. */ int path_has_submounts(const struct path *parent) { struct check_mount data = { .mnt = parent->mnt, .mounted = 0 }; read_seqlock_excl(&mount_lock); |
3a8e3611e
|
1354 |
d_walk(parent->dentry, &data, path_check_mount); |
01619491a
|
1355 1356 1357 1358 1359 |
read_sequnlock_excl(&mount_lock); return data.mounted; } EXPORT_SYMBOL(path_has_submounts); |
1da177e4c
|
1360 |
/* |
eed810076
|
1361 1362 1363 1364 |
* Called by mount code to set a mountpoint and check if the mountpoint is * reachable (e.g. NFS can unhash a directory dentry and then the complete * subtree can become unreachable). * |
1ffe46d11
|
1365 |
* Only one of d_invalidate() and d_set_mounted() must succeed. For |
eed810076
|
1366 1367 1368 1369 1370 1371 1372 1373 |
* this reason take rename_lock and d_lock on dentry and ancestors. */ int d_set_mounted(struct dentry *dentry) { struct dentry *p; int ret = -ENOENT; write_seqlock(&rename_lock); for (p = dentry->d_parent; !IS_ROOT(p); p = p->d_parent) { |
1ffe46d11
|
1374 |
/* Need exclusion wrt. d_invalidate() */ |
eed810076
|
1375 1376 1377 1378 1379 1380 1381 1382 1383 |
spin_lock(&p->d_lock); if (unlikely(d_unhashed(p))) { spin_unlock(&p->d_lock); goto out; } spin_unlock(&p->d_lock); } spin_lock(&dentry->d_lock); if (!d_unlinked(dentry)) { |
3895dbf89
|
1384 1385 1386 1387 1388 |
ret = -EBUSY; if (!d_mountpoint(dentry)) { dentry->d_flags |= DCACHE_MOUNTED; ret = 0; } |
eed810076
|
1389 1390 1391 1392 1393 1394 1395 1396 |
} spin_unlock(&dentry->d_lock); out: write_sequnlock(&rename_lock); return ret; } /* |
fd5179094
|
1397 |
* Search the dentry child list of the specified parent, |
1da177e4c
|
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 |
* and move any unused dentries to the end of the unused * list for prune_dcache(). We descend to the next level * whenever the d_subdirs list is non-empty and continue * searching. * * It returns zero iff there are no unused children, * otherwise it returns the number of children moved to * the end of the unused list. This may not be the total * number of unused children, because select_parent can * drop the lock and return early due to latency * constraints. */ |
1da177e4c
|
1410 |
|
db14fc3ab
|
1411 1412 |
struct select_data { struct dentry *start; |
9bdebc2bd
|
1413 1414 1415 1416 |
union { long found; struct dentry *victim; }; |
db14fc3ab
|
1417 |
struct list_head dispose; |
db14fc3ab
|
1418 |
}; |
230445078
|
1419 |
|
db14fc3ab
|
1420 1421 1422 1423 |
static enum d_walk_ret select_collect(void *_data, struct dentry *dentry) { struct select_data *data = _data; enum d_walk_ret ret = D_WALK_CONTINUE; |
1da177e4c
|
1424 |
|
db14fc3ab
|
1425 1426 |
if (data->start == dentry) goto out; |
2fd6b7f50
|
1427 |
|
fe91522a7
|
1428 |
if (dentry->d_flags & DCACHE_SHRINK_LIST) { |
db14fc3ab
|
1429 |
data->found++; |
fe91522a7
|
1430 1431 1432 1433 1434 1435 1436 |
} else { if (dentry->d_flags & DCACHE_LRU_LIST) d_lru_del(dentry); if (!dentry->d_lockref.count) { d_shrink_add(dentry, &data->dispose); data->found++; } |
1da177e4c
|
1437 |
} |
db14fc3ab
|
1438 1439 1440 1441 1442 |
/* * We can return to the caller if we have found some (this * ensures forward progress). We'll be coming back to find * the rest. */ |
fe91522a7
|
1443 1444 |
if (!list_empty(&data->dispose)) ret = need_resched() ? D_WALK_QUIT : D_WALK_NORETRY; |
1da177e4c
|
1445 |
out: |
db14fc3ab
|
1446 |
return ret; |
1da177e4c
|
1447 |
} |
9bdebc2bd
|
1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 |
static enum d_walk_ret select_collect2(void *_data, struct dentry *dentry) { struct select_data *data = _data; enum d_walk_ret ret = D_WALK_CONTINUE; if (data->start == dentry) goto out; if (dentry->d_flags & DCACHE_SHRINK_LIST) { if (!dentry->d_lockref.count) { rcu_read_lock(); data->victim = dentry; return D_WALK_QUIT; } } else { if (dentry->d_flags & DCACHE_LRU_LIST) d_lru_del(dentry); if (!dentry->d_lockref.count) d_shrink_add(dentry, &data->dispose); } /* * We can return to the caller if we have found some (this * ensures forward progress). We'll be coming back to find * the rest. */ if (!list_empty(&data->dispose)) ret = need_resched() ? D_WALK_QUIT : D_WALK_NORETRY; out: return ret; } |
1da177e4c
|
1478 1479 1480 1481 1482 1483 |
/** * shrink_dcache_parent - prune dcache * @parent: parent of entries to prune * * Prune the dcache to remove unused children of the parent dentry. */ |
db14fc3ab
|
1484 |
void shrink_dcache_parent(struct dentry *parent) |
1da177e4c
|
1485 |
{ |
db14fc3ab
|
1486 |
for (;;) { |
9bdebc2bd
|
1487 |
struct select_data data = {.start = parent}; |
1da177e4c
|
1488 |
|
db14fc3ab
|
1489 |
INIT_LIST_HEAD(&data.dispose); |
3a8e3611e
|
1490 |
d_walk(parent, &data, select_collect); |
4fb488714
|
1491 1492 1493 1494 1495 1496 1497 |
if (!list_empty(&data.dispose)) { shrink_dentry_list(&data.dispose); continue; } cond_resched(); |
db14fc3ab
|
1498 1499 |
if (!data.found) break; |
9bdebc2bd
|
1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 |
data.victim = NULL; d_walk(parent, &data, select_collect2); if (data.victim) { struct dentry *parent; spin_lock(&data.victim->d_lock); if (!shrink_lock_dentry(data.victim)) { spin_unlock(&data.victim->d_lock); rcu_read_unlock(); } else { rcu_read_unlock(); parent = data.victim->d_parent; if (parent != data.victim) __dput_to_list(parent, &data.dispose); __dentry_kill(data.victim); } } if (!list_empty(&data.dispose)) shrink_dentry_list(&data.dispose); |
421348f1c
|
1518 |
} |
1da177e4c
|
1519 |
} |
ec4f86059
|
1520 |
EXPORT_SYMBOL(shrink_dcache_parent); |
1da177e4c
|
1521 |
|
9c8c10e26
|
1522 |
static enum d_walk_ret umount_check(void *_data, struct dentry *dentry) |
42c326082
|
1523 |
{ |
9c8c10e26
|
1524 1525 1526 |
/* it has busy descendents; complain about those instead */ if (!list_empty(&dentry->d_subdirs)) return D_WALK_CONTINUE; |
42c326082
|
1527 |
|
9c8c10e26
|
1528 1529 1530 1531 1532 1533 1534 |
/* root with refcount 1 is fine */ if (dentry == _data && dentry->d_lockref.count == 1) return D_WALK_CONTINUE; printk(KERN_ERR "BUG: Dentry %p{i=%lx,n=%pd} " " still in use (%d) [unmount of %s %s] ", |
42c326082
|
1535 1536 1537 |
dentry, dentry->d_inode ? dentry->d_inode->i_ino : 0UL, |
9c8c10e26
|
1538 |
dentry, |
42c326082
|
1539 1540 1541 |
dentry->d_lockref.count, dentry->d_sb->s_type->name, dentry->d_sb->s_id); |
9c8c10e26
|
1542 1543 1544 1545 1546 1547 1548 |
WARN_ON(1); return D_WALK_CONTINUE; } static void do_one_tree(struct dentry *dentry) { shrink_dcache_parent(dentry); |
3a8e3611e
|
1549 |
d_walk(dentry, dentry, umount_check); |
9c8c10e26
|
1550 1551 |
d_drop(dentry); dput(dentry); |
42c326082
|
1552 1553 1554 1555 1556 1557 1558 1559 |
} /* * destroy the dentries attached to a superblock on unmounting */ void shrink_dcache_for_umount(struct super_block *sb) { struct dentry *dentry; |
9c8c10e26
|
1560 |
WARN(down_read_trylock(&sb->s_umount), "s_umount should've been locked"); |
42c326082
|
1561 1562 1563 |
dentry = sb->s_root; sb->s_root = NULL; |
9c8c10e26
|
1564 |
do_one_tree(dentry); |
42c326082
|
1565 |
|
f1ee61621
|
1566 1567 |
while (!hlist_bl_empty(&sb->s_roots)) { dentry = dget(hlist_bl_entry(hlist_bl_first(&sb->s_roots), struct dentry, d_hash)); |
9c8c10e26
|
1568 |
do_one_tree(dentry); |
42c326082
|
1569 1570 |
} } |
ff17fa561
|
1571 |
static enum d_walk_ret find_submount(void *_data, struct dentry *dentry) |
848ac114e
|
1572 |
{ |
ff17fa561
|
1573 |
struct dentry **victim = _data; |
848ac114e
|
1574 |
if (d_mountpoint(dentry)) { |
8ed936b56
|
1575 |
__dget_dlock(dentry); |
ff17fa561
|
1576 |
*victim = dentry; |
848ac114e
|
1577 1578 |
return D_WALK_QUIT; } |
ff17fa561
|
1579 |
return D_WALK_CONTINUE; |
848ac114e
|
1580 1581 1582 |
} /** |
1ffe46d11
|
1583 1584 |
* d_invalidate - detach submounts, prune dcache, and drop * @dentry: dentry to invalidate (aka detach, prune and drop) |
848ac114e
|
1585 |
*/ |
5542aa2fa
|
1586 |
void d_invalidate(struct dentry *dentry) |
848ac114e
|
1587 |
{ |
ff17fa561
|
1588 |
bool had_submounts = false; |
1ffe46d11
|
1589 1590 1591 |
spin_lock(&dentry->d_lock); if (d_unhashed(dentry)) { spin_unlock(&dentry->d_lock); |
5542aa2fa
|
1592 |
return; |
1ffe46d11
|
1593 |
} |
ff17fa561
|
1594 |
__d_drop(dentry); |
1ffe46d11
|
1595 |
spin_unlock(&dentry->d_lock); |
848ac114e
|
1596 |
/* Negative dentries can be dropped without further checks */ |
ff17fa561
|
1597 |
if (!dentry->d_inode) |
5542aa2fa
|
1598 |
return; |
848ac114e
|
1599 |
|
ff17fa561
|
1600 |
shrink_dcache_parent(dentry); |
848ac114e
|
1601 |
for (;;) { |
ff17fa561
|
1602 |
struct dentry *victim = NULL; |
3a8e3611e
|
1603 |
d_walk(dentry, &victim, find_submount); |
ff17fa561
|
1604 1605 1606 |
if (!victim) { if (had_submounts) shrink_dcache_parent(dentry); |
81be24d26
|
1607 |
return; |
8ed936b56
|
1608 |
} |
ff17fa561
|
1609 1610 1611 |
had_submounts = true; detach_mounts(victim); dput(victim); |
848ac114e
|
1612 |
} |
848ac114e
|
1613 |
} |
1ffe46d11
|
1614 |
EXPORT_SYMBOL(d_invalidate); |
848ac114e
|
1615 |
|
1da177e4c
|
1616 |
/** |
a4464dbc0
|
1617 1618 |
* __d_alloc - allocate a dcache entry * @sb: filesystem it will belong to |
1da177e4c
|
1619 1620 1621 1622 1623 1624 1625 |
* @name: qstr of the name * * Allocates a dentry. It returns %NULL if there is insufficient memory * available. On a success the dentry is returned. The name passed in is * copied and the copy passed in may be reused after this call. */ |
5c8b0dfc6
|
1626 |
static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) |
1da177e4c
|
1627 1628 1629 |
{ struct dentry *dentry; char *dname; |
285b102d3
|
1630 |
int err; |
1da177e4c
|
1631 |
|
e12ba74d8
|
1632 |
dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL); |
1da177e4c
|
1633 1634 |
if (!dentry) return NULL; |
6326c71fd
|
1635 1636 1637 1638 1639 1640 1641 |
/* * We guarantee that the inline name is always NUL-terminated. * This way the memcpy() done by the name switching in rename * will still always have a NUL at the end, even if we might * be overwriting an internal NUL character */ dentry->d_iname[DNAME_INLINE_LEN-1] = 0; |
798434bda
|
1642 |
if (unlikely(!name)) { |
cdf01226b
|
1643 |
name = &slash_name; |
798434bda
|
1644 1645 |
dname = dentry->d_iname; } else if (name->len > DNAME_INLINE_LEN-1) { |
8d85b4845
|
1646 |
size_t size = offsetof(struct external_name, name[1]); |
2e03b4bc4
|
1647 1648 1649 1650 |
struct external_name *p = kmalloc(size + name->len, GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE); if (!p) { |
1da177e4c
|
1651 1652 1653 |
kmem_cache_free(dentry_cache, dentry); return NULL; } |
2e03b4bc4
|
1654 1655 |
atomic_set(&p->u.count, 1); dname = p->name; |
1da177e4c
|
1656 1657 1658 |
} else { dname = dentry->d_iname; } |
1da177e4c
|
1659 1660 1661 1662 1663 |
dentry->d_name.len = name->len; dentry->d_name.hash = name->hash; memcpy(dname, name->name, name->len); dname[name->len] = 0; |
6326c71fd
|
1664 |
/* Make sure we always see the terminating NUL character */ |
7088efa91
|
1665 |
smp_store_release(&dentry->d_name.name, dname); /* ^^^ */ |
6326c71fd
|
1666 |
|
98474236f
|
1667 |
dentry->d_lockref.count = 1; |
dea3667bc
|
1668 |
dentry->d_flags = 0; |
1da177e4c
|
1669 |
spin_lock_init(&dentry->d_lock); |
31e6b01f4
|
1670 |
seqcount_init(&dentry->d_seq); |
1da177e4c
|
1671 |
dentry->d_inode = NULL; |
a4464dbc0
|
1672 1673 |
dentry->d_parent = dentry; dentry->d_sb = sb; |
1da177e4c
|
1674 1675 |
dentry->d_op = NULL; dentry->d_fsdata = NULL; |
ceb5bdc2d
|
1676 |
INIT_HLIST_BL_NODE(&dentry->d_hash); |
1da177e4c
|
1677 1678 |
INIT_LIST_HEAD(&dentry->d_lru); INIT_LIST_HEAD(&dentry->d_subdirs); |
946e51f2b
|
1679 1680 |
INIT_HLIST_NODE(&dentry->d_u.d_alias); INIT_LIST_HEAD(&dentry->d_child); |
a4464dbc0
|
1681 |
d_set_d_op(dentry, dentry->d_sb->s_d_op); |
1da177e4c
|
1682 |
|
285b102d3
|
1683 1684 1685 1686 1687 1688 1689 1690 1691 |
if (dentry->d_op && dentry->d_op->d_init) { err = dentry->d_op->d_init(dentry); if (err) { if (dname_external(dentry)) kfree(external_name(dentry)); kmem_cache_free(dentry_cache, dentry); return NULL; } } |
3e880fb5e
|
1692 |
this_cpu_inc(nr_dentry); |
312d3ca85
|
1693 |
|
1da177e4c
|
1694 1695 |
return dentry; } |
a4464dbc0
|
1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 |
/** * d_alloc - allocate a dcache entry * @parent: parent of entry to allocate * @name: qstr of the name * * Allocates a dentry. It returns %NULL if there is insufficient memory * available. On a success the dentry is returned. The name passed in is * copied and the copy passed in may be reused after this call. */ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) { struct dentry *dentry = __d_alloc(parent->d_sb, name); if (!dentry) return NULL; |
a4464dbc0
|
1711 1712 1713 1714 1715 1716 1717 |
spin_lock(&parent->d_lock); /* * don't need child lock because it is not subject * to concurrency here */ __dget_dlock(parent); dentry->d_parent = parent; |
946e51f2b
|
1718 |
list_add(&dentry->d_child, &parent->d_subdirs); |
a4464dbc0
|
1719 1720 1721 1722 |
spin_unlock(&parent->d_lock); return dentry; } |
ec4f86059
|
1723 |
EXPORT_SYMBOL(d_alloc); |
1da177e4c
|
1724 |
|
f9c34674b
|
1725 1726 1727 1728 1729 |
struct dentry *d_alloc_anon(struct super_block *sb) { return __d_alloc(sb, NULL); } EXPORT_SYMBOL(d_alloc_anon); |
ba65dc5ef
|
1730 1731 |
struct dentry *d_alloc_cursor(struct dentry * parent) { |
f9c34674b
|
1732 |
struct dentry *dentry = d_alloc_anon(parent->d_sb); |
ba65dc5ef
|
1733 |
if (dentry) { |
5467a68cb
|
1734 |
dentry->d_flags |= DCACHE_DENTRY_CURSOR; |
ba65dc5ef
|
1735 1736 1737 1738 |
dentry->d_parent = dget(parent); } return dentry; } |
e1a24bb0a
|
1739 1740 1741 1742 1743 1744 1745 |
/** * d_alloc_pseudo - allocate a dentry (for lookup-less filesystems) * @sb: the superblock * @name: qstr of the name * * For a filesystem that just pins its dentries in memory and never * performs lookups at all, return an unhashed IS_ROOT dentry. |
5467a68cb
|
1746 1747 1748 1749 |
* This is used for pipes, sockets et.al. - the stuff that should * never be anyone's children or parents. Unlike all other * dentries, these will not have RCU delay between dropping the * last reference and freeing them. |
ab1152dd5
|
1750 1751 1752 |
* * The only user is alloc_file_pseudo() and that's what should * be considered a public interface. Don't use directly. |
e1a24bb0a
|
1753 |
*/ |
4b936885a
|
1754 1755 |
struct dentry *d_alloc_pseudo(struct super_block *sb, const struct qstr *name) { |
5467a68cb
|
1756 1757 1758 1759 |
struct dentry *dentry = __d_alloc(sb, name); if (likely(dentry)) dentry->d_flags |= DCACHE_NORCU; return dentry; |
4b936885a
|
1760 |
} |
4b936885a
|
1761 |
|
1da177e4c
|
1762 1763 1764 1765 1766 |
struct dentry *d_alloc_name(struct dentry *parent, const char *name) { struct qstr q; q.name = name; |
8387ff257
|
1767 |
q.hash_len = hashlen_string(parent, name); |
1da177e4c
|
1768 1769 |
return d_alloc(parent, &q); } |
ef26ca97e
|
1770 |
EXPORT_SYMBOL(d_alloc_name); |
1da177e4c
|
1771 |
|
fb045adb9
|
1772 1773 |
void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op) { |
6f7f7caab
|
1774 1775 |
WARN_ON_ONCE(dentry->d_op); WARN_ON_ONCE(dentry->d_flags & (DCACHE_OP_HASH | |
fb045adb9
|
1776 1777 |
DCACHE_OP_COMPARE | DCACHE_OP_REVALIDATE | |
ecf3d1f1a
|
1778 |
DCACHE_OP_WEAK_REVALIDATE | |
4bacc9c92
|
1779 |
DCACHE_OP_DELETE | |
d101a1259
|
1780 |
DCACHE_OP_REAL)); |
fb045adb9
|
1781 1782 1783 1784 1785 1786 1787 1788 1789 |
dentry->d_op = op; if (!op) return; if (op->d_hash) dentry->d_flags |= DCACHE_OP_HASH; if (op->d_compare) dentry->d_flags |= DCACHE_OP_COMPARE; if (op->d_revalidate) dentry->d_flags |= DCACHE_OP_REVALIDATE; |
ecf3d1f1a
|
1790 1791 |
if (op->d_weak_revalidate) dentry->d_flags |= DCACHE_OP_WEAK_REVALIDATE; |
fb045adb9
|
1792 1793 |
if (op->d_delete) dentry->d_flags |= DCACHE_OP_DELETE; |
f0023bc61
|
1794 1795 |
if (op->d_prune) dentry->d_flags |= DCACHE_OP_PRUNE; |
d101a1259
|
1796 1797 |
if (op->d_real) dentry->d_flags |= DCACHE_OP_REAL; |
fb045adb9
|
1798 1799 1800 |
} EXPORT_SYMBOL(d_set_d_op); |
df1a085af
|
1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 |
/* * d_set_fallthru - Mark a dentry as falling through to a lower layer * @dentry - The dentry to mark * * Mark a dentry as falling through to the lower layer (as set with * d_pin_lower()). This flag may be recorded on the medium. */ void d_set_fallthru(struct dentry *dentry) { spin_lock(&dentry->d_lock); dentry->d_flags |= DCACHE_FALLTHRU; spin_unlock(&dentry->d_lock); } EXPORT_SYMBOL(d_set_fallthru); |
b18825a7c
|
1816 1817 |
static unsigned d_flags_for_inode(struct inode *inode) { |
44bdb5e5f
|
1818 |
unsigned add_flags = DCACHE_REGULAR_TYPE; |
b18825a7c
|
1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 |
if (!inode) return DCACHE_MISS_TYPE; if (S_ISDIR(inode->i_mode)) { add_flags = DCACHE_DIRECTORY_TYPE; if (unlikely(!(inode->i_opflags & IOP_LOOKUP))) { if (unlikely(!inode->i_op->lookup)) add_flags = DCACHE_AUTODIR_TYPE; else inode->i_opflags |= IOP_LOOKUP; } |
44bdb5e5f
|
1831 1832 1833 1834 |
goto type_determined; } if (unlikely(!(inode->i_opflags & IOP_NOFOLLOW))) { |
6b2553918
|
1835 |
if (unlikely(inode->i_op->get_link)) { |
b18825a7c
|
1836 |
add_flags = DCACHE_SYMLINK_TYPE; |
44bdb5e5f
|
1837 1838 1839 |
goto type_determined; } inode->i_opflags |= IOP_NOFOLLOW; |
b18825a7c
|
1840 |
} |
44bdb5e5f
|
1841 1842 1843 1844 |
if (unlikely(!S_ISREG(inode->i_mode))) add_flags = DCACHE_SPECIAL_TYPE; type_determined: |
b18825a7c
|
1845 1846 1847 1848 |
if (unlikely(IS_AUTOMOUNT(inode))) add_flags |= DCACHE_NEED_AUTOMOUNT; return add_flags; } |
360da9002
|
1849 1850 |
static void __d_instantiate(struct dentry *dentry, struct inode *inode) { |
b18825a7c
|
1851 |
unsigned add_flags = d_flags_for_inode(inode); |
85c7f8104
|
1852 |
WARN_ON(d_in_lookup(dentry)); |
b18825a7c
|
1853 |
|
b23fb0a60
|
1854 |
spin_lock(&dentry->d_lock); |
af0c9af1b
|
1855 1856 1857 1858 1859 |
/* * Decrement negative dentry count if it was in the LRU list. */ if (dentry->d_flags & DCACHE_LRU_LIST) this_cpu_dec(nr_dentry_negative); |
de689f5e3
|
1860 |
hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); |
a528aca7f
|
1861 |
raw_write_seqcount_begin(&dentry->d_seq); |
4bf46a272
|
1862 |
__d_set_inode_and_type(dentry, inode, add_flags); |
a528aca7f
|
1863 |
raw_write_seqcount_end(&dentry->d_seq); |
affda4841
|
1864 |
fsnotify_update_flags(dentry); |
b23fb0a60
|
1865 |
spin_unlock(&dentry->d_lock); |
360da9002
|
1866 |
} |
1da177e4c
|
1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 |
/** * d_instantiate - fill in inode information for a dentry * @entry: dentry to complete * @inode: inode to attach to this dentry * * Fill in inode information in the entry. * * This turns negative dentries into productive full members * of society. * * NOTE! This assumes that the inode count has been incremented * (or otherwise set) by the caller to indicate that it is now * in use by the dcache. */ void d_instantiate(struct dentry *entry, struct inode * inode) { |
946e51f2b
|
1884 |
BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); |
de689f5e3
|
1885 |
if (inode) { |
b96809173
|
1886 |
security_d_instantiate(entry, inode); |
873feea09
|
1887 |
spin_lock(&inode->i_lock); |
de689f5e3
|
1888 |
__d_instantiate(entry, inode); |
873feea09
|
1889 |
spin_unlock(&inode->i_lock); |
de689f5e3
|
1890 |
} |
1da177e4c
|
1891 |
} |
ec4f86059
|
1892 |
EXPORT_SYMBOL(d_instantiate); |
1da177e4c
|
1893 |
|
1e2e547a9
|
1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 |
/* * This should be equivalent to d_instantiate() + unlock_new_inode(), * with lockdep-related part of unlock_new_inode() done before * anything else. Use that instead of open-coding d_instantiate()/ * unlock_new_inode() combinations. */ void d_instantiate_new(struct dentry *entry, struct inode *inode) { BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); BUG_ON(!inode); lockdep_annotate_inode_mutex_key(inode); security_d_instantiate(entry, inode); spin_lock(&inode->i_lock); __d_instantiate(entry, inode); WARN_ON(!(inode->i_state & I_NEW)); |
c2b6d621c
|
1909 |
inode->i_state &= ~I_NEW & ~I_CREATING; |
1e2e547a9
|
1910 1911 1912 1913 1914 |
smp_mb(); wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); } EXPORT_SYMBOL(d_instantiate_new); |
adc0e91ab
|
1915 1916 1917 1918 1919 |
struct dentry *d_make_root(struct inode *root_inode) { struct dentry *res = NULL; if (root_inode) { |
f9c34674b
|
1920 |
res = d_alloc_anon(root_inode->i_sb); |
5467a68cb
|
1921 |
if (res) |
adc0e91ab
|
1922 |
d_instantiate(res, root_inode); |
5467a68cb
|
1923 |
else |
adc0e91ab
|
1924 1925 1926 1927 1928 |
iput(root_inode); } return res; } EXPORT_SYMBOL(d_make_root); |
f9c34674b
|
1929 1930 1931 |
static struct dentry *__d_instantiate_anon(struct dentry *dentry, struct inode *inode, bool disconnected) |
4ea3ada29
|
1932 |
{ |
9308a6128
|
1933 |
struct dentry *res; |
b18825a7c
|
1934 |
unsigned add_flags; |
4ea3ada29
|
1935 |
|
f9c34674b
|
1936 |
security_d_instantiate(dentry, inode); |
873feea09
|
1937 |
spin_lock(&inode->i_lock); |
d891eedbc
|
1938 |
res = __d_find_any_alias(inode); |
9308a6128
|
1939 |
if (res) { |
873feea09
|
1940 |
spin_unlock(&inode->i_lock); |
f9c34674b
|
1941 |
dput(dentry); |
9308a6128
|
1942 1943 1944 1945 |
goto out_iput; } /* attach a disconnected dentry */ |
1a0a397e4
|
1946 1947 1948 1949 |
add_flags = d_flags_for_inode(inode); if (disconnected) add_flags |= DCACHE_DISCONNECTED; |
b18825a7c
|
1950 |
|
f9c34674b
|
1951 1952 1953 |
spin_lock(&dentry->d_lock); __d_set_inode_and_type(dentry, inode, add_flags); hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); |
f1ee61621
|
1954 |
if (!disconnected) { |
139351f1f
|
1955 1956 1957 |
hlist_bl_lock(&dentry->d_sb->s_roots); hlist_bl_add_head(&dentry->d_hash, &dentry->d_sb->s_roots); hlist_bl_unlock(&dentry->d_sb->s_roots); |
f1ee61621
|
1958 |
} |
f9c34674b
|
1959 |
spin_unlock(&dentry->d_lock); |
873feea09
|
1960 |
spin_unlock(&inode->i_lock); |
9308a6128
|
1961 |
|
f9c34674b
|
1962 |
return dentry; |
9308a6128
|
1963 1964 1965 1966 |
out_iput: iput(inode); return res; |
4ea3ada29
|
1967 |
} |
1a0a397e4
|
1968 |
|
f9c34674b
|
1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 |
struct dentry *d_instantiate_anon(struct dentry *dentry, struct inode *inode) { return __d_instantiate_anon(dentry, inode, true); } EXPORT_SYMBOL(d_instantiate_anon); static struct dentry *__d_obtain_alias(struct inode *inode, bool disconnected) { struct dentry *tmp; struct dentry *res; if (!inode) return ERR_PTR(-ESTALE); if (IS_ERR(inode)) return ERR_CAST(inode); res = d_find_any_alias(inode); if (res) goto out_iput; tmp = d_alloc_anon(inode->i_sb); if (!tmp) { res = ERR_PTR(-ENOMEM); goto out_iput; } return __d_instantiate_anon(tmp, inode, disconnected); out_iput: iput(inode); return res; } |
1a0a397e4
|
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 |
/** * d_obtain_alias - find or allocate a DISCONNECTED dentry for a given inode * @inode: inode to allocate the dentry for * * Obtain a dentry for an inode resulting from NFS filehandle conversion or * similar open by handle operations. The returned dentry may be anonymous, * or may have a full name (if the inode was already in the cache). * * When called on a directory inode, we must ensure that the inode only ever * has one dentry. If a dentry is found, that is returned instead of * allocating a new one. * * On successful return, the reference to the inode has been transferred * to the dentry. In case of an error the reference on the inode is released. * To make it easier to use in export operations a %NULL or IS_ERR inode may * be passed in and the error will be propagated to the return value, * with a %NULL @inode replaced by ERR_PTR(-ESTALE). */ struct dentry *d_obtain_alias(struct inode *inode) { |
f9c34674b
|
2021 |
return __d_obtain_alias(inode, true); |
1a0a397e4
|
2022 |
} |
adc487204
|
2023 |
EXPORT_SYMBOL(d_obtain_alias); |
1da177e4c
|
2024 2025 |
/** |
1a0a397e4
|
2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 |
* d_obtain_root - find or allocate a dentry for a given inode * @inode: inode to allocate the dentry for * * Obtain an IS_ROOT dentry for the root of a filesystem. * * We must ensure that directory inodes only ever have one dentry. If a * dentry is found, that is returned instead of allocating a new one. * * On successful return, the reference to the inode has been transferred * to the dentry. In case of an error the reference on the inode is * released. A %NULL or IS_ERR inode may be passed in and will be the * error will be propagate to the return value, with a %NULL @inode * replaced by ERR_PTR(-ESTALE). */ struct dentry *d_obtain_root(struct inode *inode) { |
f9c34674b
|
2042 |
return __d_obtain_alias(inode, false); |
1a0a397e4
|
2043 2044 2045 2046 |
} EXPORT_SYMBOL(d_obtain_root); /** |
9403540c0
|
2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 |
* d_add_ci - lookup or allocate new dentry with case-exact name * @inode: the inode case-insensitive lookup has found * @dentry: the negative dentry that was passed to the parent's lookup func * @name: the case-exact name to be associated with the returned dentry * * This is to avoid filling the dcache with case-insensitive names to the * same inode, only the actual correct case is stored in the dcache for * case-insensitive filesystems. * * For a case-insensitive lookup match and if the the case-exact dentry * already exists in in the dcache, use it and return it. * * If no entry exists with the exact case name, allocate new dentry with * the exact case, and return the spliced entry. */ |
e45b590b9
|
2062 |
struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode, |
9403540c0
|
2063 2064 |
struct qstr *name) { |
d9171b934
|
2065 |
struct dentry *found, *res; |
9403540c0
|
2066 |
|
b6520c819
|
2067 2068 2069 2070 |
/* * First check if a dentry matching the name already exists, * if not go ahead and create it now. */ |
9403540c0
|
2071 |
found = d_hash_and_lookup(dentry->d_parent, name); |
d9171b934
|
2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 |
if (found) { iput(inode); return found; } if (d_in_lookup(dentry)) { found = d_alloc_parallel(dentry->d_parent, name, dentry->d_wait); if (IS_ERR(found) || !d_in_lookup(found)) { iput(inode); return found; |
9403540c0
|
2082 |
} |
d9171b934
|
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 |
} else { found = d_alloc(dentry->d_parent, name); if (!found) { iput(inode); return ERR_PTR(-ENOMEM); } } res = d_splice_alias(inode, found); if (res) { dput(found); return res; |
9403540c0
|
2094 |
} |
4f522a247
|
2095 |
return found; |
9403540c0
|
2096 |
} |
ec4f86059
|
2097 |
EXPORT_SYMBOL(d_add_ci); |
1da177e4c
|
2098 |
|
12f8ad4b0
|
2099 |
|
d4c91a8f7
|
2100 2101 2102 |
static inline bool d_same_name(const struct dentry *dentry, const struct dentry *parent, const struct qstr *name) |
12f8ad4b0
|
2103 |
{ |
d4c91a8f7
|
2104 2105 2106 2107 |
if (likely(!(parent->d_flags & DCACHE_OP_COMPARE))) { if (dentry->d_name.len != name->len) return false; return dentry_cmp(dentry, name->name, name->len) == 0; |
12f8ad4b0
|
2108 |
} |
6fa67e707
|
2109 |
return parent->d_op->d_compare(dentry, |
d4c91a8f7
|
2110 2111 |
dentry->d_name.len, dentry->d_name.name, name) == 0; |
12f8ad4b0
|
2112 |
} |
1da177e4c
|
2113 |
/** |
31e6b01f4
|
2114 2115 2116 |
* __d_lookup_rcu - search for a dentry (racy, store-free) * @parent: parent dentry * @name: qstr of name we wish to find |
1f1e6e523
|
2117 |
* @seqp: returns d_seq value at the point where the dentry was found |
31e6b01f4
|
2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 |
* Returns: dentry, or NULL * * __d_lookup_rcu is the dcache lookup function for rcu-walk name * resolution (store-free path walking) design described in * Documentation/filesystems/path-lookup.txt. * * This is not to be used outside core vfs. * * __d_lookup_rcu must only be used in rcu-walk mode, ie. with vfsmount lock * held, and rcu_read_lock held. The returned dentry must not be stored into * without taking d_lock and checking d_seq sequence count against @seq * returned here. * |
15570086b
|
2131 |
* A refcount may be taken on the found dentry with the d_rcu_to_refcount |
31e6b01f4
|
2132 2133 2134 2135 2136 2137 |
* function. * * Alternatively, __d_lookup_rcu may be called again to look up the child of * the returned dentry, so long as its parent's seqlock is checked after the * child is looked up. Thus, an interlocking stepping of sequence lock checks * is formed, giving integrity down the path walk. |
12f8ad4b0
|
2138 2139 2140 |
* * NOTE! The caller *has* to check the resulting dentry against the sequence * number we've returned before using any of the resulting dentry state! |
31e6b01f4
|
2141 |
*/ |
8966be903
|
2142 2143 |
struct dentry *__d_lookup_rcu(const struct dentry *parent, const struct qstr *name, |
da53be12b
|
2144 |
unsigned *seqp) |
31e6b01f4
|
2145 |
{ |
26fe57502
|
2146 |
u64 hashlen = name->hash_len; |
31e6b01f4
|
2147 |
const unsigned char *str = name->name; |
8387ff257
|
2148 |
struct hlist_bl_head *b = d_hash(hashlen_hash(hashlen)); |
ceb5bdc2d
|
2149 |
struct hlist_bl_node *node; |
31e6b01f4
|
2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 |
struct dentry *dentry; /* * Note: There is significant duplication with __d_lookup_rcu which is * required to prevent single threaded performance regressions * especially on architectures where smp_rmb (in seqcounts) are costly. * Keep the two functions in sync. */ /* * The hash list is protected using RCU. * * Carefully use d_seq when comparing a candidate dentry, to avoid * races with d_move(). * * It is possible that concurrent renames can mess up our list * walk here and result in missing our dentry, resulting in the * false-negative result. d_lookup() protects against concurrent * renames using rename_lock seqlock. * |
b0a4bb830
|
2170 |
* See Documentation/filesystems/path-lookup.txt for more details. |
31e6b01f4
|
2171 |
*/ |
b07ad9967
|
2172 |
hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { |
8966be903
|
2173 |
unsigned seq; |
31e6b01f4
|
2174 |
|
31e6b01f4
|
2175 |
seqretry: |
12f8ad4b0
|
2176 2177 |
/* * The dentry sequence count protects us from concurrent |
da53be12b
|
2178 |
* renames, and thus protects parent and name fields. |
12f8ad4b0
|
2179 2180 |
* * The caller must perform a seqcount check in order |
da53be12b
|
2181 |
* to do anything useful with the returned dentry. |
12f8ad4b0
|
2182 2183 2184 2185 2186 2187 2188 |
* * NOTE! We do a "raw" seqcount_begin here. That means that * we don't wait for the sequence count to stabilize if it * is in the middle of a sequence change. If we do the slow * dentry compare, we will do seqretries until it is stable, * and if we end up with a successful lookup, we actually * want to exit RCU lookup anyway. |
d4c91a8f7
|
2189 2190 2191 |
* * Note that raw_seqcount_begin still *does* smp_rmb(), so * we are still guaranteed NUL-termination of ->d_name.name. |
12f8ad4b0
|
2192 2193 |
*/ seq = raw_seqcount_begin(&dentry->d_seq); |
31e6b01f4
|
2194 2195 |
if (dentry->d_parent != parent) continue; |
2e321806b
|
2196 2197 |
if (d_unhashed(dentry)) continue; |
12f8ad4b0
|
2198 |
|
830c0f0ed
|
2199 |
if (unlikely(parent->d_flags & DCACHE_OP_COMPARE)) { |
d4c91a8f7
|
2200 2201 |
int tlen; const char *tname; |
26fe57502
|
2202 2203 |
if (dentry->d_name.hash != hashlen_hash(hashlen)) continue; |
d4c91a8f7
|
2204 2205 2206 2207 2208 |
tlen = dentry->d_name.len; tname = dentry->d_name.name; /* we want a consistent (name,len) pair */ if (read_seqcount_retry(&dentry->d_seq, seq)) { cpu_relax(); |
12f8ad4b0
|
2209 2210 |
goto seqretry; } |
6fa67e707
|
2211 |
if (parent->d_op->d_compare(dentry, |
d4c91a8f7
|
2212 2213 2214 2215 2216 2217 2218 |
tlen, tname, name) != 0) continue; } else { if (dentry->d_name.hash_len != hashlen) continue; if (dentry_cmp(dentry, str, hashlen_len(hashlen)) != 0) continue; |
31e6b01f4
|
2219 |
} |
da53be12b
|
2220 |
*seqp = seq; |
d4c91a8f7
|
2221 |
return dentry; |
31e6b01f4
|
2222 2223 2224 2225 2226 |
} return NULL; } /** |
1da177e4c
|
2227 2228 2229 |
* d_lookup - search for a dentry * @parent: parent dentry * @name: qstr of name we wish to find |
b04f784e5
|
2230 |
* Returns: dentry, or NULL |
1da177e4c
|
2231 |
* |
b04f784e5
|
2232 2233 2234 2235 |
* d_lookup searches the children of the parent dentry for the name in * question. If the dentry is found its reference count is incremented and the * dentry is returned. The caller must use dput to free the entry when it has * finished using it. %NULL is returned if the dentry does not exist. |
1da177e4c
|
2236 |
*/ |
da2d8455e
|
2237 |
struct dentry *d_lookup(const struct dentry *parent, const struct qstr *name) |
1da177e4c
|
2238 |
{ |
31e6b01f4
|
2239 |
struct dentry *dentry; |
949854d02
|
2240 |
unsigned seq; |
1da177e4c
|
2241 |
|
b8314f930
|
2242 2243 2244 2245 |
do { seq = read_seqbegin(&rename_lock); dentry = __d_lookup(parent, name); if (dentry) |
1da177e4c
|
2246 2247 2248 2249 |
break; } while (read_seqretry(&rename_lock, seq)); return dentry; } |
ec4f86059
|
2250 |
EXPORT_SYMBOL(d_lookup); |
1da177e4c
|
2251 |
|
31e6b01f4
|
2252 |
/** |
b04f784e5
|
2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 |
* __d_lookup - search for a dentry (racy) * @parent: parent dentry * @name: qstr of name we wish to find * Returns: dentry, or NULL * * __d_lookup is like d_lookup, however it may (rarely) return a * false-negative result due to unrelated rename activity. * * __d_lookup is slightly faster by avoiding rename_lock read seqlock, * however it must be used carefully, eg. with a following d_lookup in * the case of failure. * * __d_lookup callers must be commented. */ |
a713ca2ab
|
2267 |
struct dentry *__d_lookup(const struct dentry *parent, const struct qstr *name) |
1da177e4c
|
2268 |
{ |
1da177e4c
|
2269 |
unsigned int hash = name->hash; |
8387ff257
|
2270 |
struct hlist_bl_head *b = d_hash(hash); |
ceb5bdc2d
|
2271 |
struct hlist_bl_node *node; |
31e6b01f4
|
2272 |
struct dentry *found = NULL; |
665a7583f
|
2273 |
struct dentry *dentry; |
1da177e4c
|
2274 |
|
b04f784e5
|
2275 |
/* |
31e6b01f4
|
2276 2277 2278 2279 2280 2281 2282 |
* Note: There is significant duplication with __d_lookup_rcu which is * required to prevent single threaded performance regressions * especially on architectures where smp_rmb (in seqcounts) are costly. * Keep the two functions in sync. */ /* |
b04f784e5
|
2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 |
* The hash list is protected using RCU. * * Take d_lock when comparing a candidate dentry, to avoid races * with d_move(). * * It is possible that concurrent renames can mess up our list * walk here and result in missing our dentry, resulting in the * false-negative result. d_lookup() protects against concurrent * renames using rename_lock seqlock. * |
b0a4bb830
|
2293 |
* See Documentation/filesystems/path-lookup.txt for more details. |
b04f784e5
|
2294 |
*/ |
1da177e4c
|
2295 2296 |
rcu_read_lock(); |
b07ad9967
|
2297 |
hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { |
1da177e4c
|
2298 |
|
1da177e4c
|
2299 2300 |
if (dentry->d_name.hash != hash) continue; |
1da177e4c
|
2301 2302 |
spin_lock(&dentry->d_lock); |
1da177e4c
|
2303 2304 |
if (dentry->d_parent != parent) goto next; |
d0185c088
|
2305 2306 |
if (d_unhashed(dentry)) goto next; |
d4c91a8f7
|
2307 2308 |
if (!d_same_name(dentry, parent, name)) goto next; |
1da177e4c
|
2309 |
|
98474236f
|
2310 |
dentry->d_lockref.count++; |
d0185c088
|
2311 |
found = dentry; |
1da177e4c
|
2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 |
spin_unlock(&dentry->d_lock); break; next: spin_unlock(&dentry->d_lock); } rcu_read_unlock(); return found; } /** |
3e7e241f8
|
2323 2324 2325 2326 |
* d_hash_and_lookup - hash the qstr then search for a dentry * @dir: Directory to search in * @name: qstr of name we wish to find * |
4f522a247
|
2327 |
* On lookup failure NULL is returned; on bad name - ERR_PTR(-error) |
3e7e241f8
|
2328 2329 2330 |
*/ struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name) { |
3e7e241f8
|
2331 2332 2333 2334 2335 |
/* * Check for a fs-specific hash function. Note that we must * calculate the standard hash first, as the d_op->d_hash() * routine may choose to leave the hash value unchanged. */ |
8387ff257
|
2336 |
name->hash = full_name_hash(dir, name->name, name->len); |
fb045adb9
|
2337 |
if (dir->d_flags & DCACHE_OP_HASH) { |
da53be12b
|
2338 |
int err = dir->d_op->d_hash(dir, name); |
4f522a247
|
2339 2340 |
if (unlikely(err < 0)) return ERR_PTR(err); |
3e7e241f8
|
2341 |
} |
4f522a247
|
2342 |
return d_lookup(dir, name); |
3e7e241f8
|
2343 |
} |
4f522a247
|
2344 |
EXPORT_SYMBOL(d_hash_and_lookup); |
3e7e241f8
|
2345 |
|
1da177e4c
|
2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 |
/* * When a file is deleted, we have two options: * - turn this dentry into a negative dentry * - unhash this dentry and free it. * * Usually, we want to just turn this into * a negative dentry, but if anybody else is * currently using the dentry or the inode * we can't do that and we fall back on removing * it from the hash queues and waiting for * it to be deleted later when it has no users */ /** * d_delete - delete a dentry * @dentry: The dentry to delete * * Turn the dentry into a negative dentry if possible, otherwise * remove it from the hash queues so it can be deleted later */ void d_delete(struct dentry * dentry) { |
c19457f0a
|
2369 |
struct inode *inode = dentry->d_inode; |
c19457f0a
|
2370 2371 2372 |
spin_lock(&inode->i_lock); spin_lock(&dentry->d_lock); |
1da177e4c
|
2373 2374 2375 |
/* * Are we the only user? */ |
98474236f
|
2376 |
if (dentry->d_lockref.count == 1) { |
13e3c5e5b
|
2377 |
dentry->d_flags &= ~DCACHE_CANT_MOUNT; |
31e6b01f4
|
2378 |
dentry_unlink_inode(dentry); |
c19457f0a
|
2379 |
} else { |
1da177e4c
|
2380 |
__d_drop(dentry); |
c19457f0a
|
2381 2382 2383 |
spin_unlock(&dentry->d_lock); spin_unlock(&inode->i_lock); } |
1da177e4c
|
2384 |
} |
ec4f86059
|
2385 |
EXPORT_SYMBOL(d_delete); |
1da177e4c
|
2386 |
|
15d3c589f
|
2387 |
static void __d_rehash(struct dentry *entry) |
1da177e4c
|
2388 |
{ |
15d3c589f
|
2389 |
struct hlist_bl_head *b = d_hash(entry->d_name.hash); |
61647823a
|
2390 |
|
1879fd6a2
|
2391 |
hlist_bl_lock(b); |
b07ad9967
|
2392 |
hlist_bl_add_head_rcu(&entry->d_hash, b); |
1879fd6a2
|
2393 |
hlist_bl_unlock(b); |
1da177e4c
|
2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 |
} /** * d_rehash - add an entry back to the hash * @entry: dentry to add to the hash * * Adds a dentry to the hash according to its name. */ void d_rehash(struct dentry * entry) { |
1da177e4c
|
2405 |
spin_lock(&entry->d_lock); |
15d3c589f
|
2406 |
__d_rehash(entry); |
1da177e4c
|
2407 |
spin_unlock(&entry->d_lock); |
1da177e4c
|
2408 |
} |
ec4f86059
|
2409 |
EXPORT_SYMBOL(d_rehash); |
1da177e4c
|
2410 |
|
84e710da2
|
2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 |
static inline unsigned start_dir_add(struct inode *dir) { for (;;) { unsigned n = dir->i_dir_seq; if (!(n & 1) && cmpxchg(&dir->i_dir_seq, n, n + 1) == n) return n; cpu_relax(); } } static inline void end_dir_add(struct inode *dir, unsigned n) { smp_store_release(&dir->i_dir_seq, n + 2); } |
d9171b934
|
2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 |
static void d_wait_lookup(struct dentry *dentry) { if (d_in_lookup(dentry)) { DECLARE_WAITQUEUE(wait, current); add_wait_queue(dentry->d_wait, &wait); do { set_current_state(TASK_UNINTERRUPTIBLE); spin_unlock(&dentry->d_lock); schedule(); spin_lock(&dentry->d_lock); } while (d_in_lookup(dentry)); } } |
94bdd655c
|
2439 |
struct dentry *d_alloc_parallel(struct dentry *parent, |
d9171b934
|
2440 2441 |
const struct qstr *name, wait_queue_head_t *wq) |
94bdd655c
|
2442 |
{ |
94bdd655c
|
2443 |
unsigned int hash = name->hash; |
94bdd655c
|
2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 |
struct hlist_bl_head *b = in_lookup_hash(parent, hash); struct hlist_bl_node *node; struct dentry *new = d_alloc(parent, name); struct dentry *dentry; unsigned seq, r_seq, d_seq; if (unlikely(!new)) return ERR_PTR(-ENOMEM); retry: rcu_read_lock(); |
015555fd4
|
2455 |
seq = smp_load_acquire(&parent->d_inode->i_dir_seq); |
94bdd655c
|
2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 |
r_seq = read_seqbegin(&rename_lock); dentry = __d_lookup_rcu(parent, name, &d_seq); if (unlikely(dentry)) { if (!lockref_get_not_dead(&dentry->d_lockref)) { rcu_read_unlock(); goto retry; } if (read_seqcount_retry(&dentry->d_seq, d_seq)) { rcu_read_unlock(); dput(dentry); goto retry; } rcu_read_unlock(); dput(new); return dentry; } if (unlikely(read_seqretry(&rename_lock, r_seq))) { rcu_read_unlock(); goto retry; } |
015555fd4
|
2476 2477 2478 2479 2480 |
if (unlikely(seq & 1)) { rcu_read_unlock(); goto retry; } |
94bdd655c
|
2481 |
hlist_bl_lock(b); |
8cc07c808
|
2482 |
if (unlikely(READ_ONCE(parent->d_inode->i_dir_seq) != seq)) { |
94bdd655c
|
2483 2484 2485 2486 |
hlist_bl_unlock(b); rcu_read_unlock(); goto retry; } |
94bdd655c
|
2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 |
/* * No changes for the parent since the beginning of d_lookup(). * Since all removals from the chain happen with hlist_bl_lock(), * any potential in-lookup matches are going to stay here until * we unlock the chain. All fields are stable in everything * we encounter. */ hlist_bl_for_each_entry(dentry, node, b, d_u.d_in_lookup_hash) { if (dentry->d_name.hash != hash) continue; if (dentry->d_parent != parent) continue; |
d4c91a8f7
|
2499 2500 |
if (!d_same_name(dentry, parent, name)) continue; |
94bdd655c
|
2501 |
hlist_bl_unlock(b); |
e7d6ef979
|
2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 |
/* now we can try to grab a reference */ if (!lockref_get_not_dead(&dentry->d_lockref)) { rcu_read_unlock(); goto retry; } rcu_read_unlock(); /* * somebody is likely to be still doing lookup for it; * wait for them to finish */ |
d9171b934
|
2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 |
spin_lock(&dentry->d_lock); d_wait_lookup(dentry); /* * it's not in-lookup anymore; in principle we should repeat * everything from dcache lookup, but it's likely to be what * d_lookup() would've found anyway. If it is, just return it; * otherwise we really have to repeat the whole thing. */ if (unlikely(dentry->d_name.hash != hash)) goto mismatch; if (unlikely(dentry->d_parent != parent)) goto mismatch; if (unlikely(d_unhashed(dentry))) goto mismatch; |
d4c91a8f7
|
2527 2528 |
if (unlikely(!d_same_name(dentry, parent, name))) goto mismatch; |
d9171b934
|
2529 2530 |
/* OK, it *is* a hashed match; return it */ spin_unlock(&dentry->d_lock); |
94bdd655c
|
2531 2532 2533 |
dput(new); return dentry; } |
e7d6ef979
|
2534 |
rcu_read_unlock(); |
94bdd655c
|
2535 2536 |
/* we can't take ->d_lock here; it's OK, though. */ new->d_flags |= DCACHE_PAR_LOOKUP; |
d9171b934
|
2537 |
new->d_wait = wq; |
94bdd655c
|
2538 2539 2540 |
hlist_bl_add_head_rcu(&new->d_u.d_in_lookup_hash, b); hlist_bl_unlock(b); return new; |
d9171b934
|
2541 2542 2543 2544 |
mismatch: spin_unlock(&dentry->d_lock); dput(dentry); goto retry; |
94bdd655c
|
2545 2546 |
} EXPORT_SYMBOL(d_alloc_parallel); |
85c7f8104
|
2547 2548 |
void __d_lookup_done(struct dentry *dentry) { |
94bdd655c
|
2549 2550 2551 |
struct hlist_bl_head *b = in_lookup_hash(dentry->d_parent, dentry->d_name.hash); hlist_bl_lock(b); |
85c7f8104
|
2552 |
dentry->d_flags &= ~DCACHE_PAR_LOOKUP; |
94bdd655c
|
2553 |
__hlist_bl_del(&dentry->d_u.d_in_lookup_hash); |
d9171b934
|
2554 2555 |
wake_up_all(dentry->d_wait); dentry->d_wait = NULL; |
94bdd655c
|
2556 2557 |
hlist_bl_unlock(b); INIT_HLIST_NODE(&dentry->d_u.d_alias); |
d9171b934
|
2558 |
INIT_LIST_HEAD(&dentry->d_lru); |
85c7f8104
|
2559 2560 |
} EXPORT_SYMBOL(__d_lookup_done); |
ed782b5a7
|
2561 2562 2563 2564 2565 |
/* inode->i_lock held if inode is non-NULL */ static inline void __d_add(struct dentry *dentry, struct inode *inode) { |
84e710da2
|
2566 2567 |
struct inode *dir = NULL; unsigned n; |
0568d705b
|
2568 |
spin_lock(&dentry->d_lock); |
84e710da2
|
2569 2570 2571 |
if (unlikely(d_in_lookup(dentry))) { dir = dentry->d_parent->d_inode; n = start_dir_add(dir); |
85c7f8104
|
2572 |
__d_lookup_done(dentry); |
84e710da2
|
2573 |
} |
ed782b5a7
|
2574 |
if (inode) { |
0568d705b
|
2575 2576 2577 2578 2579 |
unsigned add_flags = d_flags_for_inode(inode); hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); raw_write_seqcount_begin(&dentry->d_seq); __d_set_inode_and_type(dentry, inode, add_flags); raw_write_seqcount_end(&dentry->d_seq); |
affda4841
|
2580 |
fsnotify_update_flags(dentry); |
ed782b5a7
|
2581 |
} |
15d3c589f
|
2582 |
__d_rehash(dentry); |
84e710da2
|
2583 2584 |
if (dir) end_dir_add(dir, n); |
0568d705b
|
2585 2586 2587 |
spin_unlock(&dentry->d_lock); if (inode) spin_unlock(&inode->i_lock); |
ed782b5a7
|
2588 |
} |
fb2d5b86a
|
2589 |
/** |
34d0d19dc
|
2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 |
* d_add - add dentry to hash queues * @entry: dentry to add * @inode: The inode to attach to this dentry * * This adds the entry to the hash queues and initializes @inode. * The entry was actually filled in earlier during d_alloc(). */ void d_add(struct dentry *entry, struct inode *inode) { |
b96809173
|
2600 2601 |
if (inode) { security_d_instantiate(entry, inode); |
ed782b5a7
|
2602 |
spin_lock(&inode->i_lock); |
b96809173
|
2603 |
} |
ed782b5a7
|
2604 |
__d_add(entry, inode); |
34d0d19dc
|
2605 2606 2607 2608 |
} EXPORT_SYMBOL(d_add); /** |
668d0cd56
|
2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 |
* d_exact_alias - find and hash an exact unhashed alias * @entry: dentry to add * @inode: The inode to go with this dentry * * If an unhashed dentry with the same name/parent and desired * inode already exists, hash and return it. Otherwise, return * NULL. * * Parent directory should be locked. */ struct dentry *d_exact_alias(struct dentry *entry, struct inode *inode) { struct dentry *alias; |
668d0cd56
|
2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 |
unsigned int hash = entry->d_name.hash; spin_lock(&inode->i_lock); hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) { /* * Don't need alias->d_lock here, because aliases with * d_parent == entry->d_parent are not subject to name or * parent changes, because the parent inode i_mutex is held. */ if (alias->d_name.hash != hash) continue; if (alias->d_parent != entry->d_parent) continue; |
d4c91a8f7
|
2635 |
if (!d_same_name(alias, entry->d_parent, &entry->d_name)) |
668d0cd56
|
2636 2637 2638 2639 2640 2641 2642 |
continue; spin_lock(&alias->d_lock); if (!d_unhashed(alias)) { spin_unlock(&alias->d_lock); alias = NULL; } else { __dget_dlock(alias); |
15d3c589f
|
2643 |
__d_rehash(alias); |
668d0cd56
|
2644 2645 2646 2647 2648 2649 2650 2651 2652 |
spin_unlock(&alias->d_lock); } spin_unlock(&inode->i_lock); return alias; } spin_unlock(&inode->i_lock); return NULL; } EXPORT_SYMBOL(d_exact_alias); |
8d85b4845
|
2653 |
static void swap_names(struct dentry *dentry, struct dentry *target) |
1da177e4c
|
2654 |
{ |
8d85b4845
|
2655 2656 |
if (unlikely(dname_external(target))) { if (unlikely(dname_external(dentry))) { |
1da177e4c
|
2657 2658 2659 |
/* * Both external: swap the pointers */ |
9a8d5bb4a
|
2660 |
swap(target->d_name.name, dentry->d_name.name); |
1da177e4c
|
2661 2662 2663 2664 2665 |
} else { /* * dentry:internal, target:external. Steal target's * storage and make target internal. */ |
321bcf921
|
2666 2667 |
memcpy(target->d_iname, dentry->d_name.name, dentry->d_name.len + 1); |
1da177e4c
|
2668 2669 2670 2671 |
dentry->d_name.name = target->d_name.name; target->d_name.name = target->d_iname; } } else { |
8d85b4845
|
2672 |
if (unlikely(dname_external(dentry))) { |
1da177e4c
|
2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 |
/* * dentry:external, target:internal. Give dentry's * storage to target and make dentry internal */ memcpy(dentry->d_iname, target->d_name.name, target->d_name.len + 1); target->d_name.name = dentry->d_name.name; dentry->d_name.name = dentry->d_iname; } else { /* |
da1ce0670
|
2683 |
* Both are internal. |
1da177e4c
|
2684 |
*/ |
da1ce0670
|
2685 2686 2687 2688 2689 2690 |
unsigned int i; BUILD_BUG_ON(!IS_ALIGNED(DNAME_INLINE_LEN, sizeof(long))); for (i = 0; i < DNAME_INLINE_LEN / sizeof(long); i++) { swap(((long *) &dentry->d_iname)[i], ((long *) &target->d_iname)[i]); } |
1da177e4c
|
2691 2692 |
} } |
a28ddb87c
|
2693 |
swap(dentry->d_name.hash_len, target->d_name.hash_len); |
1da177e4c
|
2694 |
} |
8d85b4845
|
2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 |
static void copy_name(struct dentry *dentry, struct dentry *target) { struct external_name *old_name = NULL; if (unlikely(dname_external(dentry))) old_name = external_name(dentry); if (unlikely(dname_external(target))) { atomic_inc(&external_name(target)->u.count); dentry->d_name = target->d_name; } else { memcpy(dentry->d_iname, target->d_name.name, target->d_name.len + 1); dentry->d_name.name = dentry->d_iname; dentry->d_name.hash_len = target->d_name.hash_len; } if (old_name && likely(atomic_dec_and_test(&old_name->u.count))) |
2e03b4bc4
|
2710 |
kfree_rcu(old_name, u.head); |
8d85b4845
|
2711 |
} |
1da177e4c
|
2712 |
/* |
183675011
|
2713 |
* __d_move - move a dentry |
1da177e4c
|
2714 2715 |
* @dentry: entry to move * @target: new dentry |
da1ce0670
|
2716 |
* @exchange: exchange the two dentries |
1da177e4c
|
2717 2718 |
* * Update the dcache to reflect the move of a file name. Negative |
c46c88774
|
2719 2720 2721 |
* dcache entries should not be moved in this way. Caller must hold * rename_lock, the i_mutex of the source and target directories, * and the sb->s_vfs_rename_mutex if they differ. See lock_rename(). |
1da177e4c
|
2722 |
*/ |
da1ce0670
|
2723 2724 |
static void __d_move(struct dentry *dentry, struct dentry *target, bool exchange) |
1da177e4c
|
2725 |
{ |
42177007a
|
2726 |
struct dentry *old_parent, *p; |
84e710da2
|
2727 2728 |
struct inode *dir = NULL; unsigned n; |
1da177e4c
|
2729 |
|
42177007a
|
2730 2731 2732 |
WARN_ON(!dentry->d_inode); if (WARN_ON(dentry == target)) return; |
2fd6b7f50
|
2733 |
BUG_ON(d_ancestor(target, dentry)); |
42177007a
|
2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 |
old_parent = dentry->d_parent; p = d_ancestor(old_parent, target); if (IS_ROOT(dentry)) { BUG_ON(p); spin_lock(&target->d_parent->d_lock); } else if (!p) { /* target is not a descendent of dentry->d_parent */ spin_lock(&target->d_parent->d_lock); spin_lock_nested(&old_parent->d_lock, DENTRY_D_LOCK_NESTED); } else { BUG_ON(p == dentry); spin_lock(&old_parent->d_lock); if (p != target) spin_lock_nested(&target->d_parent->d_lock, DENTRY_D_LOCK_NESTED); } spin_lock_nested(&dentry->d_lock, 2); spin_lock_nested(&target->d_lock, 3); |
2fd6b7f50
|
2752 |
|
84e710da2
|
2753 2754 2755 |
if (unlikely(d_in_lookup(target))) { dir = target->d_parent->d_inode; n = start_dir_add(dir); |
85c7f8104
|
2756 |
__d_lookup_done(target); |
84e710da2
|
2757 |
} |
1da177e4c
|
2758 |
|
31e6b01f4
|
2759 |
write_seqcount_begin(&dentry->d_seq); |
1ca7d67cf
|
2760 |
write_seqcount_begin_nested(&target->d_seq, DENTRY_D_LOCK_NESTED); |
31e6b01f4
|
2761 |
|
15d3c589f
|
2762 |
/* unhash both */ |
0632a9ac7
|
2763 2764 2765 2766 |
if (!d_unhashed(dentry)) ___d_drop(dentry); if (!d_unhashed(target)) ___d_drop(target); |
1da177e4c
|
2767 |
|
076515fc9
|
2768 2769 2770 |
/* ... and switch them in the tree */ dentry->d_parent = target->d_parent; if (!exchange) { |
8d85b4845
|
2771 |
copy_name(dentry, target); |
61647823a
|
2772 |
target->d_hash.pprev = NULL; |
076515fc9
|
2773 |
dentry->d_parent->d_lockref.count++; |
5467a68cb
|
2774 |
if (dentry != old_parent) /* wasn't IS_ROOT */ |
076515fc9
|
2775 |
WARN_ON(!--old_parent->d_lockref.count); |
1da177e4c
|
2776 |
} else { |
076515fc9
|
2777 2778 |
target->d_parent = old_parent; swap_names(dentry, target); |
946e51f2b
|
2779 |
list_move(&target->d_child, &target->d_parent->d_subdirs); |
076515fc9
|
2780 2781 |
__d_rehash(target); fsnotify_update_flags(target); |
1da177e4c
|
2782 |
} |
076515fc9
|
2783 2784 2785 |
list_move(&dentry->d_child, &dentry->d_parent->d_subdirs); __d_rehash(dentry); fsnotify_update_flags(dentry); |
0bf3d5c16
|
2786 |
fscrypt_handle_d_move(dentry); |
1da177e4c
|
2787 |
|
31e6b01f4
|
2788 2789 |
write_seqcount_end(&target->d_seq); write_seqcount_end(&dentry->d_seq); |
84e710da2
|
2790 2791 |
if (dir) end_dir_add(dir, n); |
076515fc9
|
2792 2793 2794 2795 2796 2797 2798 |
if (dentry->d_parent != old_parent) spin_unlock(&dentry->d_parent->d_lock); if (dentry != old_parent) spin_unlock(&old_parent->d_lock); spin_unlock(&target->d_lock); spin_unlock(&dentry->d_lock); |
183675011
|
2799 2800 2801 2802 2803 2804 2805 2806 |
} /* * d_move - move a dentry * @dentry: entry to move * @target: new dentry * * Update the dcache to reflect the move of a file name. Negative |
c46c88774
|
2807 2808 |
* dcache entries should not be moved in this way. See the locking * requirements for __d_move. |
183675011
|
2809 2810 2811 2812 |
*/ void d_move(struct dentry *dentry, struct dentry *target) { write_seqlock(&rename_lock); |
da1ce0670
|
2813 |
__d_move(dentry, target, false); |
1da177e4c
|
2814 |
write_sequnlock(&rename_lock); |
9eaef27b3
|
2815 |
} |
ec4f86059
|
2816 |
EXPORT_SYMBOL(d_move); |
1da177e4c
|
2817 |
|
da1ce0670
|
2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 |
/* * d_exchange - exchange two dentries * @dentry1: first dentry * @dentry2: second dentry */ void d_exchange(struct dentry *dentry1, struct dentry *dentry2) { write_seqlock(&rename_lock); WARN_ON(!dentry1->d_inode); WARN_ON(!dentry2->d_inode); WARN_ON(IS_ROOT(dentry1)); WARN_ON(IS_ROOT(dentry2)); __d_move(dentry1, dentry2, true); write_sequnlock(&rename_lock); } |
e2761a116
|
2836 2837 2838 2839 2840 2841 2842 |
/** * d_ancestor - search for an ancestor * @p1: ancestor dentry * @p2: child dentry * * Returns the ancestor dentry of p2 which is a child of p1, if p1 is * an ancestor of p2, else NULL. |
9eaef27b3
|
2843 |
*/ |
e2761a116
|
2844 |
struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2) |
9eaef27b3
|
2845 2846 |
{ struct dentry *p; |
871c0067d
|
2847 |
for (p = p2; !IS_ROOT(p); p = p->d_parent) { |
9eaef27b3
|
2848 |
if (p->d_parent == p1) |
e2761a116
|
2849 |
return p; |
9eaef27b3
|
2850 |
} |
e2761a116
|
2851 |
return NULL; |
9eaef27b3
|
2852 2853 2854 2855 2856 2857 |
} /* * This helper attempts to cope with remotely renamed directories * * It assumes that the caller is already holding |
a03e283bf
|
2858 |
* dentry->d_parent->d_inode->i_mutex, and rename_lock |
9eaef27b3
|
2859 2860 2861 |
* * Note: If ever the locking in lock_rename() changes, then please * remember to update this too... |
9eaef27b3
|
2862 |
*/ |
b5ae6b15b
|
2863 |
static int __d_unalias(struct inode *inode, |
873feea09
|
2864 |
struct dentry *dentry, struct dentry *alias) |
9eaef27b3
|
2865 |
{ |
9902af79c
|
2866 2867 |
struct mutex *m1 = NULL; struct rw_semaphore *m2 = NULL; |
3d330dc17
|
2868 |
int ret = -ESTALE; |
9eaef27b3
|
2869 2870 2871 2872 |
/* If alias and dentry share a parent, then no extra locks required */ if (alias->d_parent == dentry->d_parent) goto out_unalias; |
9eaef27b3
|
2873 |
/* See lock_rename() */ |
9eaef27b3
|
2874 2875 2876 |
if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex)) goto out_err; m1 = &dentry->d_sb->s_vfs_rename_mutex; |
9902af79c
|
2877 |
if (!inode_trylock_shared(alias->d_parent->d_inode)) |
9eaef27b3
|
2878 |
goto out_err; |
9902af79c
|
2879 |
m2 = &alias->d_parent->d_inode->i_rwsem; |
9eaef27b3
|
2880 |
out_unalias: |
8ed936b56
|
2881 |
__d_move(alias, dentry, false); |
b5ae6b15b
|
2882 |
ret = 0; |
9eaef27b3
|
2883 |
out_err: |
9eaef27b3
|
2884 |
if (m2) |
9902af79c
|
2885 |
up_read(m2); |
9eaef27b3
|
2886 2887 2888 2889 |
if (m1) mutex_unlock(m1); return ret; } |
770bfad84
|
2890 |
/** |
3f70bd51c
|
2891 2892 2893 2894 |
* d_splice_alias - splice a disconnected dentry into the tree if one exists * @inode: the inode which may have a disconnected dentry * @dentry: a negative dentry which we want to point to the inode. * |
da093a9b7
|
2895 2896 2897 |
* If inode is a directory and has an IS_ROOT alias, then d_move that in * place of the given dentry and return it, else simply d_add the inode * to the dentry and return NULL. |
3f70bd51c
|
2898 |
* |
908790fa3
|
2899 2900 2901 |
* If a non-IS_ROOT directory is found, the filesystem is corrupt, and * we should error out: directories can't have multiple aliases. * |
3f70bd51c
|
2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 |
* This is needed in the lookup routine of any filesystem that is exportable * (via knfsd) so that we can build dcache paths to directories effectively. * * If a dentry was found and moved, then it is returned. Otherwise NULL * is returned. This matches the expected return value of ->lookup. * * Cluster filesystems may call this function with a negative, hashed dentry. * In that case, we know that the inode will be a regular file, and also this * will only occur during atomic_open. So we need to check for the dentry * being already hashed only in the final case. */ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) { |
3f70bd51c
|
2915 2916 |
if (IS_ERR(inode)) return ERR_CAST(inode); |
770bfad84
|
2917 |
BUG_ON(!d_unhashed(dentry)); |
de689f5e3
|
2918 |
if (!inode) |
b5ae6b15b
|
2919 |
goto out; |
de689f5e3
|
2920 |
|
b96809173
|
2921 |
security_d_instantiate(dentry, inode); |
873feea09
|
2922 |
spin_lock(&inode->i_lock); |
9eaef27b3
|
2923 |
if (S_ISDIR(inode->i_mode)) { |
b5ae6b15b
|
2924 2925 |
struct dentry *new = __d_find_any_alias(inode); if (unlikely(new)) { |
a03e283bf
|
2926 2927 |
/* The reference to new ensures it remains an alias */ spin_unlock(&inode->i_lock); |
183675011
|
2928 |
write_seqlock(&rename_lock); |
b5ae6b15b
|
2929 2930 |
if (unlikely(d_ancestor(new, dentry))) { write_sequnlock(&rename_lock); |
b5ae6b15b
|
2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 |
dput(new); new = ERR_PTR(-ELOOP); pr_warn_ratelimited( "VFS: Lookup of '%s' in %s %s" " would have caused loop ", dentry->d_name.name, inode->i_sb->s_type->name, inode->i_sb->s_id); } else if (!IS_ROOT(new)) { |
076515fc9
|
2941 |
struct dentry *old_parent = dget(new->d_parent); |
b5ae6b15b
|
2942 |
int err = __d_unalias(inode, dentry, new); |
183675011
|
2943 |
write_sequnlock(&rename_lock); |
b5ae6b15b
|
2944 2945 2946 2947 |
if (err) { dput(new); new = ERR_PTR(err); } |
076515fc9
|
2948 |
dput(old_parent); |
183675011
|
2949 |
} else { |
b5ae6b15b
|
2950 2951 |
__d_move(new, dentry, false); write_sequnlock(&rename_lock); |
dd179946d
|
2952 |
} |
b5ae6b15b
|
2953 2954 |
iput(inode); return new; |
9eaef27b3
|
2955 |
} |
770bfad84
|
2956 |
} |
b5ae6b15b
|
2957 |
out: |
ed782b5a7
|
2958 |
__d_add(dentry, inode); |
b5ae6b15b
|
2959 |
return NULL; |
770bfad84
|
2960 |
} |
b5ae6b15b
|
2961 |
EXPORT_SYMBOL(d_splice_alias); |
770bfad84
|
2962 |
|
1da177e4c
|
2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 |
/* * Test whether new_dentry is a subdirectory of old_dentry. * * Trivially implemented using the dcache structure */ /** * is_subdir - is new dentry a subdirectory of old_dentry * @new_dentry: new dentry * @old_dentry: old dentry * |
a6e5787fc
|
2974 2975 |
* Returns true if new_dentry is a subdirectory of the parent (at any depth). * Returns false otherwise. |
1da177e4c
|
2976 2977 2978 |
* Caller must ensure that "new_dentry" is pinned before calling is_subdir() */ |
a6e5787fc
|
2979 |
bool is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) |
1da177e4c
|
2980 |
{ |
a6e5787fc
|
2981 |
bool result; |
949854d02
|
2982 |
unsigned seq; |
1da177e4c
|
2983 |
|
e2761a116
|
2984 |
if (new_dentry == old_dentry) |
a6e5787fc
|
2985 |
return true; |
e2761a116
|
2986 |
|
e2761a116
|
2987 |
do { |
1da177e4c
|
2988 |
/* for restarting inner loop in case of seq retry */ |
1da177e4c
|
2989 |
seq = read_seqbegin(&rename_lock); |
949854d02
|
2990 2991 2992 2993 2994 |
/* * Need rcu_readlock to protect against the d_parent trashing * due to d_move */ rcu_read_lock(); |
e2761a116
|
2995 |
if (d_ancestor(old_dentry, new_dentry)) |
a6e5787fc
|
2996 |
result = true; |
e2761a116
|
2997 |
else |
a6e5787fc
|
2998 |
result = false; |
949854d02
|
2999 |
rcu_read_unlock(); |
1da177e4c
|
3000 |
} while (read_seqretry(&rename_lock, seq)); |
1da177e4c
|
3001 3002 3003 |
return result; } |
e8f9e5b78
|
3004 |
EXPORT_SYMBOL(is_subdir); |
1da177e4c
|
3005 |
|
db14fc3ab
|
3006 |
static enum d_walk_ret d_genocide_kill(void *data, struct dentry *dentry) |
1da177e4c
|
3007 |
{ |
db14fc3ab
|
3008 3009 3010 3011 |
struct dentry *root = data; if (dentry != root) { if (d_unhashed(dentry) || !dentry->d_inode) return D_WALK_SKIP; |
1da177e4c
|
3012 |
|
01ddc4ede
|
3013 3014 3015 3016 |
if (!(dentry->d_flags & DCACHE_GENOCIDE)) { dentry->d_flags |= DCACHE_GENOCIDE; dentry->d_lockref.count--; } |
1da177e4c
|
3017 |
} |
db14fc3ab
|
3018 3019 |
return D_WALK_CONTINUE; } |
58db63d08
|
3020 |
|
db14fc3ab
|
3021 3022 |
void d_genocide(struct dentry *parent) { |
3a8e3611e
|
3023 |
d_walk(parent, parent, d_genocide_kill); |
1da177e4c
|
3024 |
} |
cbd4a5bcb
|
3025 |
EXPORT_SYMBOL(d_genocide); |
60545d0d4
|
3026 |
void d_tmpfile(struct dentry *dentry, struct inode *inode) |
1da177e4c
|
3027 |
{ |
60545d0d4
|
3028 3029 |
inode_dec_link_count(inode); BUG_ON(dentry->d_name.name != dentry->d_iname || |
946e51f2b
|
3030 |
!hlist_unhashed(&dentry->d_u.d_alias) || |
60545d0d4
|
3031 3032 3033 3034 3035 3036 3037 3038 |
!d_unlinked(dentry)); spin_lock(&dentry->d_parent->d_lock); spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); dentry->d_name.len = sprintf(dentry->d_iname, "#%llu", (unsigned long long)inode->i_ino); spin_unlock(&dentry->d_lock); spin_unlock(&dentry->d_parent->d_lock); d_instantiate(dentry, inode); |
1da177e4c
|
3039 |
} |
60545d0d4
|
3040 |
EXPORT_SYMBOL(d_tmpfile); |
1da177e4c
|
3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 |
static __initdata unsigned long dhash_entries; static int __init set_dhash_entries(char *str) { if (!str) return 0; dhash_entries = simple_strtoul(str, &str, 0); return 1; } __setup("dhash_entries=", set_dhash_entries); static void __init dcache_init_early(void) { |
1da177e4c
|
3054 3055 3056 3057 3058 3059 3060 3061 |
/* If hashes are distributed across NUMA nodes, defer * hash allocation until vmalloc space is available. */ if (hashdist) return; dentry_hashtable = alloc_large_system_hash("Dentry cache", |
b07ad9967
|
3062 |
sizeof(struct hlist_bl_head), |
1da177e4c
|
3063 3064 |
dhash_entries, 13, |
3d375d785
|
3065 |
HASH_EARLY | HASH_ZERO, |
1da177e4c
|
3066 |
&d_hash_shift, |
b35d786b6
|
3067 |
NULL, |
31fe62b95
|
3068 |
0, |
1da177e4c
|
3069 |
0); |
854d3e634
|
3070 |
d_hash_shift = 32 - d_hash_shift; |
1da177e4c
|
3071 |
} |
74bf17cff
|
3072 |
static void __init dcache_init(void) |
1da177e4c
|
3073 |
{ |
3d375d785
|
3074 |
/* |
1da177e4c
|
3075 3076 |
* A constructor could be added for stable state like the lists, * but it is probably not worth it because of the cache nature |
3d375d785
|
3077 |
* of the dcache. |
1da177e4c
|
3078 |
*/ |
80344266c
|
3079 3080 3081 |
dentry_cache = KMEM_CACHE_USERCOPY(dentry, SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|SLAB_MEM_SPREAD|SLAB_ACCOUNT, d_iname); |
1da177e4c
|
3082 3083 3084 3085 3086 3087 3088 |
/* Hash may have been set up in dcache_init_early */ if (!hashdist) return; dentry_hashtable = alloc_large_system_hash("Dentry cache", |
b07ad9967
|
3089 |
sizeof(struct hlist_bl_head), |
1da177e4c
|
3090 3091 |
dhash_entries, 13, |
3d375d785
|
3092 |
HASH_ZERO, |
1da177e4c
|
3093 |
&d_hash_shift, |
b35d786b6
|
3094 |
NULL, |
31fe62b95
|
3095 |
0, |
1da177e4c
|
3096 |
0); |
854d3e634
|
3097 |
d_hash_shift = 32 - d_hash_shift; |
1da177e4c
|
3098 3099 3100 |
} /* SLAB cache for __getname() consumers */ |
e18b890bb
|
3101 |
struct kmem_cache *names_cachep __read_mostly; |
ec4f86059
|
3102 |
EXPORT_SYMBOL(names_cachep); |
1da177e4c
|
3103 |
|
1da177e4c
|
3104 3105 |
void __init vfs_caches_init_early(void) { |
6916363f3
|
3106 3107 3108 3109 |
int i; for (i = 0; i < ARRAY_SIZE(in_lookup_hashtable); i++) INIT_HLIST_BL_HEAD(&in_lookup_hashtable[i]); |
1da177e4c
|
3110 3111 3112 |
dcache_init_early(); inode_init_early(); } |
4248b0da4
|
3113 |
void __init vfs_caches_init(void) |
1da177e4c
|
3114 |
{ |
6a9b88204
|
3115 3116 |
names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, 0, PATH_MAX, NULL); |
1da177e4c
|
3117 |
|
74bf17cff
|
3118 3119 |
dcache_init(); inode_init(); |
4248b0da4
|
3120 3121 |
files_init(); files_maxfiles_init(); |
74bf17cff
|
3122 |
mnt_init(); |
1da177e4c
|
3123 3124 3125 |
bdev_cache_init(); chrdev_init(); } |