]>
Commit | Line | Data |
---|---|---|
1 | /* | |
2 | * fs/dcache.c | |
3 | * | |
4 | * Complete reimplementation | |
5 | * (C) 1997 Thomas Schoebel-Theuer, | |
6 | * with heavy changes by Linus Torvalds | |
7 | */ | |
8 | ||
9 | /* | |
10 | * Notes on the allocation strategy: | |
11 | * | |
12 | * The dcache is a master of the icache - whenever a dcache entry | |
13 | * exists, the inode will always exist. "iput()" is done either when | |
14 | * the dcache entry is deleted or garbage collected. | |
15 | */ | |
16 | ||
17 | #include <linux/syscalls.h> | |
18 | #include <linux/string.h> | |
19 | #include <linux/mm.h> | |
20 | #include <linux/fs.h> | |
21 | #include <linux/fsnotify.h> | |
22 | #include <linux/slab.h> | |
23 | #include <linux/init.h> | |
24 | #include <linux/hash.h> | |
25 | #include <linux/cache.h> | |
26 | #include <linux/export.h> | |
27 | #include <linux/mount.h> | |
28 | #include <linux/file.h> | |
29 | #include <linux/uaccess.h> | |
30 | #include <linux/security.h> | |
31 | #include <linux/seqlock.h> | |
32 | #include <linux/swap.h> | |
33 | #include <linux/bootmem.h> | |
34 | #include <linux/fs_struct.h> | |
35 | #include <linux/hardirq.h> | |
36 | #include <linux/bit_spinlock.h> | |
37 | #include <linux/rculist_bl.h> | |
38 | #include <linux/prefetch.h> | |
39 | #include <linux/ratelimit.h> | |
40 | #include <linux/list_lru.h> | |
41 | #include <linux/kasan.h> | |
42 | ||
43 | #include "internal.h" | |
44 | #include "mount.h" | |
45 | ||
46 | /* | |
47 | * Usage: | |
48 | * dcache->d_inode->i_lock protects: | |
49 | * - i_dentry, d_u.d_alias, d_inode of aliases | |
50 | * dcache_hash_bucket lock protects: | |
51 | * - the dcache hash table | |
52 | * s_anon bl list spinlock protects: | |
53 | * - the s_anon list (see __d_drop) | |
54 | * dentry->d_sb->s_dentry_lru_lock protects: | |
55 | * - the dcache lru lists and counters | |
56 | * d_lock protects: | |
57 | * - d_flags | |
58 | * - d_name | |
59 | * - d_lru | |
60 | * - d_count | |
61 | * - d_unhashed() | |
62 | * - d_parent and d_subdirs | |
63 | * - childrens' d_child and d_parent | |
64 | * - d_u.d_alias, d_inode | |
65 | * | |
66 | * Ordering: | |
67 | * dentry->d_inode->i_lock | |
68 | * dentry->d_lock | |
69 | * dentry->d_sb->s_dentry_lru_lock | |
70 | * dcache_hash_bucket lock | |
71 | * s_anon lock | |
72 | * | |
73 | * If there is an ancestor relationship: | |
74 | * dentry->d_parent->...->d_parent->d_lock | |
75 | * ... | |
76 | * dentry->d_parent->d_lock | |
77 | * dentry->d_lock | |
78 | * | |
79 | * If no ancestor relationship: | |
80 | * if (dentry1 < dentry2) | |
81 | * dentry1->d_lock | |
82 | * dentry2->d_lock | |
83 | */ | |
84 | int sysctl_vfs_cache_pressure __read_mostly = 100; | |
85 | EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure); | |
86 | ||
87 | __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); | |
88 | ||
89 | EXPORT_SYMBOL(rename_lock); | |
90 | ||
91 | static struct kmem_cache *dentry_cache __read_mostly; | |
92 | ||
93 | /* | |
94 | * This is the single most critical data structure when it comes | |
95 | * to the dcache: the hashtable for lookups. Somebody should try | |
96 | * to make this good - I've just made it work. | |
97 | * | |
98 | * This hash-function tries to avoid losing too many bits of hash | |
99 | * information, yet avoid using a prime hash-size or similar. | |
100 | */ | |
101 | ||
102 | static unsigned int d_hash_mask __read_mostly; | |
103 | static unsigned int d_hash_shift __read_mostly; | |
104 | ||
105 | static struct hlist_bl_head *dentry_hashtable __read_mostly; | |
106 | ||
107 | static inline struct hlist_bl_head *d_hash(unsigned int hash) | |
108 | { | |
109 | return dentry_hashtable + (hash >> (32 - d_hash_shift)); | |
110 | } | |
111 | ||
112 | #define IN_LOOKUP_SHIFT 10 | |
113 | static struct hlist_bl_head in_lookup_hashtable[1 << IN_LOOKUP_SHIFT]; | |
114 | ||
115 | static inline struct hlist_bl_head *in_lookup_hash(const struct dentry *parent, | |
116 | unsigned int hash) | |
117 | { | |
118 | hash += (unsigned long) parent / L1_CACHE_BYTES; | |
119 | return in_lookup_hashtable + hash_32(hash, IN_LOOKUP_SHIFT); | |
120 | } | |
121 | ||
122 | ||
123 | /* Statistics gathering. */ | |
124 | struct dentry_stat_t dentry_stat = { | |
125 | .age_limit = 45, | |
126 | }; | |
127 | ||
128 | static DEFINE_PER_CPU(long, nr_dentry); | |
129 | static DEFINE_PER_CPU(long, nr_dentry_unused); | |
130 | ||
131 | #if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) | |
132 | ||
133 | /* | |
134 | * Here we resort to our own counters instead of using generic per-cpu counters | |
135 | * for consistency with what the vfs inode code does. We are expected to harvest | |
136 | * better code and performance by having our own specialized counters. | |
137 | * | |
138 | * Please note that the loop is done over all possible CPUs, not over all online | |
139 | * CPUs. The reason for this is that we don't want to play games with CPUs going | |
140 | * on and off. If one of them goes off, we will just keep their counters. | |
141 | * | |
142 | * glommer: See cffbc8a for details, and if you ever intend to change this, | |
143 | * please update all vfs counters to match. | |
144 | */ | |
145 | static long get_nr_dentry(void) | |
146 | { | |
147 | int i; | |
148 | long sum = 0; | |
149 | for_each_possible_cpu(i) | |
150 | sum += per_cpu(nr_dentry, i); | |
151 | return sum < 0 ? 0 : sum; | |
152 | } | |
153 | ||
154 | static long get_nr_dentry_unused(void) | |
155 | { | |
156 | int i; | |
157 | long sum = 0; | |
158 | for_each_possible_cpu(i) | |
159 | sum += per_cpu(nr_dentry_unused, i); | |
160 | return sum < 0 ? 0 : sum; | |
161 | } | |
162 | ||
163 | int proc_nr_dentry(struct ctl_table *table, int write, void __user *buffer, | |
164 | size_t *lenp, loff_t *ppos) | |
165 | { | |
166 | dentry_stat.nr_dentry = get_nr_dentry(); | |
167 | dentry_stat.nr_unused = get_nr_dentry_unused(); | |
168 | return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); | |
169 | } | |
170 | #endif | |
171 | ||
172 | /* | |
173 | * Compare 2 name strings, return 0 if they match, otherwise non-zero. | |
174 | * The strings are both count bytes long, and count is non-zero. | |
175 | */ | |
176 | #ifdef CONFIG_DCACHE_WORD_ACCESS | |
177 | ||
178 | #include <asm/word-at-a-time.h> | |
179 | /* | |
180 | * NOTE! 'cs' and 'scount' come from a dentry, so it has a | |
181 | * aligned allocation for this particular component. We don't | |
182 | * strictly need the load_unaligned_zeropad() safety, but it | |
183 | * doesn't hurt either. | |
184 | * | |
185 | * In contrast, 'ct' and 'tcount' can be from a pathname, and do | |
186 | * need the careful unaligned handling. | |
187 | */ | |
188 | static inline int dentry_string_cmp(const unsigned char *cs, const unsigned char *ct, unsigned tcount) | |
189 | { | |
190 | unsigned long a,b,mask; | |
191 | ||
192 | for (;;) { | |
193 | a = *(unsigned long *)cs; | |
194 | b = load_unaligned_zeropad(ct); | |
195 | if (tcount < sizeof(unsigned long)) | |
196 | break; | |
197 | if (unlikely(a != b)) | |
198 | return 1; | |
199 | cs += sizeof(unsigned long); | |
200 | ct += sizeof(unsigned long); | |
201 | tcount -= sizeof(unsigned long); | |
202 | if (!tcount) | |
203 | return 0; | |
204 | } | |
205 | mask = bytemask_from_count(tcount); | |
206 | return unlikely(!!((a ^ b) & mask)); | |
207 | } | |
208 | ||
209 | #else | |
210 | ||
211 | static inline int dentry_string_cmp(const unsigned char *cs, const unsigned char *ct, unsigned tcount) | |
212 | { | |
213 | do { | |
214 | if (*cs != *ct) | |
215 | return 1; | |
216 | cs++; | |
217 | ct++; | |
218 | tcount--; | |
219 | } while (tcount); | |
220 | return 0; | |
221 | } | |
222 | ||
223 | #endif | |
224 | ||
225 | static inline int dentry_cmp(const struct dentry *dentry, const unsigned char *ct, unsigned tcount) | |
226 | { | |
227 | /* | |
228 | * Be careful about RCU walk racing with rename: | |
229 | * use 'lockless_dereference' to fetch the name pointer. | |
230 | * | |
231 | * NOTE! Even if a rename will mean that the length | |
232 | * was not loaded atomically, we don't care. The | |
233 | * RCU walk will check the sequence count eventually, | |
234 | * and catch it. And we won't overrun the buffer, | |
235 | * because we're reading the name pointer atomically, | |
236 | * and a dentry name is guaranteed to be properly | |
237 | * terminated with a NUL byte. | |
238 | * | |
239 | * End result: even if 'len' is wrong, we'll exit | |
240 | * early because the data cannot match (there can | |
241 | * be no NUL in the ct/tcount data) | |
242 | */ | |
243 | const unsigned char *cs = lockless_dereference(dentry->d_name.name); | |
244 | ||
245 | return dentry_string_cmp(cs, ct, tcount); | |
246 | } | |
247 | ||
248 | struct external_name { | |
249 | union { | |
250 | atomic_t count; | |
251 | struct rcu_head head; | |
252 | } u; | |
253 | unsigned char name[]; | |
254 | }; | |
255 | ||
256 | static inline struct external_name *external_name(struct dentry *dentry) | |
257 | { | |
258 | return container_of(dentry->d_name.name, struct external_name, name[0]); | |
259 | } | |
260 | ||
261 | static void __d_free(struct rcu_head *head) | |
262 | { | |
263 | struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); | |
264 | ||
265 | kmem_cache_free(dentry_cache, dentry); | |
266 | } | |
267 | ||
268 | static void __d_free_external(struct rcu_head *head) | |
269 | { | |
270 | struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); | |
271 | kfree(external_name(dentry)); | |
272 | kmem_cache_free(dentry_cache, dentry); | |
273 | } | |
274 | ||
275 | static inline int dname_external(const struct dentry *dentry) | |
276 | { | |
277 | return dentry->d_name.name != dentry->d_iname; | |
278 | } | |
279 | ||
280 | static inline void __d_set_inode_and_type(struct dentry *dentry, | |
281 | struct inode *inode, | |
282 | unsigned type_flags) | |
283 | { | |
284 | unsigned flags; | |
285 | ||
286 | dentry->d_inode = inode; | |
287 | flags = READ_ONCE(dentry->d_flags); | |
288 | flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); | |
289 | flags |= type_flags; | |
290 | WRITE_ONCE(dentry->d_flags, flags); | |
291 | } | |
292 | ||
293 | static inline void __d_clear_type_and_inode(struct dentry *dentry) | |
294 | { | |
295 | unsigned flags = READ_ONCE(dentry->d_flags); | |
296 | ||
297 | flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); | |
298 | WRITE_ONCE(dentry->d_flags, flags); | |
299 | dentry->d_inode = NULL; | |
300 | } | |
301 | ||
302 | static void dentry_free(struct dentry *dentry) | |
303 | { | |
304 | WARN_ON(!hlist_unhashed(&dentry->d_u.d_alias)); | |
305 | if (unlikely(dname_external(dentry))) { | |
306 | struct external_name *p = external_name(dentry); | |
307 | if (likely(atomic_dec_and_test(&p->u.count))) { | |
308 | call_rcu(&dentry->d_u.d_rcu, __d_free_external); | |
309 | return; | |
310 | } | |
311 | } | |
312 | /* if dentry was never visible to RCU, immediate free is OK */ | |
313 | if (!(dentry->d_flags & DCACHE_RCUACCESS)) | |
314 | __d_free(&dentry->d_u.d_rcu); | |
315 | else | |
316 | call_rcu(&dentry->d_u.d_rcu, __d_free); | |
317 | } | |
318 | ||
319 | /* | |
320 | * Release the dentry's inode, using the filesystem | |
321 | * d_iput() operation if defined. | |
322 | */ | |
323 | static void dentry_unlink_inode(struct dentry * dentry) | |
324 | __releases(dentry->d_lock) | |
325 | __releases(dentry->d_inode->i_lock) | |
326 | { | |
327 | struct inode *inode = dentry->d_inode; | |
328 | bool hashed = !d_unhashed(dentry); | |
329 | ||
330 | if (hashed) | |
331 | raw_write_seqcount_begin(&dentry->d_seq); | |
332 | __d_clear_type_and_inode(dentry); | |
333 | hlist_del_init(&dentry->d_u.d_alias); | |
334 | if (hashed) | |
335 | raw_write_seqcount_end(&dentry->d_seq); | |
336 | spin_unlock(&dentry->d_lock); | |
337 | spin_unlock(&inode->i_lock); | |
338 | if (!inode->i_nlink) | |
339 | fsnotify_inoderemove(inode); | |
340 | if (dentry->d_op && dentry->d_op->d_iput) | |
341 | dentry->d_op->d_iput(dentry, inode); | |
342 | else | |
343 | iput(inode); | |
344 | } | |
345 | ||
346 | /* | |
347 | * The DCACHE_LRU_LIST bit is set whenever the 'd_lru' entry | |
348 | * is in use - which includes both the "real" per-superblock | |
349 | * LRU list _and_ the DCACHE_SHRINK_LIST use. | |
350 | * | |
351 | * The DCACHE_SHRINK_LIST bit is set whenever the dentry is | |
352 | * on the shrink list (ie not on the superblock LRU list). | |
353 | * | |
354 | * The per-cpu "nr_dentry_unused" counters are updated with | |
355 | * the DCACHE_LRU_LIST bit. | |
356 | * | |
357 | * These helper functions make sure we always follow the | |
358 | * rules. d_lock must be held by the caller. | |
359 | */ | |
360 | #define D_FLAG_VERIFY(dentry,x) WARN_ON_ONCE(((dentry)->d_flags & (DCACHE_LRU_LIST | DCACHE_SHRINK_LIST)) != (x)) | |
361 | static void d_lru_add(struct dentry *dentry) | |
362 | { | |
363 | D_FLAG_VERIFY(dentry, 0); | |
364 | dentry->d_flags |= DCACHE_LRU_LIST; | |
365 | this_cpu_inc(nr_dentry_unused); | |
366 | WARN_ON_ONCE(!list_lru_add(&dentry->d_sb->s_dentry_lru, &dentry->d_lru)); | |
367 | } | |
368 | ||
369 | static void d_lru_del(struct dentry *dentry) | |
370 | { | |
371 | D_FLAG_VERIFY(dentry, DCACHE_LRU_LIST); | |
372 | dentry->d_flags &= ~DCACHE_LRU_LIST; | |
373 | this_cpu_dec(nr_dentry_unused); | |
374 | WARN_ON_ONCE(!list_lru_del(&dentry->d_sb->s_dentry_lru, &dentry->d_lru)); | |
375 | } | |
376 | ||
377 | static void d_shrink_del(struct dentry *dentry) | |
378 | { | |
379 | D_FLAG_VERIFY(dentry, DCACHE_SHRINK_LIST | DCACHE_LRU_LIST); | |
380 | list_del_init(&dentry->d_lru); | |
381 | dentry->d_flags &= ~(DCACHE_SHRINK_LIST | DCACHE_LRU_LIST); | |
382 | this_cpu_dec(nr_dentry_unused); | |
383 | } | |
384 | ||
385 | static void d_shrink_add(struct dentry *dentry, struct list_head *list) | |
386 | { | |
387 | D_FLAG_VERIFY(dentry, 0); | |
388 | list_add(&dentry->d_lru, list); | |
389 | dentry->d_flags |= DCACHE_SHRINK_LIST | DCACHE_LRU_LIST; | |
390 | this_cpu_inc(nr_dentry_unused); | |
391 | } | |
392 | ||
393 | /* | |
394 | * These can only be called under the global LRU lock, ie during the | |
395 | * callback for freeing the LRU list. "isolate" removes it from the | |
396 | * LRU lists entirely, while shrink_move moves it to the indicated | |
397 | * private list. | |
398 | */ | |
399 | static void d_lru_isolate(struct list_lru_one *lru, struct dentry *dentry) | |
400 | { | |
401 | D_FLAG_VERIFY(dentry, DCACHE_LRU_LIST); | |
402 | dentry->d_flags &= ~DCACHE_LRU_LIST; | |
403 | this_cpu_dec(nr_dentry_unused); | |
404 | list_lru_isolate(lru, &dentry->d_lru); | |
405 | } | |
406 | ||
407 | static void d_lru_shrink_move(struct list_lru_one *lru, struct dentry *dentry, | |
408 | struct list_head *list) | |
409 | { | |
410 | D_FLAG_VERIFY(dentry, DCACHE_LRU_LIST); | |
411 | dentry->d_flags |= DCACHE_SHRINK_LIST; | |
412 | list_lru_isolate_move(lru, &dentry->d_lru, list); | |
413 | } | |
414 | ||
415 | /* | |
416 | * dentry_lru_(add|del)_list) must be called with d_lock held. | |
417 | */ | |
418 | static void dentry_lru_add(struct dentry *dentry) | |
419 | { | |
420 | if (unlikely(!(dentry->d_flags & DCACHE_LRU_LIST))) | |
421 | d_lru_add(dentry); | |
422 | } | |
423 | ||
424 | /** | |
425 | * d_drop - drop a dentry | |
426 | * @dentry: dentry to drop | |
427 | * | |
428 | * d_drop() unhashes the entry from the parent dentry hashes, so that it won't | |
429 | * be found through a VFS lookup any more. Note that this is different from | |
430 | * deleting the dentry - d_delete will try to mark the dentry negative if | |
431 | * possible, giving a successful _negative_ lookup, while d_drop will | |
432 | * just make the cache lookup fail. | |
433 | * | |
434 | * d_drop() is used mainly for stuff that wants to invalidate a dentry for some | |
435 | * reason (NFS timeouts or autofs deletes). | |
436 | * | |
437 | * __d_drop requires dentry->d_lock. | |
438 | */ | |
439 | void __d_drop(struct dentry *dentry) | |
440 | { | |
441 | if (!d_unhashed(dentry)) { | |
442 | struct hlist_bl_head *b; | |
443 | /* | |
444 | * Hashed dentries are normally on the dentry hashtable, | |
445 | * with the exception of those newly allocated by | |
446 | * d_obtain_alias, which are always IS_ROOT: | |
447 | */ | |
448 | if (unlikely(IS_ROOT(dentry))) | |
449 | b = &dentry->d_sb->s_anon; | |
450 | else | |
451 | b = d_hash(dentry->d_name.hash); | |
452 | ||
453 | hlist_bl_lock(b); | |
454 | __hlist_bl_del(&dentry->d_hash); | |
455 | dentry->d_hash.pprev = NULL; | |
456 | hlist_bl_unlock(b); | |
457 | /* After this call, in-progress rcu-walk path lookup will fail. */ | |
458 | write_seqcount_invalidate(&dentry->d_seq); | |
459 | } | |
460 | } | |
461 | EXPORT_SYMBOL(__d_drop); | |
462 | ||
463 | void d_drop(struct dentry *dentry) | |
464 | { | |
465 | spin_lock(&dentry->d_lock); | |
466 | __d_drop(dentry); | |
467 | spin_unlock(&dentry->d_lock); | |
468 | } | |
469 | EXPORT_SYMBOL(d_drop); | |
470 | ||
471 | static inline void dentry_unlist(struct dentry *dentry, struct dentry *parent) | |
472 | { | |
473 | struct dentry *next; | |
474 | /* | |
475 | * Inform d_walk() and shrink_dentry_list() that we are no longer | |
476 | * attached to the dentry tree | |
477 | */ | |
478 | dentry->d_flags |= DCACHE_DENTRY_KILLED; | |
479 | if (unlikely(list_empty(&dentry->d_child))) | |
480 | return; | |
481 | __list_del_entry(&dentry->d_child); | |
482 | /* | |
483 | * Cursors can move around the list of children. While we'd been | |
484 | * a normal list member, it didn't matter - ->d_child.next would've | |
485 | * been updated. However, from now on it won't be and for the | |
486 | * things like d_walk() it might end up with a nasty surprise. | |
487 | * Normally d_walk() doesn't care about cursors moving around - | |
488 | * ->d_lock on parent prevents that and since a cursor has no children | |
489 | * of its own, we get through it without ever unlocking the parent. | |
490 | * There is one exception, though - if we ascend from a child that | |
491 | * gets killed as soon as we unlock it, the next sibling is found | |
492 | * using the value left in its ->d_child.next. And if _that_ | |
493 | * pointed to a cursor, and cursor got moved (e.g. by lseek()) | |
494 | * before d_walk() regains parent->d_lock, we'll end up skipping | |
495 | * everything the cursor had been moved past. | |
496 | * | |
497 | * Solution: make sure that the pointer left behind in ->d_child.next | |
498 | * points to something that won't be moving around. I.e. skip the | |
499 | * cursors. | |
500 | */ | |
501 | while (dentry->d_child.next != &parent->d_subdirs) { | |
502 | next = list_entry(dentry->d_child.next, struct dentry, d_child); | |
503 | if (likely(!(next->d_flags & DCACHE_DENTRY_CURSOR))) | |
504 | break; | |
505 | dentry->d_child.next = next->d_child.next; | |
506 | } | |
507 | } | |
508 | ||
509 | static void __dentry_kill(struct dentry *dentry) | |
510 | { | |
511 | struct dentry *parent = NULL; | |
512 | bool can_free = true; | |
513 | if (!IS_ROOT(dentry)) | |
514 | parent = dentry->d_parent; | |
515 | ||
516 | /* | |
517 | * The dentry is now unrecoverably dead to the world. | |
518 | */ | |
519 | lockref_mark_dead(&dentry->d_lockref); | |
520 | ||
521 | /* | |
522 | * inform the fs via d_prune that this dentry is about to be | |
523 | * unhashed and destroyed. | |
524 | */ | |
525 | if (dentry->d_flags & DCACHE_OP_PRUNE) | |
526 | dentry->d_op->d_prune(dentry); | |
527 | ||
528 | if (dentry->d_flags & DCACHE_LRU_LIST) { | |
529 | if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) | |
530 | d_lru_del(dentry); | |
531 | } | |
532 | /* if it was on the hash then remove it */ | |
533 | __d_drop(dentry); | |
534 | dentry_unlist(dentry, parent); | |
535 | if (parent) | |
536 | spin_unlock(&parent->d_lock); | |
537 | if (dentry->d_inode) | |
538 | dentry_unlink_inode(dentry); | |
539 | else | |
540 | spin_unlock(&dentry->d_lock); | |
541 | this_cpu_dec(nr_dentry); | |
542 | if (dentry->d_op && dentry->d_op->d_release) | |
543 | dentry->d_op->d_release(dentry); | |
544 | ||
545 | spin_lock(&dentry->d_lock); | |
546 | if (dentry->d_flags & DCACHE_SHRINK_LIST) { | |
547 | dentry->d_flags |= DCACHE_MAY_FREE; | |
548 | can_free = false; | |
549 | } | |
550 | spin_unlock(&dentry->d_lock); | |
551 | if (likely(can_free)) | |
552 | dentry_free(dentry); | |
553 | } | |
554 | ||
555 | /* | |
556 | * Finish off a dentry we've decided to kill. | |
557 | * dentry->d_lock must be held, returns with it unlocked. | |
558 | * If ref is non-zero, then decrement the refcount too. | |
559 | * Returns dentry requiring refcount drop, or NULL if we're done. | |
560 | */ | |
561 | static struct dentry *dentry_kill(struct dentry *dentry) | |
562 | __releases(dentry->d_lock) | |
563 | { | |
564 | struct inode *inode = dentry->d_inode; | |
565 | struct dentry *parent = NULL; | |
566 | ||
567 | if (inode && unlikely(!spin_trylock(&inode->i_lock))) | |
568 | goto failed; | |
569 | ||
570 | if (!IS_ROOT(dentry)) { | |
571 | parent = dentry->d_parent; | |
572 | if (unlikely(!spin_trylock(&parent->d_lock))) { | |
573 | if (inode) | |
574 | spin_unlock(&inode->i_lock); | |
575 | goto failed; | |
576 | } | |
577 | } | |
578 | ||
579 | __dentry_kill(dentry); | |
580 | return parent; | |
581 | ||
582 | failed: | |
583 | spin_unlock(&dentry->d_lock); | |
584 | return dentry; /* try again with same dentry */ | |
585 | } | |
586 | ||
587 | static inline struct dentry *lock_parent(struct dentry *dentry) | |
588 | { | |
589 | struct dentry *parent = dentry->d_parent; | |
590 | if (IS_ROOT(dentry)) | |
591 | return NULL; | |
592 | if (unlikely(dentry->d_lockref.count < 0)) | |
593 | return NULL; | |
594 | if (likely(spin_trylock(&parent->d_lock))) | |
595 | return parent; | |
596 | rcu_read_lock(); | |
597 | spin_unlock(&dentry->d_lock); | |
598 | again: | |
599 | parent = ACCESS_ONCE(dentry->d_parent); | |
600 | spin_lock(&parent->d_lock); | |
601 | /* | |
602 | * We can't blindly lock dentry until we are sure | |
603 | * that we won't violate the locking order. | |
604 | * Any changes of dentry->d_parent must have | |
605 | * been done with parent->d_lock held, so | |
606 | * spin_lock() above is enough of a barrier | |
607 | * for checking if it's still our child. | |
608 | */ | |
609 | if (unlikely(parent != dentry->d_parent)) { | |
610 | spin_unlock(&parent->d_lock); | |
611 | goto again; | |
612 | } | |
613 | rcu_read_unlock(); | |
614 | if (parent != dentry) | |
615 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); | |
616 | else | |
617 | parent = NULL; | |
618 | return parent; | |
619 | } | |
620 | ||
621 | /* | |
622 | * Try to do a lockless dput(), and return whether that was successful. | |
623 | * | |
624 | * If unsuccessful, we return false, having already taken the dentry lock. | |
625 | * | |
626 | * The caller needs to hold the RCU read lock, so that the dentry is | |
627 | * guaranteed to stay around even if the refcount goes down to zero! | |
628 | */ | |
629 | static inline bool fast_dput(struct dentry *dentry) | |
630 | { | |
631 | int ret; | |
632 | unsigned int d_flags; | |
633 | ||
634 | /* | |
635 | * If we have a d_op->d_delete() operation, we sould not | |
636 | * let the dentry count go to zero, so use "put_or_lock". | |
637 | */ | |
638 | if (unlikely(dentry->d_flags & DCACHE_OP_DELETE)) | |
639 | return lockref_put_or_lock(&dentry->d_lockref); | |
640 | ||
641 | /* | |
642 | * .. otherwise, we can try to just decrement the | |
643 | * lockref optimistically. | |
644 | */ | |
645 | ret = lockref_put_return(&dentry->d_lockref); | |
646 | ||
647 | /* | |
648 | * If the lockref_put_return() failed due to the lock being held | |
649 | * by somebody else, the fast path has failed. We will need to | |
650 | * get the lock, and then check the count again. | |
651 | */ | |
652 | if (unlikely(ret < 0)) { | |
653 | spin_lock(&dentry->d_lock); | |
654 | if (dentry->d_lockref.count > 1) { | |
655 | dentry->d_lockref.count--; | |
656 | spin_unlock(&dentry->d_lock); | |
657 | return 1; | |
658 | } | |
659 | return 0; | |
660 | } | |
661 | ||
662 | /* | |
663 | * If we weren't the last ref, we're done. | |
664 | */ | |
665 | if (ret) | |
666 | return 1; | |
667 | ||
668 | /* | |
669 | * Careful, careful. The reference count went down | |
670 | * to zero, but we don't hold the dentry lock, so | |
671 | * somebody else could get it again, and do another | |
672 | * dput(), and we need to not race with that. | |
673 | * | |
674 | * However, there is a very special and common case | |
675 | * where we don't care, because there is nothing to | |
676 | * do: the dentry is still hashed, it does not have | |
677 | * a 'delete' op, and it's referenced and already on | |
678 | * the LRU list. | |
679 | * | |
680 | * NOTE! Since we aren't locked, these values are | |
681 | * not "stable". However, it is sufficient that at | |
682 | * some point after we dropped the reference the | |
683 | * dentry was hashed and the flags had the proper | |
684 | * value. Other dentry users may have re-gotten | |
685 | * a reference to the dentry and change that, but | |
686 | * our work is done - we can leave the dentry | |
687 | * around with a zero refcount. | |
688 | */ | |
689 | smp_rmb(); | |
690 | d_flags = ACCESS_ONCE(dentry->d_flags); | |
691 | d_flags &= DCACHE_REFERENCED | DCACHE_LRU_LIST | DCACHE_DISCONNECTED; | |
692 | ||
693 | /* Nothing to do? Dropping the reference was all we needed? */ | |
694 | if (d_flags == (DCACHE_REFERENCED | DCACHE_LRU_LIST) && !d_unhashed(dentry)) | |
695 | return 1; | |
696 | ||
697 | /* | |
698 | * Not the fast normal case? Get the lock. We've already decremented | |
699 | * the refcount, but we'll need to re-check the situation after | |
700 | * getting the lock. | |
701 | */ | |
702 | spin_lock(&dentry->d_lock); | |
703 | ||
704 | /* | |
705 | * Did somebody else grab a reference to it in the meantime, and | |
706 | * we're no longer the last user after all? Alternatively, somebody | |
707 | * else could have killed it and marked it dead. Either way, we | |
708 | * don't need to do anything else. | |
709 | */ | |
710 | if (dentry->d_lockref.count) { | |
711 | spin_unlock(&dentry->d_lock); | |
712 | return 1; | |
713 | } | |
714 | ||
715 | /* | |
716 | * Re-get the reference we optimistically dropped. We hold the | |
717 | * lock, and we just tested that it was zero, so we can just | |
718 | * set it to 1. | |
719 | */ | |
720 | dentry->d_lockref.count = 1; | |
721 | return 0; | |
722 | } | |
723 | ||
724 | ||
725 | /* | |
726 | * This is dput | |
727 | * | |
728 | * This is complicated by the fact that we do not want to put | |
729 | * dentries that are no longer on any hash chain on the unused | |
730 | * list: we'd much rather just get rid of them immediately. | |
731 | * | |
732 | * However, that implies that we have to traverse the dentry | |
733 | * tree upwards to the parents which might _also_ now be | |
734 | * scheduled for deletion (it may have been only waiting for | |
735 | * its last child to go away). | |
736 | * | |
737 | * This tail recursion is done by hand as we don't want to depend | |
738 | * on the compiler to always get this right (gcc generally doesn't). | |
739 | * Real recursion would eat up our stack space. | |
740 | */ | |
741 | ||
742 | /* | |
743 | * dput - release a dentry | |
744 | * @dentry: dentry to release | |
745 | * | |
746 | * Release a dentry. This will drop the usage count and if appropriate | |
747 | * call the dentry unlink method as well as removing it from the queues and | |
748 | * releasing its resources. If the parent dentries were scheduled for release | |
749 | * they too may now get deleted. | |
750 | */ | |
751 | void dput(struct dentry *dentry) | |
752 | { | |
753 | if (unlikely(!dentry)) | |
754 | return; | |
755 | ||
756 | repeat: | |
757 | might_sleep(); | |
758 | ||
759 | rcu_read_lock(); | |
760 | if (likely(fast_dput(dentry))) { | |
761 | rcu_read_unlock(); | |
762 | return; | |
763 | } | |
764 | ||
765 | /* Slow case: now with the dentry lock held */ | |
766 | rcu_read_unlock(); | |
767 | ||
768 | WARN_ON(d_in_lookup(dentry)); | |
769 | ||
770 | /* Unreachable? Get rid of it */ | |
771 | if (unlikely(d_unhashed(dentry))) | |
772 | goto kill_it; | |
773 | ||
774 | if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) | |
775 | goto kill_it; | |
776 | ||
777 | if (unlikely(dentry->d_flags & DCACHE_OP_DELETE)) { | |
778 | if (dentry->d_op->d_delete(dentry)) | |
779 | goto kill_it; | |
780 | } | |
781 | ||
782 | if (!(dentry->d_flags & DCACHE_REFERENCED)) | |
783 | dentry->d_flags |= DCACHE_REFERENCED; | |
784 | dentry_lru_add(dentry); | |
785 | ||
786 | dentry->d_lockref.count--; | |
787 | spin_unlock(&dentry->d_lock); | |
788 | return; | |
789 | ||
790 | kill_it: | |
791 | dentry = dentry_kill(dentry); | |
792 | if (dentry) { | |
793 | cond_resched(); | |
794 | goto repeat; | |
795 | } | |
796 | } | |
797 | EXPORT_SYMBOL(dput); | |
798 | ||
799 | ||
800 | /* This must be called with d_lock held */ | |
801 | static inline void __dget_dlock(struct dentry *dentry) | |
802 | { | |
803 | dentry->d_lockref.count++; | |
804 | } | |
805 | ||
806 | static inline void __dget(struct dentry *dentry) | |
807 | { | |
808 | lockref_get(&dentry->d_lockref); | |
809 | } | |
810 | ||
811 | struct dentry *dget_parent(struct dentry *dentry) | |
812 | { | |
813 | int gotref; | |
814 | struct dentry *ret; | |
815 | ||
816 | /* | |
817 | * Do optimistic parent lookup without any | |
818 | * locking. | |
819 | */ | |
820 | rcu_read_lock(); | |
821 | ret = ACCESS_ONCE(dentry->d_parent); | |
822 | gotref = lockref_get_not_zero(&ret->d_lockref); | |
823 | rcu_read_unlock(); | |
824 | if (likely(gotref)) { | |
825 | if (likely(ret == ACCESS_ONCE(dentry->d_parent))) | |
826 | return ret; | |
827 | dput(ret); | |
828 | } | |
829 | ||
830 | repeat: | |
831 | /* | |
832 | * Don't need rcu_dereference because we re-check it was correct under | |
833 | * the lock. | |
834 | */ | |
835 | rcu_read_lock(); | |
836 | ret = dentry->d_parent; | |
837 | spin_lock(&ret->d_lock); | |
838 | if (unlikely(ret != dentry->d_parent)) { | |
839 | spin_unlock(&ret->d_lock); | |
840 | rcu_read_unlock(); | |
841 | goto repeat; | |
842 | } | |
843 | rcu_read_unlock(); | |
844 | BUG_ON(!ret->d_lockref.count); | |
845 | ret->d_lockref.count++; | |
846 | spin_unlock(&ret->d_lock); | |
847 | return ret; | |
848 | } | |
849 | EXPORT_SYMBOL(dget_parent); | |
850 | ||
851 | /** | |
852 | * d_find_alias - grab a hashed alias of inode | |
853 | * @inode: inode in question | |
854 | * | |
855 | * If inode has a hashed alias, or is a directory and has any alias, | |
856 | * acquire the reference to alias and return it. Otherwise return NULL. | |
857 | * Notice that if inode is a directory there can be only one alias and | |
858 | * it can be unhashed only if it has no children, or if it is the root | |
859 | * of a filesystem, or if the directory was renamed and d_revalidate | |
860 | * was the first vfs operation to notice. | |
861 | * | |
862 | * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer | |
863 | * any other hashed alias over that one. | |
864 | */ | |
865 | static struct dentry *__d_find_alias(struct inode *inode) | |
866 | { | |
867 | struct dentry *alias, *discon_alias; | |
868 | ||
869 | again: | |
870 | discon_alias = NULL; | |
871 | hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) { | |
872 | spin_lock(&alias->d_lock); | |
873 | if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { | |
874 | if (IS_ROOT(alias) && | |
875 | (alias->d_flags & DCACHE_DISCONNECTED)) { | |
876 | discon_alias = alias; | |
877 | } else { | |
878 | __dget_dlock(alias); | |
879 | spin_unlock(&alias->d_lock); | |
880 | return alias; | |
881 | } | |
882 | } | |
883 | spin_unlock(&alias->d_lock); | |
884 | } | |
885 | if (discon_alias) { | |
886 | alias = discon_alias; | |
887 | spin_lock(&alias->d_lock); | |
888 | if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { | |
889 | __dget_dlock(alias); | |
890 | spin_unlock(&alias->d_lock); | |
891 | return alias; | |
892 | } | |
893 | spin_unlock(&alias->d_lock); | |
894 | goto again; | |
895 | } | |
896 | return NULL; | |
897 | } | |
898 | ||
899 | struct dentry *d_find_alias(struct inode *inode) | |
900 | { | |
901 | struct dentry *de = NULL; | |
902 | ||
903 | if (!hlist_empty(&inode->i_dentry)) { | |
904 | spin_lock(&inode->i_lock); | |
905 | de = __d_find_alias(inode); | |
906 | spin_unlock(&inode->i_lock); | |
907 | } | |
908 | return de; | |
909 | } | |
910 | EXPORT_SYMBOL(d_find_alias); | |
911 | ||
912 | /* | |
913 | * Try to kill dentries associated with this inode. | |
914 | * WARNING: you must own a reference to inode. | |
915 | */ | |
916 | void d_prune_aliases(struct inode *inode) | |
917 | { | |
918 | struct dentry *dentry; | |
919 | restart: | |
920 | spin_lock(&inode->i_lock); | |
921 | hlist_for_each_entry(dentry, &inode->i_dentry, d_u.d_alias) { | |
922 | spin_lock(&dentry->d_lock); | |
923 | if (!dentry->d_lockref.count) { | |
924 | struct dentry *parent = lock_parent(dentry); | |
925 | if (likely(!dentry->d_lockref.count)) { | |
926 | __dentry_kill(dentry); | |
927 | dput(parent); | |
928 | goto restart; | |
929 | } | |
930 | if (parent) | |
931 | spin_unlock(&parent->d_lock); | |
932 | } | |
933 | spin_unlock(&dentry->d_lock); | |
934 | } | |
935 | spin_unlock(&inode->i_lock); | |
936 | } | |
937 | EXPORT_SYMBOL(d_prune_aliases); | |
938 | ||
939 | static void shrink_dentry_list(struct list_head *list) | |
940 | { | |
941 | struct dentry *dentry, *parent; | |
942 | ||
943 | while (!list_empty(list)) { | |
944 | struct inode *inode; | |
945 | dentry = list_entry(list->prev, struct dentry, d_lru); | |
946 | spin_lock(&dentry->d_lock); | |
947 | parent = lock_parent(dentry); | |
948 | ||
949 | /* | |
950 | * The dispose list is isolated and dentries are not accounted | |
951 | * to the LRU here, so we can simply remove it from the list | |
952 | * here regardless of whether it is referenced or not. | |
953 | */ | |
954 | d_shrink_del(dentry); | |
955 | ||
956 | /* | |
957 | * We found an inuse dentry which was not removed from | |
958 | * the LRU because of laziness during lookup. Do not free it. | |
959 | */ | |
960 | if (dentry->d_lockref.count > 0) { | |
961 | spin_unlock(&dentry->d_lock); | |
962 | if (parent) | |
963 | spin_unlock(&parent->d_lock); | |
964 | continue; | |
965 | } | |
966 | ||
967 | ||
968 | if (unlikely(dentry->d_flags & DCACHE_DENTRY_KILLED)) { | |
969 | bool can_free = dentry->d_flags & DCACHE_MAY_FREE; | |
970 | spin_unlock(&dentry->d_lock); | |
971 | if (parent) | |
972 | spin_unlock(&parent->d_lock); | |
973 | if (can_free) | |
974 | dentry_free(dentry); | |
975 | continue; | |
976 | } | |
977 | ||
978 | inode = dentry->d_inode; | |
979 | if (inode && unlikely(!spin_trylock(&inode->i_lock))) { | |
980 | d_shrink_add(dentry, list); | |
981 | spin_unlock(&dentry->d_lock); | |
982 | if (parent) | |
983 | spin_unlock(&parent->d_lock); | |
984 | continue; | |
985 | } | |
986 | ||
987 | __dentry_kill(dentry); | |
988 | ||
989 | /* | |
990 | * We need to prune ancestors too. This is necessary to prevent | |
991 | * quadratic behavior of shrink_dcache_parent(), but is also | |
992 | * expected to be beneficial in reducing dentry cache | |
993 | * fragmentation. | |
994 | */ | |
995 | dentry = parent; | |
996 | while (dentry && !lockref_put_or_lock(&dentry->d_lockref)) { | |
997 | parent = lock_parent(dentry); | |
998 | if (dentry->d_lockref.count != 1) { | |
999 | dentry->d_lockref.count--; | |
1000 | spin_unlock(&dentry->d_lock); | |
1001 | if (parent) | |
1002 | spin_unlock(&parent->d_lock); | |
1003 | break; | |
1004 | } | |
1005 | inode = dentry->d_inode; /* can't be NULL */ | |
1006 | if (unlikely(!spin_trylock(&inode->i_lock))) { | |
1007 | spin_unlock(&dentry->d_lock); | |
1008 | if (parent) | |
1009 | spin_unlock(&parent->d_lock); | |
1010 | cpu_relax(); | |
1011 | continue; | |
1012 | } | |
1013 | __dentry_kill(dentry); | |
1014 | dentry = parent; | |
1015 | } | |
1016 | } | |
1017 | } | |
1018 | ||
1019 | static enum lru_status dentry_lru_isolate(struct list_head *item, | |
1020 | struct list_lru_one *lru, spinlock_t *lru_lock, void *arg) | |
1021 | { | |
1022 | struct list_head *freeable = arg; | |
1023 | struct dentry *dentry = container_of(item, struct dentry, d_lru); | |
1024 | ||
1025 | ||
1026 | /* | |
1027 | * we are inverting the lru lock/dentry->d_lock here, | |
1028 | * so use a trylock. If we fail to get the lock, just skip | |
1029 | * it | |
1030 | */ | |
1031 | if (!spin_trylock(&dentry->d_lock)) | |
1032 | return LRU_SKIP; | |
1033 | ||
1034 | /* | |
1035 | * Referenced dentries are still in use. If they have active | |
1036 | * counts, just remove them from the LRU. Otherwise give them | |
1037 | * another pass through the LRU. | |
1038 | */ | |
1039 | if (dentry->d_lockref.count) { | |
1040 | d_lru_isolate(lru, dentry); | |
1041 | spin_unlock(&dentry->d_lock); | |
1042 | return LRU_REMOVED; | |
1043 | } | |
1044 | ||
1045 | if (dentry->d_flags & DCACHE_REFERENCED) { | |
1046 | dentry->d_flags &= ~DCACHE_REFERENCED; | |
1047 | spin_unlock(&dentry->d_lock); | |
1048 | ||
1049 | /* | |
1050 | * The list move itself will be made by the common LRU code. At | |
1051 | * this point, we've dropped the dentry->d_lock but keep the | |
1052 | * lru lock. This is safe to do, since every list movement is | |
1053 | * protected by the lru lock even if both locks are held. | |
1054 | * | |
1055 | * This is guaranteed by the fact that all LRU management | |
1056 | * functions are intermediated by the LRU API calls like | |
1057 | * list_lru_add and list_lru_del. List movement in this file | |
1058 | * only ever occur through this functions or through callbacks | |
1059 | * like this one, that are called from the LRU API. | |
1060 | * | |
1061 | * The only exceptions to this are functions like | |
1062 | * shrink_dentry_list, and code that first checks for the | |
1063 | * DCACHE_SHRINK_LIST flag. Those are guaranteed to be | |
1064 | * operating only with stack provided lists after they are | |
1065 | * properly isolated from the main list. It is thus, always a | |
1066 | * local access. | |
1067 | */ | |
1068 | return LRU_ROTATE; | |
1069 | } | |
1070 | ||
1071 | d_lru_shrink_move(lru, dentry, freeable); | |
1072 | spin_unlock(&dentry->d_lock); | |
1073 | ||
1074 | return LRU_REMOVED; | |
1075 | } | |
1076 | ||
1077 | /** | |
1078 | * prune_dcache_sb - shrink the dcache | |
1079 | * @sb: superblock | |
1080 | * @sc: shrink control, passed to list_lru_shrink_walk() | |
1081 | * | |
1082 | * Attempt to shrink the superblock dcache LRU by @sc->nr_to_scan entries. This | |
1083 | * is done when we need more memory and called from the superblock shrinker | |
1084 | * function. | |
1085 | * | |
1086 | * This function may fail to free any resources if all the dentries are in | |
1087 | * use. | |
1088 | */ | |
1089 | long prune_dcache_sb(struct super_block *sb, struct shrink_control *sc) | |
1090 | { | |
1091 | LIST_HEAD(dispose); | |
1092 | long freed; | |
1093 | ||
1094 | freed = list_lru_shrink_walk(&sb->s_dentry_lru, sc, | |
1095 | dentry_lru_isolate, &dispose); | |
1096 | shrink_dentry_list(&dispose); | |
1097 | return freed; | |
1098 | } | |
1099 | ||
1100 | static enum lru_status dentry_lru_isolate_shrink(struct list_head *item, | |
1101 | struct list_lru_one *lru, spinlock_t *lru_lock, void *arg) | |
1102 | { | |
1103 | struct list_head *freeable = arg; | |
1104 | struct dentry *dentry = container_of(item, struct dentry, d_lru); | |
1105 | ||
1106 | /* | |
1107 | * we are inverting the lru lock/dentry->d_lock here, | |
1108 | * so use a trylock. If we fail to get the lock, just skip | |
1109 | * it | |
1110 | */ | |
1111 | if (!spin_trylock(&dentry->d_lock)) | |
1112 | return LRU_SKIP; | |
1113 | ||
1114 | d_lru_shrink_move(lru, dentry, freeable); | |
1115 | spin_unlock(&dentry->d_lock); | |
1116 | ||
1117 | return LRU_REMOVED; | |
1118 | } | |
1119 | ||
1120 | ||
1121 | /** | |
1122 | * shrink_dcache_sb - shrink dcache for a superblock | |
1123 | * @sb: superblock | |
1124 | * | |
1125 | * Shrink the dcache for the specified super block. This is used to free | |
1126 | * the dcache before unmounting a file system. | |
1127 | */ | |
1128 | void shrink_dcache_sb(struct super_block *sb) | |
1129 | { | |
1130 | long freed; | |
1131 | ||
1132 | do { | |
1133 | LIST_HEAD(dispose); | |
1134 | ||
1135 | freed = list_lru_walk(&sb->s_dentry_lru, | |
1136 | dentry_lru_isolate_shrink, &dispose, UINT_MAX); | |
1137 | ||
1138 | this_cpu_sub(nr_dentry_unused, freed); | |
1139 | shrink_dentry_list(&dispose); | |
1140 | } while (freed > 0); | |
1141 | } | |
1142 | EXPORT_SYMBOL(shrink_dcache_sb); | |
1143 | ||
1144 | /** | |
1145 | * enum d_walk_ret - action to talke during tree walk | |
1146 | * @D_WALK_CONTINUE: contrinue walk | |
1147 | * @D_WALK_QUIT: quit walk | |
1148 | * @D_WALK_NORETRY: quit when retry is needed | |
1149 | * @D_WALK_SKIP: skip this dentry and its children | |
1150 | */ | |
1151 | enum d_walk_ret { | |
1152 | D_WALK_CONTINUE, | |
1153 | D_WALK_QUIT, | |
1154 | D_WALK_NORETRY, | |
1155 | D_WALK_SKIP, | |
1156 | }; | |
1157 | ||
1158 | /** | |
1159 | * d_walk - walk the dentry tree | |
1160 | * @parent: start of walk | |
1161 | * @data: data passed to @enter() and @finish() | |
1162 | * @enter: callback when first entering the dentry | |
1163 | * @finish: callback when successfully finished the walk | |
1164 | * | |
1165 | * The @enter() and @finish() callbacks are called with d_lock held. | |
1166 | */ | |
1167 | static void d_walk(struct dentry *parent, void *data, | |
1168 | enum d_walk_ret (*enter)(void *, struct dentry *), | |
1169 | void (*finish)(void *)) | |
1170 | { | |
1171 | struct dentry *this_parent; | |
1172 | struct list_head *next; | |
1173 | unsigned seq = 0; | |
1174 | enum d_walk_ret ret; | |
1175 | bool retry = true; | |
1176 | ||
1177 | again: | |
1178 | read_seqbegin_or_lock(&rename_lock, &seq); | |
1179 | this_parent = parent; | |
1180 | spin_lock(&this_parent->d_lock); | |
1181 | ||
1182 | ret = enter(data, this_parent); | |
1183 | switch (ret) { | |
1184 | case D_WALK_CONTINUE: | |
1185 | break; | |
1186 | case D_WALK_QUIT: | |
1187 | case D_WALK_SKIP: | |
1188 | goto out_unlock; | |
1189 | case D_WALK_NORETRY: | |
1190 | retry = false; | |
1191 | break; | |
1192 | } | |
1193 | repeat: | |
1194 | next = this_parent->d_subdirs.next; | |
1195 | resume: | |
1196 | while (next != &this_parent->d_subdirs) { | |
1197 | struct list_head *tmp = next; | |
1198 | struct dentry *dentry = list_entry(tmp, struct dentry, d_child); | |
1199 | next = tmp->next; | |
1200 | ||
1201 | if (unlikely(dentry->d_flags & DCACHE_DENTRY_CURSOR)) | |
1202 | continue; | |
1203 | ||
1204 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); | |
1205 | ||
1206 | ret = enter(data, dentry); | |
1207 | switch (ret) { | |
1208 | case D_WALK_CONTINUE: | |
1209 | break; | |
1210 | case D_WALK_QUIT: | |
1211 | spin_unlock(&dentry->d_lock); | |
1212 | goto out_unlock; | |
1213 | case D_WALK_NORETRY: | |
1214 | retry = false; | |
1215 | break; | |
1216 | case D_WALK_SKIP: | |
1217 | spin_unlock(&dentry->d_lock); | |
1218 | continue; | |
1219 | } | |
1220 | ||
1221 | if (!list_empty(&dentry->d_subdirs)) { | |
1222 | spin_unlock(&this_parent->d_lock); | |
1223 | spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); | |
1224 | this_parent = dentry; | |
1225 | spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); | |
1226 | goto repeat; | |
1227 | } | |
1228 | spin_unlock(&dentry->d_lock); | |
1229 | } | |
1230 | /* | |
1231 | * All done at this level ... ascend and resume the search. | |
1232 | */ | |
1233 | rcu_read_lock(); | |
1234 | ascend: | |
1235 | if (this_parent != parent) { | |
1236 | struct dentry *child = this_parent; | |
1237 | this_parent = child->d_parent; | |
1238 | ||
1239 | spin_unlock(&child->d_lock); | |
1240 | spin_lock(&this_parent->d_lock); | |
1241 | ||
1242 | /* might go back up the wrong parent if we have had a rename. */ | |
1243 | if (need_seqretry(&rename_lock, seq)) | |
1244 | goto rename_retry; | |
1245 | /* go into the first sibling still alive */ | |
1246 | do { | |
1247 | next = child->d_child.next; | |
1248 | if (next == &this_parent->d_subdirs) | |
1249 | goto ascend; | |
1250 | child = list_entry(next, struct dentry, d_child); | |
1251 | } while (unlikely(child->d_flags & DCACHE_DENTRY_KILLED)); | |
1252 | rcu_read_unlock(); | |
1253 | goto resume; | |
1254 | } | |
1255 | if (need_seqretry(&rename_lock, seq)) | |
1256 | goto rename_retry; | |
1257 | rcu_read_unlock(); | |
1258 | if (finish) | |
1259 | finish(data); | |
1260 | ||
1261 | out_unlock: | |
1262 | spin_unlock(&this_parent->d_lock); | |
1263 | done_seqretry(&rename_lock, seq); | |
1264 | return; | |
1265 | ||
1266 | rename_retry: | |
1267 | spin_unlock(&this_parent->d_lock); | |
1268 | rcu_read_unlock(); | |
1269 | BUG_ON(seq & 1); | |
1270 | if (!retry) | |
1271 | return; | |
1272 | seq = 1; | |
1273 | goto again; | |
1274 | } | |
1275 | ||
1276 | struct check_mount { | |
1277 | struct vfsmount *mnt; | |
1278 | unsigned int mounted; | |
1279 | }; | |
1280 | ||
1281 | static enum d_walk_ret path_check_mount(void *data, struct dentry *dentry) | |
1282 | { | |
1283 | struct check_mount *info = data; | |
1284 | struct path path = { .mnt = info->mnt, .dentry = dentry }; | |
1285 | ||
1286 | if (likely(!d_mountpoint(dentry))) | |
1287 | return D_WALK_CONTINUE; | |
1288 | if (__path_is_mountpoint(&path)) { | |
1289 | info->mounted = 1; | |
1290 | return D_WALK_QUIT; | |
1291 | } | |
1292 | return D_WALK_CONTINUE; | |
1293 | } | |
1294 | ||
1295 | /** | |
1296 | * path_has_submounts - check for mounts over a dentry in the | |
1297 | * current namespace. | |
1298 | * @parent: path to check. | |
1299 | * | |
1300 | * Return true if the parent or its subdirectories contain | |
1301 | * a mount point in the current namespace. | |
1302 | */ | |
1303 | int path_has_submounts(const struct path *parent) | |
1304 | { | |
1305 | struct check_mount data = { .mnt = parent->mnt, .mounted = 0 }; | |
1306 | ||
1307 | read_seqlock_excl(&mount_lock); | |
1308 | d_walk(parent->dentry, &data, path_check_mount, NULL); | |
1309 | read_sequnlock_excl(&mount_lock); | |
1310 | ||
1311 | return data.mounted; | |
1312 | } | |
1313 | EXPORT_SYMBOL(path_has_submounts); | |
1314 | ||
1315 | /* | |
1316 | * Called by mount code to set a mountpoint and check if the mountpoint is | |
1317 | * reachable (e.g. NFS can unhash a directory dentry and then the complete | |
1318 | * subtree can become unreachable). | |
1319 | * | |
1320 | * Only one of d_invalidate() and d_set_mounted() must succeed. For | |
1321 | * this reason take rename_lock and d_lock on dentry and ancestors. | |
1322 | */ | |
1323 | int d_set_mounted(struct dentry *dentry) | |
1324 | { | |
1325 | struct dentry *p; | |
1326 | int ret = -ENOENT; | |
1327 | write_seqlock(&rename_lock); | |
1328 | for (p = dentry->d_parent; !IS_ROOT(p); p = p->d_parent) { | |
1329 | /* Need exclusion wrt. d_invalidate() */ | |
1330 | spin_lock(&p->d_lock); | |
1331 | if (unlikely(d_unhashed(p))) { | |
1332 | spin_unlock(&p->d_lock); | |
1333 | goto out; | |
1334 | } | |
1335 | spin_unlock(&p->d_lock); | |
1336 | } | |
1337 | spin_lock(&dentry->d_lock); | |
1338 | if (!d_unlinked(dentry)) { | |
1339 | ret = -EBUSY; | |
1340 | if (!d_mountpoint(dentry)) { | |
1341 | dentry->d_flags |= DCACHE_MOUNTED; | |
1342 | ret = 0; | |
1343 | } | |
1344 | } | |
1345 | spin_unlock(&dentry->d_lock); | |
1346 | out: | |
1347 | write_sequnlock(&rename_lock); | |
1348 | return ret; | |
1349 | } | |
1350 | ||
1351 | /* | |
1352 | * Search the dentry child list of the specified parent, | |
1353 | * and move any unused dentries to the end of the unused | |
1354 | * list for prune_dcache(). We descend to the next level | |
1355 | * whenever the d_subdirs list is non-empty and continue | |
1356 | * searching. | |
1357 | * | |
1358 | * It returns zero iff there are no unused children, | |
1359 | * otherwise it returns the number of children moved to | |
1360 | * the end of the unused list. This may not be the total | |
1361 | * number of unused children, because select_parent can | |
1362 | * drop the lock and return early due to latency | |
1363 | * constraints. | |
1364 | */ | |
1365 | ||
1366 | struct select_data { | |
1367 | struct dentry *start; | |
1368 | struct list_head dispose; | |
1369 | int found; | |
1370 | }; | |
1371 | ||
1372 | static enum d_walk_ret select_collect(void *_data, struct dentry *dentry) | |
1373 | { | |
1374 | struct select_data *data = _data; | |
1375 | enum d_walk_ret ret = D_WALK_CONTINUE; | |
1376 | ||
1377 | if (data->start == dentry) | |
1378 | goto out; | |
1379 | ||
1380 | if (dentry->d_flags & DCACHE_SHRINK_LIST) { | |
1381 | data->found++; | |
1382 | } else { | |
1383 | if (dentry->d_flags & DCACHE_LRU_LIST) | |
1384 | d_lru_del(dentry); | |
1385 | if (!dentry->d_lockref.count) { | |
1386 | d_shrink_add(dentry, &data->dispose); | |
1387 | data->found++; | |
1388 | } | |
1389 | } | |
1390 | /* | |
1391 | * We can return to the caller if we have found some (this | |
1392 | * ensures forward progress). We'll be coming back to find | |
1393 | * the rest. | |
1394 | */ | |
1395 | if (!list_empty(&data->dispose)) | |
1396 | ret = need_resched() ? D_WALK_QUIT : D_WALK_NORETRY; | |
1397 | out: | |
1398 | return ret; | |
1399 | } | |
1400 | ||
1401 | /** | |
1402 | * shrink_dcache_parent - prune dcache | |
1403 | * @parent: parent of entries to prune | |
1404 | * | |
1405 | * Prune the dcache to remove unused children of the parent dentry. | |
1406 | */ | |
1407 | void shrink_dcache_parent(struct dentry *parent) | |
1408 | { | |
1409 | for (;;) { | |
1410 | struct select_data data; | |
1411 | ||
1412 | INIT_LIST_HEAD(&data.dispose); | |
1413 | data.start = parent; | |
1414 | data.found = 0; | |
1415 | ||
1416 | d_walk(parent, &data, select_collect, NULL); | |
1417 | if (!data.found) | |
1418 | break; | |
1419 | ||
1420 | shrink_dentry_list(&data.dispose); | |
1421 | cond_resched(); | |
1422 | } | |
1423 | } | |
1424 | EXPORT_SYMBOL(shrink_dcache_parent); | |
1425 | ||
1426 | static enum d_walk_ret umount_check(void *_data, struct dentry *dentry) | |
1427 | { | |
1428 | /* it has busy descendents; complain about those instead */ | |
1429 | if (!list_empty(&dentry->d_subdirs)) | |
1430 | return D_WALK_CONTINUE; | |
1431 | ||
1432 | /* root with refcount 1 is fine */ | |
1433 | if (dentry == _data && dentry->d_lockref.count == 1) | |
1434 | return D_WALK_CONTINUE; | |
1435 | ||
1436 | printk(KERN_ERR "BUG: Dentry %p{i=%lx,n=%pd} " | |
1437 | " still in use (%d) [unmount of %s %s]\n", | |
1438 | dentry, | |
1439 | dentry->d_inode ? | |
1440 | dentry->d_inode->i_ino : 0UL, | |
1441 | dentry, | |
1442 | dentry->d_lockref.count, | |
1443 | dentry->d_sb->s_type->name, | |
1444 | dentry->d_sb->s_id); | |
1445 | WARN_ON(1); | |
1446 | return D_WALK_CONTINUE; | |
1447 | } | |
1448 | ||
1449 | static void do_one_tree(struct dentry *dentry) | |
1450 | { | |
1451 | shrink_dcache_parent(dentry); | |
1452 | d_walk(dentry, dentry, umount_check, NULL); | |
1453 | d_drop(dentry); | |
1454 | dput(dentry); | |
1455 | } | |
1456 | ||
1457 | /* | |
1458 | * destroy the dentries attached to a superblock on unmounting | |
1459 | */ | |
1460 | void shrink_dcache_for_umount(struct super_block *sb) | |
1461 | { | |
1462 | struct dentry *dentry; | |
1463 | ||
1464 | WARN(down_read_trylock(&sb->s_umount), "s_umount should've been locked"); | |
1465 | ||
1466 | dentry = sb->s_root; | |
1467 | sb->s_root = NULL; | |
1468 | do_one_tree(dentry); | |
1469 | ||
1470 | while (!hlist_bl_empty(&sb->s_anon)) { | |
1471 | dentry = dget(hlist_bl_entry(hlist_bl_first(&sb->s_anon), struct dentry, d_hash)); | |
1472 | do_one_tree(dentry); | |
1473 | } | |
1474 | } | |
1475 | ||
1476 | struct detach_data { | |
1477 | struct select_data select; | |
1478 | struct dentry *mountpoint; | |
1479 | }; | |
1480 | static enum d_walk_ret detach_and_collect(void *_data, struct dentry *dentry) | |
1481 | { | |
1482 | struct detach_data *data = _data; | |
1483 | ||
1484 | if (d_mountpoint(dentry)) { | |
1485 | __dget_dlock(dentry); | |
1486 | data->mountpoint = dentry; | |
1487 | return D_WALK_QUIT; | |
1488 | } | |
1489 | ||
1490 | return select_collect(&data->select, dentry); | |
1491 | } | |
1492 | ||
1493 | static void check_and_drop(void *_data) | |
1494 | { | |
1495 | struct detach_data *data = _data; | |
1496 | ||
1497 | if (!data->mountpoint && !data->select.found) | |
1498 | __d_drop(data->select.start); | |
1499 | } | |
1500 | ||
1501 | /** | |
1502 | * d_invalidate - detach submounts, prune dcache, and drop | |
1503 | * @dentry: dentry to invalidate (aka detach, prune and drop) | |
1504 | * | |
1505 | * no dcache lock. | |
1506 | * | |
1507 | * The final d_drop is done as an atomic operation relative to | |
1508 | * rename_lock ensuring there are no races with d_set_mounted. This | |
1509 | * ensures there are no unhashed dentries on the path to a mountpoint. | |
1510 | */ | |
1511 | void d_invalidate(struct dentry *dentry) | |
1512 | { | |
1513 | /* | |
1514 | * If it's already been dropped, return OK. | |
1515 | */ | |
1516 | spin_lock(&dentry->d_lock); | |
1517 | if (d_unhashed(dentry)) { | |
1518 | spin_unlock(&dentry->d_lock); | |
1519 | return; | |
1520 | } | |
1521 | spin_unlock(&dentry->d_lock); | |
1522 | ||
1523 | /* Negative dentries can be dropped without further checks */ | |
1524 | if (!dentry->d_inode) { | |
1525 | d_drop(dentry); | |
1526 | return; | |
1527 | } | |
1528 | ||
1529 | for (;;) { | |
1530 | struct detach_data data; | |
1531 | ||
1532 | data.mountpoint = NULL; | |
1533 | INIT_LIST_HEAD(&data.select.dispose); | |
1534 | data.select.start = dentry; | |
1535 | data.select.found = 0; | |
1536 | ||
1537 | d_walk(dentry, &data, detach_and_collect, check_and_drop); | |
1538 | ||
1539 | if (data.select.found) | |
1540 | shrink_dentry_list(&data.select.dispose); | |
1541 | ||
1542 | if (data.mountpoint) { | |
1543 | detach_mounts(data.mountpoint); | |
1544 | dput(data.mountpoint); | |
1545 | } | |
1546 | ||
1547 | if (!data.mountpoint && !data.select.found) | |
1548 | break; | |
1549 | ||
1550 | cond_resched(); | |
1551 | } | |
1552 | } | |
1553 | EXPORT_SYMBOL(d_invalidate); | |
1554 | ||
1555 | /** | |
1556 | * __d_alloc - allocate a dcache entry | |
1557 | * @sb: filesystem it will belong to | |
1558 | * @name: qstr of the name | |
1559 | * | |
1560 | * Allocates a dentry. It returns %NULL if there is insufficient memory | |
1561 | * available. On a success the dentry is returned. The name passed in is | |
1562 | * copied and the copy passed in may be reused after this call. | |
1563 | */ | |
1564 | ||
1565 | struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) | |
1566 | { | |
1567 | struct dentry *dentry; | |
1568 | char *dname; | |
1569 | int err; | |
1570 | ||
1571 | dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL); | |
1572 | if (!dentry) | |
1573 | return NULL; | |
1574 | ||
1575 | /* | |
1576 | * We guarantee that the inline name is always NUL-terminated. | |
1577 | * This way the memcpy() done by the name switching in rename | |
1578 | * will still always have a NUL at the end, even if we might | |
1579 | * be overwriting an internal NUL character | |
1580 | */ | |
1581 | dentry->d_iname[DNAME_INLINE_LEN-1] = 0; | |
1582 | if (unlikely(!name)) { | |
1583 | static const struct qstr anon = QSTR_INIT("/", 1); | |
1584 | name = &anon; | |
1585 | dname = dentry->d_iname; | |
1586 | } else if (name->len > DNAME_INLINE_LEN-1) { | |
1587 | size_t size = offsetof(struct external_name, name[1]); | |
1588 | struct external_name *p = kmalloc(size + name->len, | |
1589 | GFP_KERNEL_ACCOUNT); | |
1590 | if (!p) { | |
1591 | kmem_cache_free(dentry_cache, dentry); | |
1592 | return NULL; | |
1593 | } | |
1594 | atomic_set(&p->u.count, 1); | |
1595 | dname = p->name; | |
1596 | if (IS_ENABLED(CONFIG_DCACHE_WORD_ACCESS)) | |
1597 | kasan_unpoison_shadow(dname, | |
1598 | round_up(name->len + 1, sizeof(unsigned long))); | |
1599 | } else { | |
1600 | dname = dentry->d_iname; | |
1601 | } | |
1602 | ||
1603 | dentry->d_name.len = name->len; | |
1604 | dentry->d_name.hash = name->hash; | |
1605 | memcpy(dname, name->name, name->len); | |
1606 | dname[name->len] = 0; | |
1607 | ||
1608 | /* Make sure we always see the terminating NUL character */ | |
1609 | smp_wmb(); | |
1610 | dentry->d_name.name = dname; | |
1611 | ||
1612 | dentry->d_lockref.count = 1; | |
1613 | dentry->d_flags = 0; | |
1614 | spin_lock_init(&dentry->d_lock); | |
1615 | seqcount_init(&dentry->d_seq); | |
1616 | dentry->d_inode = NULL; | |
1617 | dentry->d_parent = dentry; | |
1618 | dentry->d_sb = sb; | |
1619 | dentry->d_op = NULL; | |
1620 | dentry->d_fsdata = NULL; | |
1621 | INIT_HLIST_BL_NODE(&dentry->d_hash); | |
1622 | INIT_LIST_HEAD(&dentry->d_lru); | |
1623 | INIT_LIST_HEAD(&dentry->d_subdirs); | |
1624 | INIT_HLIST_NODE(&dentry->d_u.d_alias); | |
1625 | INIT_LIST_HEAD(&dentry->d_child); | |
1626 | d_set_d_op(dentry, dentry->d_sb->s_d_op); | |
1627 | ||
1628 | if (dentry->d_op && dentry->d_op->d_init) { | |
1629 | err = dentry->d_op->d_init(dentry); | |
1630 | if (err) { | |
1631 | if (dname_external(dentry)) | |
1632 | kfree(external_name(dentry)); | |
1633 | kmem_cache_free(dentry_cache, dentry); | |
1634 | return NULL; | |
1635 | } | |
1636 | } | |
1637 | ||
1638 | this_cpu_inc(nr_dentry); | |
1639 | ||
1640 | return dentry; | |
1641 | } | |
1642 | ||
1643 | /** | |
1644 | * d_alloc - allocate a dcache entry | |
1645 | * @parent: parent of entry to allocate | |
1646 | * @name: qstr of the name | |
1647 | * | |
1648 | * Allocates a dentry. It returns %NULL if there is insufficient memory | |
1649 | * available. On a success the dentry is returned. The name passed in is | |
1650 | * copied and the copy passed in may be reused after this call. | |
1651 | */ | |
1652 | struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) | |
1653 | { | |
1654 | struct dentry *dentry = __d_alloc(parent->d_sb, name); | |
1655 | if (!dentry) | |
1656 | return NULL; | |
1657 | dentry->d_flags |= DCACHE_RCUACCESS; | |
1658 | spin_lock(&parent->d_lock); | |
1659 | /* | |
1660 | * don't need child lock because it is not subject | |
1661 | * to concurrency here | |
1662 | */ | |
1663 | __dget_dlock(parent); | |
1664 | dentry->d_parent = parent; | |
1665 | list_add(&dentry->d_child, &parent->d_subdirs); | |
1666 | spin_unlock(&parent->d_lock); | |
1667 | ||
1668 | return dentry; | |
1669 | } | |
1670 | EXPORT_SYMBOL(d_alloc); | |
1671 | ||
1672 | struct dentry *d_alloc_cursor(struct dentry * parent) | |
1673 | { | |
1674 | struct dentry *dentry = __d_alloc(parent->d_sb, NULL); | |
1675 | if (dentry) { | |
1676 | dentry->d_flags |= DCACHE_RCUACCESS | DCACHE_DENTRY_CURSOR; | |
1677 | dentry->d_parent = dget(parent); | |
1678 | } | |
1679 | return dentry; | |
1680 | } | |
1681 | ||
1682 | /** | |
1683 | * d_alloc_pseudo - allocate a dentry (for lookup-less filesystems) | |
1684 | * @sb: the superblock | |
1685 | * @name: qstr of the name | |
1686 | * | |
1687 | * For a filesystem that just pins its dentries in memory and never | |
1688 | * performs lookups at all, return an unhashed IS_ROOT dentry. | |
1689 | */ | |
1690 | struct dentry *d_alloc_pseudo(struct super_block *sb, const struct qstr *name) | |
1691 | { | |
1692 | return __d_alloc(sb, name); | |
1693 | } | |
1694 | EXPORT_SYMBOL(d_alloc_pseudo); | |
1695 | ||
1696 | struct dentry *d_alloc_name(struct dentry *parent, const char *name) | |
1697 | { | |
1698 | struct qstr q; | |
1699 | ||
1700 | q.name = name; | |
1701 | q.hash_len = hashlen_string(parent, name); | |
1702 | return d_alloc(parent, &q); | |
1703 | } | |
1704 | EXPORT_SYMBOL(d_alloc_name); | |
1705 | ||
1706 | void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op) | |
1707 | { | |
1708 | WARN_ON_ONCE(dentry->d_op); | |
1709 | WARN_ON_ONCE(dentry->d_flags & (DCACHE_OP_HASH | | |
1710 | DCACHE_OP_COMPARE | | |
1711 | DCACHE_OP_REVALIDATE | | |
1712 | DCACHE_OP_WEAK_REVALIDATE | | |
1713 | DCACHE_OP_DELETE | | |
1714 | DCACHE_OP_REAL)); | |
1715 | dentry->d_op = op; | |
1716 | if (!op) | |
1717 | return; | |
1718 | if (op->d_hash) | |
1719 | dentry->d_flags |= DCACHE_OP_HASH; | |
1720 | if (op->d_compare) | |
1721 | dentry->d_flags |= DCACHE_OP_COMPARE; | |
1722 | if (op->d_revalidate) | |
1723 | dentry->d_flags |= DCACHE_OP_REVALIDATE; | |
1724 | if (op->d_weak_revalidate) | |
1725 | dentry->d_flags |= DCACHE_OP_WEAK_REVALIDATE; | |
1726 | if (op->d_delete) | |
1727 | dentry->d_flags |= DCACHE_OP_DELETE; | |
1728 | if (op->d_prune) | |
1729 | dentry->d_flags |= DCACHE_OP_PRUNE; | |
1730 | if (op->d_real) | |
1731 | dentry->d_flags |= DCACHE_OP_REAL; | |
1732 | ||
1733 | } | |
1734 | EXPORT_SYMBOL(d_set_d_op); | |
1735 | ||
1736 | ||
1737 | /* | |
1738 | * d_set_fallthru - Mark a dentry as falling through to a lower layer | |
1739 | * @dentry - The dentry to mark | |
1740 | * | |
1741 | * Mark a dentry as falling through to the lower layer (as set with | |
1742 | * d_pin_lower()). This flag may be recorded on the medium. | |
1743 | */ | |
1744 | void d_set_fallthru(struct dentry *dentry) | |
1745 | { | |
1746 | spin_lock(&dentry->d_lock); | |
1747 | dentry->d_flags |= DCACHE_FALLTHRU; | |
1748 | spin_unlock(&dentry->d_lock); | |
1749 | } | |
1750 | EXPORT_SYMBOL(d_set_fallthru); | |
1751 | ||
1752 | static unsigned d_flags_for_inode(struct inode *inode) | |
1753 | { | |
1754 | unsigned add_flags = DCACHE_REGULAR_TYPE; | |
1755 | ||
1756 | if (!inode) | |
1757 | return DCACHE_MISS_TYPE; | |
1758 | ||
1759 | if (S_ISDIR(inode->i_mode)) { | |
1760 | add_flags = DCACHE_DIRECTORY_TYPE; | |
1761 | if (unlikely(!(inode->i_opflags & IOP_LOOKUP))) { | |
1762 | if (unlikely(!inode->i_op->lookup)) | |
1763 | add_flags = DCACHE_AUTODIR_TYPE; | |
1764 | else | |
1765 | inode->i_opflags |= IOP_LOOKUP; | |
1766 | } | |
1767 | goto type_determined; | |
1768 | } | |
1769 | ||
1770 | if (unlikely(!(inode->i_opflags & IOP_NOFOLLOW))) { | |
1771 | if (unlikely(inode->i_op->get_link)) { | |
1772 | add_flags = DCACHE_SYMLINK_TYPE; | |
1773 | goto type_determined; | |
1774 | } | |
1775 | inode->i_opflags |= IOP_NOFOLLOW; | |
1776 | } | |
1777 | ||
1778 | if (unlikely(!S_ISREG(inode->i_mode))) | |
1779 | add_flags = DCACHE_SPECIAL_TYPE; | |
1780 | ||
1781 | type_determined: | |
1782 | if (unlikely(IS_AUTOMOUNT(inode))) | |
1783 | add_flags |= DCACHE_NEED_AUTOMOUNT; | |
1784 | return add_flags; | |
1785 | } | |
1786 | ||
1787 | static void __d_instantiate(struct dentry *dentry, struct inode *inode) | |
1788 | { | |
1789 | unsigned add_flags = d_flags_for_inode(inode); | |
1790 | WARN_ON(d_in_lookup(dentry)); | |
1791 | ||
1792 | spin_lock(&dentry->d_lock); | |
1793 | hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); | |
1794 | raw_write_seqcount_begin(&dentry->d_seq); | |
1795 | __d_set_inode_and_type(dentry, inode, add_flags); | |
1796 | raw_write_seqcount_end(&dentry->d_seq); | |
1797 | fsnotify_update_flags(dentry); | |
1798 | spin_unlock(&dentry->d_lock); | |
1799 | } | |
1800 | ||
1801 | /** | |
1802 | * d_instantiate - fill in inode information for a dentry | |
1803 | * @entry: dentry to complete | |
1804 | * @inode: inode to attach to this dentry | |
1805 | * | |
1806 | * Fill in inode information in the entry. | |
1807 | * | |
1808 | * This turns negative dentries into productive full members | |
1809 | * of society. | |
1810 | * | |
1811 | * NOTE! This assumes that the inode count has been incremented | |
1812 | * (or otherwise set) by the caller to indicate that it is now | |
1813 | * in use by the dcache. | |
1814 | */ | |
1815 | ||
1816 | void d_instantiate(struct dentry *entry, struct inode * inode) | |
1817 | { | |
1818 | BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); | |
1819 | if (inode) { | |
1820 | security_d_instantiate(entry, inode); | |
1821 | spin_lock(&inode->i_lock); | |
1822 | __d_instantiate(entry, inode); | |
1823 | spin_unlock(&inode->i_lock); | |
1824 | } | |
1825 | } | |
1826 | EXPORT_SYMBOL(d_instantiate); | |
1827 | ||
1828 | /** | |
1829 | * d_instantiate_no_diralias - instantiate a non-aliased dentry | |
1830 | * @entry: dentry to complete | |
1831 | * @inode: inode to attach to this dentry | |
1832 | * | |
1833 | * Fill in inode information in the entry. If a directory alias is found, then | |
1834 | * return an error (and drop inode). Together with d_materialise_unique() this | |
1835 | * guarantees that a directory inode may never have more than one alias. | |
1836 | */ | |
1837 | int d_instantiate_no_diralias(struct dentry *entry, struct inode *inode) | |
1838 | { | |
1839 | BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); | |
1840 | ||
1841 | security_d_instantiate(entry, inode); | |
1842 | spin_lock(&inode->i_lock); | |
1843 | if (S_ISDIR(inode->i_mode) && !hlist_empty(&inode->i_dentry)) { | |
1844 | spin_unlock(&inode->i_lock); | |
1845 | iput(inode); | |
1846 | return -EBUSY; | |
1847 | } | |
1848 | __d_instantiate(entry, inode); | |
1849 | spin_unlock(&inode->i_lock); | |
1850 | ||
1851 | return 0; | |
1852 | } | |
1853 | EXPORT_SYMBOL(d_instantiate_no_diralias); | |
1854 | ||
1855 | struct dentry *d_make_root(struct inode *root_inode) | |
1856 | { | |
1857 | struct dentry *res = NULL; | |
1858 | ||
1859 | if (root_inode) { | |
1860 | res = __d_alloc(root_inode->i_sb, NULL); | |
1861 | if (res) | |
1862 | d_instantiate(res, root_inode); | |
1863 | else | |
1864 | iput(root_inode); | |
1865 | } | |
1866 | return res; | |
1867 | } | |
1868 | EXPORT_SYMBOL(d_make_root); | |
1869 | ||
1870 | static struct dentry * __d_find_any_alias(struct inode *inode) | |
1871 | { | |
1872 | struct dentry *alias; | |
1873 | ||
1874 | if (hlist_empty(&inode->i_dentry)) | |
1875 | return NULL; | |
1876 | alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias); | |
1877 | __dget(alias); | |
1878 | return alias; | |
1879 | } | |
1880 | ||
1881 | /** | |
1882 | * d_find_any_alias - find any alias for a given inode | |
1883 | * @inode: inode to find an alias for | |
1884 | * | |
1885 | * If any aliases exist for the given inode, take and return a | |
1886 | * reference for one of them. If no aliases exist, return %NULL. | |
1887 | */ | |
1888 | struct dentry *d_find_any_alias(struct inode *inode) | |
1889 | { | |
1890 | struct dentry *de; | |
1891 | ||
1892 | spin_lock(&inode->i_lock); | |
1893 | de = __d_find_any_alias(inode); | |
1894 | spin_unlock(&inode->i_lock); | |
1895 | return de; | |
1896 | } | |
1897 | EXPORT_SYMBOL(d_find_any_alias); | |
1898 | ||
1899 | static struct dentry *__d_obtain_alias(struct inode *inode, int disconnected) | |
1900 | { | |
1901 | struct dentry *tmp; | |
1902 | struct dentry *res; | |
1903 | unsigned add_flags; | |
1904 | ||
1905 | if (!inode) | |
1906 | return ERR_PTR(-ESTALE); | |
1907 | if (IS_ERR(inode)) | |
1908 | return ERR_CAST(inode); | |
1909 | ||
1910 | res = d_find_any_alias(inode); | |
1911 | if (res) | |
1912 | goto out_iput; | |
1913 | ||
1914 | tmp = __d_alloc(inode->i_sb, NULL); | |
1915 | if (!tmp) { | |
1916 | res = ERR_PTR(-ENOMEM); | |
1917 | goto out_iput; | |
1918 | } | |
1919 | ||
1920 | security_d_instantiate(tmp, inode); | |
1921 | spin_lock(&inode->i_lock); | |
1922 | res = __d_find_any_alias(inode); | |
1923 | if (res) { | |
1924 | spin_unlock(&inode->i_lock); | |
1925 | dput(tmp); | |
1926 | goto out_iput; | |
1927 | } | |
1928 | ||
1929 | /* attach a disconnected dentry */ | |
1930 | add_flags = d_flags_for_inode(inode); | |
1931 | ||
1932 | if (disconnected) | |
1933 | add_flags |= DCACHE_DISCONNECTED; | |
1934 | ||
1935 | spin_lock(&tmp->d_lock); | |
1936 | __d_set_inode_and_type(tmp, inode, add_flags); | |
1937 | hlist_add_head(&tmp->d_u.d_alias, &inode->i_dentry); | |
1938 | hlist_bl_lock(&tmp->d_sb->s_anon); | |
1939 | hlist_bl_add_head(&tmp->d_hash, &tmp->d_sb->s_anon); | |
1940 | hlist_bl_unlock(&tmp->d_sb->s_anon); | |
1941 | spin_unlock(&tmp->d_lock); | |
1942 | spin_unlock(&inode->i_lock); | |
1943 | ||
1944 | return tmp; | |
1945 | ||
1946 | out_iput: | |
1947 | iput(inode); | |
1948 | return res; | |
1949 | } | |
1950 | ||
1951 | /** | |
1952 | * d_obtain_alias - find or allocate a DISCONNECTED dentry for a given inode | |
1953 | * @inode: inode to allocate the dentry for | |
1954 | * | |
1955 | * Obtain a dentry for an inode resulting from NFS filehandle conversion or | |
1956 | * similar open by handle operations. The returned dentry may be anonymous, | |
1957 | * or may have a full name (if the inode was already in the cache). | |
1958 | * | |
1959 | * When called on a directory inode, we must ensure that the inode only ever | |
1960 | * has one dentry. If a dentry is found, that is returned instead of | |
1961 | * allocating a new one. | |
1962 | * | |
1963 | * On successful return, the reference to the inode has been transferred | |
1964 | * to the dentry. In case of an error the reference on the inode is released. | |
1965 | * To make it easier to use in export operations a %NULL or IS_ERR inode may | |
1966 | * be passed in and the error will be propagated to the return value, | |
1967 | * with a %NULL @inode replaced by ERR_PTR(-ESTALE). | |
1968 | */ | |
1969 | struct dentry *d_obtain_alias(struct inode *inode) | |
1970 | { | |
1971 | return __d_obtain_alias(inode, 1); | |
1972 | } | |
1973 | EXPORT_SYMBOL(d_obtain_alias); | |
1974 | ||
1975 | /** | |
1976 | * d_obtain_root - find or allocate a dentry for a given inode | |
1977 | * @inode: inode to allocate the dentry for | |
1978 | * | |
1979 | * Obtain an IS_ROOT dentry for the root of a filesystem. | |
1980 | * | |
1981 | * We must ensure that directory inodes only ever have one dentry. If a | |
1982 | * dentry is found, that is returned instead of allocating a new one. | |
1983 | * | |
1984 | * On successful return, the reference to the inode has been transferred | |
1985 | * to the dentry. In case of an error the reference on the inode is | |
1986 | * released. A %NULL or IS_ERR inode may be passed in and will be the | |
1987 | * error will be propagate to the return value, with a %NULL @inode | |
1988 | * replaced by ERR_PTR(-ESTALE). | |
1989 | */ | |
1990 | struct dentry *d_obtain_root(struct inode *inode) | |
1991 | { | |
1992 | return __d_obtain_alias(inode, 0); | |
1993 | } | |
1994 | EXPORT_SYMBOL(d_obtain_root); | |
1995 | ||
1996 | /** | |
1997 | * d_add_ci - lookup or allocate new dentry with case-exact name | |
1998 | * @inode: the inode case-insensitive lookup has found | |
1999 | * @dentry: the negative dentry that was passed to the parent's lookup func | |
2000 | * @name: the case-exact name to be associated with the returned dentry | |
2001 | * | |
2002 | * This is to avoid filling the dcache with case-insensitive names to the | |
2003 | * same inode, only the actual correct case is stored in the dcache for | |
2004 | * case-insensitive filesystems. | |
2005 | * | |
2006 | * For a case-insensitive lookup match and if the the case-exact dentry | |
2007 | * already exists in in the dcache, use it and return it. | |
2008 | * | |
2009 | * If no entry exists with the exact case name, allocate new dentry with | |
2010 | * the exact case, and return the spliced entry. | |
2011 | */ | |
2012 | struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode, | |
2013 | struct qstr *name) | |
2014 | { | |
2015 | struct dentry *found, *res; | |
2016 | ||
2017 | /* | |
2018 | * First check if a dentry matching the name already exists, | |
2019 | * if not go ahead and create it now. | |
2020 | */ | |
2021 | found = d_hash_and_lookup(dentry->d_parent, name); | |
2022 | if (found) { | |
2023 | iput(inode); | |
2024 | return found; | |
2025 | } | |
2026 | if (d_in_lookup(dentry)) { | |
2027 | found = d_alloc_parallel(dentry->d_parent, name, | |
2028 | dentry->d_wait); | |
2029 | if (IS_ERR(found) || !d_in_lookup(found)) { | |
2030 | iput(inode); | |
2031 | return found; | |
2032 | } | |
2033 | } else { | |
2034 | found = d_alloc(dentry->d_parent, name); | |
2035 | if (!found) { | |
2036 | iput(inode); | |
2037 | return ERR_PTR(-ENOMEM); | |
2038 | } | |
2039 | } | |
2040 | res = d_splice_alias(inode, found); | |
2041 | if (res) { | |
2042 | dput(found); | |
2043 | return res; | |
2044 | } | |
2045 | return found; | |
2046 | } | |
2047 | EXPORT_SYMBOL(d_add_ci); | |
2048 | ||
2049 | ||
2050 | static inline bool d_same_name(const struct dentry *dentry, | |
2051 | const struct dentry *parent, | |
2052 | const struct qstr *name) | |
2053 | { | |
2054 | if (likely(!(parent->d_flags & DCACHE_OP_COMPARE))) { | |
2055 | if (dentry->d_name.len != name->len) | |
2056 | return false; | |
2057 | return dentry_cmp(dentry, name->name, name->len) == 0; | |
2058 | } | |
2059 | return parent->d_op->d_compare(dentry, | |
2060 | dentry->d_name.len, dentry->d_name.name, | |
2061 | name) == 0; | |
2062 | } | |
2063 | ||
2064 | /** | |
2065 | * __d_lookup_rcu - search for a dentry (racy, store-free) | |
2066 | * @parent: parent dentry | |
2067 | * @name: qstr of name we wish to find | |
2068 | * @seqp: returns d_seq value at the point where the dentry was found | |
2069 | * Returns: dentry, or NULL | |
2070 | * | |
2071 | * __d_lookup_rcu is the dcache lookup function for rcu-walk name | |
2072 | * resolution (store-free path walking) design described in | |
2073 | * Documentation/filesystems/path-lookup.txt. | |
2074 | * | |
2075 | * This is not to be used outside core vfs. | |
2076 | * | |
2077 | * __d_lookup_rcu must only be used in rcu-walk mode, ie. with vfsmount lock | |
2078 | * held, and rcu_read_lock held. The returned dentry must not be stored into | |
2079 | * without taking d_lock and checking d_seq sequence count against @seq | |
2080 | * returned here. | |
2081 | * | |
2082 | * A refcount may be taken on the found dentry with the d_rcu_to_refcount | |
2083 | * function. | |
2084 | * | |
2085 | * Alternatively, __d_lookup_rcu may be called again to look up the child of | |
2086 | * the returned dentry, so long as its parent's seqlock is checked after the | |
2087 | * child is looked up. Thus, an interlocking stepping of sequence lock checks | |
2088 | * is formed, giving integrity down the path walk. | |
2089 | * | |
2090 | * NOTE! The caller *has* to check the resulting dentry against the sequence | |
2091 | * number we've returned before using any of the resulting dentry state! | |
2092 | */ | |
2093 | struct dentry *__d_lookup_rcu(const struct dentry *parent, | |
2094 | const struct qstr *name, | |
2095 | unsigned *seqp) | |
2096 | { | |
2097 | u64 hashlen = name->hash_len; | |
2098 | const unsigned char *str = name->name; | |
2099 | struct hlist_bl_head *b = d_hash(hashlen_hash(hashlen)); | |
2100 | struct hlist_bl_node *node; | |
2101 | struct dentry *dentry; | |
2102 | ||
2103 | /* | |
2104 | * Note: There is significant duplication with __d_lookup_rcu which is | |
2105 | * required to prevent single threaded performance regressions | |
2106 | * especially on architectures where smp_rmb (in seqcounts) are costly. | |
2107 | * Keep the two functions in sync. | |
2108 | */ | |
2109 | ||
2110 | /* | |
2111 | * The hash list is protected using RCU. | |
2112 | * | |
2113 | * Carefully use d_seq when comparing a candidate dentry, to avoid | |
2114 | * races with d_move(). | |
2115 | * | |
2116 | * It is possible that concurrent renames can mess up our list | |
2117 | * walk here and result in missing our dentry, resulting in the | |
2118 | * false-negative result. d_lookup() protects against concurrent | |
2119 | * renames using rename_lock seqlock. | |
2120 | * | |
2121 | * See Documentation/filesystems/path-lookup.txt for more details. | |
2122 | */ | |
2123 | hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { | |
2124 | unsigned seq; | |
2125 | ||
2126 | seqretry: | |
2127 | /* | |
2128 | * The dentry sequence count protects us from concurrent | |
2129 | * renames, and thus protects parent and name fields. | |
2130 | * | |
2131 | * The caller must perform a seqcount check in order | |
2132 | * to do anything useful with the returned dentry. | |
2133 | * | |
2134 | * NOTE! We do a "raw" seqcount_begin here. That means that | |
2135 | * we don't wait for the sequence count to stabilize if it | |
2136 | * is in the middle of a sequence change. If we do the slow | |
2137 | * dentry compare, we will do seqretries until it is stable, | |
2138 | * and if we end up with a successful lookup, we actually | |
2139 | * want to exit RCU lookup anyway. | |
2140 | * | |
2141 | * Note that raw_seqcount_begin still *does* smp_rmb(), so | |
2142 | * we are still guaranteed NUL-termination of ->d_name.name. | |
2143 | */ | |
2144 | seq = raw_seqcount_begin(&dentry->d_seq); | |
2145 | if (dentry->d_parent != parent) | |
2146 | continue; | |
2147 | if (d_unhashed(dentry)) | |
2148 | continue; | |
2149 | ||
2150 | if (unlikely(parent->d_flags & DCACHE_OP_COMPARE)) { | |
2151 | int tlen; | |
2152 | const char *tname; | |
2153 | if (dentry->d_name.hash != hashlen_hash(hashlen)) | |
2154 | continue; | |
2155 | tlen = dentry->d_name.len; | |
2156 | tname = dentry->d_name.name; | |
2157 | /* we want a consistent (name,len) pair */ | |
2158 | if (read_seqcount_retry(&dentry->d_seq, seq)) { | |
2159 | cpu_relax(); | |
2160 | goto seqretry; | |
2161 | } | |
2162 | if (parent->d_op->d_compare(dentry, | |
2163 | tlen, tname, name) != 0) | |
2164 | continue; | |
2165 | } else { | |
2166 | if (dentry->d_name.hash_len != hashlen) | |
2167 | continue; | |
2168 | if (dentry_cmp(dentry, str, hashlen_len(hashlen)) != 0) | |
2169 | continue; | |
2170 | } | |
2171 | *seqp = seq; | |
2172 | return dentry; | |
2173 | } | |
2174 | return NULL; | |
2175 | } | |
2176 | ||
2177 | /** | |
2178 | * d_lookup - search for a dentry | |
2179 | * @parent: parent dentry | |
2180 | * @name: qstr of name we wish to find | |
2181 | * Returns: dentry, or NULL | |
2182 | * | |
2183 | * d_lookup searches the children of the parent dentry for the name in | |
2184 | * question. If the dentry is found its reference count is incremented and the | |
2185 | * dentry is returned. The caller must use dput to free the entry when it has | |
2186 | * finished using it. %NULL is returned if the dentry does not exist. | |
2187 | */ | |
2188 | struct dentry *d_lookup(const struct dentry *parent, const struct qstr *name) | |
2189 | { | |
2190 | struct dentry *dentry; | |
2191 | unsigned seq; | |
2192 | ||
2193 | do { | |
2194 | seq = read_seqbegin(&rename_lock); | |
2195 | dentry = __d_lookup(parent, name); | |
2196 | if (dentry) | |
2197 | break; | |
2198 | } while (read_seqretry(&rename_lock, seq)); | |
2199 | return dentry; | |
2200 | } | |
2201 | EXPORT_SYMBOL(d_lookup); | |
2202 | ||
2203 | /** | |
2204 | * __d_lookup - search for a dentry (racy) | |
2205 | * @parent: parent dentry | |
2206 | * @name: qstr of name we wish to find | |
2207 | * Returns: dentry, or NULL | |
2208 | * | |
2209 | * __d_lookup is like d_lookup, however it may (rarely) return a | |
2210 | * false-negative result due to unrelated rename activity. | |
2211 | * | |
2212 | * __d_lookup is slightly faster by avoiding rename_lock read seqlock, | |
2213 | * however it must be used carefully, eg. with a following d_lookup in | |
2214 | * the case of failure. | |
2215 | * | |
2216 | * __d_lookup callers must be commented. | |
2217 | */ | |
2218 | struct dentry *__d_lookup(const struct dentry *parent, const struct qstr *name) | |
2219 | { | |
2220 | unsigned int hash = name->hash; | |
2221 | struct hlist_bl_head *b = d_hash(hash); | |
2222 | struct hlist_bl_node *node; | |
2223 | struct dentry *found = NULL; | |
2224 | struct dentry *dentry; | |
2225 | ||
2226 | /* | |
2227 | * Note: There is significant duplication with __d_lookup_rcu which is | |
2228 | * required to prevent single threaded performance regressions | |
2229 | * especially on architectures where smp_rmb (in seqcounts) are costly. | |
2230 | * Keep the two functions in sync. | |
2231 | */ | |
2232 | ||
2233 | /* | |
2234 | * The hash list is protected using RCU. | |
2235 | * | |
2236 | * Take d_lock when comparing a candidate dentry, to avoid races | |
2237 | * with d_move(). | |
2238 | * | |
2239 | * It is possible that concurrent renames can mess up our list | |
2240 | * walk here and result in missing our dentry, resulting in the | |
2241 | * false-negative result. d_lookup() protects against concurrent | |
2242 | * renames using rename_lock seqlock. | |
2243 | * | |
2244 | * See Documentation/filesystems/path-lookup.txt for more details. | |
2245 | */ | |
2246 | rcu_read_lock(); | |
2247 | ||
2248 | hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { | |
2249 | ||
2250 | if (dentry->d_name.hash != hash) | |
2251 | continue; | |
2252 | ||
2253 | spin_lock(&dentry->d_lock); | |
2254 | if (dentry->d_parent != parent) | |
2255 | goto next; | |
2256 | if (d_unhashed(dentry)) | |
2257 | goto next; | |
2258 | ||
2259 | if (!d_same_name(dentry, parent, name)) | |
2260 | goto next; | |
2261 | ||
2262 | dentry->d_lockref.count++; | |
2263 | found = dentry; | |
2264 | spin_unlock(&dentry->d_lock); | |
2265 | break; | |
2266 | next: | |
2267 | spin_unlock(&dentry->d_lock); | |
2268 | } | |
2269 | rcu_read_unlock(); | |
2270 | ||
2271 | return found; | |
2272 | } | |
2273 | ||
2274 | /** | |
2275 | * d_hash_and_lookup - hash the qstr then search for a dentry | |
2276 | * @dir: Directory to search in | |
2277 | * @name: qstr of name we wish to find | |
2278 | * | |
2279 | * On lookup failure NULL is returned; on bad name - ERR_PTR(-error) | |
2280 | */ | |
2281 | struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name) | |
2282 | { | |
2283 | /* | |
2284 | * Check for a fs-specific hash function. Note that we must | |
2285 | * calculate the standard hash first, as the d_op->d_hash() | |
2286 | * routine may choose to leave the hash value unchanged. | |
2287 | */ | |
2288 | name->hash = full_name_hash(dir, name->name, name->len); | |
2289 | if (dir->d_flags & DCACHE_OP_HASH) { | |
2290 | int err = dir->d_op->d_hash(dir, name); | |
2291 | if (unlikely(err < 0)) | |
2292 | return ERR_PTR(err); | |
2293 | } | |
2294 | return d_lookup(dir, name); | |
2295 | } | |
2296 | EXPORT_SYMBOL(d_hash_and_lookup); | |
2297 | ||
2298 | /* | |
2299 | * When a file is deleted, we have two options: | |
2300 | * - turn this dentry into a negative dentry | |
2301 | * - unhash this dentry and free it. | |
2302 | * | |
2303 | * Usually, we want to just turn this into | |
2304 | * a negative dentry, but if anybody else is | |
2305 | * currently using the dentry or the inode | |
2306 | * we can't do that and we fall back on removing | |
2307 | * it from the hash queues and waiting for | |
2308 | * it to be deleted later when it has no users | |
2309 | */ | |
2310 | ||
2311 | /** | |
2312 | * d_delete - delete a dentry | |
2313 | * @dentry: The dentry to delete | |
2314 | * | |
2315 | * Turn the dentry into a negative dentry if possible, otherwise | |
2316 | * remove it from the hash queues so it can be deleted later | |
2317 | */ | |
2318 | ||
2319 | void d_delete(struct dentry * dentry) | |
2320 | { | |
2321 | struct inode *inode; | |
2322 | int isdir = 0; | |
2323 | /* | |
2324 | * Are we the only user? | |
2325 | */ | |
2326 | again: | |
2327 | spin_lock(&dentry->d_lock); | |
2328 | inode = dentry->d_inode; | |
2329 | isdir = S_ISDIR(inode->i_mode); | |
2330 | if (dentry->d_lockref.count == 1) { | |
2331 | if (!spin_trylock(&inode->i_lock)) { | |
2332 | spin_unlock(&dentry->d_lock); | |
2333 | cpu_relax(); | |
2334 | goto again; | |
2335 | } | |
2336 | dentry->d_flags &= ~DCACHE_CANT_MOUNT; | |
2337 | dentry_unlink_inode(dentry); | |
2338 | fsnotify_nameremove(dentry, isdir); | |
2339 | return; | |
2340 | } | |
2341 | ||
2342 | if (!d_unhashed(dentry)) | |
2343 | __d_drop(dentry); | |
2344 | ||
2345 | spin_unlock(&dentry->d_lock); | |
2346 | ||
2347 | fsnotify_nameremove(dentry, isdir); | |
2348 | } | |
2349 | EXPORT_SYMBOL(d_delete); | |
2350 | ||
2351 | static void __d_rehash(struct dentry *entry) | |
2352 | { | |
2353 | struct hlist_bl_head *b = d_hash(entry->d_name.hash); | |
2354 | BUG_ON(!d_unhashed(entry)); | |
2355 | hlist_bl_lock(b); | |
2356 | hlist_bl_add_head_rcu(&entry->d_hash, b); | |
2357 | hlist_bl_unlock(b); | |
2358 | } | |
2359 | ||
2360 | /** | |
2361 | * d_rehash - add an entry back to the hash | |
2362 | * @entry: dentry to add to the hash | |
2363 | * | |
2364 | * Adds a dentry to the hash according to its name. | |
2365 | */ | |
2366 | ||
2367 | void d_rehash(struct dentry * entry) | |
2368 | { | |
2369 | spin_lock(&entry->d_lock); | |
2370 | __d_rehash(entry); | |
2371 | spin_unlock(&entry->d_lock); | |
2372 | } | |
2373 | EXPORT_SYMBOL(d_rehash); | |
2374 | ||
2375 | static inline unsigned start_dir_add(struct inode *dir) | |
2376 | { | |
2377 | ||
2378 | for (;;) { | |
2379 | unsigned n = dir->i_dir_seq; | |
2380 | if (!(n & 1) && cmpxchg(&dir->i_dir_seq, n, n + 1) == n) | |
2381 | return n; | |
2382 | cpu_relax(); | |
2383 | } | |
2384 | } | |
2385 | ||
2386 | static inline void end_dir_add(struct inode *dir, unsigned n) | |
2387 | { | |
2388 | smp_store_release(&dir->i_dir_seq, n + 2); | |
2389 | } | |
2390 | ||
2391 | static void d_wait_lookup(struct dentry *dentry) | |
2392 | { | |
2393 | if (d_in_lookup(dentry)) { | |
2394 | DECLARE_WAITQUEUE(wait, current); | |
2395 | add_wait_queue(dentry->d_wait, &wait); | |
2396 | do { | |
2397 | set_current_state(TASK_UNINTERRUPTIBLE); | |
2398 | spin_unlock(&dentry->d_lock); | |
2399 | schedule(); | |
2400 | spin_lock(&dentry->d_lock); | |
2401 | } while (d_in_lookup(dentry)); | |
2402 | } | |
2403 | } | |
2404 | ||
2405 | struct dentry *d_alloc_parallel(struct dentry *parent, | |
2406 | const struct qstr *name, | |
2407 | wait_queue_head_t *wq) | |
2408 | { | |
2409 | unsigned int hash = name->hash; | |
2410 | struct hlist_bl_head *b = in_lookup_hash(parent, hash); | |
2411 | struct hlist_bl_node *node; | |
2412 | struct dentry *new = d_alloc(parent, name); | |
2413 | struct dentry *dentry; | |
2414 | unsigned seq, r_seq, d_seq; | |
2415 | ||
2416 | if (unlikely(!new)) | |
2417 | return ERR_PTR(-ENOMEM); | |
2418 | ||
2419 | retry: | |
2420 | rcu_read_lock(); | |
2421 | seq = smp_load_acquire(&parent->d_inode->i_dir_seq) & ~1; | |
2422 | r_seq = read_seqbegin(&rename_lock); | |
2423 | dentry = __d_lookup_rcu(parent, name, &d_seq); | |
2424 | if (unlikely(dentry)) { | |
2425 | if (!lockref_get_not_dead(&dentry->d_lockref)) { | |
2426 | rcu_read_unlock(); | |
2427 | goto retry; | |
2428 | } | |
2429 | if (read_seqcount_retry(&dentry->d_seq, d_seq)) { | |
2430 | rcu_read_unlock(); | |
2431 | dput(dentry); | |
2432 | goto retry; | |
2433 | } | |
2434 | rcu_read_unlock(); | |
2435 | dput(new); | |
2436 | return dentry; | |
2437 | } | |
2438 | if (unlikely(read_seqretry(&rename_lock, r_seq))) { | |
2439 | rcu_read_unlock(); | |
2440 | goto retry; | |
2441 | } | |
2442 | hlist_bl_lock(b); | |
2443 | if (unlikely(parent->d_inode->i_dir_seq != seq)) { | |
2444 | hlist_bl_unlock(b); | |
2445 | rcu_read_unlock(); | |
2446 | goto retry; | |
2447 | } | |
2448 | /* | |
2449 | * No changes for the parent since the beginning of d_lookup(). | |
2450 | * Since all removals from the chain happen with hlist_bl_lock(), | |
2451 | * any potential in-lookup matches are going to stay here until | |
2452 | * we unlock the chain. All fields are stable in everything | |
2453 | * we encounter. | |
2454 | */ | |
2455 | hlist_bl_for_each_entry(dentry, node, b, d_u.d_in_lookup_hash) { | |
2456 | if (dentry->d_name.hash != hash) | |
2457 | continue; | |
2458 | if (dentry->d_parent != parent) | |
2459 | continue; | |
2460 | if (!d_same_name(dentry, parent, name)) | |
2461 | continue; | |
2462 | hlist_bl_unlock(b); | |
2463 | /* now we can try to grab a reference */ | |
2464 | if (!lockref_get_not_dead(&dentry->d_lockref)) { | |
2465 | rcu_read_unlock(); | |
2466 | goto retry; | |
2467 | } | |
2468 | ||
2469 | rcu_read_unlock(); | |
2470 | /* | |
2471 | * somebody is likely to be still doing lookup for it; | |
2472 | * wait for them to finish | |
2473 | */ | |
2474 | spin_lock(&dentry->d_lock); | |
2475 | d_wait_lookup(dentry); | |
2476 | /* | |
2477 | * it's not in-lookup anymore; in principle we should repeat | |
2478 | * everything from dcache lookup, but it's likely to be what | |
2479 | * d_lookup() would've found anyway. If it is, just return it; | |
2480 | * otherwise we really have to repeat the whole thing. | |
2481 | */ | |
2482 | if (unlikely(dentry->d_name.hash != hash)) | |
2483 | goto mismatch; | |
2484 | if (unlikely(dentry->d_parent != parent)) | |
2485 | goto mismatch; | |
2486 | if (unlikely(d_unhashed(dentry))) | |
2487 | goto mismatch; | |
2488 | if (unlikely(!d_same_name(dentry, parent, name))) | |
2489 | goto mismatch; | |
2490 | /* OK, it *is* a hashed match; return it */ | |
2491 | spin_unlock(&dentry->d_lock); | |
2492 | dput(new); | |
2493 | return dentry; | |
2494 | } | |
2495 | rcu_read_unlock(); | |
2496 | /* we can't take ->d_lock here; it's OK, though. */ | |
2497 | new->d_flags |= DCACHE_PAR_LOOKUP; | |
2498 | new->d_wait = wq; | |
2499 | hlist_bl_add_head_rcu(&new->d_u.d_in_lookup_hash, b); | |
2500 | hlist_bl_unlock(b); | |
2501 | return new; | |
2502 | mismatch: | |
2503 | spin_unlock(&dentry->d_lock); | |
2504 | dput(dentry); | |
2505 | goto retry; | |
2506 | } | |
2507 | EXPORT_SYMBOL(d_alloc_parallel); | |
2508 | ||
2509 | void __d_lookup_done(struct dentry *dentry) | |
2510 | { | |
2511 | struct hlist_bl_head *b = in_lookup_hash(dentry->d_parent, | |
2512 | dentry->d_name.hash); | |
2513 | hlist_bl_lock(b); | |
2514 | dentry->d_flags &= ~DCACHE_PAR_LOOKUP; | |
2515 | __hlist_bl_del(&dentry->d_u.d_in_lookup_hash); | |
2516 | wake_up_all(dentry->d_wait); | |
2517 | dentry->d_wait = NULL; | |
2518 | hlist_bl_unlock(b); | |
2519 | INIT_HLIST_NODE(&dentry->d_u.d_alias); | |
2520 | INIT_LIST_HEAD(&dentry->d_lru); | |
2521 | } | |
2522 | EXPORT_SYMBOL(__d_lookup_done); | |
2523 | ||
2524 | /* inode->i_lock held if inode is non-NULL */ | |
2525 | ||
2526 | static inline void __d_add(struct dentry *dentry, struct inode *inode) | |
2527 | { | |
2528 | struct inode *dir = NULL; | |
2529 | unsigned n; | |
2530 | spin_lock(&dentry->d_lock); | |
2531 | if (unlikely(d_in_lookup(dentry))) { | |
2532 | dir = dentry->d_parent->d_inode; | |
2533 | n = start_dir_add(dir); | |
2534 | __d_lookup_done(dentry); | |
2535 | } | |
2536 | if (inode) { | |
2537 | unsigned add_flags = d_flags_for_inode(inode); | |
2538 | hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); | |
2539 | raw_write_seqcount_begin(&dentry->d_seq); | |
2540 | __d_set_inode_and_type(dentry, inode, add_flags); | |
2541 | raw_write_seqcount_end(&dentry->d_seq); | |
2542 | fsnotify_update_flags(dentry); | |
2543 | } | |
2544 | __d_rehash(dentry); | |
2545 | if (dir) | |
2546 | end_dir_add(dir, n); | |
2547 | spin_unlock(&dentry->d_lock); | |
2548 | if (inode) | |
2549 | spin_unlock(&inode->i_lock); | |
2550 | } | |
2551 | ||
2552 | /** | |
2553 | * d_add - add dentry to hash queues | |
2554 | * @entry: dentry to add | |
2555 | * @inode: The inode to attach to this dentry | |
2556 | * | |
2557 | * This adds the entry to the hash queues and initializes @inode. | |
2558 | * The entry was actually filled in earlier during d_alloc(). | |
2559 | */ | |
2560 | ||
2561 | void d_add(struct dentry *entry, struct inode *inode) | |
2562 | { | |
2563 | if (inode) { | |
2564 | security_d_instantiate(entry, inode); | |
2565 | spin_lock(&inode->i_lock); | |
2566 | } | |
2567 | __d_add(entry, inode); | |
2568 | } | |
2569 | EXPORT_SYMBOL(d_add); | |
2570 | ||
2571 | /** | |
2572 | * d_exact_alias - find and hash an exact unhashed alias | |
2573 | * @entry: dentry to add | |
2574 | * @inode: The inode to go with this dentry | |
2575 | * | |
2576 | * If an unhashed dentry with the same name/parent and desired | |
2577 | * inode already exists, hash and return it. Otherwise, return | |
2578 | * NULL. | |
2579 | * | |
2580 | * Parent directory should be locked. | |
2581 | */ | |
2582 | struct dentry *d_exact_alias(struct dentry *entry, struct inode *inode) | |
2583 | { | |
2584 | struct dentry *alias; | |
2585 | unsigned int hash = entry->d_name.hash; | |
2586 | ||
2587 | spin_lock(&inode->i_lock); | |
2588 | hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) { | |
2589 | /* | |
2590 | * Don't need alias->d_lock here, because aliases with | |
2591 | * d_parent == entry->d_parent are not subject to name or | |
2592 | * parent changes, because the parent inode i_mutex is held. | |
2593 | */ | |
2594 | if (alias->d_name.hash != hash) | |
2595 | continue; | |
2596 | if (alias->d_parent != entry->d_parent) | |
2597 | continue; | |
2598 | if (!d_same_name(alias, entry->d_parent, &entry->d_name)) | |
2599 | continue; | |
2600 | spin_lock(&alias->d_lock); | |
2601 | if (!d_unhashed(alias)) { | |
2602 | spin_unlock(&alias->d_lock); | |
2603 | alias = NULL; | |
2604 | } else { | |
2605 | __dget_dlock(alias); | |
2606 | __d_rehash(alias); | |
2607 | spin_unlock(&alias->d_lock); | |
2608 | } | |
2609 | spin_unlock(&inode->i_lock); | |
2610 | return alias; | |
2611 | } | |
2612 | spin_unlock(&inode->i_lock); | |
2613 | return NULL; | |
2614 | } | |
2615 | EXPORT_SYMBOL(d_exact_alias); | |
2616 | ||
2617 | /** | |
2618 | * dentry_update_name_case - update case insensitive dentry with a new name | |
2619 | * @dentry: dentry to be updated | |
2620 | * @name: new name | |
2621 | * | |
2622 | * Update a case insensitive dentry with new case of name. | |
2623 | * | |
2624 | * dentry must have been returned by d_lookup with name @name. Old and new | |
2625 | * name lengths must match (ie. no d_compare which allows mismatched name | |
2626 | * lengths). | |
2627 | * | |
2628 | * Parent inode i_mutex must be held over d_lookup and into this call (to | |
2629 | * keep renames and concurrent inserts, and readdir(2) away). | |
2630 | */ | |
2631 | void dentry_update_name_case(struct dentry *dentry, const struct qstr *name) | |
2632 | { | |
2633 | BUG_ON(!inode_is_locked(dentry->d_parent->d_inode)); | |
2634 | BUG_ON(dentry->d_name.len != name->len); /* d_lookup gives this */ | |
2635 | ||
2636 | spin_lock(&dentry->d_lock); | |
2637 | write_seqcount_begin(&dentry->d_seq); | |
2638 | memcpy((unsigned char *)dentry->d_name.name, name->name, name->len); | |
2639 | write_seqcount_end(&dentry->d_seq); | |
2640 | spin_unlock(&dentry->d_lock); | |
2641 | } | |
2642 | EXPORT_SYMBOL(dentry_update_name_case); | |
2643 | ||
2644 | static void swap_names(struct dentry *dentry, struct dentry *target) | |
2645 | { | |
2646 | if (unlikely(dname_external(target))) { | |
2647 | if (unlikely(dname_external(dentry))) { | |
2648 | /* | |
2649 | * Both external: swap the pointers | |
2650 | */ | |
2651 | swap(target->d_name.name, dentry->d_name.name); | |
2652 | } else { | |
2653 | /* | |
2654 | * dentry:internal, target:external. Steal target's | |
2655 | * storage and make target internal. | |
2656 | */ | |
2657 | memcpy(target->d_iname, dentry->d_name.name, | |
2658 | dentry->d_name.len + 1); | |
2659 | dentry->d_name.name = target->d_name.name; | |
2660 | target->d_name.name = target->d_iname; | |
2661 | } | |
2662 | } else { | |
2663 | if (unlikely(dname_external(dentry))) { | |
2664 | /* | |
2665 | * dentry:external, target:internal. Give dentry's | |
2666 | * storage to target and make dentry internal | |
2667 | */ | |
2668 | memcpy(dentry->d_iname, target->d_name.name, | |
2669 | target->d_name.len + 1); | |
2670 | target->d_name.name = dentry->d_name.name; | |
2671 | dentry->d_name.name = dentry->d_iname; | |
2672 | } else { | |
2673 | /* | |
2674 | * Both are internal. | |
2675 | */ | |
2676 | unsigned int i; | |
2677 | BUILD_BUG_ON(!IS_ALIGNED(DNAME_INLINE_LEN, sizeof(long))); | |
2678 | kmemcheck_mark_initialized(dentry->d_iname, DNAME_INLINE_LEN); | |
2679 | kmemcheck_mark_initialized(target->d_iname, DNAME_INLINE_LEN); | |
2680 | for (i = 0; i < DNAME_INLINE_LEN / sizeof(long); i++) { | |
2681 | swap(((long *) &dentry->d_iname)[i], | |
2682 | ((long *) &target->d_iname)[i]); | |
2683 | } | |
2684 | } | |
2685 | } | |
2686 | swap(dentry->d_name.hash_len, target->d_name.hash_len); | |
2687 | } | |
2688 | ||
2689 | static void copy_name(struct dentry *dentry, struct dentry *target) | |
2690 | { | |
2691 | struct external_name *old_name = NULL; | |
2692 | if (unlikely(dname_external(dentry))) | |
2693 | old_name = external_name(dentry); | |
2694 | if (unlikely(dname_external(target))) { | |
2695 | atomic_inc(&external_name(target)->u.count); | |
2696 | dentry->d_name = target->d_name; | |
2697 | } else { | |
2698 | memcpy(dentry->d_iname, target->d_name.name, | |
2699 | target->d_name.len + 1); | |
2700 | dentry->d_name.name = dentry->d_iname; | |
2701 | dentry->d_name.hash_len = target->d_name.hash_len; | |
2702 | } | |
2703 | if (old_name && likely(atomic_dec_and_test(&old_name->u.count))) | |
2704 | kfree_rcu(old_name, u.head); | |
2705 | } | |
2706 | ||
2707 | static void dentry_lock_for_move(struct dentry *dentry, struct dentry *target) | |
2708 | { | |
2709 | /* | |
2710 | * XXXX: do we really need to take target->d_lock? | |
2711 | */ | |
2712 | if (IS_ROOT(dentry) || dentry->d_parent == target->d_parent) | |
2713 | spin_lock(&target->d_parent->d_lock); | |
2714 | else { | |
2715 | if (d_ancestor(dentry->d_parent, target->d_parent)) { | |
2716 | spin_lock(&dentry->d_parent->d_lock); | |
2717 | spin_lock_nested(&target->d_parent->d_lock, | |
2718 | DENTRY_D_LOCK_NESTED); | |
2719 | } else { | |
2720 | spin_lock(&target->d_parent->d_lock); | |
2721 | spin_lock_nested(&dentry->d_parent->d_lock, | |
2722 | DENTRY_D_LOCK_NESTED); | |
2723 | } | |
2724 | } | |
2725 | if (target < dentry) { | |
2726 | spin_lock_nested(&target->d_lock, 2); | |
2727 | spin_lock_nested(&dentry->d_lock, 3); | |
2728 | } else { | |
2729 | spin_lock_nested(&dentry->d_lock, 2); | |
2730 | spin_lock_nested(&target->d_lock, 3); | |
2731 | } | |
2732 | } | |
2733 | ||
2734 | static void dentry_unlock_for_move(struct dentry *dentry, struct dentry *target) | |
2735 | { | |
2736 | if (target->d_parent != dentry->d_parent) | |
2737 | spin_unlock(&dentry->d_parent->d_lock); | |
2738 | if (target->d_parent != target) | |
2739 | spin_unlock(&target->d_parent->d_lock); | |
2740 | spin_unlock(&target->d_lock); | |
2741 | spin_unlock(&dentry->d_lock); | |
2742 | } | |
2743 | ||
2744 | /* | |
2745 | * When switching names, the actual string doesn't strictly have to | |
2746 | * be preserved in the target - because we're dropping the target | |
2747 | * anyway. As such, we can just do a simple memcpy() to copy over | |
2748 | * the new name before we switch, unless we are going to rehash | |
2749 | * it. Note that if we *do* unhash the target, we are not allowed | |
2750 | * to rehash it without giving it a new name/hash key - whether | |
2751 | * we swap or overwrite the names here, resulting name won't match | |
2752 | * the reality in filesystem; it's only there for d_path() purposes. | |
2753 | * Note that all of this is happening under rename_lock, so the | |
2754 | * any hash lookup seeing it in the middle of manipulations will | |
2755 | * be discarded anyway. So we do not care what happens to the hash | |
2756 | * key in that case. | |
2757 | */ | |
2758 | /* | |
2759 | * __d_move - move a dentry | |
2760 | * @dentry: entry to move | |
2761 | * @target: new dentry | |
2762 | * @exchange: exchange the two dentries | |
2763 | * | |
2764 | * Update the dcache to reflect the move of a file name. Negative | |
2765 | * dcache entries should not be moved in this way. Caller must hold | |
2766 | * rename_lock, the i_mutex of the source and target directories, | |
2767 | * and the sb->s_vfs_rename_mutex if they differ. See lock_rename(). | |
2768 | */ | |
2769 | static void __d_move(struct dentry *dentry, struct dentry *target, | |
2770 | bool exchange) | |
2771 | { | |
2772 | struct inode *dir = NULL; | |
2773 | unsigned n; | |
2774 | if (!dentry->d_inode) | |
2775 | printk(KERN_WARNING "VFS: moving negative dcache entry\n"); | |
2776 | ||
2777 | BUG_ON(d_ancestor(dentry, target)); | |
2778 | BUG_ON(d_ancestor(target, dentry)); | |
2779 | ||
2780 | dentry_lock_for_move(dentry, target); | |
2781 | if (unlikely(d_in_lookup(target))) { | |
2782 | dir = target->d_parent->d_inode; | |
2783 | n = start_dir_add(dir); | |
2784 | __d_lookup_done(target); | |
2785 | } | |
2786 | ||
2787 | write_seqcount_begin(&dentry->d_seq); | |
2788 | write_seqcount_begin_nested(&target->d_seq, DENTRY_D_LOCK_NESTED); | |
2789 | ||
2790 | /* unhash both */ | |
2791 | /* __d_drop does write_seqcount_barrier, but they're OK to nest. */ | |
2792 | __d_drop(dentry); | |
2793 | __d_drop(target); | |
2794 | ||
2795 | /* Switch the names.. */ | |
2796 | if (exchange) | |
2797 | swap_names(dentry, target); | |
2798 | else | |
2799 | copy_name(dentry, target); | |
2800 | ||
2801 | /* rehash in new place(s) */ | |
2802 | __d_rehash(dentry); | |
2803 | if (exchange) | |
2804 | __d_rehash(target); | |
2805 | ||
2806 | /* ... and switch them in the tree */ | |
2807 | if (IS_ROOT(dentry)) { | |
2808 | /* splicing a tree */ | |
2809 | dentry->d_flags |= DCACHE_RCUACCESS; | |
2810 | dentry->d_parent = target->d_parent; | |
2811 | target->d_parent = target; | |
2812 | list_del_init(&target->d_child); | |
2813 | list_move(&dentry->d_child, &dentry->d_parent->d_subdirs); | |
2814 | } else { | |
2815 | /* swapping two dentries */ | |
2816 | swap(dentry->d_parent, target->d_parent); | |
2817 | list_move(&target->d_child, &target->d_parent->d_subdirs); | |
2818 | list_move(&dentry->d_child, &dentry->d_parent->d_subdirs); | |
2819 | if (exchange) | |
2820 | fsnotify_update_flags(target); | |
2821 | fsnotify_update_flags(dentry); | |
2822 | } | |
2823 | ||
2824 | write_seqcount_end(&target->d_seq); | |
2825 | write_seqcount_end(&dentry->d_seq); | |
2826 | ||
2827 | if (dir) | |
2828 | end_dir_add(dir, n); | |
2829 | dentry_unlock_for_move(dentry, target); | |
2830 | } | |
2831 | ||
2832 | /* | |
2833 | * d_move - move a dentry | |
2834 | * @dentry: entry to move | |
2835 | * @target: new dentry | |
2836 | * | |
2837 | * Update the dcache to reflect the move of a file name. Negative | |
2838 | * dcache entries should not be moved in this way. See the locking | |
2839 | * requirements for __d_move. | |
2840 | */ | |
2841 | void d_move(struct dentry *dentry, struct dentry *target) | |
2842 | { | |
2843 | write_seqlock(&rename_lock); | |
2844 | __d_move(dentry, target, false); | |
2845 | write_sequnlock(&rename_lock); | |
2846 | } | |
2847 | EXPORT_SYMBOL(d_move); | |
2848 | ||
2849 | /* | |
2850 | * d_exchange - exchange two dentries | |
2851 | * @dentry1: first dentry | |
2852 | * @dentry2: second dentry | |
2853 | */ | |
2854 | void d_exchange(struct dentry *dentry1, struct dentry *dentry2) | |
2855 | { | |
2856 | write_seqlock(&rename_lock); | |
2857 | ||
2858 | WARN_ON(!dentry1->d_inode); | |
2859 | WARN_ON(!dentry2->d_inode); | |
2860 | WARN_ON(IS_ROOT(dentry1)); | |
2861 | WARN_ON(IS_ROOT(dentry2)); | |
2862 | ||
2863 | __d_move(dentry1, dentry2, true); | |
2864 | ||
2865 | write_sequnlock(&rename_lock); | |
2866 | } | |
2867 | ||
2868 | /** | |
2869 | * d_ancestor - search for an ancestor | |
2870 | * @p1: ancestor dentry | |
2871 | * @p2: child dentry | |
2872 | * | |
2873 | * Returns the ancestor dentry of p2 which is a child of p1, if p1 is | |
2874 | * an ancestor of p2, else NULL. | |
2875 | */ | |
2876 | struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2) | |
2877 | { | |
2878 | struct dentry *p; | |
2879 | ||
2880 | for (p = p2; !IS_ROOT(p); p = p->d_parent) { | |
2881 | if (p->d_parent == p1) | |
2882 | return p; | |
2883 | } | |
2884 | return NULL; | |
2885 | } | |
2886 | ||
2887 | /* | |
2888 | * This helper attempts to cope with remotely renamed directories | |
2889 | * | |
2890 | * It assumes that the caller is already holding | |
2891 | * dentry->d_parent->d_inode->i_mutex, and rename_lock | |
2892 | * | |
2893 | * Note: If ever the locking in lock_rename() changes, then please | |
2894 | * remember to update this too... | |
2895 | */ | |
2896 | static int __d_unalias(struct inode *inode, | |
2897 | struct dentry *dentry, struct dentry *alias) | |
2898 | { | |
2899 | struct mutex *m1 = NULL; | |
2900 | struct rw_semaphore *m2 = NULL; | |
2901 | int ret = -ESTALE; | |
2902 | ||
2903 | /* If alias and dentry share a parent, then no extra locks required */ | |
2904 | if (alias->d_parent == dentry->d_parent) | |
2905 | goto out_unalias; | |
2906 | ||
2907 | /* See lock_rename() */ | |
2908 | if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex)) | |
2909 | goto out_err; | |
2910 | m1 = &dentry->d_sb->s_vfs_rename_mutex; | |
2911 | if (!inode_trylock_shared(alias->d_parent->d_inode)) | |
2912 | goto out_err; | |
2913 | m2 = &alias->d_parent->d_inode->i_rwsem; | |
2914 | out_unalias: | |
2915 | __d_move(alias, dentry, false); | |
2916 | ret = 0; | |
2917 | out_err: | |
2918 | if (m2) | |
2919 | up_read(m2); | |
2920 | if (m1) | |
2921 | mutex_unlock(m1); | |
2922 | return ret; | |
2923 | } | |
2924 | ||
2925 | /** | |
2926 | * d_splice_alias - splice a disconnected dentry into the tree if one exists | |
2927 | * @inode: the inode which may have a disconnected dentry | |
2928 | * @dentry: a negative dentry which we want to point to the inode. | |
2929 | * | |
2930 | * If inode is a directory and has an IS_ROOT alias, then d_move that in | |
2931 | * place of the given dentry and return it, else simply d_add the inode | |
2932 | * to the dentry and return NULL. | |
2933 | * | |
2934 | * If a non-IS_ROOT directory is found, the filesystem is corrupt, and | |
2935 | * we should error out: directories can't have multiple aliases. | |
2936 | * | |
2937 | * This is needed in the lookup routine of any filesystem that is exportable | |
2938 | * (via knfsd) so that we can build dcache paths to directories effectively. | |
2939 | * | |
2940 | * If a dentry was found and moved, then it is returned. Otherwise NULL | |
2941 | * is returned. This matches the expected return value of ->lookup. | |
2942 | * | |
2943 | * Cluster filesystems may call this function with a negative, hashed dentry. | |
2944 | * In that case, we know that the inode will be a regular file, and also this | |
2945 | * will only occur during atomic_open. So we need to check for the dentry | |
2946 | * being already hashed only in the final case. | |
2947 | */ | |
2948 | struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) | |
2949 | { | |
2950 | if (IS_ERR(inode)) | |
2951 | return ERR_CAST(inode); | |
2952 | ||
2953 | BUG_ON(!d_unhashed(dentry)); | |
2954 | ||
2955 | if (!inode) | |
2956 | goto out; | |
2957 | ||
2958 | security_d_instantiate(dentry, inode); | |
2959 | spin_lock(&inode->i_lock); | |
2960 | if (S_ISDIR(inode->i_mode)) { | |
2961 | struct dentry *new = __d_find_any_alias(inode); | |
2962 | if (unlikely(new)) { | |
2963 | /* The reference to new ensures it remains an alias */ | |
2964 | spin_unlock(&inode->i_lock); | |
2965 | write_seqlock(&rename_lock); | |
2966 | if (unlikely(d_ancestor(new, dentry))) { | |
2967 | write_sequnlock(&rename_lock); | |
2968 | dput(new); | |
2969 | new = ERR_PTR(-ELOOP); | |
2970 | pr_warn_ratelimited( | |
2971 | "VFS: Lookup of '%s' in %s %s" | |
2972 | " would have caused loop\n", | |
2973 | dentry->d_name.name, | |
2974 | inode->i_sb->s_type->name, | |
2975 | inode->i_sb->s_id); | |
2976 | } else if (!IS_ROOT(new)) { | |
2977 | int err = __d_unalias(inode, dentry, new); | |
2978 | write_sequnlock(&rename_lock); | |
2979 | if (err) { | |
2980 | dput(new); | |
2981 | new = ERR_PTR(err); | |
2982 | } | |
2983 | } else { | |
2984 | __d_move(new, dentry, false); | |
2985 | write_sequnlock(&rename_lock); | |
2986 | } | |
2987 | iput(inode); | |
2988 | return new; | |
2989 | } | |
2990 | } | |
2991 | out: | |
2992 | __d_add(dentry, inode); | |
2993 | return NULL; | |
2994 | } | |
2995 | EXPORT_SYMBOL(d_splice_alias); | |
2996 | ||
2997 | static int prepend(char **buffer, int *buflen, const char *str, int namelen) | |
2998 | { | |
2999 | *buflen -= namelen; | |
3000 | if (*buflen < 0) | |
3001 | return -ENAMETOOLONG; | |
3002 | *buffer -= namelen; | |
3003 | memcpy(*buffer, str, namelen); | |
3004 | return 0; | |
3005 | } | |
3006 | ||
3007 | /** | |
3008 | * prepend_name - prepend a pathname in front of current buffer pointer | |
3009 | * @buffer: buffer pointer | |
3010 | * @buflen: allocated length of the buffer | |
3011 | * @name: name string and length qstr structure | |
3012 | * | |
3013 | * With RCU path tracing, it may race with d_move(). Use ACCESS_ONCE() to | |
3014 | * make sure that either the old or the new name pointer and length are | |
3015 | * fetched. However, there may be mismatch between length and pointer. | |
3016 | * The length cannot be trusted, we need to copy it byte-by-byte until | |
3017 | * the length is reached or a null byte is found. It also prepends "/" at | |
3018 | * the beginning of the name. The sequence number check at the caller will | |
3019 | * retry it again when a d_move() does happen. So any garbage in the buffer | |
3020 | * due to mismatched pointer and length will be discarded. | |
3021 | * | |
3022 | * Data dependency barrier is needed to make sure that we see that terminating | |
3023 | * NUL. Alpha strikes again, film at 11... | |
3024 | */ | |
3025 | static int prepend_name(char **buffer, int *buflen, const struct qstr *name) | |
3026 | { | |
3027 | const char *dname = ACCESS_ONCE(name->name); | |
3028 | u32 dlen = ACCESS_ONCE(name->len); | |
3029 | char *p; | |
3030 | ||
3031 | smp_read_barrier_depends(); | |
3032 | ||
3033 | *buflen -= dlen + 1; | |
3034 | if (*buflen < 0) | |
3035 | return -ENAMETOOLONG; | |
3036 | p = *buffer -= dlen + 1; | |
3037 | *p++ = '/'; | |
3038 | while (dlen--) { | |
3039 | char c = *dname++; | |
3040 | if (!c) | |
3041 | break; | |
3042 | *p++ = c; | |
3043 | } | |
3044 | return 0; | |
3045 | } | |
3046 | ||
3047 | /** | |
3048 | * prepend_path - Prepend path string to a buffer | |
3049 | * @path: the dentry/vfsmount to report | |
3050 | * @root: root vfsmnt/dentry | |
3051 | * @buffer: pointer to the end of the buffer | |
3052 | * @buflen: pointer to buffer length | |
3053 | * | |
3054 | * The function will first try to write out the pathname without taking any | |
3055 | * lock other than the RCU read lock to make sure that dentries won't go away. | |
3056 | * It only checks the sequence number of the global rename_lock as any change | |
3057 | * in the dentry's d_seq will be preceded by changes in the rename_lock | |
3058 | * sequence number. If the sequence number had been changed, it will restart | |
3059 | * the whole pathname back-tracing sequence again by taking the rename_lock. | |
3060 | * In this case, there is no need to take the RCU read lock as the recursive | |
3061 | * parent pointer references will keep the dentry chain alive as long as no | |
3062 | * rename operation is performed. | |
3063 | */ | |
3064 | static int prepend_path(const struct path *path, | |
3065 | const struct path *root, | |
3066 | char **buffer, int *buflen) | |
3067 | { | |
3068 | struct dentry *dentry; | |
3069 | struct vfsmount *vfsmnt; | |
3070 | struct mount *mnt; | |
3071 | int error = 0; | |
3072 | unsigned seq, m_seq = 0; | |
3073 | char *bptr; | |
3074 | int blen; | |
3075 | ||
3076 | rcu_read_lock(); | |
3077 | restart_mnt: | |
3078 | read_seqbegin_or_lock(&mount_lock, &m_seq); | |
3079 | seq = 0; | |
3080 | rcu_read_lock(); | |
3081 | restart: | |
3082 | bptr = *buffer; | |
3083 | blen = *buflen; | |
3084 | error = 0; | |
3085 | dentry = path->dentry; | |
3086 | vfsmnt = path->mnt; | |
3087 | mnt = real_mount(vfsmnt); | |
3088 | read_seqbegin_or_lock(&rename_lock, &seq); | |
3089 | while (dentry != root->dentry || vfsmnt != root->mnt) { | |
3090 | struct dentry * parent; | |
3091 | ||
3092 | if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { | |
3093 | struct mount *parent = ACCESS_ONCE(mnt->mnt_parent); | |
3094 | /* Escaped? */ | |
3095 | if (dentry != vfsmnt->mnt_root) { | |
3096 | bptr = *buffer; | |
3097 | blen = *buflen; | |
3098 | error = 3; | |
3099 | break; | |
3100 | } | |
3101 | /* Global root? */ | |
3102 | if (mnt != parent) { | |
3103 | dentry = ACCESS_ONCE(mnt->mnt_mountpoint); | |
3104 | mnt = parent; | |
3105 | vfsmnt = &mnt->mnt; | |
3106 | continue; | |
3107 | } | |
3108 | if (!error) | |
3109 | error = is_mounted(vfsmnt) ? 1 : 2; | |
3110 | break; | |
3111 | } | |
3112 | parent = dentry->d_parent; | |
3113 | prefetch(parent); | |
3114 | error = prepend_name(&bptr, &blen, &dentry->d_name); | |
3115 | if (error) | |
3116 | break; | |
3117 | ||
3118 | dentry = parent; | |
3119 | } | |
3120 | if (!(seq & 1)) | |
3121 | rcu_read_unlock(); | |
3122 | if (need_seqretry(&rename_lock, seq)) { | |
3123 | seq = 1; | |
3124 | goto restart; | |
3125 | } | |
3126 | done_seqretry(&rename_lock, seq); | |
3127 | ||
3128 | if (!(m_seq & 1)) | |
3129 | rcu_read_unlock(); | |
3130 | if (need_seqretry(&mount_lock, m_seq)) { | |
3131 | m_seq = 1; | |
3132 | goto restart_mnt; | |
3133 | } | |
3134 | done_seqretry(&mount_lock, m_seq); | |
3135 | ||
3136 | if (error >= 0 && bptr == *buffer) { | |
3137 | if (--blen < 0) | |
3138 | error = -ENAMETOOLONG; | |
3139 | else | |
3140 | *--bptr = '/'; | |
3141 | } | |
3142 | *buffer = bptr; | |
3143 | *buflen = blen; | |
3144 | return error; | |
3145 | } | |
3146 | ||
3147 | /** | |
3148 | * __d_path - return the path of a dentry | |
3149 | * @path: the dentry/vfsmount to report | |
3150 | * @root: root vfsmnt/dentry | |
3151 | * @buf: buffer to return value in | |
3152 | * @buflen: buffer length | |
3153 | * | |
3154 | * Convert a dentry into an ASCII path name. | |
3155 | * | |
3156 | * Returns a pointer into the buffer or an error code if the | |
3157 | * path was too long. | |
3158 | * | |
3159 | * "buflen" should be positive. | |
3160 | * | |
3161 | * If the path is not reachable from the supplied root, return %NULL. | |
3162 | */ | |
3163 | char *__d_path(const struct path *path, | |
3164 | const struct path *root, | |
3165 | char *buf, int buflen) | |
3166 | { | |
3167 | char *res = buf + buflen; | |
3168 | int error; | |
3169 | ||
3170 | prepend(&res, &buflen, "\0", 1); | |
3171 | error = prepend_path(path, root, &res, &buflen); | |
3172 | ||
3173 | if (error < 0) | |
3174 | return ERR_PTR(error); | |
3175 | if (error > 0) | |
3176 | return NULL; | |
3177 | return res; | |
3178 | } | |
3179 | ||
3180 | char *d_absolute_path(const struct path *path, | |
3181 | char *buf, int buflen) | |
3182 | { | |
3183 | struct path root = {}; | |
3184 | char *res = buf + buflen; | |
3185 | int error; | |
3186 | ||
3187 | prepend(&res, &buflen, "\0", 1); | |
3188 | error = prepend_path(path, &root, &res, &buflen); | |
3189 | ||
3190 | if (error > 1) | |
3191 | error = -EINVAL; | |
3192 | if (error < 0) | |
3193 | return ERR_PTR(error); | |
3194 | return res; | |
3195 | } | |
3196 | ||
3197 | /* | |
3198 | * same as __d_path but appends "(deleted)" for unlinked files. | |
3199 | */ | |
3200 | static int path_with_deleted(const struct path *path, | |
3201 | const struct path *root, | |
3202 | char **buf, int *buflen) | |
3203 | { | |
3204 | prepend(buf, buflen, "\0", 1); | |
3205 | if (d_unlinked(path->dentry)) { | |
3206 | int error = prepend(buf, buflen, " (deleted)", 10); | |
3207 | if (error) | |
3208 | return error; | |
3209 | } | |
3210 | ||
3211 | return prepend_path(path, root, buf, buflen); | |
3212 | } | |
3213 | ||
3214 | static int prepend_unreachable(char **buffer, int *buflen) | |
3215 | { | |
3216 | return prepend(buffer, buflen, "(unreachable)", 13); | |
3217 | } | |
3218 | ||
3219 | static void get_fs_root_rcu(struct fs_struct *fs, struct path *root) | |
3220 | { | |
3221 | unsigned seq; | |
3222 | ||
3223 | do { | |
3224 | seq = read_seqcount_begin(&fs->seq); | |
3225 | *root = fs->root; | |
3226 | } while (read_seqcount_retry(&fs->seq, seq)); | |
3227 | } | |
3228 | ||
3229 | /** | |
3230 | * d_path - return the path of a dentry | |
3231 | * @path: path to report | |
3232 | * @buf: buffer to return value in | |
3233 | * @buflen: buffer length | |
3234 | * | |
3235 | * Convert a dentry into an ASCII path name. If the entry has been deleted | |
3236 | * the string " (deleted)" is appended. Note that this is ambiguous. | |
3237 | * | |
3238 | * Returns a pointer into the buffer or an error code if the path was | |
3239 | * too long. Note: Callers should use the returned pointer, not the passed | |
3240 | * in buffer, to use the name! The implementation often starts at an offset | |
3241 | * into the buffer, and may leave 0 bytes at the start. | |
3242 | * | |
3243 | * "buflen" should be positive. | |
3244 | */ | |
3245 | char *d_path(const struct path *path, char *buf, int buflen) | |
3246 | { | |
3247 | char *res = buf + buflen; | |
3248 | struct path root; | |
3249 | int error; | |
3250 | ||
3251 | /* | |
3252 | * We have various synthetic filesystems that never get mounted. On | |
3253 | * these filesystems dentries are never used for lookup purposes, and | |
3254 | * thus don't need to be hashed. They also don't need a name until a | |
3255 | * user wants to identify the object in /proc/pid/fd/. The little hack | |
3256 | * below allows us to generate a name for these objects on demand: | |
3257 | * | |
3258 | * Some pseudo inodes are mountable. When they are mounted | |
3259 | * path->dentry == path->mnt->mnt_root. In that case don't call d_dname | |
3260 | * and instead have d_path return the mounted path. | |
3261 | */ | |
3262 | if (path->dentry->d_op && path->dentry->d_op->d_dname && | |
3263 | (!IS_ROOT(path->dentry) || path->dentry != path->mnt->mnt_root)) | |
3264 | return path->dentry->d_op->d_dname(path->dentry, buf, buflen); | |
3265 | ||
3266 | rcu_read_lock(); | |
3267 | get_fs_root_rcu(current->fs, &root); | |
3268 | error = path_with_deleted(path, &root, &res, &buflen); | |
3269 | rcu_read_unlock(); | |
3270 | ||
3271 | if (error < 0) | |
3272 | res = ERR_PTR(error); | |
3273 | return res; | |
3274 | } | |
3275 | EXPORT_SYMBOL(d_path); | |
3276 | ||
3277 | /* | |
3278 | * Helper function for dentry_operations.d_dname() members | |
3279 | */ | |
3280 | char *dynamic_dname(struct dentry *dentry, char *buffer, int buflen, | |
3281 | const char *fmt, ...) | |
3282 | { | |
3283 | va_list args; | |
3284 | char temp[64]; | |
3285 | int sz; | |
3286 | ||
3287 | va_start(args, fmt); | |
3288 | sz = vsnprintf(temp, sizeof(temp), fmt, args) + 1; | |
3289 | va_end(args); | |
3290 | ||
3291 | if (sz > sizeof(temp) || sz > buflen) | |
3292 | return ERR_PTR(-ENAMETOOLONG); | |
3293 | ||
3294 | buffer += buflen - sz; | |
3295 | return memcpy(buffer, temp, sz); | |
3296 | } | |
3297 | ||
3298 | char *simple_dname(struct dentry *dentry, char *buffer, int buflen) | |
3299 | { | |
3300 | char *end = buffer + buflen; | |
3301 | /* these dentries are never renamed, so d_lock is not needed */ | |
3302 | if (prepend(&end, &buflen, " (deleted)", 11) || | |
3303 | prepend(&end, &buflen, dentry->d_name.name, dentry->d_name.len) || | |
3304 | prepend(&end, &buflen, "/", 1)) | |
3305 | end = ERR_PTR(-ENAMETOOLONG); | |
3306 | return end; | |
3307 | } | |
3308 | EXPORT_SYMBOL(simple_dname); | |
3309 | ||
3310 | /* | |
3311 | * Write full pathname from the root of the filesystem into the buffer. | |
3312 | */ | |
3313 | static char *__dentry_path(struct dentry *d, char *buf, int buflen) | |
3314 | { | |
3315 | struct dentry *dentry; | |
3316 | char *end, *retval; | |
3317 | int len, seq = 0; | |
3318 | int error = 0; | |
3319 | ||
3320 | if (buflen < 2) | |
3321 | goto Elong; | |
3322 | ||
3323 | rcu_read_lock(); | |
3324 | restart: | |
3325 | dentry = d; | |
3326 | end = buf + buflen; | |
3327 | len = buflen; | |
3328 | prepend(&end, &len, "\0", 1); | |
3329 | /* Get '/' right */ | |
3330 | retval = end-1; | |
3331 | *retval = '/'; | |
3332 | read_seqbegin_or_lock(&rename_lock, &seq); | |
3333 | while (!IS_ROOT(dentry)) { | |
3334 | struct dentry *parent = dentry->d_parent; | |
3335 | ||
3336 | prefetch(parent); | |
3337 | error = prepend_name(&end, &len, &dentry->d_name); | |
3338 | if (error) | |
3339 | break; | |
3340 | ||
3341 | retval = end; | |
3342 | dentry = parent; | |
3343 | } | |
3344 | if (!(seq & 1)) | |
3345 | rcu_read_unlock(); | |
3346 | if (need_seqretry(&rename_lock, seq)) { | |
3347 | seq = 1; | |
3348 | goto restart; | |
3349 | } | |
3350 | done_seqretry(&rename_lock, seq); | |
3351 | if (error) | |
3352 | goto Elong; | |
3353 | return retval; | |
3354 | Elong: | |
3355 | return ERR_PTR(-ENAMETOOLONG); | |
3356 | } | |
3357 | ||
3358 | char *dentry_path_raw(struct dentry *dentry, char *buf, int buflen) | |
3359 | { | |
3360 | return __dentry_path(dentry, buf, buflen); | |
3361 | } | |
3362 | EXPORT_SYMBOL(dentry_path_raw); | |
3363 | ||
3364 | char *dentry_path(struct dentry *dentry, char *buf, int buflen) | |
3365 | { | |
3366 | char *p = NULL; | |
3367 | char *retval; | |
3368 | ||
3369 | if (d_unlinked(dentry)) { | |
3370 | p = buf + buflen; | |
3371 | if (prepend(&p, &buflen, "//deleted", 10) != 0) | |
3372 | goto Elong; | |
3373 | buflen++; | |
3374 | } | |
3375 | retval = __dentry_path(dentry, buf, buflen); | |
3376 | if (!IS_ERR(retval) && p) | |
3377 | *p = '/'; /* restore '/' overriden with '\0' */ | |
3378 | return retval; | |
3379 | Elong: | |
3380 | return ERR_PTR(-ENAMETOOLONG); | |
3381 | } | |
3382 | ||
3383 | static void get_fs_root_and_pwd_rcu(struct fs_struct *fs, struct path *root, | |
3384 | struct path *pwd) | |
3385 | { | |
3386 | unsigned seq; | |
3387 | ||
3388 | do { | |
3389 | seq = read_seqcount_begin(&fs->seq); | |
3390 | *root = fs->root; | |
3391 | *pwd = fs->pwd; | |
3392 | } while (read_seqcount_retry(&fs->seq, seq)); | |
3393 | } | |
3394 | ||
3395 | /* | |
3396 | * NOTE! The user-level library version returns a | |
3397 | * character pointer. The kernel system call just | |
3398 | * returns the length of the buffer filled (which | |
3399 | * includes the ending '\0' character), or a negative | |
3400 | * error value. So libc would do something like | |
3401 | * | |
3402 | * char *getcwd(char * buf, size_t size) | |
3403 | * { | |
3404 | * int retval; | |
3405 | * | |
3406 | * retval = sys_getcwd(buf, size); | |
3407 | * if (retval >= 0) | |
3408 | * return buf; | |
3409 | * errno = -retval; | |
3410 | * return NULL; | |
3411 | * } | |
3412 | */ | |
3413 | SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) | |
3414 | { | |
3415 | int error; | |
3416 | struct path pwd, root; | |
3417 | char *page = __getname(); | |
3418 | ||
3419 | if (!page) | |
3420 | return -ENOMEM; | |
3421 | ||
3422 | rcu_read_lock(); | |
3423 | get_fs_root_and_pwd_rcu(current->fs, &root, &pwd); | |
3424 | ||
3425 | error = -ENOENT; | |
3426 | if (!d_unlinked(pwd.dentry)) { | |
3427 | unsigned long len; | |
3428 | char *cwd = page + PATH_MAX; | |
3429 | int buflen = PATH_MAX; | |
3430 | ||
3431 | prepend(&cwd, &buflen, "\0", 1); | |
3432 | error = prepend_path(&pwd, &root, &cwd, &buflen); | |
3433 | rcu_read_unlock(); | |
3434 | ||
3435 | if (error < 0) | |
3436 | goto out; | |
3437 | ||
3438 | /* Unreachable from current root */ | |
3439 | if (error > 0) { | |
3440 | error = prepend_unreachable(&cwd, &buflen); | |
3441 | if (error) | |
3442 | goto out; | |
3443 | } | |
3444 | ||
3445 | error = -ERANGE; | |
3446 | len = PATH_MAX + page - cwd; | |
3447 | if (len <= size) { | |
3448 | error = len; | |
3449 | if (copy_to_user(buf, cwd, len)) | |
3450 | error = -EFAULT; | |
3451 | } | |
3452 | } else { | |
3453 | rcu_read_unlock(); | |
3454 | } | |
3455 | ||
3456 | out: | |
3457 | __putname(page); | |
3458 | return error; | |
3459 | } | |
3460 | ||
3461 | /* | |
3462 | * Test whether new_dentry is a subdirectory of old_dentry. | |
3463 | * | |
3464 | * Trivially implemented using the dcache structure | |
3465 | */ | |
3466 | ||
3467 | /** | |
3468 | * is_subdir - is new dentry a subdirectory of old_dentry | |
3469 | * @new_dentry: new dentry | |
3470 | * @old_dentry: old dentry | |
3471 | * | |
3472 | * Returns true if new_dentry is a subdirectory of the parent (at any depth). | |
3473 | * Returns false otherwise. | |
3474 | * Caller must ensure that "new_dentry" is pinned before calling is_subdir() | |
3475 | */ | |
3476 | ||
3477 | bool is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) | |
3478 | { | |
3479 | bool result; | |
3480 | unsigned seq; | |
3481 | ||
3482 | if (new_dentry == old_dentry) | |
3483 | return true; | |
3484 | ||
3485 | do { | |
3486 | /* for restarting inner loop in case of seq retry */ | |
3487 | seq = read_seqbegin(&rename_lock); | |
3488 | /* | |
3489 | * Need rcu_readlock to protect against the d_parent trashing | |
3490 | * due to d_move | |
3491 | */ | |
3492 | rcu_read_lock(); | |
3493 | if (d_ancestor(old_dentry, new_dentry)) | |
3494 | result = true; | |
3495 | else | |
3496 | result = false; | |
3497 | rcu_read_unlock(); | |
3498 | } while (read_seqretry(&rename_lock, seq)); | |
3499 | ||
3500 | return result; | |
3501 | } | |
3502 | ||
3503 | static enum d_walk_ret d_genocide_kill(void *data, struct dentry *dentry) | |
3504 | { | |
3505 | struct dentry *root = data; | |
3506 | if (dentry != root) { | |
3507 | if (d_unhashed(dentry) || !dentry->d_inode) | |
3508 | return D_WALK_SKIP; | |
3509 | ||
3510 | if (!(dentry->d_flags & DCACHE_GENOCIDE)) { | |
3511 | dentry->d_flags |= DCACHE_GENOCIDE; | |
3512 | dentry->d_lockref.count--; | |
3513 | } | |
3514 | } | |
3515 | return D_WALK_CONTINUE; | |
3516 | } | |
3517 | ||
3518 | void d_genocide(struct dentry *parent) | |
3519 | { | |
3520 | d_walk(parent, parent, d_genocide_kill, NULL); | |
3521 | } | |
3522 | ||
3523 | void d_tmpfile(struct dentry *dentry, struct inode *inode) | |
3524 | { | |
3525 | inode_dec_link_count(inode); | |
3526 | BUG_ON(dentry->d_name.name != dentry->d_iname || | |
3527 | !hlist_unhashed(&dentry->d_u.d_alias) || | |
3528 | !d_unlinked(dentry)); | |
3529 | spin_lock(&dentry->d_parent->d_lock); | |
3530 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); | |
3531 | dentry->d_name.len = sprintf(dentry->d_iname, "#%llu", | |
3532 | (unsigned long long)inode->i_ino); | |
3533 | spin_unlock(&dentry->d_lock); | |
3534 | spin_unlock(&dentry->d_parent->d_lock); | |
3535 | d_instantiate(dentry, inode); | |
3536 | } | |
3537 | EXPORT_SYMBOL(d_tmpfile); | |
3538 | ||
3539 | static __initdata unsigned long dhash_entries; | |
3540 | static int __init set_dhash_entries(char *str) | |
3541 | { | |
3542 | if (!str) | |
3543 | return 0; | |
3544 | dhash_entries = simple_strtoul(str, &str, 0); | |
3545 | return 1; | |
3546 | } | |
3547 | __setup("dhash_entries=", set_dhash_entries); | |
3548 | ||
3549 | static void __init dcache_init_early(void) | |
3550 | { | |
3551 | unsigned int loop; | |
3552 | ||
3553 | /* If hashes are distributed across NUMA nodes, defer | |
3554 | * hash allocation until vmalloc space is available. | |
3555 | */ | |
3556 | if (hashdist) | |
3557 | return; | |
3558 | ||
3559 | dentry_hashtable = | |
3560 | alloc_large_system_hash("Dentry cache", | |
3561 | sizeof(struct hlist_bl_head), | |
3562 | dhash_entries, | |
3563 | 13, | |
3564 | HASH_EARLY, | |
3565 | &d_hash_shift, | |
3566 | &d_hash_mask, | |
3567 | 0, | |
3568 | 0); | |
3569 | ||
3570 | for (loop = 0; loop < (1U << d_hash_shift); loop++) | |
3571 | INIT_HLIST_BL_HEAD(dentry_hashtable + loop); | |
3572 | } | |
3573 | ||
3574 | static void __init dcache_init(void) | |
3575 | { | |
3576 | unsigned int loop; | |
3577 | ||
3578 | /* | |
3579 | * A constructor could be added for stable state like the lists, | |
3580 | * but it is probably not worth it because of the cache nature | |
3581 | * of the dcache. | |
3582 | */ | |
3583 | dentry_cache = KMEM_CACHE(dentry, | |
3584 | SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|SLAB_MEM_SPREAD|SLAB_ACCOUNT); | |
3585 | ||
3586 | /* Hash may have been set up in dcache_init_early */ | |
3587 | if (!hashdist) | |
3588 | return; | |
3589 | ||
3590 | dentry_hashtable = | |
3591 | alloc_large_system_hash("Dentry cache", | |
3592 | sizeof(struct hlist_bl_head), | |
3593 | dhash_entries, | |
3594 | 13, | |
3595 | 0, | |
3596 | &d_hash_shift, | |
3597 | &d_hash_mask, | |
3598 | 0, | |
3599 | 0); | |
3600 | ||
3601 | for (loop = 0; loop < (1U << d_hash_shift); loop++) | |
3602 | INIT_HLIST_BL_HEAD(dentry_hashtable + loop); | |
3603 | } | |
3604 | ||
3605 | /* SLAB cache for __getname() consumers */ | |
3606 | struct kmem_cache *names_cachep __read_mostly; | |
3607 | EXPORT_SYMBOL(names_cachep); | |
3608 | ||
3609 | EXPORT_SYMBOL(d_genocide); | |
3610 | ||
3611 | void __init vfs_caches_init_early(void) | |
3612 | { | |
3613 | dcache_init_early(); | |
3614 | inode_init_early(); | |
3615 | } | |
3616 | ||
3617 | void __init vfs_caches_init(void) | |
3618 | { | |
3619 | names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0, | |
3620 | SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); | |
3621 | ||
3622 | dcache_init(); | |
3623 | inode_init(); | |
3624 | files_init(); | |
3625 | files_maxfiles_init(); | |
3626 | mnt_init(); | |
3627 | bdev_cache_init(); | |
3628 | chrdev_init(); | |
3629 | } |