Root/
Source at commit be977234bfb4a6dca8a39e7c52165e4cd536ad71 created 12 years 9 months ago. By Lars-Peter Clausen, jz4740: Fix compile error | |
---|---|
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/module.h> |
27 | #include <linux/mount.h> |
28 | #include <linux/file.h> |
29 | #include <asm/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 "internal.h" |
39 | |
40 | /* |
41 | * Usage: |
42 | * dcache->d_inode->i_lock protects: |
43 | * - i_dentry, d_alias, d_inode of aliases |
44 | * dcache_hash_bucket lock protects: |
45 | * - the dcache hash table |
46 | * s_anon bl list spinlock protects: |
47 | * - the s_anon list (see __d_drop) |
48 | * dcache_lru_lock protects: |
49 | * - the dcache lru lists and counters |
50 | * d_lock protects: |
51 | * - d_flags |
52 | * - d_name |
53 | * - d_lru |
54 | * - d_count |
55 | * - d_unhashed() |
56 | * - d_parent and d_subdirs |
57 | * - childrens' d_child and d_parent |
58 | * - d_alias, d_inode |
59 | * |
60 | * Ordering: |
61 | * dentry->d_inode->i_lock |
62 | * dentry->d_lock |
63 | * dcache_lru_lock |
64 | * dcache_hash_bucket lock |
65 | * s_anon lock |
66 | * |
67 | * If there is an ancestor relationship: |
68 | * dentry->d_parent->...->d_parent->d_lock |
69 | * ... |
70 | * dentry->d_parent->d_lock |
71 | * dentry->d_lock |
72 | * |
73 | * If no ancestor relationship: |
74 | * if (dentry1 < dentry2) |
75 | * dentry1->d_lock |
76 | * dentry2->d_lock |
77 | */ |
78 | int sysctl_vfs_cache_pressure __read_mostly = 100; |
79 | EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure); |
80 | |
81 | static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lru_lock); |
82 | __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); |
83 | |
84 | EXPORT_SYMBOL(rename_lock); |
85 | |
86 | static struct kmem_cache *dentry_cache __read_mostly; |
87 | |
88 | /* |
89 | * This is the single most critical data structure when it comes |
90 | * to the dcache: the hashtable for lookups. Somebody should try |
91 | * to make this good - I've just made it work. |
92 | * |
93 | * This hash-function tries to avoid losing too many bits of hash |
94 | * information, yet avoid using a prime hash-size or similar. |
95 | */ |
96 | #define D_HASHBITS d_hash_shift |
97 | #define D_HASHMASK d_hash_mask |
98 | |
99 | static unsigned int d_hash_mask __read_mostly; |
100 | static unsigned int d_hash_shift __read_mostly; |
101 | |
102 | static struct hlist_bl_head *dentry_hashtable __read_mostly; |
103 | |
104 | static inline struct hlist_bl_head *d_hash(struct dentry *parent, |
105 | unsigned long hash) |
106 | { |
107 | hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES; |
108 | hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS); |
109 | return dentry_hashtable + (hash & D_HASHMASK); |
110 | } |
111 | |
112 | /* Statistics gathering. */ |
113 | struct dentry_stat_t dentry_stat = { |
114 | .age_limit = 45, |
115 | }; |
116 | |
117 | static DEFINE_PER_CPU(unsigned int, nr_dentry); |
118 | |
119 | #if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) |
120 | static int get_nr_dentry(void) |
121 | { |
122 | int i; |
123 | int sum = 0; |
124 | for_each_possible_cpu(i) |
125 | sum += per_cpu(nr_dentry, i); |
126 | return sum < 0 ? 0 : sum; |
127 | } |
128 | |
129 | int proc_nr_dentry(ctl_table *table, int write, void __user *buffer, |
130 | size_t *lenp, loff_t *ppos) |
131 | { |
132 | dentry_stat.nr_dentry = get_nr_dentry(); |
133 | return proc_dointvec(table, write, buffer, lenp, ppos); |
134 | } |
135 | #endif |
136 | |
137 | static void __d_free(struct rcu_head *head) |
138 | { |
139 | struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); |
140 | |
141 | WARN_ON(!list_empty(&dentry->d_alias)); |
142 | if (dname_external(dentry)) |
143 | kfree(dentry->d_name.name); |
144 | kmem_cache_free(dentry_cache, dentry); |
145 | } |
146 | |
147 | /* |
148 | * no locks, please. |
149 | */ |
150 | static void d_free(struct dentry *dentry) |
151 | { |
152 | BUG_ON(dentry->d_count); |
153 | this_cpu_dec(nr_dentry); |
154 | if (dentry->d_op && dentry->d_op->d_release) |
155 | dentry->d_op->d_release(dentry); |
156 | |
157 | /* if dentry was never visible to RCU, immediate free is OK */ |
158 | if (!(dentry->d_flags & DCACHE_RCUACCESS)) |
159 | __d_free(&dentry->d_u.d_rcu); |
160 | else |
161 | call_rcu(&dentry->d_u.d_rcu, __d_free); |
162 | } |
163 | |
164 | /** |
165 | * dentry_rcuwalk_barrier - invalidate in-progress rcu-walk lookups |
166 | * @dentry: the target dentry |
167 | * After this call, in-progress rcu-walk path lookup will fail. This |
168 | * should be called after unhashing, and after changing d_inode (if |
169 | * the dentry has not already been unhashed). |
170 | */ |
171 | static inline void dentry_rcuwalk_barrier(struct dentry *dentry) |
172 | { |
173 | assert_spin_locked(&dentry->d_lock); |
174 | /* Go through a barrier */ |
175 | write_seqcount_barrier(&dentry->d_seq); |
176 | } |
177 | |
178 | /* |
179 | * Release the dentry's inode, using the filesystem |
180 | * d_iput() operation if defined. Dentry has no refcount |
181 | * and is unhashed. |
182 | */ |
183 | static void dentry_iput(struct dentry * dentry) |
184 | __releases(dentry->d_lock) |
185 | __releases(dentry->d_inode->i_lock) |
186 | { |
187 | struct inode *inode = dentry->d_inode; |
188 | if (inode) { |
189 | dentry->d_inode = NULL; |
190 | list_del_init(&dentry->d_alias); |
191 | spin_unlock(&dentry->d_lock); |
192 | spin_unlock(&inode->i_lock); |
193 | if (!inode->i_nlink) |
194 | fsnotify_inoderemove(inode); |
195 | if (dentry->d_op && dentry->d_op->d_iput) |
196 | dentry->d_op->d_iput(dentry, inode); |
197 | else |
198 | iput(inode); |
199 | } else { |
200 | spin_unlock(&dentry->d_lock); |
201 | } |
202 | } |
203 | |
204 | /* |
205 | * Release the dentry's inode, using the filesystem |
206 | * d_iput() operation if defined. dentry remains in-use. |
207 | */ |
208 | static void dentry_unlink_inode(struct dentry * dentry) |
209 | __releases(dentry->d_lock) |
210 | __releases(dentry->d_inode->i_lock) |
211 | { |
212 | struct inode *inode = dentry->d_inode; |
213 | dentry->d_inode = NULL; |
214 | list_del_init(&dentry->d_alias); |
215 | dentry_rcuwalk_barrier(dentry); |
216 | spin_unlock(&dentry->d_lock); |
217 | spin_unlock(&inode->i_lock); |
218 | if (!inode->i_nlink) |
219 | fsnotify_inoderemove(inode); |
220 | if (dentry->d_op && dentry->d_op->d_iput) |
221 | dentry->d_op->d_iput(dentry, inode); |
222 | else |
223 | iput(inode); |
224 | } |
225 | |
226 | /* |
227 | * dentry_lru_(add|del|move_tail) must be called with d_lock held. |
228 | */ |
229 | static void dentry_lru_add(struct dentry *dentry) |
230 | { |
231 | if (list_empty(&dentry->d_lru)) { |
232 | spin_lock(&dcache_lru_lock); |
233 | list_add(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); |
234 | dentry->d_sb->s_nr_dentry_unused++; |
235 | dentry_stat.nr_unused++; |
236 | spin_unlock(&dcache_lru_lock); |
237 | } |
238 | } |
239 | |
240 | static void __dentry_lru_del(struct dentry *dentry) |
241 | { |
242 | list_del_init(&dentry->d_lru); |
243 | dentry->d_sb->s_nr_dentry_unused--; |
244 | dentry_stat.nr_unused--; |
245 | } |
246 | |
247 | static void dentry_lru_del(struct dentry *dentry) |
248 | { |
249 | if (!list_empty(&dentry->d_lru)) { |
250 | spin_lock(&dcache_lru_lock); |
251 | __dentry_lru_del(dentry); |
252 | spin_unlock(&dcache_lru_lock); |
253 | } |
254 | } |
255 | |
256 | static void dentry_lru_move_tail(struct dentry *dentry) |
257 | { |
258 | spin_lock(&dcache_lru_lock); |
259 | if (list_empty(&dentry->d_lru)) { |
260 | list_add_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); |
261 | dentry->d_sb->s_nr_dentry_unused++; |
262 | dentry_stat.nr_unused++; |
263 | } else { |
264 | list_move_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); |
265 | } |
266 | spin_unlock(&dcache_lru_lock); |
267 | } |
268 | |
269 | /** |
270 | * d_kill - kill dentry and return parent |
271 | * @dentry: dentry to kill |
272 | * @parent: parent dentry |
273 | * |
274 | * The dentry must already be unhashed and removed from the LRU. |
275 | * |
276 | * If this is the root of the dentry tree, return NULL. |
277 | * |
278 | * dentry->d_lock and parent->d_lock must be held by caller, and are dropped by |
279 | * d_kill. |
280 | */ |
281 | static struct dentry *d_kill(struct dentry *dentry, struct dentry *parent) |
282 | __releases(dentry->d_lock) |
283 | __releases(parent->d_lock) |
284 | __releases(dentry->d_inode->i_lock) |
285 | { |
286 | list_del(&dentry->d_u.d_child); |
287 | /* |
288 | * Inform try_to_ascend() that we are no longer attached to the |
289 | * dentry tree |
290 | */ |
291 | dentry->d_flags |= DCACHE_DISCONNECTED; |
292 | if (parent) |
293 | spin_unlock(&parent->d_lock); |
294 | dentry_iput(dentry); |
295 | /* |
296 | * dentry_iput drops the locks, at which point nobody (except |
297 | * transient RCU lookups) can reach this dentry. |
298 | */ |
299 | d_free(dentry); |
300 | return parent; |
301 | } |
302 | |
303 | /** |
304 | * d_drop - drop a dentry |
305 | * @dentry: dentry to drop |
306 | * |
307 | * d_drop() unhashes the entry from the parent dentry hashes, so that it won't |
308 | * be found through a VFS lookup any more. Note that this is different from |
309 | * deleting the dentry - d_delete will try to mark the dentry negative if |
310 | * possible, giving a successful _negative_ lookup, while d_drop will |
311 | * just make the cache lookup fail. |
312 | * |
313 | * d_drop() is used mainly for stuff that wants to invalidate a dentry for some |
314 | * reason (NFS timeouts or autofs deletes). |
315 | * |
316 | * __d_drop requires dentry->d_lock. |
317 | */ |
318 | void __d_drop(struct dentry *dentry) |
319 | { |
320 | if (!d_unhashed(dentry)) { |
321 | struct hlist_bl_head *b; |
322 | if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) |
323 | b = &dentry->d_sb->s_anon; |
324 | else |
325 | b = d_hash(dentry->d_parent, dentry->d_name.hash); |
326 | |
327 | hlist_bl_lock(b); |
328 | __hlist_bl_del(&dentry->d_hash); |
329 | dentry->d_hash.pprev = NULL; |
330 | hlist_bl_unlock(b); |
331 | |
332 | dentry_rcuwalk_barrier(dentry); |
333 | } |
334 | } |
335 | EXPORT_SYMBOL(__d_drop); |
336 | |
337 | void d_drop(struct dentry *dentry) |
338 | { |
339 | spin_lock(&dentry->d_lock); |
340 | __d_drop(dentry); |
341 | spin_unlock(&dentry->d_lock); |
342 | } |
343 | EXPORT_SYMBOL(d_drop); |
344 | |
345 | /* |
346 | * Finish off a dentry we've decided to kill. |
347 | * dentry->d_lock must be held, returns with it unlocked. |
348 | * If ref is non-zero, then decrement the refcount too. |
349 | * Returns dentry requiring refcount drop, or NULL if we're done. |
350 | */ |
351 | static inline struct dentry *dentry_kill(struct dentry *dentry, int ref) |
352 | __releases(dentry->d_lock) |
353 | { |
354 | struct inode *inode; |
355 | struct dentry *parent; |
356 | |
357 | inode = dentry->d_inode; |
358 | if (inode && !spin_trylock(&inode->i_lock)) { |
359 | relock: |
360 | spin_unlock(&dentry->d_lock); |
361 | cpu_relax(); |
362 | return dentry; /* try again with same dentry */ |
363 | } |
364 | if (IS_ROOT(dentry)) |
365 | parent = NULL; |
366 | else |
367 | parent = dentry->d_parent; |
368 | if (parent && !spin_trylock(&parent->d_lock)) { |
369 | if (inode) |
370 | spin_unlock(&inode->i_lock); |
371 | goto relock; |
372 | } |
373 | |
374 | if (ref) |
375 | dentry->d_count--; |
376 | /* if dentry was on the d_lru list delete it from there */ |
377 | dentry_lru_del(dentry); |
378 | /* if it was on the hash then remove it */ |
379 | __d_drop(dentry); |
380 | return d_kill(dentry, parent); |
381 | } |
382 | |
383 | /* |
384 | * This is dput |
385 | * |
386 | * This is complicated by the fact that we do not want to put |
387 | * dentries that are no longer on any hash chain on the unused |
388 | * list: we'd much rather just get rid of them immediately. |
389 | * |
390 | * However, that implies that we have to traverse the dentry |
391 | * tree upwards to the parents which might _also_ now be |
392 | * scheduled for deletion (it may have been only waiting for |
393 | * its last child to go away). |
394 | * |
395 | * This tail recursion is done by hand as we don't want to depend |
396 | * on the compiler to always get this right (gcc generally doesn't). |
397 | * Real recursion would eat up our stack space. |
398 | */ |
399 | |
400 | /* |
401 | * dput - release a dentry |
402 | * @dentry: dentry to release |
403 | * |
404 | * Release a dentry. This will drop the usage count and if appropriate |
405 | * call the dentry unlink method as well as removing it from the queues and |
406 | * releasing its resources. If the parent dentries were scheduled for release |
407 | * they too may now get deleted. |
408 | */ |
409 | void dput(struct dentry *dentry) |
410 | { |
411 | if (!dentry) |
412 | return; |
413 | |
414 | repeat: |
415 | if (dentry->d_count == 1) |
416 | might_sleep(); |
417 | spin_lock(&dentry->d_lock); |
418 | BUG_ON(!dentry->d_count); |
419 | if (dentry->d_count > 1) { |
420 | dentry->d_count--; |
421 | spin_unlock(&dentry->d_lock); |
422 | return; |
423 | } |
424 | |
425 | if (dentry->d_flags & DCACHE_OP_DELETE) { |
426 | if (dentry->d_op->d_delete(dentry)) |
427 | goto kill_it; |
428 | } |
429 | |
430 | /* Unreachable? Get rid of it */ |
431 | if (d_unhashed(dentry)) |
432 | goto kill_it; |
433 | |
434 | /* Otherwise leave it cached and ensure it's on the LRU */ |
435 | dentry->d_flags |= DCACHE_REFERENCED; |
436 | dentry_lru_add(dentry); |
437 | |
438 | dentry->d_count--; |
439 | spin_unlock(&dentry->d_lock); |
440 | return; |
441 | |
442 | kill_it: |
443 | dentry = dentry_kill(dentry, 1); |
444 | if (dentry) |
445 | goto repeat; |
446 | } |
447 | EXPORT_SYMBOL(dput); |
448 | |
449 | /** |
450 | * d_invalidate - invalidate a dentry |
451 | * @dentry: dentry to invalidate |
452 | * |
453 | * Try to invalidate the dentry if it turns out to be |
454 | * possible. If there are other dentries that can be |
455 | * reached through this one we can't delete it and we |
456 | * return -EBUSY. On success we return 0. |
457 | * |
458 | * no dcache lock. |
459 | */ |
460 | |
461 | int d_invalidate(struct dentry * dentry) |
462 | { |
463 | /* |
464 | * If it's already been dropped, return OK. |
465 | */ |
466 | spin_lock(&dentry->d_lock); |
467 | if (d_unhashed(dentry)) { |
468 | spin_unlock(&dentry->d_lock); |
469 | return 0; |
470 | } |
471 | /* |
472 | * Check whether to do a partial shrink_dcache |
473 | * to get rid of unused child entries. |
474 | */ |
475 | if (!list_empty(&dentry->d_subdirs)) { |
476 | spin_unlock(&dentry->d_lock); |
477 | shrink_dcache_parent(dentry); |
478 | spin_lock(&dentry->d_lock); |
479 | } |
480 | |
481 | /* |
482 | * Somebody else still using it? |
483 | * |
484 | * If it's a directory, we can't drop it |
485 | * for fear of somebody re-populating it |
486 | * with children (even though dropping it |
487 | * would make it unreachable from the root, |
488 | * we might still populate it if it was a |
489 | * working directory or similar). |
490 | */ |
491 | if (dentry->d_count > 1) { |
492 | if (dentry->d_inode && S_ISDIR(dentry->d_inode->i_mode)) { |
493 | spin_unlock(&dentry->d_lock); |
494 | return -EBUSY; |
495 | } |
496 | } |
497 | |
498 | __d_drop(dentry); |
499 | spin_unlock(&dentry->d_lock); |
500 | return 0; |
501 | } |
502 | EXPORT_SYMBOL(d_invalidate); |
503 | |
504 | /* This must be called with d_lock held */ |
505 | static inline void __dget_dlock(struct dentry *dentry) |
506 | { |
507 | dentry->d_count++; |
508 | } |
509 | |
510 | static inline void __dget(struct dentry *dentry) |
511 | { |
512 | spin_lock(&dentry->d_lock); |
513 | __dget_dlock(dentry); |
514 | spin_unlock(&dentry->d_lock); |
515 | } |
516 | |
517 | struct dentry *dget_parent(struct dentry *dentry) |
518 | { |
519 | struct dentry *ret; |
520 | |
521 | repeat: |
522 | /* |
523 | * Don't need rcu_dereference because we re-check it was correct under |
524 | * the lock. |
525 | */ |
526 | rcu_read_lock(); |
527 | ret = dentry->d_parent; |
528 | if (!ret) { |
529 | rcu_read_unlock(); |
530 | goto out; |
531 | } |
532 | spin_lock(&ret->d_lock); |
533 | if (unlikely(ret != dentry->d_parent)) { |
534 | spin_unlock(&ret->d_lock); |
535 | rcu_read_unlock(); |
536 | goto repeat; |
537 | } |
538 | rcu_read_unlock(); |
539 | BUG_ON(!ret->d_count); |
540 | ret->d_count++; |
541 | spin_unlock(&ret->d_lock); |
542 | out: |
543 | return ret; |
544 | } |
545 | EXPORT_SYMBOL(dget_parent); |
546 | |
547 | /** |
548 | * d_find_alias - grab a hashed alias of inode |
549 | * @inode: inode in question |
550 | * @want_discon: flag, used by d_splice_alias, to request |
551 | * that only a DISCONNECTED alias be returned. |
552 | * |
553 | * If inode has a hashed alias, or is a directory and has any alias, |
554 | * acquire the reference to alias and return it. Otherwise return NULL. |
555 | * Notice that if inode is a directory there can be only one alias and |
556 | * it can be unhashed only if it has no children, or if it is the root |
557 | * of a filesystem. |
558 | * |
559 | * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer |
560 | * any other hashed alias over that one unless @want_discon is set, |
561 | * in which case only return an IS_ROOT, DCACHE_DISCONNECTED alias. |
562 | */ |
563 | static struct dentry *__d_find_alias(struct inode *inode, int want_discon) |
564 | { |
565 | struct dentry *alias, *discon_alias; |
566 | |
567 | again: |
568 | discon_alias = NULL; |
569 | list_for_each_entry(alias, &inode->i_dentry, d_alias) { |
570 | spin_lock(&alias->d_lock); |
571 | if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { |
572 | if (IS_ROOT(alias) && |
573 | (alias->d_flags & DCACHE_DISCONNECTED)) { |
574 | discon_alias = alias; |
575 | } else if (!want_discon) { |
576 | __dget_dlock(alias); |
577 | spin_unlock(&alias->d_lock); |
578 | return alias; |
579 | } |
580 | } |
581 | spin_unlock(&alias->d_lock); |
582 | } |
583 | if (discon_alias) { |
584 | alias = discon_alias; |
585 | spin_lock(&alias->d_lock); |
586 | if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { |
587 | if (IS_ROOT(alias) && |
588 | (alias->d_flags & DCACHE_DISCONNECTED)) { |
589 | __dget_dlock(alias); |
590 | spin_unlock(&alias->d_lock); |
591 | return alias; |
592 | } |
593 | } |
594 | spin_unlock(&alias->d_lock); |
595 | goto again; |
596 | } |
597 | return NULL; |
598 | } |
599 | |
600 | struct dentry *d_find_alias(struct inode *inode) |
601 | { |
602 | struct dentry *de = NULL; |
603 | |
604 | if (!list_empty(&inode->i_dentry)) { |
605 | spin_lock(&inode->i_lock); |
606 | de = __d_find_alias(inode, 0); |
607 | spin_unlock(&inode->i_lock); |
608 | } |
609 | return de; |
610 | } |
611 | EXPORT_SYMBOL(d_find_alias); |
612 | |
613 | /* |
614 | * Try to kill dentries associated with this inode. |
615 | * WARNING: you must own a reference to inode. |
616 | */ |
617 | void d_prune_aliases(struct inode *inode) |
618 | { |
619 | struct dentry *dentry; |
620 | restart: |
621 | spin_lock(&inode->i_lock); |
622 | list_for_each_entry(dentry, &inode->i_dentry, d_alias) { |
623 | spin_lock(&dentry->d_lock); |
624 | if (!dentry->d_count) { |
625 | __dget_dlock(dentry); |
626 | __d_drop(dentry); |
627 | spin_unlock(&dentry->d_lock); |
628 | spin_unlock(&inode->i_lock); |
629 | dput(dentry); |
630 | goto restart; |
631 | } |
632 | spin_unlock(&dentry->d_lock); |
633 | } |
634 | spin_unlock(&inode->i_lock); |
635 | } |
636 | EXPORT_SYMBOL(d_prune_aliases); |
637 | |
638 | /* |
639 | * Try to throw away a dentry - free the inode, dput the parent. |
640 | * Requires dentry->d_lock is held, and dentry->d_count == 0. |
641 | * Releases dentry->d_lock. |
642 | * |
643 | * This may fail if locks cannot be acquired no problem, just try again. |
644 | */ |
645 | static void try_prune_one_dentry(struct dentry *dentry) |
646 | __releases(dentry->d_lock) |
647 | { |
648 | struct dentry *parent; |
649 | |
650 | parent = dentry_kill(dentry, 0); |
651 | /* |
652 | * If dentry_kill returns NULL, we have nothing more to do. |
653 | * if it returns the same dentry, trylocks failed. In either |
654 | * case, just loop again. |
655 | * |
656 | * Otherwise, we need to prune ancestors too. This is necessary |
657 | * to prevent quadratic behavior of shrink_dcache_parent(), but |
658 | * is also expected to be beneficial in reducing dentry cache |
659 | * fragmentation. |
660 | */ |
661 | if (!parent) |
662 | return; |
663 | if (parent == dentry) |
664 | return; |
665 | |
666 | /* Prune ancestors. */ |
667 | dentry = parent; |
668 | while (dentry) { |
669 | spin_lock(&dentry->d_lock); |
670 | if (dentry->d_count > 1) { |
671 | dentry->d_count--; |
672 | spin_unlock(&dentry->d_lock); |
673 | return; |
674 | } |
675 | dentry = dentry_kill(dentry, 1); |
676 | } |
677 | } |
678 | |
679 | static void shrink_dentry_list(struct list_head *list) |
680 | { |
681 | struct dentry *dentry; |
682 | |
683 | rcu_read_lock(); |
684 | for (;;) { |
685 | dentry = list_entry_rcu(list->prev, struct dentry, d_lru); |
686 | if (&dentry->d_lru == list) |
687 | break; /* empty */ |
688 | spin_lock(&dentry->d_lock); |
689 | if (dentry != list_entry(list->prev, struct dentry, d_lru)) { |
690 | spin_unlock(&dentry->d_lock); |
691 | continue; |
692 | } |
693 | |
694 | /* |
695 | * We found an inuse dentry which was not removed from |
696 | * the LRU because of laziness during lookup. Do not free |
697 | * it - just keep it off the LRU list. |
698 | */ |
699 | if (dentry->d_count) { |
700 | dentry_lru_del(dentry); |
701 | spin_unlock(&dentry->d_lock); |
702 | continue; |
703 | } |
704 | |
705 | rcu_read_unlock(); |
706 | |
707 | try_prune_one_dentry(dentry); |
708 | |
709 | rcu_read_lock(); |
710 | } |
711 | rcu_read_unlock(); |
712 | } |
713 | |
714 | /** |
715 | * __shrink_dcache_sb - shrink the dentry LRU on a given superblock |
716 | * @sb: superblock to shrink dentry LRU. |
717 | * @count: number of entries to prune |
718 | * @flags: flags to control the dentry processing |
719 | * |
720 | * If flags contains DCACHE_REFERENCED reference dentries will not be pruned. |
721 | */ |
722 | static void __shrink_dcache_sb(struct super_block *sb, int *count, int flags) |
723 | { |
724 | /* called from prune_dcache() and shrink_dcache_parent() */ |
725 | struct dentry *dentry; |
726 | LIST_HEAD(referenced); |
727 | LIST_HEAD(tmp); |
728 | int cnt = *count; |
729 | |
730 | relock: |
731 | spin_lock(&dcache_lru_lock); |
732 | while (!list_empty(&sb->s_dentry_lru)) { |
733 | dentry = list_entry(sb->s_dentry_lru.prev, |
734 | struct dentry, d_lru); |
735 | BUG_ON(dentry->d_sb != sb); |
736 | |
737 | if (!spin_trylock(&dentry->d_lock)) { |
738 | spin_unlock(&dcache_lru_lock); |
739 | cpu_relax(); |
740 | goto relock; |
741 | } |
742 | |
743 | /* |
744 | * If we are honouring the DCACHE_REFERENCED flag and the |
745 | * dentry has this flag set, don't free it. Clear the flag |
746 | * and put it back on the LRU. |
747 | */ |
748 | if (flags & DCACHE_REFERENCED && |
749 | dentry->d_flags & DCACHE_REFERENCED) { |
750 | dentry->d_flags &= ~DCACHE_REFERENCED; |
751 | list_move(&dentry->d_lru, &referenced); |
752 | spin_unlock(&dentry->d_lock); |
753 | } else { |
754 | list_move_tail(&dentry->d_lru, &tmp); |
755 | spin_unlock(&dentry->d_lock); |
756 | if (!--cnt) |
757 | break; |
758 | } |
759 | cond_resched_lock(&dcache_lru_lock); |
760 | } |
761 | if (!list_empty(&referenced)) |
762 | list_splice(&referenced, &sb->s_dentry_lru); |
763 | spin_unlock(&dcache_lru_lock); |
764 | |
765 | shrink_dentry_list(&tmp); |
766 | |
767 | *count = cnt; |
768 | } |
769 | |
770 | /** |
771 | * prune_dcache - shrink the dcache |
772 | * @count: number of entries to try to free |
773 | * |
774 | * Shrink the dcache. This is done when we need more memory, or simply when we |
775 | * need to unmount something (at which point we need to unuse all dentries). |
776 | * |
777 | * This function may fail to free any resources if all the dentries are in use. |
778 | */ |
779 | static void prune_dcache(int count) |
780 | { |
781 | struct super_block *sb, *p = NULL; |
782 | int w_count; |
783 | int unused = dentry_stat.nr_unused; |
784 | int prune_ratio; |
785 | int pruned; |
786 | |
787 | if (unused == 0 || count == 0) |
788 | return; |
789 | if (count >= unused) |
790 | prune_ratio = 1; |
791 | else |
792 | prune_ratio = unused / count; |
793 | spin_lock(&sb_lock); |
794 | list_for_each_entry(sb, &super_blocks, s_list) { |
795 | if (list_empty(&sb->s_instances)) |
796 | continue; |
797 | if (sb->s_nr_dentry_unused == 0) |
798 | continue; |
799 | sb->s_count++; |
800 | /* Now, we reclaim unused dentrins with fairness. |
801 | * We reclaim them same percentage from each superblock. |
802 | * We calculate number of dentries to scan on this sb |
803 | * as follows, but the implementation is arranged to avoid |
804 | * overflows: |
805 | * number of dentries to scan on this sb = |
806 | * count * (number of dentries on this sb / |
807 | * number of dentries in the machine) |
808 | */ |
809 | spin_unlock(&sb_lock); |
810 | if (prune_ratio != 1) |
811 | w_count = (sb->s_nr_dentry_unused / prune_ratio) + 1; |
812 | else |
813 | w_count = sb->s_nr_dentry_unused; |
814 | pruned = w_count; |
815 | /* |
816 | * We need to be sure this filesystem isn't being unmounted, |
817 | * otherwise we could race with generic_shutdown_super(), and |
818 | * end up holding a reference to an inode while the filesystem |
819 | * is unmounted. So we try to get s_umount, and make sure |
820 | * s_root isn't NULL. |
821 | */ |
822 | if (down_read_trylock(&sb->s_umount)) { |
823 | if ((sb->s_root != NULL) && |
824 | (!list_empty(&sb->s_dentry_lru))) { |
825 | __shrink_dcache_sb(sb, &w_count, |
826 | DCACHE_REFERENCED); |
827 | pruned -= w_count; |
828 | } |
829 | up_read(&sb->s_umount); |
830 | } |
831 | spin_lock(&sb_lock); |
832 | if (p) |
833 | __put_super(p); |
834 | count -= pruned; |
835 | p = sb; |
836 | /* more work left to do? */ |
837 | if (count <= 0) |
838 | break; |
839 | } |
840 | if (p) |
841 | __put_super(p); |
842 | spin_unlock(&sb_lock); |
843 | } |
844 | |
845 | /** |
846 | * shrink_dcache_sb - shrink dcache for a superblock |
847 | * @sb: superblock |
848 | * |
849 | * Shrink the dcache for the specified super block. This is used to free |
850 | * the dcache before unmounting a file system. |
851 | */ |
852 | void shrink_dcache_sb(struct super_block *sb) |
853 | { |
854 | LIST_HEAD(tmp); |
855 | |
856 | spin_lock(&dcache_lru_lock); |
857 | while (!list_empty(&sb->s_dentry_lru)) { |
858 | list_splice_init(&sb->s_dentry_lru, &tmp); |
859 | spin_unlock(&dcache_lru_lock); |
860 | shrink_dentry_list(&tmp); |
861 | spin_lock(&dcache_lru_lock); |
862 | } |
863 | spin_unlock(&dcache_lru_lock); |
864 | } |
865 | EXPORT_SYMBOL(shrink_dcache_sb); |
866 | |
867 | /* |
868 | * destroy a single subtree of dentries for unmount |
869 | * - see the comments on shrink_dcache_for_umount() for a description of the |
870 | * locking |
871 | */ |
872 | static void shrink_dcache_for_umount_subtree(struct dentry *dentry) |
873 | { |
874 | struct dentry *parent; |
875 | unsigned detached = 0; |
876 | |
877 | BUG_ON(!IS_ROOT(dentry)); |
878 | |
879 | /* detach this root from the system */ |
880 | spin_lock(&dentry->d_lock); |
881 | dentry_lru_del(dentry); |
882 | __d_drop(dentry); |
883 | spin_unlock(&dentry->d_lock); |
884 | |
885 | for (;;) { |
886 | /* descend to the first leaf in the current subtree */ |
887 | while (!list_empty(&dentry->d_subdirs)) { |
888 | struct dentry *loop; |
889 | |
890 | /* this is a branch with children - detach all of them |
891 | * from the system in one go */ |
892 | spin_lock(&dentry->d_lock); |
893 | list_for_each_entry(loop, &dentry->d_subdirs, |
894 | d_u.d_child) { |
895 | spin_lock_nested(&loop->d_lock, |
896 | DENTRY_D_LOCK_NESTED); |
897 | dentry_lru_del(loop); |
898 | __d_drop(loop); |
899 | spin_unlock(&loop->d_lock); |
900 | } |
901 | spin_unlock(&dentry->d_lock); |
902 | |
903 | /* move to the first child */ |
904 | dentry = list_entry(dentry->d_subdirs.next, |
905 | struct dentry, d_u.d_child); |
906 | } |
907 | |
908 | /* consume the dentries from this leaf up through its parents |
909 | * until we find one with children or run out altogether */ |
910 | do { |
911 | struct inode *inode; |
912 | |
913 | if (dentry->d_count != 0) { |
914 | printk(KERN_ERR |
915 | "BUG: Dentry %p{i=%lx,n=%s}" |
916 | " still in use (%d)" |
917 | " [unmount of %s %s]\n", |
918 | dentry, |
919 | dentry->d_inode ? |
920 | dentry->d_inode->i_ino : 0UL, |
921 | dentry->d_name.name, |
922 | dentry->d_count, |
923 | dentry->d_sb->s_type->name, |
924 | dentry->d_sb->s_id); |
925 | BUG(); |
926 | } |
927 | |
928 | if (IS_ROOT(dentry)) { |
929 | parent = NULL; |
930 | list_del(&dentry->d_u.d_child); |
931 | } else { |
932 | parent = dentry->d_parent; |
933 | spin_lock(&parent->d_lock); |
934 | parent->d_count--; |
935 | list_del(&dentry->d_u.d_child); |
936 | spin_unlock(&parent->d_lock); |
937 | } |
938 | |
939 | detached++; |
940 | |
941 | inode = dentry->d_inode; |
942 | if (inode) { |
943 | dentry->d_inode = NULL; |
944 | list_del_init(&dentry->d_alias); |
945 | if (dentry->d_op && dentry->d_op->d_iput) |
946 | dentry->d_op->d_iput(dentry, inode); |
947 | else |
948 | iput(inode); |
949 | } |
950 | |
951 | d_free(dentry); |
952 | |
953 | /* finished when we fall off the top of the tree, |
954 | * otherwise we ascend to the parent and move to the |
955 | * next sibling if there is one */ |
956 | if (!parent) |
957 | return; |
958 | dentry = parent; |
959 | } while (list_empty(&dentry->d_subdirs)); |
960 | |
961 | dentry = list_entry(dentry->d_subdirs.next, |
962 | struct dentry, d_u.d_child); |
963 | } |
964 | } |
965 | |
966 | /* |
967 | * destroy the dentries attached to a superblock on unmounting |
968 | * - we don't need to use dentry->d_lock because: |
969 | * - the superblock is detached from all mountings and open files, so the |
970 | * dentry trees will not be rearranged by the VFS |
971 | * - s_umount is write-locked, so the memory pressure shrinker will ignore |
972 | * any dentries belonging to this superblock that it comes across |
973 | * - the filesystem itself is no longer permitted to rearrange the dentries |
974 | * in this superblock |
975 | */ |
976 | void shrink_dcache_for_umount(struct super_block *sb) |
977 | { |
978 | struct dentry *dentry; |
979 | |
980 | if (down_read_trylock(&sb->s_umount)) |
981 | BUG(); |
982 | |
983 | dentry = sb->s_root; |
984 | sb->s_root = NULL; |
985 | spin_lock(&dentry->d_lock); |
986 | dentry->d_count--; |
987 | spin_unlock(&dentry->d_lock); |
988 | shrink_dcache_for_umount_subtree(dentry); |
989 | |
990 | while (!hlist_bl_empty(&sb->s_anon)) { |
991 | dentry = hlist_bl_entry(hlist_bl_first(&sb->s_anon), struct dentry, d_hash); |
992 | shrink_dcache_for_umount_subtree(dentry); |
993 | } |
994 | } |
995 | |
996 | /* |
997 | * This tries to ascend one level of parenthood, but |
998 | * we can race with renaming, so we need to re-check |
999 | * the parenthood after dropping the lock and check |
1000 | * that the sequence number still matches. |
1001 | */ |
1002 | static struct dentry *try_to_ascend(struct dentry *old, int locked, unsigned seq) |
1003 | { |
1004 | struct dentry *new = old->d_parent; |
1005 | |
1006 | rcu_read_lock(); |
1007 | spin_unlock(&old->d_lock); |
1008 | spin_lock(&new->d_lock); |
1009 | |
1010 | /* |
1011 | * might go back up the wrong parent if we have had a rename |
1012 | * or deletion |
1013 | */ |
1014 | if (new != old->d_parent || |
1015 | (old->d_flags & DCACHE_DISCONNECTED) || |
1016 | (!locked && read_seqretry(&rename_lock, seq))) { |
1017 | spin_unlock(&new->d_lock); |
1018 | new = NULL; |
1019 | } |
1020 | rcu_read_unlock(); |
1021 | return new; |
1022 | } |
1023 | |
1024 | |
1025 | /* |
1026 | * Search for at least 1 mount point in the dentry's subdirs. |
1027 | * We descend to the next level whenever the d_subdirs |
1028 | * list is non-empty and continue searching. |
1029 | */ |
1030 | |
1031 | /** |
1032 | * have_submounts - check for mounts over a dentry |
1033 | * @parent: dentry to check. |
1034 | * |
1035 | * Return true if the parent or its subdirectories contain |
1036 | * a mount point |
1037 | */ |
1038 | int have_submounts(struct dentry *parent) |
1039 | { |
1040 | struct dentry *this_parent; |
1041 | struct list_head *next; |
1042 | unsigned seq; |
1043 | int locked = 0; |
1044 | |
1045 | seq = read_seqbegin(&rename_lock); |
1046 | again: |
1047 | this_parent = parent; |
1048 | |
1049 | if (d_mountpoint(parent)) |
1050 | goto positive; |
1051 | spin_lock(&this_parent->d_lock); |
1052 | repeat: |
1053 | next = this_parent->d_subdirs.next; |
1054 | resume: |
1055 | while (next != &this_parent->d_subdirs) { |
1056 | struct list_head *tmp = next; |
1057 | struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); |
1058 | next = tmp->next; |
1059 | |
1060 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
1061 | /* Have we found a mount point ? */ |
1062 | if (d_mountpoint(dentry)) { |
1063 | spin_unlock(&dentry->d_lock); |
1064 | spin_unlock(&this_parent->d_lock); |
1065 | goto positive; |
1066 | } |
1067 | if (!list_empty(&dentry->d_subdirs)) { |
1068 | spin_unlock(&this_parent->d_lock); |
1069 | spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); |
1070 | this_parent = dentry; |
1071 | spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); |
1072 | goto repeat; |
1073 | } |
1074 | spin_unlock(&dentry->d_lock); |
1075 | } |
1076 | /* |
1077 | * All done at this level ... ascend and resume the search. |
1078 | */ |
1079 | if (this_parent != parent) { |
1080 | struct dentry *child = this_parent; |
1081 | this_parent = try_to_ascend(this_parent, locked, seq); |
1082 | if (!this_parent) |
1083 | goto rename_retry; |
1084 | next = child->d_u.d_child.next; |
1085 | goto resume; |
1086 | } |
1087 | spin_unlock(&this_parent->d_lock); |
1088 | if (!locked && read_seqretry(&rename_lock, seq)) |
1089 | goto rename_retry; |
1090 | if (locked) |
1091 | write_sequnlock(&rename_lock); |
1092 | return 0; /* No mount points found in tree */ |
1093 | positive: |
1094 | if (!locked && read_seqretry(&rename_lock, seq)) |
1095 | goto rename_retry; |
1096 | if (locked) |
1097 | write_sequnlock(&rename_lock); |
1098 | return 1; |
1099 | |
1100 | rename_retry: |
1101 | locked = 1; |
1102 | write_seqlock(&rename_lock); |
1103 | goto again; |
1104 | } |
1105 | EXPORT_SYMBOL(have_submounts); |
1106 | |
1107 | /* |
1108 | * Search the dentry child list for the specified parent, |
1109 | * and move any unused dentries to the end of the unused |
1110 | * list for prune_dcache(). We descend to the next level |
1111 | * whenever the d_subdirs list is non-empty and continue |
1112 | * searching. |
1113 | * |
1114 | * It returns zero iff there are no unused children, |
1115 | * otherwise it returns the number of children moved to |
1116 | * the end of the unused list. This may not be the total |
1117 | * number of unused children, because select_parent can |
1118 | * drop the lock and return early due to latency |
1119 | * constraints. |
1120 | */ |
1121 | static int select_parent(struct dentry * parent) |
1122 | { |
1123 | struct dentry *this_parent; |
1124 | struct list_head *next; |
1125 | unsigned seq; |
1126 | int found = 0; |
1127 | int locked = 0; |
1128 | |
1129 | seq = read_seqbegin(&rename_lock); |
1130 | again: |
1131 | this_parent = parent; |
1132 | spin_lock(&this_parent->d_lock); |
1133 | repeat: |
1134 | next = this_parent->d_subdirs.next; |
1135 | resume: |
1136 | while (next != &this_parent->d_subdirs) { |
1137 | struct list_head *tmp = next; |
1138 | struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); |
1139 | next = tmp->next; |
1140 | |
1141 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
1142 | |
1143 | /* |
1144 | * move only zero ref count dentries to the end |
1145 | * of the unused list for prune_dcache |
1146 | */ |
1147 | if (!dentry->d_count) { |
1148 | dentry_lru_move_tail(dentry); |
1149 | found++; |
1150 | } else { |
1151 | dentry_lru_del(dentry); |
1152 | } |
1153 | |
1154 | /* |
1155 | * We can return to the caller if we have found some (this |
1156 | * ensures forward progress). We'll be coming back to find |
1157 | * the rest. |
1158 | */ |
1159 | if (found && need_resched()) { |
1160 | spin_unlock(&dentry->d_lock); |
1161 | goto out; |
1162 | } |
1163 | |
1164 | /* |
1165 | * Descend a level if the d_subdirs list is non-empty. |
1166 | */ |
1167 | if (!list_empty(&dentry->d_subdirs)) { |
1168 | spin_unlock(&this_parent->d_lock); |
1169 | spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); |
1170 | this_parent = dentry; |
1171 | spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); |
1172 | goto repeat; |
1173 | } |
1174 | |
1175 | spin_unlock(&dentry->d_lock); |
1176 | } |
1177 | /* |
1178 | * All done at this level ... ascend and resume the search. |
1179 | */ |
1180 | if (this_parent != parent) { |
1181 | struct dentry *child = this_parent; |
1182 | this_parent = try_to_ascend(this_parent, locked, seq); |
1183 | if (!this_parent) |
1184 | goto rename_retry; |
1185 | next = child->d_u.d_child.next; |
1186 | goto resume; |
1187 | } |
1188 | out: |
1189 | spin_unlock(&this_parent->d_lock); |
1190 | if (!locked && read_seqretry(&rename_lock, seq)) |
1191 | goto rename_retry; |
1192 | if (locked) |
1193 | write_sequnlock(&rename_lock); |
1194 | return found; |
1195 | |
1196 | rename_retry: |
1197 | if (found) |
1198 | return found; |
1199 | locked = 1; |
1200 | write_seqlock(&rename_lock); |
1201 | goto again; |
1202 | } |
1203 | |
1204 | /** |
1205 | * shrink_dcache_parent - prune dcache |
1206 | * @parent: parent of entries to prune |
1207 | * |
1208 | * Prune the dcache to remove unused children of the parent dentry. |
1209 | */ |
1210 | |
1211 | void shrink_dcache_parent(struct dentry * parent) |
1212 | { |
1213 | struct super_block *sb = parent->d_sb; |
1214 | int found; |
1215 | |
1216 | while ((found = select_parent(parent)) != 0) |
1217 | __shrink_dcache_sb(sb, &found, 0); |
1218 | } |
1219 | EXPORT_SYMBOL(shrink_dcache_parent); |
1220 | |
1221 | /* |
1222 | * Scan `nr' dentries and return the number which remain. |
1223 | * |
1224 | * We need to avoid reentering the filesystem if the caller is performing a |
1225 | * GFP_NOFS allocation attempt. One example deadlock is: |
1226 | * |
1227 | * ext2_new_block->getblk->GFP->shrink_dcache_memory->prune_dcache-> |
1228 | * prune_one_dentry->dput->dentry_iput->iput->inode->i_sb->s_op->put_inode-> |
1229 | * ext2_discard_prealloc->ext2_free_blocks->lock_super->DEADLOCK. |
1230 | * |
1231 | * In this case we return -1 to tell the caller that we baled. |
1232 | */ |
1233 | static int shrink_dcache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) |
1234 | { |
1235 | if (nr) { |
1236 | if (!(gfp_mask & __GFP_FS)) |
1237 | return -1; |
1238 | prune_dcache(nr); |
1239 | } |
1240 | |
1241 | return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure; |
1242 | } |
1243 | |
1244 | static struct shrinker dcache_shrinker = { |
1245 | .shrink = shrink_dcache_memory, |
1246 | .seeks = DEFAULT_SEEKS, |
1247 | }; |
1248 | |
1249 | /** |
1250 | * d_alloc - allocate a dcache entry |
1251 | * @parent: parent of entry to allocate |
1252 | * @name: qstr of the name |
1253 | * |
1254 | * Allocates a dentry. It returns %NULL if there is insufficient memory |
1255 | * available. On a success the dentry is returned. The name passed in is |
1256 | * copied and the copy passed in may be reused after this call. |
1257 | */ |
1258 | |
1259 | struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) |
1260 | { |
1261 | struct dentry *dentry; |
1262 | char *dname; |
1263 | |
1264 | dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL); |
1265 | if (!dentry) |
1266 | return NULL; |
1267 | |
1268 | if (name->len > DNAME_INLINE_LEN-1) { |
1269 | dname = kmalloc(name->len + 1, GFP_KERNEL); |
1270 | if (!dname) { |
1271 | kmem_cache_free(dentry_cache, dentry); |
1272 | return NULL; |
1273 | } |
1274 | } else { |
1275 | dname = dentry->d_iname; |
1276 | } |
1277 | dentry->d_name.name = dname; |
1278 | |
1279 | dentry->d_name.len = name->len; |
1280 | dentry->d_name.hash = name->hash; |
1281 | memcpy(dname, name->name, name->len); |
1282 | dname[name->len] = 0; |
1283 | |
1284 | dentry->d_count = 1; |
1285 | dentry->d_flags = 0; |
1286 | spin_lock_init(&dentry->d_lock); |
1287 | seqcount_init(&dentry->d_seq); |
1288 | dentry->d_inode = NULL; |
1289 | dentry->d_parent = NULL; |
1290 | dentry->d_sb = NULL; |
1291 | dentry->d_op = NULL; |
1292 | dentry->d_fsdata = NULL; |
1293 | INIT_HLIST_BL_NODE(&dentry->d_hash); |
1294 | INIT_LIST_HEAD(&dentry->d_lru); |
1295 | INIT_LIST_HEAD(&dentry->d_subdirs); |
1296 | INIT_LIST_HEAD(&dentry->d_alias); |
1297 | INIT_LIST_HEAD(&dentry->d_u.d_child); |
1298 | |
1299 | if (parent) { |
1300 | spin_lock(&parent->d_lock); |
1301 | /* |
1302 | * don't need child lock because it is not subject |
1303 | * to concurrency here |
1304 | */ |
1305 | __dget_dlock(parent); |
1306 | dentry->d_parent = parent; |
1307 | dentry->d_sb = parent->d_sb; |
1308 | d_set_d_op(dentry, dentry->d_sb->s_d_op); |
1309 | list_add(&dentry->d_u.d_child, &parent->d_subdirs); |
1310 | spin_unlock(&parent->d_lock); |
1311 | } |
1312 | |
1313 | this_cpu_inc(nr_dentry); |
1314 | |
1315 | return dentry; |
1316 | } |
1317 | EXPORT_SYMBOL(d_alloc); |
1318 | |
1319 | struct dentry *d_alloc_pseudo(struct super_block *sb, const struct qstr *name) |
1320 | { |
1321 | struct dentry *dentry = d_alloc(NULL, name); |
1322 | if (dentry) { |
1323 | dentry->d_sb = sb; |
1324 | d_set_d_op(dentry, dentry->d_sb->s_d_op); |
1325 | dentry->d_parent = dentry; |
1326 | dentry->d_flags |= DCACHE_DISCONNECTED; |
1327 | } |
1328 | return dentry; |
1329 | } |
1330 | EXPORT_SYMBOL(d_alloc_pseudo); |
1331 | |
1332 | struct dentry *d_alloc_name(struct dentry *parent, const char *name) |
1333 | { |
1334 | struct qstr q; |
1335 | |
1336 | q.name = name; |
1337 | q.len = strlen(name); |
1338 | q.hash = full_name_hash(q.name, q.len); |
1339 | return d_alloc(parent, &q); |
1340 | } |
1341 | EXPORT_SYMBOL(d_alloc_name); |
1342 | |
1343 | void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op) |
1344 | { |
1345 | WARN_ON_ONCE(dentry->d_op); |
1346 | WARN_ON_ONCE(dentry->d_flags & (DCACHE_OP_HASH | |
1347 | DCACHE_OP_COMPARE | |
1348 | DCACHE_OP_REVALIDATE | |
1349 | DCACHE_OP_DELETE )); |
1350 | dentry->d_op = op; |
1351 | if (!op) |
1352 | return; |
1353 | if (op->d_hash) |
1354 | dentry->d_flags |= DCACHE_OP_HASH; |
1355 | if (op->d_compare) |
1356 | dentry->d_flags |= DCACHE_OP_COMPARE; |
1357 | if (op->d_revalidate) |
1358 | dentry->d_flags |= DCACHE_OP_REVALIDATE; |
1359 | if (op->d_delete) |
1360 | dentry->d_flags |= DCACHE_OP_DELETE; |
1361 | |
1362 | } |
1363 | EXPORT_SYMBOL(d_set_d_op); |
1364 | |
1365 | static void __d_instantiate(struct dentry *dentry, struct inode *inode) |
1366 | { |
1367 | spin_lock(&dentry->d_lock); |
1368 | if (inode) { |
1369 | if (unlikely(IS_AUTOMOUNT(inode))) |
1370 | dentry->d_flags |= DCACHE_NEED_AUTOMOUNT; |
1371 | list_add(&dentry->d_alias, &inode->i_dentry); |
1372 | } |
1373 | dentry->d_inode = inode; |
1374 | dentry_rcuwalk_barrier(dentry); |
1375 | spin_unlock(&dentry->d_lock); |
1376 | fsnotify_d_instantiate(dentry, inode); |
1377 | } |
1378 | |
1379 | /** |
1380 | * d_instantiate - fill in inode information for a dentry |
1381 | * @entry: dentry to complete |
1382 | * @inode: inode to attach to this dentry |
1383 | * |
1384 | * Fill in inode information in the entry. |
1385 | * |
1386 | * This turns negative dentries into productive full members |
1387 | * of society. |
1388 | * |
1389 | * NOTE! This assumes that the inode count has been incremented |
1390 | * (or otherwise set) by the caller to indicate that it is now |
1391 | * in use by the dcache. |
1392 | */ |
1393 | |
1394 | void d_instantiate(struct dentry *entry, struct inode * inode) |
1395 | { |
1396 | BUG_ON(!list_empty(&entry->d_alias)); |
1397 | if (inode) |
1398 | spin_lock(&inode->i_lock); |
1399 | __d_instantiate(entry, inode); |
1400 | if (inode) |
1401 | spin_unlock(&inode->i_lock); |
1402 | security_d_instantiate(entry, inode); |
1403 | } |
1404 | EXPORT_SYMBOL(d_instantiate); |
1405 | |
1406 | /** |
1407 | * d_instantiate_unique - instantiate a non-aliased dentry |
1408 | * @entry: dentry to instantiate |
1409 | * @inode: inode to attach to this dentry |
1410 | * |
1411 | * Fill in inode information in the entry. On success, it returns NULL. |
1412 | * If an unhashed alias of "entry" already exists, then we return the |
1413 | * aliased dentry instead and drop one reference to inode. |
1414 | * |
1415 | * Note that in order to avoid conflicts with rename() etc, the caller |
1416 | * had better be holding the parent directory semaphore. |
1417 | * |
1418 | * This also assumes that the inode count has been incremented |
1419 | * (or otherwise set) by the caller to indicate that it is now |
1420 | * in use by the dcache. |
1421 | */ |
1422 | static struct dentry *__d_instantiate_unique(struct dentry *entry, |
1423 | struct inode *inode) |
1424 | { |
1425 | struct dentry *alias; |
1426 | int len = entry->d_name.len; |
1427 | const char *name = entry->d_name.name; |
1428 | unsigned int hash = entry->d_name.hash; |
1429 | |
1430 | if (!inode) { |
1431 | __d_instantiate(entry, NULL); |
1432 | return NULL; |
1433 | } |
1434 | |
1435 | list_for_each_entry(alias, &inode->i_dentry, d_alias) { |
1436 | struct qstr *qstr = &alias->d_name; |
1437 | |
1438 | /* |
1439 | * Don't need alias->d_lock here, because aliases with |
1440 | * d_parent == entry->d_parent are not subject to name or |
1441 | * parent changes, because the parent inode i_mutex is held. |
1442 | */ |
1443 | if (qstr->hash != hash) |
1444 | continue; |
1445 | if (alias->d_parent != entry->d_parent) |
1446 | continue; |
1447 | if (dentry_cmp(qstr->name, qstr->len, name, len)) |
1448 | continue; |
1449 | __dget(alias); |
1450 | return alias; |
1451 | } |
1452 | |
1453 | __d_instantiate(entry, inode); |
1454 | return NULL; |
1455 | } |
1456 | |
1457 | struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode) |
1458 | { |
1459 | struct dentry *result; |
1460 | |
1461 | BUG_ON(!list_empty(&entry->d_alias)); |
1462 | |
1463 | if (inode) |
1464 | spin_lock(&inode->i_lock); |
1465 | result = __d_instantiate_unique(entry, inode); |
1466 | if (inode) |
1467 | spin_unlock(&inode->i_lock); |
1468 | |
1469 | if (!result) { |
1470 | security_d_instantiate(entry, inode); |
1471 | return NULL; |
1472 | } |
1473 | |
1474 | BUG_ON(!d_unhashed(result)); |
1475 | iput(inode); |
1476 | return result; |
1477 | } |
1478 | |
1479 | EXPORT_SYMBOL(d_instantiate_unique); |
1480 | |
1481 | /** |
1482 | * d_alloc_root - allocate root dentry |
1483 | * @root_inode: inode to allocate the root for |
1484 | * |
1485 | * Allocate a root ("/") dentry for the inode given. The inode is |
1486 | * instantiated and returned. %NULL is returned if there is insufficient |
1487 | * memory or the inode passed is %NULL. |
1488 | */ |
1489 | |
1490 | struct dentry * d_alloc_root(struct inode * root_inode) |
1491 | { |
1492 | struct dentry *res = NULL; |
1493 | |
1494 | if (root_inode) { |
1495 | static const struct qstr name = { .name = "/", .len = 1 }; |
1496 | |
1497 | res = d_alloc(NULL, &name); |
1498 | if (res) { |
1499 | res->d_sb = root_inode->i_sb; |
1500 | d_set_d_op(res, res->d_sb->s_d_op); |
1501 | res->d_parent = res; |
1502 | d_instantiate(res, root_inode); |
1503 | } |
1504 | } |
1505 | return res; |
1506 | } |
1507 | EXPORT_SYMBOL(d_alloc_root); |
1508 | |
1509 | static struct dentry * __d_find_any_alias(struct inode *inode) |
1510 | { |
1511 | struct dentry *alias; |
1512 | |
1513 | if (list_empty(&inode->i_dentry)) |
1514 | return NULL; |
1515 | alias = list_first_entry(&inode->i_dentry, struct dentry, d_alias); |
1516 | __dget(alias); |
1517 | return alias; |
1518 | } |
1519 | |
1520 | static struct dentry * d_find_any_alias(struct inode *inode) |
1521 | { |
1522 | struct dentry *de; |
1523 | |
1524 | spin_lock(&inode->i_lock); |
1525 | de = __d_find_any_alias(inode); |
1526 | spin_unlock(&inode->i_lock); |
1527 | return de; |
1528 | } |
1529 | |
1530 | |
1531 | /** |
1532 | * d_obtain_alias - find or allocate a dentry for a given inode |
1533 | * @inode: inode to allocate the dentry for |
1534 | * |
1535 | * Obtain a dentry for an inode resulting from NFS filehandle conversion or |
1536 | * similar open by handle operations. The returned dentry may be anonymous, |
1537 | * or may have a full name (if the inode was already in the cache). |
1538 | * |
1539 | * When called on a directory inode, we must ensure that the inode only ever |
1540 | * has one dentry. If a dentry is found, that is returned instead of |
1541 | * allocating a new one. |
1542 | * |
1543 | * On successful return, the reference to the inode has been transferred |
1544 | * to the dentry. In case of an error the reference on the inode is released. |
1545 | * To make it easier to use in export operations a %NULL or IS_ERR inode may |
1546 | * be passed in and will be the error will be propagate to the return value, |
1547 | * with a %NULL @inode replaced by ERR_PTR(-ESTALE). |
1548 | */ |
1549 | struct dentry *d_obtain_alias(struct inode *inode) |
1550 | { |
1551 | static const struct qstr anonstring = { .name = "" }; |
1552 | struct dentry *tmp; |
1553 | struct dentry *res; |
1554 | |
1555 | if (!inode) |
1556 | return ERR_PTR(-ESTALE); |
1557 | if (IS_ERR(inode)) |
1558 | return ERR_CAST(inode); |
1559 | |
1560 | res = d_find_any_alias(inode); |
1561 | if (res) |
1562 | goto out_iput; |
1563 | |
1564 | tmp = d_alloc(NULL, &anonstring); |
1565 | if (!tmp) { |
1566 | res = ERR_PTR(-ENOMEM); |
1567 | goto out_iput; |
1568 | } |
1569 | tmp->d_parent = tmp; /* make sure dput doesn't croak */ |
1570 | |
1571 | |
1572 | spin_lock(&inode->i_lock); |
1573 | res = __d_find_any_alias(inode); |
1574 | if (res) { |
1575 | spin_unlock(&inode->i_lock); |
1576 | dput(tmp); |
1577 | goto out_iput; |
1578 | } |
1579 | |
1580 | /* attach a disconnected dentry */ |
1581 | spin_lock(&tmp->d_lock); |
1582 | tmp->d_sb = inode->i_sb; |
1583 | d_set_d_op(tmp, tmp->d_sb->s_d_op); |
1584 | tmp->d_inode = inode; |
1585 | tmp->d_flags |= DCACHE_DISCONNECTED; |
1586 | list_add(&tmp->d_alias, &inode->i_dentry); |
1587 | hlist_bl_lock(&tmp->d_sb->s_anon); |
1588 | hlist_bl_add_head(&tmp->d_hash, &tmp->d_sb->s_anon); |
1589 | hlist_bl_unlock(&tmp->d_sb->s_anon); |
1590 | spin_unlock(&tmp->d_lock); |
1591 | spin_unlock(&inode->i_lock); |
1592 | security_d_instantiate(tmp, inode); |
1593 | |
1594 | return tmp; |
1595 | |
1596 | out_iput: |
1597 | if (res && !IS_ERR(res)) |
1598 | security_d_instantiate(res, inode); |
1599 | iput(inode); |
1600 | return res; |
1601 | } |
1602 | EXPORT_SYMBOL(d_obtain_alias); |
1603 | |
1604 | /** |
1605 | * d_splice_alias - splice a disconnected dentry into the tree if one exists |
1606 | * @inode: the inode which may have a disconnected dentry |
1607 | * @dentry: a negative dentry which we want to point to the inode. |
1608 | * |
1609 | * If inode is a directory and has a 'disconnected' dentry (i.e. IS_ROOT and |
1610 | * DCACHE_DISCONNECTED), then d_move that in place of the given dentry |
1611 | * and return it, else simply d_add the inode to the dentry and return NULL. |
1612 | * |
1613 | * This is needed in the lookup routine of any filesystem that is exportable |
1614 | * (via knfsd) so that we can build dcache paths to directories effectively. |
1615 | * |
1616 | * If a dentry was found and moved, then it is returned. Otherwise NULL |
1617 | * is returned. This matches the expected return value of ->lookup. |
1618 | * |
1619 | */ |
1620 | struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) |
1621 | { |
1622 | struct dentry *new = NULL; |
1623 | |
1624 | if (inode && S_ISDIR(inode->i_mode)) { |
1625 | spin_lock(&inode->i_lock); |
1626 | new = __d_find_alias(inode, 1); |
1627 | if (new) { |
1628 | BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED)); |
1629 | spin_unlock(&inode->i_lock); |
1630 | security_d_instantiate(new, inode); |
1631 | d_move(new, dentry); |
1632 | iput(inode); |
1633 | } else { |
1634 | /* already taking inode->i_lock, so d_add() by hand */ |
1635 | __d_instantiate(dentry, inode); |
1636 | spin_unlock(&inode->i_lock); |
1637 | security_d_instantiate(dentry, inode); |
1638 | d_rehash(dentry); |
1639 | } |
1640 | } else |
1641 | d_add(dentry, inode); |
1642 | return new; |
1643 | } |
1644 | EXPORT_SYMBOL(d_splice_alias); |
1645 | |
1646 | /** |
1647 | * d_add_ci - lookup or allocate new dentry with case-exact name |
1648 | * @inode: the inode case-insensitive lookup has found |
1649 | * @dentry: the negative dentry that was passed to the parent's lookup func |
1650 | * @name: the case-exact name to be associated with the returned dentry |
1651 | * |
1652 | * This is to avoid filling the dcache with case-insensitive names to the |
1653 | * same inode, only the actual correct case is stored in the dcache for |
1654 | * case-insensitive filesystems. |
1655 | * |
1656 | * For a case-insensitive lookup match and if the the case-exact dentry |
1657 | * already exists in in the dcache, use it and return it. |
1658 | * |
1659 | * If no entry exists with the exact case name, allocate new dentry with |
1660 | * the exact case, and return the spliced entry. |
1661 | */ |
1662 | struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode, |
1663 | struct qstr *name) |
1664 | { |
1665 | int error; |
1666 | struct dentry *found; |
1667 | struct dentry *new; |
1668 | |
1669 | /* |
1670 | * First check if a dentry matching the name already exists, |
1671 | * if not go ahead and create it now. |
1672 | */ |
1673 | found = d_hash_and_lookup(dentry->d_parent, name); |
1674 | if (!found) { |
1675 | new = d_alloc(dentry->d_parent, name); |
1676 | if (!new) { |
1677 | error = -ENOMEM; |
1678 | goto err_out; |
1679 | } |
1680 | |
1681 | found = d_splice_alias(inode, new); |
1682 | if (found) { |
1683 | dput(new); |
1684 | return found; |
1685 | } |
1686 | return new; |
1687 | } |
1688 | |
1689 | /* |
1690 | * If a matching dentry exists, and it's not negative use it. |
1691 | * |
1692 | * Decrement the reference count to balance the iget() done |
1693 | * earlier on. |
1694 | */ |
1695 | if (found->d_inode) { |
1696 | if (unlikely(found->d_inode != inode)) { |
1697 | /* This can't happen because bad inodes are unhashed. */ |
1698 | BUG_ON(!is_bad_inode(inode)); |
1699 | BUG_ON(!is_bad_inode(found->d_inode)); |
1700 | } |
1701 | iput(inode); |
1702 | return found; |
1703 | } |
1704 | |
1705 | /* |
1706 | * Negative dentry: instantiate it unless the inode is a directory and |
1707 | * already has a dentry. |
1708 | */ |
1709 | spin_lock(&inode->i_lock); |
1710 | if (!S_ISDIR(inode->i_mode) || list_empty(&inode->i_dentry)) { |
1711 | __d_instantiate(found, inode); |
1712 | spin_unlock(&inode->i_lock); |
1713 | security_d_instantiate(found, inode); |
1714 | return found; |
1715 | } |
1716 | |
1717 | /* |
1718 | * In case a directory already has a (disconnected) entry grab a |
1719 | * reference to it, move it in place and use it. |
1720 | */ |
1721 | new = list_entry(inode->i_dentry.next, struct dentry, d_alias); |
1722 | __dget(new); |
1723 | spin_unlock(&inode->i_lock); |
1724 | security_d_instantiate(found, inode); |
1725 | d_move(new, found); |
1726 | iput(inode); |
1727 | dput(found); |
1728 | return new; |
1729 | |
1730 | err_out: |
1731 | iput(inode); |
1732 | return ERR_PTR(error); |
1733 | } |
1734 | EXPORT_SYMBOL(d_add_ci); |
1735 | |
1736 | /** |
1737 | * __d_lookup_rcu - search for a dentry (racy, store-free) |
1738 | * @parent: parent dentry |
1739 | * @name: qstr of name we wish to find |
1740 | * @seq: returns d_seq value at the point where the dentry was found |
1741 | * @inode: returns dentry->d_inode when the inode was found valid. |
1742 | * Returns: dentry, or NULL |
1743 | * |
1744 | * __d_lookup_rcu is the dcache lookup function for rcu-walk name |
1745 | * resolution (store-free path walking) design described in |
1746 | * Documentation/filesystems/path-lookup.txt. |
1747 | * |
1748 | * This is not to be used outside core vfs. |
1749 | * |
1750 | * __d_lookup_rcu must only be used in rcu-walk mode, ie. with vfsmount lock |
1751 | * held, and rcu_read_lock held. The returned dentry must not be stored into |
1752 | * without taking d_lock and checking d_seq sequence count against @seq |
1753 | * returned here. |
1754 | * |
1755 | * A refcount may be taken on the found dentry with the __d_rcu_to_refcount |
1756 | * function. |
1757 | * |
1758 | * Alternatively, __d_lookup_rcu may be called again to look up the child of |
1759 | * the returned dentry, so long as its parent's seqlock is checked after the |
1760 | * child is looked up. Thus, an interlocking stepping of sequence lock checks |
1761 | * is formed, giving integrity down the path walk. |
1762 | */ |
1763 | struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, |
1764 | unsigned *seq, struct inode **inode) |
1765 | { |
1766 | unsigned int len = name->len; |
1767 | unsigned int hash = name->hash; |
1768 | const unsigned char *str = name->name; |
1769 | struct hlist_bl_head *b = d_hash(parent, hash); |
1770 | struct hlist_bl_node *node; |
1771 | struct dentry *dentry; |
1772 | |
1773 | /* |
1774 | * Note: There is significant duplication with __d_lookup_rcu which is |
1775 | * required to prevent single threaded performance regressions |
1776 | * especially on architectures where smp_rmb (in seqcounts) are costly. |
1777 | * Keep the two functions in sync. |
1778 | */ |
1779 | |
1780 | /* |
1781 | * The hash list is protected using RCU. |
1782 | * |
1783 | * Carefully use d_seq when comparing a candidate dentry, to avoid |
1784 | * races with d_move(). |
1785 | * |
1786 | * It is possible that concurrent renames can mess up our list |
1787 | * walk here and result in missing our dentry, resulting in the |
1788 | * false-negative result. d_lookup() protects against concurrent |
1789 | * renames using rename_lock seqlock. |
1790 | * |
1791 | * See Documentation/filesystems/path-lookup.txt for more details. |
1792 | */ |
1793 | hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { |
1794 | struct inode *i; |
1795 | const char *tname; |
1796 | int tlen; |
1797 | |
1798 | if (dentry->d_name.hash != hash) |
1799 | continue; |
1800 | |
1801 | seqretry: |
1802 | *seq = read_seqcount_begin(&dentry->d_seq); |
1803 | if (dentry->d_parent != parent) |
1804 | continue; |
1805 | if (d_unhashed(dentry)) |
1806 | continue; |
1807 | tlen = dentry->d_name.len; |
1808 | tname = dentry->d_name.name; |
1809 | i = dentry->d_inode; |
1810 | prefetch(tname); |
1811 | if (i) |
1812 | prefetch(i); |
1813 | /* |
1814 | * This seqcount check is required to ensure name and |
1815 | * len are loaded atomically, so as not to walk off the |
1816 | * edge of memory when walking. If we could load this |
1817 | * atomically some other way, we could drop this check. |
1818 | */ |
1819 | if (read_seqcount_retry(&dentry->d_seq, *seq)) |
1820 | goto seqretry; |
1821 | if (parent->d_flags & DCACHE_OP_COMPARE) { |
1822 | if (parent->d_op->d_compare(parent, *inode, |
1823 | dentry, i, |
1824 | tlen, tname, name)) |
1825 | continue; |
1826 | } else { |
1827 | if (dentry_cmp(tname, tlen, str, len)) |
1828 | continue; |
1829 | } |
1830 | /* |
1831 | * No extra seqcount check is required after the name |
1832 | * compare. The caller must perform a seqcount check in |
1833 | * order to do anything useful with the returned dentry |
1834 | * anyway. |
1835 | */ |
1836 | *inode = i; |
1837 | return dentry; |
1838 | } |
1839 | return NULL; |
1840 | } |
1841 | |
1842 | /** |
1843 | * d_lookup - search for a dentry |
1844 | * @parent: parent dentry |
1845 | * @name: qstr of name we wish to find |
1846 | * Returns: dentry, or NULL |
1847 | * |
1848 | * d_lookup searches the children of the parent dentry for the name in |
1849 | * question. If the dentry is found its reference count is incremented and the |
1850 | * dentry is returned. The caller must use dput to free the entry when it has |
1851 | * finished using it. %NULL is returned if the dentry does not exist. |
1852 | */ |
1853 | struct dentry *d_lookup(struct dentry *parent, struct qstr *name) |
1854 | { |
1855 | struct dentry *dentry; |
1856 | unsigned seq; |
1857 | |
1858 | do { |
1859 | seq = read_seqbegin(&rename_lock); |
1860 | dentry = __d_lookup(parent, name); |
1861 | if (dentry) |
1862 | break; |
1863 | } while (read_seqretry(&rename_lock, seq)); |
1864 | return dentry; |
1865 | } |
1866 | EXPORT_SYMBOL(d_lookup); |
1867 | |
1868 | /** |
1869 | * __d_lookup - search for a dentry (racy) |
1870 | * @parent: parent dentry |
1871 | * @name: qstr of name we wish to find |
1872 | * Returns: dentry, or NULL |
1873 | * |
1874 | * __d_lookup is like d_lookup, however it may (rarely) return a |
1875 | * false-negative result due to unrelated rename activity. |
1876 | * |
1877 | * __d_lookup is slightly faster by avoiding rename_lock read seqlock, |
1878 | * however it must be used carefully, eg. with a following d_lookup in |
1879 | * the case of failure. |
1880 | * |
1881 | * __d_lookup callers must be commented. |
1882 | */ |
1883 | struct dentry *__d_lookup(struct dentry *parent, struct qstr *name) |
1884 | { |
1885 | unsigned int len = name->len; |
1886 | unsigned int hash = name->hash; |
1887 | const unsigned char *str = name->name; |
1888 | struct hlist_bl_head *b = d_hash(parent, hash); |
1889 | struct hlist_bl_node *node; |
1890 | struct dentry *found = NULL; |
1891 | struct dentry *dentry; |
1892 | |
1893 | /* |
1894 | * Note: There is significant duplication with __d_lookup_rcu which is |
1895 | * required to prevent single threaded performance regressions |
1896 | * especially on architectures where smp_rmb (in seqcounts) are costly. |
1897 | * Keep the two functions in sync. |
1898 | */ |
1899 | |
1900 | /* |
1901 | * The hash list is protected using RCU. |
1902 | * |
1903 | * Take d_lock when comparing a candidate dentry, to avoid races |
1904 | * with d_move(). |
1905 | * |
1906 | * It is possible that concurrent renames can mess up our list |
1907 | * walk here and result in missing our dentry, resulting in the |
1908 | * false-negative result. d_lookup() protects against concurrent |
1909 | * renames using rename_lock seqlock. |
1910 | * |
1911 | * See Documentation/filesystems/path-lookup.txt for more details. |
1912 | */ |
1913 | rcu_read_lock(); |
1914 | |
1915 | hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { |
1916 | const char *tname; |
1917 | int tlen; |
1918 | |
1919 | if (dentry->d_name.hash != hash) |
1920 | continue; |
1921 | |
1922 | spin_lock(&dentry->d_lock); |
1923 | if (dentry->d_parent != parent) |
1924 | goto next; |
1925 | if (d_unhashed(dentry)) |
1926 | goto next; |
1927 | |
1928 | /* |
1929 | * It is safe to compare names since d_move() cannot |
1930 | * change the qstr (protected by d_lock). |
1931 | */ |
1932 | tlen = dentry->d_name.len; |
1933 | tname = dentry->d_name.name; |
1934 | if (parent->d_flags & DCACHE_OP_COMPARE) { |
1935 | if (parent->d_op->d_compare(parent, parent->d_inode, |
1936 | dentry, dentry->d_inode, |
1937 | tlen, tname, name)) |
1938 | goto next; |
1939 | } else { |
1940 | if (dentry_cmp(tname, tlen, str, len)) |
1941 | goto next; |
1942 | } |
1943 | |
1944 | dentry->d_count++; |
1945 | found = dentry; |
1946 | spin_unlock(&dentry->d_lock); |
1947 | break; |
1948 | next: |
1949 | spin_unlock(&dentry->d_lock); |
1950 | } |
1951 | rcu_read_unlock(); |
1952 | |
1953 | return found; |
1954 | } |
1955 | |
1956 | /** |
1957 | * d_hash_and_lookup - hash the qstr then search for a dentry |
1958 | * @dir: Directory to search in |
1959 | * @name: qstr of name we wish to find |
1960 | * |
1961 | * On hash failure or on lookup failure NULL is returned. |
1962 | */ |
1963 | struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name) |
1964 | { |
1965 | struct dentry *dentry = NULL; |
1966 | |
1967 | /* |
1968 | * Check for a fs-specific hash function. Note that we must |
1969 | * calculate the standard hash first, as the d_op->d_hash() |
1970 | * routine may choose to leave the hash value unchanged. |
1971 | */ |
1972 | name->hash = full_name_hash(name->name, name->len); |
1973 | if (dir->d_flags & DCACHE_OP_HASH) { |
1974 | if (dir->d_op->d_hash(dir, dir->d_inode, name) < 0) |
1975 | goto out; |
1976 | } |
1977 | dentry = d_lookup(dir, name); |
1978 | out: |
1979 | return dentry; |
1980 | } |
1981 | |
1982 | /** |
1983 | * d_validate - verify dentry provided from insecure source (deprecated) |
1984 | * @dentry: The dentry alleged to be valid child of @dparent |
1985 | * @dparent: The parent dentry (known to be valid) |
1986 | * |
1987 | * An insecure source has sent us a dentry, here we verify it and dget() it. |
1988 | * This is used by ncpfs in its readdir implementation. |
1989 | * Zero is returned in the dentry is invalid. |
1990 | * |
1991 | * This function is slow for big directories, and deprecated, do not use it. |
1992 | */ |
1993 | int d_validate(struct dentry *dentry, struct dentry *dparent) |
1994 | { |
1995 | struct dentry *child; |
1996 | |
1997 | spin_lock(&dparent->d_lock); |
1998 | list_for_each_entry(child, &dparent->d_subdirs, d_u.d_child) { |
1999 | if (dentry == child) { |
2000 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
2001 | __dget_dlock(dentry); |
2002 | spin_unlock(&dentry->d_lock); |
2003 | spin_unlock(&dparent->d_lock); |
2004 | return 1; |
2005 | } |
2006 | } |
2007 | spin_unlock(&dparent->d_lock); |
2008 | |
2009 | return 0; |
2010 | } |
2011 | EXPORT_SYMBOL(d_validate); |
2012 | |
2013 | /* |
2014 | * When a file is deleted, we have two options: |
2015 | * - turn this dentry into a negative dentry |
2016 | * - unhash this dentry and free it. |
2017 | * |
2018 | * Usually, we want to just turn this into |
2019 | * a negative dentry, but if anybody else is |
2020 | * currently using the dentry or the inode |
2021 | * we can't do that and we fall back on removing |
2022 | * it from the hash queues and waiting for |
2023 | * it to be deleted later when it has no users |
2024 | */ |
2025 | |
2026 | /** |
2027 | * d_delete - delete a dentry |
2028 | * @dentry: The dentry to delete |
2029 | * |
2030 | * Turn the dentry into a negative dentry if possible, otherwise |
2031 | * remove it from the hash queues so it can be deleted later |
2032 | */ |
2033 | |
2034 | void d_delete(struct dentry * dentry) |
2035 | { |
2036 | struct inode *inode; |
2037 | int isdir = 0; |
2038 | /* |
2039 | * Are we the only user? |
2040 | */ |
2041 | again: |
2042 | spin_lock(&dentry->d_lock); |
2043 | inode = dentry->d_inode; |
2044 | isdir = S_ISDIR(inode->i_mode); |
2045 | if (dentry->d_count == 1) { |
2046 | if (inode && !spin_trylock(&inode->i_lock)) { |
2047 | spin_unlock(&dentry->d_lock); |
2048 | cpu_relax(); |
2049 | goto again; |
2050 | } |
2051 | dentry->d_flags &= ~DCACHE_CANT_MOUNT; |
2052 | dentry_unlink_inode(dentry); |
2053 | fsnotify_nameremove(dentry, isdir); |
2054 | return; |
2055 | } |
2056 | |
2057 | if (!d_unhashed(dentry)) |
2058 | __d_drop(dentry); |
2059 | |
2060 | spin_unlock(&dentry->d_lock); |
2061 | |
2062 | fsnotify_nameremove(dentry, isdir); |
2063 | } |
2064 | EXPORT_SYMBOL(d_delete); |
2065 | |
2066 | static void __d_rehash(struct dentry * entry, struct hlist_bl_head *b) |
2067 | { |
2068 | BUG_ON(!d_unhashed(entry)); |
2069 | hlist_bl_lock(b); |
2070 | entry->d_flags |= DCACHE_RCUACCESS; |
2071 | hlist_bl_add_head_rcu(&entry->d_hash, b); |
2072 | hlist_bl_unlock(b); |
2073 | } |
2074 | |
2075 | static void _d_rehash(struct dentry * entry) |
2076 | { |
2077 | __d_rehash(entry, d_hash(entry->d_parent, entry->d_name.hash)); |
2078 | } |
2079 | |
2080 | /** |
2081 | * d_rehash - add an entry back to the hash |
2082 | * @entry: dentry to add to the hash |
2083 | * |
2084 | * Adds a dentry to the hash according to its name. |
2085 | */ |
2086 | |
2087 | void d_rehash(struct dentry * entry) |
2088 | { |
2089 | spin_lock(&entry->d_lock); |
2090 | _d_rehash(entry); |
2091 | spin_unlock(&entry->d_lock); |
2092 | } |
2093 | EXPORT_SYMBOL(d_rehash); |
2094 | |
2095 | /** |
2096 | * dentry_update_name_case - update case insensitive dentry with a new name |
2097 | * @dentry: dentry to be updated |
2098 | * @name: new name |
2099 | * |
2100 | * Update a case insensitive dentry with new case of name. |
2101 | * |
2102 | * dentry must have been returned by d_lookup with name @name. Old and new |
2103 | * name lengths must match (ie. no d_compare which allows mismatched name |
2104 | * lengths). |
2105 | * |
2106 | * Parent inode i_mutex must be held over d_lookup and into this call (to |
2107 | * keep renames and concurrent inserts, and readdir(2) away). |
2108 | */ |
2109 | void dentry_update_name_case(struct dentry *dentry, struct qstr *name) |
2110 | { |
2111 | BUG_ON(!mutex_is_locked(&dentry->d_parent->d_inode->i_mutex)); |
2112 | BUG_ON(dentry->d_name.len != name->len); /* d_lookup gives this */ |
2113 | |
2114 | spin_lock(&dentry->d_lock); |
2115 | write_seqcount_begin(&dentry->d_seq); |
2116 | memcpy((unsigned char *)dentry->d_name.name, name->name, name->len); |
2117 | write_seqcount_end(&dentry->d_seq); |
2118 | spin_unlock(&dentry->d_lock); |
2119 | } |
2120 | EXPORT_SYMBOL(dentry_update_name_case); |
2121 | |
2122 | static void switch_names(struct dentry *dentry, struct dentry *target) |
2123 | { |
2124 | if (dname_external(target)) { |
2125 | if (dname_external(dentry)) { |
2126 | /* |
2127 | * Both external: swap the pointers |
2128 | */ |
2129 | swap(target->d_name.name, dentry->d_name.name); |
2130 | } else { |
2131 | /* |
2132 | * dentry:internal, target:external. Steal target's |
2133 | * storage and make target internal. |
2134 | */ |
2135 | memcpy(target->d_iname, dentry->d_name.name, |
2136 | dentry->d_name.len + 1); |
2137 | dentry->d_name.name = target->d_name.name; |
2138 | target->d_name.name = target->d_iname; |
2139 | } |
2140 | } else { |
2141 | if (dname_external(dentry)) { |
2142 | /* |
2143 | * dentry:external, target:internal. Give dentry's |
2144 | * storage to target and make dentry internal |
2145 | */ |
2146 | memcpy(dentry->d_iname, target->d_name.name, |
2147 | target->d_name.len + 1); |
2148 | target->d_name.name = dentry->d_name.name; |
2149 | dentry->d_name.name = dentry->d_iname; |
2150 | } else { |
2151 | /* |
2152 | * Both are internal. Just copy target to dentry |
2153 | */ |
2154 | memcpy(dentry->d_iname, target->d_name.name, |
2155 | target->d_name.len + 1); |
2156 | dentry->d_name.len = target->d_name.len; |
2157 | return; |
2158 | } |
2159 | } |
2160 | swap(dentry->d_name.len, target->d_name.len); |
2161 | } |
2162 | |
2163 | static void dentry_lock_for_move(struct dentry *dentry, struct dentry *target) |
2164 | { |
2165 | /* |
2166 | * XXXX: do we really need to take target->d_lock? |
2167 | */ |
2168 | if (IS_ROOT(dentry) || dentry->d_parent == target->d_parent) |
2169 | spin_lock(&target->d_parent->d_lock); |
2170 | else { |
2171 | if (d_ancestor(dentry->d_parent, target->d_parent)) { |
2172 | spin_lock(&dentry->d_parent->d_lock); |
2173 | spin_lock_nested(&target->d_parent->d_lock, |
2174 | DENTRY_D_LOCK_NESTED); |
2175 | } else { |
2176 | spin_lock(&target->d_parent->d_lock); |
2177 | spin_lock_nested(&dentry->d_parent->d_lock, |
2178 | DENTRY_D_LOCK_NESTED); |
2179 | } |
2180 | } |
2181 | if (target < dentry) { |
2182 | spin_lock_nested(&target->d_lock, 2); |
2183 | spin_lock_nested(&dentry->d_lock, 3); |
2184 | } else { |
2185 | spin_lock_nested(&dentry->d_lock, 2); |
2186 | spin_lock_nested(&target->d_lock, 3); |
2187 | } |
2188 | } |
2189 | |
2190 | static void dentry_unlock_parents_for_move(struct dentry *dentry, |
2191 | struct dentry *target) |
2192 | { |
2193 | if (target->d_parent != dentry->d_parent) |
2194 | spin_unlock(&dentry->d_parent->d_lock); |
2195 | if (target->d_parent != target) |
2196 | spin_unlock(&target->d_parent->d_lock); |
2197 | } |
2198 | |
2199 | /* |
2200 | * When switching names, the actual string doesn't strictly have to |
2201 | * be preserved in the target - because we're dropping the target |
2202 | * anyway. As such, we can just do a simple memcpy() to copy over |
2203 | * the new name before we switch. |
2204 | * |
2205 | * Note that we have to be a lot more careful about getting the hash |
2206 | * switched - we have to switch the hash value properly even if it |
2207 | * then no longer matches the actual (corrupted) string of the target. |
2208 | * The hash value has to match the hash queue that the dentry is on.. |
2209 | */ |
2210 | /* |
2211 | * d_move - move a dentry |
2212 | * @dentry: entry to move |
2213 | * @target: new dentry |
2214 | * |
2215 | * Update the dcache to reflect the move of a file name. Negative |
2216 | * dcache entries should not be moved in this way. |
2217 | */ |
2218 | void d_move(struct dentry * dentry, struct dentry * target) |
2219 | { |
2220 | if (!dentry->d_inode) |
2221 | printk(KERN_WARNING "VFS: moving negative dcache entry\n"); |
2222 | |
2223 | BUG_ON(d_ancestor(dentry, target)); |
2224 | BUG_ON(d_ancestor(target, dentry)); |
2225 | |
2226 | write_seqlock(&rename_lock); |
2227 | |
2228 | dentry_lock_for_move(dentry, target); |
2229 | |
2230 | write_seqcount_begin(&dentry->d_seq); |
2231 | write_seqcount_begin(&target->d_seq); |
2232 | |
2233 | /* __d_drop does write_seqcount_barrier, but they're OK to nest. */ |
2234 | |
2235 | /* |
2236 | * Move the dentry to the target hash queue. Don't bother checking |
2237 | * for the same hash queue because of how unlikely it is. |
2238 | */ |
2239 | __d_drop(dentry); |
2240 | __d_rehash(dentry, d_hash(target->d_parent, target->d_name.hash)); |
2241 | |
2242 | /* Unhash the target: dput() will then get rid of it */ |
2243 | __d_drop(target); |
2244 | |
2245 | list_del(&dentry->d_u.d_child); |
2246 | list_del(&target->d_u.d_child); |
2247 | |
2248 | /* Switch the names.. */ |
2249 | switch_names(dentry, target); |
2250 | swap(dentry->d_name.hash, target->d_name.hash); |
2251 | |
2252 | /* ... and switch the parents */ |
2253 | if (IS_ROOT(dentry)) { |
2254 | dentry->d_parent = target->d_parent; |
2255 | target->d_parent = target; |
2256 | INIT_LIST_HEAD(&target->d_u.d_child); |
2257 | } else { |
2258 | swap(dentry->d_parent, target->d_parent); |
2259 | |
2260 | /* And add them back to the (new) parent lists */ |
2261 | list_add(&target->d_u.d_child, &target->d_parent->d_subdirs); |
2262 | } |
2263 | |
2264 | list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs); |
2265 | |
2266 | write_seqcount_end(&target->d_seq); |
2267 | write_seqcount_end(&dentry->d_seq); |
2268 | |
2269 | dentry_unlock_parents_for_move(dentry, target); |
2270 | spin_unlock(&target->d_lock); |
2271 | fsnotify_d_move(dentry); |
2272 | spin_unlock(&dentry->d_lock); |
2273 | write_sequnlock(&rename_lock); |
2274 | } |
2275 | EXPORT_SYMBOL(d_move); |
2276 | |
2277 | /** |
2278 | * d_ancestor - search for an ancestor |
2279 | * @p1: ancestor dentry |
2280 | * @p2: child dentry |
2281 | * |
2282 | * Returns the ancestor dentry of p2 which is a child of p1, if p1 is |
2283 | * an ancestor of p2, else NULL. |
2284 | */ |
2285 | struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2) |
2286 | { |
2287 | struct dentry *p; |
2288 | |
2289 | for (p = p2; !IS_ROOT(p); p = p->d_parent) { |
2290 | if (p->d_parent == p1) |
2291 | return p; |
2292 | } |
2293 | return NULL; |
2294 | } |
2295 | |
2296 | /* |
2297 | * This helper attempts to cope with remotely renamed directories |
2298 | * |
2299 | * It assumes that the caller is already holding |
2300 | * dentry->d_parent->d_inode->i_mutex and the inode->i_lock |
2301 | * |
2302 | * Note: If ever the locking in lock_rename() changes, then please |
2303 | * remember to update this too... |
2304 | */ |
2305 | static struct dentry *__d_unalias(struct inode *inode, |
2306 | struct dentry *dentry, struct dentry *alias) |
2307 | { |
2308 | struct mutex *m1 = NULL, *m2 = NULL; |
2309 | struct dentry *ret; |
2310 | |
2311 | /* If alias and dentry share a parent, then no extra locks required */ |
2312 | if (alias->d_parent == dentry->d_parent) |
2313 | goto out_unalias; |
2314 | |
2315 | /* Check for loops */ |
2316 | ret = ERR_PTR(-ELOOP); |
2317 | if (d_ancestor(alias, dentry)) |
2318 | goto out_err; |
2319 | |
2320 | /* See lock_rename() */ |
2321 | ret = ERR_PTR(-EBUSY); |
2322 | if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex)) |
2323 | goto out_err; |
2324 | m1 = &dentry->d_sb->s_vfs_rename_mutex; |
2325 | if (!mutex_trylock(&alias->d_parent->d_inode->i_mutex)) |
2326 | goto out_err; |
2327 | m2 = &alias->d_parent->d_inode->i_mutex; |
2328 | out_unalias: |
2329 | d_move(alias, dentry); |
2330 | ret = alias; |
2331 | out_err: |
2332 | spin_unlock(&inode->i_lock); |
2333 | if (m2) |
2334 | mutex_unlock(m2); |
2335 | if (m1) |
2336 | mutex_unlock(m1); |
2337 | return ret; |
2338 | } |
2339 | |
2340 | /* |
2341 | * Prepare an anonymous dentry for life in the superblock's dentry tree as a |
2342 | * named dentry in place of the dentry to be replaced. |
2343 | * returns with anon->d_lock held! |
2344 | */ |
2345 | static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon) |
2346 | { |
2347 | struct dentry *dparent, *aparent; |
2348 | |
2349 | dentry_lock_for_move(anon, dentry); |
2350 | |
2351 | write_seqcount_begin(&dentry->d_seq); |
2352 | write_seqcount_begin(&anon->d_seq); |
2353 | |
2354 | dparent = dentry->d_parent; |
2355 | aparent = anon->d_parent; |
2356 | |
2357 | switch_names(dentry, anon); |
2358 | swap(dentry->d_name.hash, anon->d_name.hash); |
2359 | |
2360 | dentry->d_parent = (aparent == anon) ? dentry : aparent; |
2361 | list_del(&dentry->d_u.d_child); |
2362 | if (!IS_ROOT(dentry)) |
2363 | list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs); |
2364 | else |
2365 | INIT_LIST_HEAD(&dentry->d_u.d_child); |
2366 | |
2367 | anon->d_parent = (dparent == dentry) ? anon : dparent; |
2368 | list_del(&anon->d_u.d_child); |
2369 | if (!IS_ROOT(anon)) |
2370 | list_add(&anon->d_u.d_child, &anon->d_parent->d_subdirs); |
2371 | else |
2372 | INIT_LIST_HEAD(&anon->d_u.d_child); |
2373 | |
2374 | write_seqcount_end(&dentry->d_seq); |
2375 | write_seqcount_end(&anon->d_seq); |
2376 | |
2377 | dentry_unlock_parents_for_move(anon, dentry); |
2378 | spin_unlock(&dentry->d_lock); |
2379 | |
2380 | /* anon->d_lock still locked, returns locked */ |
2381 | anon->d_flags &= ~DCACHE_DISCONNECTED; |
2382 | } |
2383 | |
2384 | /** |
2385 | * d_materialise_unique - introduce an inode into the tree |
2386 | * @dentry: candidate dentry |
2387 | * @inode: inode to bind to the dentry, to which aliases may be attached |
2388 | * |
2389 | * Introduces an dentry into the tree, substituting an extant disconnected |
2390 | * root directory alias in its place if there is one |
2391 | */ |
2392 | struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) |
2393 | { |
2394 | struct dentry *actual; |
2395 | |
2396 | BUG_ON(!d_unhashed(dentry)); |
2397 | |
2398 | if (!inode) { |
2399 | actual = dentry; |
2400 | __d_instantiate(dentry, NULL); |
2401 | d_rehash(actual); |
2402 | goto out_nolock; |
2403 | } |
2404 | |
2405 | spin_lock(&inode->i_lock); |
2406 | |
2407 | if (S_ISDIR(inode->i_mode)) { |
2408 | struct dentry *alias; |
2409 | |
2410 | /* Does an aliased dentry already exist? */ |
2411 | alias = __d_find_alias(inode, 0); |
2412 | if (alias) { |
2413 | actual = alias; |
2414 | /* Is this an anonymous mountpoint that we could splice |
2415 | * into our tree? */ |
2416 | if (IS_ROOT(alias)) { |
2417 | __d_materialise_dentry(dentry, alias); |
2418 | __d_drop(alias); |
2419 | goto found; |
2420 | } |
2421 | /* Nope, but we must(!) avoid directory aliasing */ |
2422 | actual = __d_unalias(inode, dentry, alias); |
2423 | if (IS_ERR(actual)) |
2424 | dput(alias); |
2425 | goto out_nolock; |
2426 | } |
2427 | } |
2428 | |
2429 | /* Add a unique reference */ |
2430 | actual = __d_instantiate_unique(dentry, inode); |
2431 | if (!actual) |
2432 | actual = dentry; |
2433 | else |
2434 | BUG_ON(!d_unhashed(actual)); |
2435 | |
2436 | spin_lock(&actual->d_lock); |
2437 | found: |
2438 | _d_rehash(actual); |
2439 | spin_unlock(&actual->d_lock); |
2440 | spin_unlock(&inode->i_lock); |
2441 | out_nolock: |
2442 | if (actual == dentry) { |
2443 | security_d_instantiate(dentry, inode); |
2444 | return NULL; |
2445 | } |
2446 | |
2447 | iput(inode); |
2448 | return actual; |
2449 | } |
2450 | EXPORT_SYMBOL_GPL(d_materialise_unique); |
2451 | |
2452 | static int prepend(char **buffer, int *buflen, const char *str, int namelen) |
2453 | { |
2454 | *buflen -= namelen; |
2455 | if (*buflen < 0) |
2456 | return -ENAMETOOLONG; |
2457 | *buffer -= namelen; |
2458 | memcpy(*buffer, str, namelen); |
2459 | return 0; |
2460 | } |
2461 | |
2462 | static int prepend_name(char **buffer, int *buflen, struct qstr *name) |
2463 | { |
2464 | return prepend(buffer, buflen, name->name, name->len); |
2465 | } |
2466 | |
2467 | /** |
2468 | * prepend_path - Prepend path string to a buffer |
2469 | * @path: the dentry/vfsmount to report |
2470 | * @root: root vfsmnt/dentry (may be modified by this function) |
2471 | * @buffer: pointer to the end of the buffer |
2472 | * @buflen: pointer to buffer length |
2473 | * |
2474 | * Caller holds the rename_lock. |
2475 | * |
2476 | * If path is not reachable from the supplied root, then the value of |
2477 | * root is changed (without modifying refcounts). |
2478 | */ |
2479 | static int prepend_path(const struct path *path, struct path *root, |
2480 | char **buffer, int *buflen) |
2481 | { |
2482 | struct dentry *dentry = path->dentry; |
2483 | struct vfsmount *vfsmnt = path->mnt; |
2484 | bool slash = false; |
2485 | int error = 0; |
2486 | |
2487 | br_read_lock(vfsmount_lock); |
2488 | while (dentry != root->dentry || vfsmnt != root->mnt) { |
2489 | struct dentry * parent; |
2490 | |
2491 | if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { |
2492 | /* Global root? */ |
2493 | if (vfsmnt->mnt_parent == vfsmnt) { |
2494 | goto global_root; |
2495 | } |
2496 | dentry = vfsmnt->mnt_mountpoint; |
2497 | vfsmnt = vfsmnt->mnt_parent; |
2498 | continue; |
2499 | } |
2500 | parent = dentry->d_parent; |
2501 | prefetch(parent); |
2502 | spin_lock(&dentry->d_lock); |
2503 | error = prepend_name(buffer, buflen, &dentry->d_name); |
2504 | spin_unlock(&dentry->d_lock); |
2505 | if (!error) |
2506 | error = prepend(buffer, buflen, "/", 1); |
2507 | if (error) |
2508 | break; |
2509 | |
2510 | slash = true; |
2511 | dentry = parent; |
2512 | } |
2513 | |
2514 | out: |
2515 | if (!error && !slash) |
2516 | error = prepend(buffer, buflen, "/", 1); |
2517 | |
2518 | br_read_unlock(vfsmount_lock); |
2519 | return error; |
2520 | |
2521 | global_root: |
2522 | /* |
2523 | * Filesystems needing to implement special "root names" |
2524 | * should do so with ->d_dname() |
2525 | */ |
2526 | if (IS_ROOT(dentry) && |
2527 | (dentry->d_name.len != 1 || dentry->d_name.name[0] != '/')) { |
2528 | WARN(1, "Root dentry has weird name <%.*s>\n", |
2529 | (int) dentry->d_name.len, dentry->d_name.name); |
2530 | } |
2531 | root->mnt = vfsmnt; |
2532 | root->dentry = dentry; |
2533 | goto out; |
2534 | } |
2535 | |
2536 | /** |
2537 | * __d_path - return the path of a dentry |
2538 | * @path: the dentry/vfsmount to report |
2539 | * @root: root vfsmnt/dentry (may be modified by this function) |
2540 | * @buf: buffer to return value in |
2541 | * @buflen: buffer length |
2542 | * |
2543 | * Convert a dentry into an ASCII path name. |
2544 | * |
2545 | * Returns a pointer into the buffer or an error code if the |
2546 | * path was too long. |
2547 | * |
2548 | * "buflen" should be positive. |
2549 | * |
2550 | * If path is not reachable from the supplied root, then the value of |
2551 | * root is changed (without modifying refcounts). |
2552 | */ |
2553 | char *__d_path(const struct path *path, struct path *root, |
2554 | char *buf, int buflen) |
2555 | { |
2556 | char *res = buf + buflen; |
2557 | int error; |
2558 | |
2559 | prepend(&res, &buflen, "\0", 1); |
2560 | write_seqlock(&rename_lock); |
2561 | error = prepend_path(path, root, &res, &buflen); |
2562 | write_sequnlock(&rename_lock); |
2563 | |
2564 | if (error) |
2565 | return ERR_PTR(error); |
2566 | return res; |
2567 | } |
2568 | |
2569 | /* |
2570 | * same as __d_path but appends "(deleted)" for unlinked files. |
2571 | */ |
2572 | static int path_with_deleted(const struct path *path, struct path *root, |
2573 | char **buf, int *buflen) |
2574 | { |
2575 | prepend(buf, buflen, "\0", 1); |
2576 | if (d_unlinked(path->dentry)) { |
2577 | int error = prepend(buf, buflen, " (deleted)", 10); |
2578 | if (error) |
2579 | return error; |
2580 | } |
2581 | |
2582 | return prepend_path(path, root, buf, buflen); |
2583 | } |
2584 | |
2585 | static int prepend_unreachable(char **buffer, int *buflen) |
2586 | { |
2587 | return prepend(buffer, buflen, "(unreachable)", 13); |
2588 | } |
2589 | |
2590 | /** |
2591 | * d_path - return the path of a dentry |
2592 | * @path: path to report |
2593 | * @buf: buffer to return value in |
2594 | * @buflen: buffer length |
2595 | * |
2596 | * Convert a dentry into an ASCII path name. If the entry has been deleted |
2597 | * the string " (deleted)" is appended. Note that this is ambiguous. |
2598 | * |
2599 | * Returns a pointer into the buffer or an error code if the path was |
2600 | * too long. Note: Callers should use the returned pointer, not the passed |
2601 | * in buffer, to use the name! The implementation often starts at an offset |
2602 | * into the buffer, and may leave 0 bytes at the start. |
2603 | * |
2604 | * "buflen" should be positive. |
2605 | */ |
2606 | char *d_path(const struct path *path, char *buf, int buflen) |
2607 | { |
2608 | char *res = buf + buflen; |
2609 | struct path root; |
2610 | struct path tmp; |
2611 | int error; |
2612 | |
2613 | /* |
2614 | * We have various synthetic filesystems that never get mounted. On |
2615 | * these filesystems dentries are never used for lookup purposes, and |
2616 | * thus don't need to be hashed. They also don't need a name until a |
2617 | * user wants to identify the object in /proc/pid/fd/. The little hack |
2618 | * below allows us to generate a name for these objects on demand: |
2619 | */ |
2620 | if (path->dentry->d_op && path->dentry->d_op->d_dname) |
2621 | return path->dentry->d_op->d_dname(path->dentry, buf, buflen); |
2622 | |
2623 | get_fs_root(current->fs, &root); |
2624 | write_seqlock(&rename_lock); |
2625 | tmp = root; |
2626 | error = path_with_deleted(path, &tmp, &res, &buflen); |
2627 | if (error) |
2628 | res = ERR_PTR(error); |
2629 | write_sequnlock(&rename_lock); |
2630 | path_put(&root); |
2631 | return res; |
2632 | } |
2633 | EXPORT_SYMBOL(d_path); |
2634 | |
2635 | /** |
2636 | * d_path_with_unreachable - return the path of a dentry |
2637 | * @path: path to report |
2638 | * @buf: buffer to return value in |
2639 | * @buflen: buffer length |
2640 | * |
2641 | * The difference from d_path() is that this prepends "(unreachable)" |
2642 | * to paths which are unreachable from the current process' root. |
2643 | */ |
2644 | char *d_path_with_unreachable(const struct path *path, char *buf, int buflen) |
2645 | { |
2646 | char *res = buf + buflen; |
2647 | struct path root; |
2648 | struct path tmp; |
2649 | int error; |
2650 | |
2651 | if (path->dentry->d_op && path->dentry->d_op->d_dname) |
2652 | return path->dentry->d_op->d_dname(path->dentry, buf, buflen); |
2653 | |
2654 | get_fs_root(current->fs, &root); |
2655 | write_seqlock(&rename_lock); |
2656 | tmp = root; |
2657 | error = path_with_deleted(path, &tmp, &res, &buflen); |
2658 | if (!error && !path_equal(&tmp, &root)) |
2659 | error = prepend_unreachable(&res, &buflen); |
2660 | write_sequnlock(&rename_lock); |
2661 | path_put(&root); |
2662 | if (error) |
2663 | res = ERR_PTR(error); |
2664 | |
2665 | return res; |
2666 | } |
2667 | |
2668 | /* |
2669 | * Helper function for dentry_operations.d_dname() members |
2670 | */ |
2671 | char *dynamic_dname(struct dentry *dentry, char *buffer, int buflen, |
2672 | const char *fmt, ...) |
2673 | { |
2674 | va_list args; |
2675 | char temp[64]; |
2676 | int sz; |
2677 | |
2678 | va_start(args, fmt); |
2679 | sz = vsnprintf(temp, sizeof(temp), fmt, args) + 1; |
2680 | va_end(args); |
2681 | |
2682 | if (sz > sizeof(temp) || sz > buflen) |
2683 | return ERR_PTR(-ENAMETOOLONG); |
2684 | |
2685 | buffer += buflen - sz; |
2686 | return memcpy(buffer, temp, sz); |
2687 | } |
2688 | |
2689 | /* |
2690 | * Write full pathname from the root of the filesystem into the buffer. |
2691 | */ |
2692 | static char *__dentry_path(struct dentry *dentry, char *buf, int buflen) |
2693 | { |
2694 | char *end = buf + buflen; |
2695 | char *retval; |
2696 | |
2697 | prepend(&end, &buflen, "\0", 1); |
2698 | if (buflen < 1) |
2699 | goto Elong; |
2700 | /* Get '/' right */ |
2701 | retval = end-1; |
2702 | *retval = '/'; |
2703 | |
2704 | while (!IS_ROOT(dentry)) { |
2705 | struct dentry *parent = dentry->d_parent; |
2706 | int error; |
2707 | |
2708 | prefetch(parent); |
2709 | spin_lock(&dentry->d_lock); |
2710 | error = prepend_name(&end, &buflen, &dentry->d_name); |
2711 | spin_unlock(&dentry->d_lock); |
2712 | if (error != 0 || prepend(&end, &buflen, "/", 1) != 0) |
2713 | goto Elong; |
2714 | |
2715 | retval = end; |
2716 | dentry = parent; |
2717 | } |
2718 | return retval; |
2719 | Elong: |
2720 | return ERR_PTR(-ENAMETOOLONG); |
2721 | } |
2722 | |
2723 | char *dentry_path_raw(struct dentry *dentry, char *buf, int buflen) |
2724 | { |
2725 | char *retval; |
2726 | |
2727 | write_seqlock(&rename_lock); |
2728 | retval = __dentry_path(dentry, buf, buflen); |
2729 | write_sequnlock(&rename_lock); |
2730 | |
2731 | return retval; |
2732 | } |
2733 | EXPORT_SYMBOL(dentry_path_raw); |
2734 | |
2735 | char *dentry_path(struct dentry *dentry, char *buf, int buflen) |
2736 | { |
2737 | char *p = NULL; |
2738 | char *retval; |
2739 | |
2740 | write_seqlock(&rename_lock); |
2741 | if (d_unlinked(dentry)) { |
2742 | p = buf + buflen; |
2743 | if (prepend(&p, &buflen, "//deleted", 10) != 0) |
2744 | goto Elong; |
2745 | buflen++; |
2746 | } |
2747 | retval = __dentry_path(dentry, buf, buflen); |
2748 | write_sequnlock(&rename_lock); |
2749 | if (!IS_ERR(retval) && p) |
2750 | *p = '/'; /* restore '/' overriden with '\0' */ |
2751 | return retval; |
2752 | Elong: |
2753 | return ERR_PTR(-ENAMETOOLONG); |
2754 | } |
2755 | |
2756 | /* |
2757 | * NOTE! The user-level library version returns a |
2758 | * character pointer. The kernel system call just |
2759 | * returns the length of the buffer filled (which |
2760 | * includes the ending '\0' character), or a negative |
2761 | * error value. So libc would do something like |
2762 | * |
2763 | * char *getcwd(char * buf, size_t size) |
2764 | * { |
2765 | * int retval; |
2766 | * |
2767 | * retval = sys_getcwd(buf, size); |
2768 | * if (retval >= 0) |
2769 | * return buf; |
2770 | * errno = -retval; |
2771 | * return NULL; |
2772 | * } |
2773 | */ |
2774 | SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) |
2775 | { |
2776 | int error; |
2777 | struct path pwd, root; |
2778 | char *page = (char *) __get_free_page(GFP_USER); |
2779 | |
2780 | if (!page) |
2781 | return -ENOMEM; |
2782 | |
2783 | get_fs_root_and_pwd(current->fs, &root, &pwd); |
2784 | |
2785 | error = -ENOENT; |
2786 | write_seqlock(&rename_lock); |
2787 | if (!d_unlinked(pwd.dentry)) { |
2788 | unsigned long len; |
2789 | struct path tmp = root; |
2790 | char *cwd = page + PAGE_SIZE; |
2791 | int buflen = PAGE_SIZE; |
2792 | |
2793 | prepend(&cwd, &buflen, "\0", 1); |
2794 | error = prepend_path(&pwd, &tmp, &cwd, &buflen); |
2795 | write_sequnlock(&rename_lock); |
2796 | |
2797 | if (error) |
2798 | goto out; |
2799 | |
2800 | /* Unreachable from current root */ |
2801 | if (!path_equal(&tmp, &root)) { |
2802 | error = prepend_unreachable(&cwd, &buflen); |
2803 | if (error) |
2804 | goto out; |
2805 | } |
2806 | |
2807 | error = -ERANGE; |
2808 | len = PAGE_SIZE + page - cwd; |
2809 | if (len <= size) { |
2810 | error = len; |
2811 | if (copy_to_user(buf, cwd, len)) |
2812 | error = -EFAULT; |
2813 | } |
2814 | } else { |
2815 | write_sequnlock(&rename_lock); |
2816 | } |
2817 | |
2818 | out: |
2819 | path_put(&pwd); |
2820 | path_put(&root); |
2821 | free_page((unsigned long) page); |
2822 | return error; |
2823 | } |
2824 | |
2825 | /* |
2826 | * Test whether new_dentry is a subdirectory of old_dentry. |
2827 | * |
2828 | * Trivially implemented using the dcache structure |
2829 | */ |
2830 | |
2831 | /** |
2832 | * is_subdir - is new dentry a subdirectory of old_dentry |
2833 | * @new_dentry: new dentry |
2834 | * @old_dentry: old dentry |
2835 | * |
2836 | * Returns 1 if new_dentry is a subdirectory of the parent (at any depth). |
2837 | * Returns 0 otherwise. |
2838 | * Caller must ensure that "new_dentry" is pinned before calling is_subdir() |
2839 | */ |
2840 | |
2841 | int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) |
2842 | { |
2843 | int result; |
2844 | unsigned seq; |
2845 | |
2846 | if (new_dentry == old_dentry) |
2847 | return 1; |
2848 | |
2849 | do { |
2850 | /* for restarting inner loop in case of seq retry */ |
2851 | seq = read_seqbegin(&rename_lock); |
2852 | /* |
2853 | * Need rcu_readlock to protect against the d_parent trashing |
2854 | * due to d_move |
2855 | */ |
2856 | rcu_read_lock(); |
2857 | if (d_ancestor(old_dentry, new_dentry)) |
2858 | result = 1; |
2859 | else |
2860 | result = 0; |
2861 | rcu_read_unlock(); |
2862 | } while (read_seqretry(&rename_lock, seq)); |
2863 | |
2864 | return result; |
2865 | } |
2866 | |
2867 | int path_is_under(struct path *path1, struct path *path2) |
2868 | { |
2869 | struct vfsmount *mnt = path1->mnt; |
2870 | struct dentry *dentry = path1->dentry; |
2871 | int res; |
2872 | |
2873 | br_read_lock(vfsmount_lock); |
2874 | if (mnt != path2->mnt) { |
2875 | for (;;) { |
2876 | if (mnt->mnt_parent == mnt) { |
2877 | br_read_unlock(vfsmount_lock); |
2878 | return 0; |
2879 | } |
2880 | if (mnt->mnt_parent == path2->mnt) |
2881 | break; |
2882 | mnt = mnt->mnt_parent; |
2883 | } |
2884 | dentry = mnt->mnt_mountpoint; |
2885 | } |
2886 | res = is_subdir(dentry, path2->dentry); |
2887 | br_read_unlock(vfsmount_lock); |
2888 | return res; |
2889 | } |
2890 | EXPORT_SYMBOL(path_is_under); |
2891 | |
2892 | void d_genocide(struct dentry *root) |
2893 | { |
2894 | struct dentry *this_parent; |
2895 | struct list_head *next; |
2896 | unsigned seq; |
2897 | int locked = 0; |
2898 | |
2899 | seq = read_seqbegin(&rename_lock); |
2900 | again: |
2901 | this_parent = root; |
2902 | spin_lock(&this_parent->d_lock); |
2903 | repeat: |
2904 | next = this_parent->d_subdirs.next; |
2905 | resume: |
2906 | while (next != &this_parent->d_subdirs) { |
2907 | struct list_head *tmp = next; |
2908 | struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); |
2909 | next = tmp->next; |
2910 | |
2911 | spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); |
2912 | if (d_unhashed(dentry) || !dentry->d_inode) { |
2913 | spin_unlock(&dentry->d_lock); |
2914 | continue; |
2915 | } |
2916 | if (!list_empty(&dentry->d_subdirs)) { |
2917 | spin_unlock(&this_parent->d_lock); |
2918 | spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); |
2919 | this_parent = dentry; |
2920 | spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); |
2921 | goto repeat; |
2922 | } |
2923 | if (!(dentry->d_flags & DCACHE_GENOCIDE)) { |
2924 | dentry->d_flags |= DCACHE_GENOCIDE; |
2925 | dentry->d_count--; |
2926 | } |
2927 | spin_unlock(&dentry->d_lock); |
2928 | } |
2929 | if (this_parent != root) { |
2930 | struct dentry *child = this_parent; |
2931 | if (!(this_parent->d_flags & DCACHE_GENOCIDE)) { |
2932 | this_parent->d_flags |= DCACHE_GENOCIDE; |
2933 | this_parent->d_count--; |
2934 | } |
2935 | this_parent = try_to_ascend(this_parent, locked, seq); |
2936 | if (!this_parent) |
2937 | goto rename_retry; |
2938 | next = child->d_u.d_child.next; |
2939 | goto resume; |
2940 | } |
2941 | spin_unlock(&this_parent->d_lock); |
2942 | if (!locked && read_seqretry(&rename_lock, seq)) |
2943 | goto rename_retry; |
2944 | if (locked) |
2945 | write_sequnlock(&rename_lock); |
2946 | return; |
2947 | |
2948 | rename_retry: |
2949 | locked = 1; |
2950 | write_seqlock(&rename_lock); |
2951 | goto again; |
2952 | } |
2953 | |
2954 | /** |
2955 | * find_inode_number - check for dentry with name |
2956 | * @dir: directory to check |
2957 | * @name: Name to find. |
2958 | * |
2959 | * Check whether a dentry already exists for the given name, |
2960 | * and return the inode number if it has an inode. Otherwise |
2961 | * 0 is returned. |
2962 | * |
2963 | * This routine is used to post-process directory listings for |
2964 | * filesystems using synthetic inode numbers, and is necessary |
2965 | * to keep getcwd() working. |
2966 | */ |
2967 | |
2968 | ino_t find_inode_number(struct dentry *dir, struct qstr *name) |
2969 | { |
2970 | struct dentry * dentry; |
2971 | ino_t ino = 0; |
2972 | |
2973 | dentry = d_hash_and_lookup(dir, name); |
2974 | if (dentry) { |
2975 | if (dentry->d_inode) |
2976 | ino = dentry->d_inode->i_ino; |
2977 | dput(dentry); |
2978 | } |
2979 | return ino; |
2980 | } |
2981 | EXPORT_SYMBOL(find_inode_number); |
2982 | |
2983 | static __initdata unsigned long dhash_entries; |
2984 | static int __init set_dhash_entries(char *str) |
2985 | { |
2986 | if (!str) |
2987 | return 0; |
2988 | dhash_entries = simple_strtoul(str, &str, 0); |
2989 | return 1; |
2990 | } |
2991 | __setup("dhash_entries=", set_dhash_entries); |
2992 | |
2993 | static void __init dcache_init_early(void) |
2994 | { |
2995 | int loop; |
2996 | |
2997 | /* If hashes are distributed across NUMA nodes, defer |
2998 | * hash allocation until vmalloc space is available. |
2999 | */ |
3000 | if (hashdist) |
3001 | return; |
3002 | |
3003 | dentry_hashtable = |
3004 | alloc_large_system_hash("Dentry cache", |
3005 | sizeof(struct hlist_bl_head), |
3006 | dhash_entries, |
3007 | 13, |
3008 | HASH_EARLY, |
3009 | &d_hash_shift, |
3010 | &d_hash_mask, |
3011 | 0); |
3012 | |
3013 | for (loop = 0; loop < (1 << d_hash_shift); loop++) |
3014 | INIT_HLIST_BL_HEAD(dentry_hashtable + loop); |
3015 | } |
3016 | |
3017 | static void __init dcache_init(void) |
3018 | { |
3019 | int loop; |
3020 | |
3021 | /* |
3022 | * A constructor could be added for stable state like the lists, |
3023 | * but it is probably not worth it because of the cache nature |
3024 | * of the dcache. |
3025 | */ |
3026 | dentry_cache = KMEM_CACHE(dentry, |
3027 | SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|SLAB_MEM_SPREAD); |
3028 | |
3029 | register_shrinker(&dcache_shrinker); |
3030 | |
3031 | /* Hash may have been set up in dcache_init_early */ |
3032 | if (!hashdist) |
3033 | return; |
3034 | |
3035 | dentry_hashtable = |
3036 | alloc_large_system_hash("Dentry cache", |
3037 | sizeof(struct hlist_bl_head), |
3038 | dhash_entries, |
3039 | 13, |
3040 | 0, |
3041 | &d_hash_shift, |
3042 | &d_hash_mask, |
3043 | 0); |
3044 | |
3045 | for (loop = 0; loop < (1 << d_hash_shift); loop++) |
3046 | INIT_HLIST_BL_HEAD(dentry_hashtable + loop); |
3047 | } |
3048 | |
3049 | /* SLAB cache for __getname() consumers */ |
3050 | struct kmem_cache *names_cachep __read_mostly; |
3051 | EXPORT_SYMBOL(names_cachep); |
3052 | |
3053 | EXPORT_SYMBOL(d_genocide); |
3054 | |
3055 | void __init vfs_caches_init_early(void) |
3056 | { |
3057 | dcache_init_early(); |
3058 | inode_init_early(); |
3059 | } |
3060 | |
3061 | void __init vfs_caches_init(unsigned long mempages) |
3062 | { |
3063 | unsigned long reserve; |
3064 | |
3065 | /* Base hash sizes on available memory, with a reserve equal to |
3066 | 150% of current kernel size */ |
3067 | |
3068 | reserve = min((mempages - nr_free_pages()) * 3/2, mempages - 1); |
3069 | mempages -= reserve; |
3070 | |
3071 | names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0, |
3072 | SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); |
3073 | |
3074 | dcache_init(); |
3075 | inode_init(); |
3076 | files_init(mempages); |
3077 | mnt_init(); |
3078 | bdev_cache_init(); |
3079 | chrdev_init(); |
3080 | } |
3081 |
Branches:
ben-wpan
ben-wpan-stefan
javiroman/ks7010
jz-2.6.34
jz-2.6.34-rc5
jz-2.6.34-rc6
jz-2.6.34-rc7
jz-2.6.35
jz-2.6.36
jz-2.6.37
jz-2.6.38
jz-2.6.39
jz-3.0
jz-3.1
jz-3.11
jz-3.12
jz-3.13
jz-3.15
jz-3.16
jz-3.18-dt
jz-3.2
jz-3.3
jz-3.4
jz-3.5
jz-3.6
jz-3.6-rc2-pwm
jz-3.9
jz-3.9-clk
jz-3.9-rc8
jz47xx
jz47xx-2.6.38
master
Tags:
od-2011-09-04
od-2011-09-18
v2.6.34-rc5
v2.6.34-rc6
v2.6.34-rc7
v3.9