dcache.c 59.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/*
 * fs/dcache.c
 *
 * Complete reimplementation
 * (C) 1997 Thomas Schoebel-Theuer,
 * with heavy changes by Linus Torvalds
 */

/*
 * Notes on the allocation strategy:
 *
 * The dcache is a master of the icache - whenever a dcache entry
 * exists, the inode will always exist. "iput()" is done either when
 * the dcache entry is deleted or garbage collected.
 */

#include <linux/syscalls.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/fs.h>
21
#include <linux/fsnotify.h>
L
Linus Torvalds 已提交
22 23 24 25 26 27 28 29 30 31 32 33
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/hash.h>
#include <linux/cache.h>
#include <linux/module.h>
#include <linux/mount.h>
#include <linux/file.h>
#include <asm/uaccess.h>
#include <linux/security.h>
#include <linux/seqlock.h>
#include <linux/swap.h>
#include <linux/bootmem.h>
34
#include <linux/fs_struct.h>
35
#include <linux/hardirq.h>
36
#include "internal.h"
L
Linus Torvalds 已提交
37

38
int sysctl_vfs_cache_pressure __read_mostly = 100;
L
Linus Torvalds 已提交
39 40 41
EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure);

 __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lock);
A
Al Viro 已提交
42
__cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock);
L
Linus Torvalds 已提交
43 44 45

EXPORT_SYMBOL(dcache_lock);

46
static struct kmem_cache *dentry_cache __read_mostly;
L
Linus Torvalds 已提交
47 48 49 50 51 52 53 54 55 56 57 58 59 60

#define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname))

/*
 * This is the single most critical data structure when it comes
 * to the dcache: the hashtable for lookups. Somebody should try
 * to make this good - I've just made it work.
 *
 * This hash-function tries to avoid losing too many bits of hash
 * information, yet avoid using a prime hash-size or similar.
 */
#define D_HASHBITS     d_hash_shift
#define D_HASHMASK     d_hash_mask

61 62 63
static unsigned int d_hash_mask __read_mostly;
static unsigned int d_hash_shift __read_mostly;
static struct hlist_head *dentry_hashtable __read_mostly;
L
Linus Torvalds 已提交
64 65 66 67 68 69

/* Statistics gathering. */
struct dentry_stat_t dentry_stat = {
	.age_limit = 45,
};

70
static void __d_free(struct dentry *dentry)
L
Linus Torvalds 已提交
71
{
72
	WARN_ON(!list_empty(&dentry->d_alias));
L
Linus Torvalds 已提交
73 74 75 76 77
	if (dname_external(dentry))
		kfree(dentry->d_name.name);
	kmem_cache_free(dentry_cache, dentry); 
}

78 79 80 81 82 83
static void d_callback(struct rcu_head *head)
{
	struct dentry * dentry = container_of(head, struct dentry, d_u.d_rcu);
	__d_free(dentry);
}

L
Linus Torvalds 已提交
84 85 86 87 88 89 90 91
/*
 * no dcache_lock, please.  The caller must decrement dentry_stat.nr_dentry
 * inside dcache_lock.
 */
static void d_free(struct dentry *dentry)
{
	if (dentry->d_op && dentry->d_op->d_release)
		dentry->d_op->d_release(dentry);
92
	/* if dentry was never inserted into hash, immediate free is OK */
A
Akinobu Mita 已提交
93
	if (hlist_unhashed(&dentry->d_hash))
94 95 96
		__d_free(dentry);
	else
		call_rcu(&dentry->d_u.d_rcu, d_callback);
L
Linus Torvalds 已提交
97 98 99 100 101 102
}

/*
 * Release the dentry's inode, using the filesystem
 * d_iput() operation if defined.
 */
103
static void dentry_iput(struct dentry * dentry)
104 105
	__releases(dentry->d_lock)
	__releases(dcache_lock)
L
Linus Torvalds 已提交
106 107 108 109 110 111 112
{
	struct inode *inode = dentry->d_inode;
	if (inode) {
		dentry->d_inode = NULL;
		list_del_init(&dentry->d_alias);
		spin_unlock(&dentry->d_lock);
		spin_unlock(&dcache_lock);
113 114
		if (!inode->i_nlink)
			fsnotify_inoderemove(inode);
L
Linus Torvalds 已提交
115 116 117 118 119 120 121 122 123 124
		if (dentry->d_op && dentry->d_op->d_iput)
			dentry->d_op->d_iput(dentry, inode);
		else
			iput(inode);
	} else {
		spin_unlock(&dentry->d_lock);
		spin_unlock(&dcache_lock);
	}
}

125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
/*
 * dentry_lru_(add|add_tail|del|del_init) must be called with dcache_lock held.
 */
static void dentry_lru_add(struct dentry *dentry)
{
	list_add(&dentry->d_lru, &dentry->d_sb->s_dentry_lru);
	dentry->d_sb->s_nr_dentry_unused++;
	dentry_stat.nr_unused++;
}

static void dentry_lru_add_tail(struct dentry *dentry)
{
	list_add_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru);
	dentry->d_sb->s_nr_dentry_unused++;
	dentry_stat.nr_unused++;
}

static void dentry_lru_del(struct dentry *dentry)
{
	if (!list_empty(&dentry->d_lru)) {
		list_del(&dentry->d_lru);
		dentry->d_sb->s_nr_dentry_unused--;
		dentry_stat.nr_unused--;
	}
}

static void dentry_lru_del_init(struct dentry *dentry)
{
	if (likely(!list_empty(&dentry->d_lru))) {
		list_del_init(&dentry->d_lru);
		dentry->d_sb->s_nr_dentry_unused--;
		dentry_stat.nr_unused--;
	}
}

160 161 162 163
/**
 * d_kill - kill dentry and return parent
 * @dentry: dentry to kill
 *
164
 * The dentry must already be unhashed and removed from the LRU.
165 166 167 168
 *
 * If this is the root of the dentry tree, return NULL.
 */
static struct dentry *d_kill(struct dentry *dentry)
169 170
	__releases(dentry->d_lock)
	__releases(dcache_lock)
171 172 173 174 175 176 177
{
	struct dentry *parent;

	list_del(&dentry->d_u.d_child);
	dentry_stat.nr_dentry--;	/* For d_free, below */
	/*drops the locks, at that point nobody can reach this dentry */
	dentry_iput(dentry);
178 179 180 181
	if (IS_ROOT(dentry))
		parent = NULL;
	else
		parent = dentry->d_parent;
182
	d_free(dentry);
183
	return parent;
184 185
}

L
Linus Torvalds 已提交
186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
/* 
 * This is dput
 *
 * This is complicated by the fact that we do not want to put
 * dentries that are no longer on any hash chain on the unused
 * list: we'd much rather just get rid of them immediately.
 *
 * However, that implies that we have to traverse the dentry
 * tree upwards to the parents which might _also_ now be
 * scheduled for deletion (it may have been only waiting for
 * its last child to go away).
 *
 * This tail recursion is done by hand as we don't want to depend
 * on the compiler to always get this right (gcc generally doesn't).
 * Real recursion would eat up our stack space.
 */

/*
 * dput - release a dentry
 * @dentry: dentry to release 
 *
 * Release a dentry. This will drop the usage count and if appropriate
 * call the dentry unlink method as well as removing it from the queues and
 * releasing its resources. If the parent dentries were scheduled for release
 * they too may now get deleted.
 *
 * no dcache lock, please.
 */

void dput(struct dentry *dentry)
{
	if (!dentry)
		return;

repeat:
	if (atomic_read(&dentry->d_count) == 1)
		might_sleep();
	if (!atomic_dec_and_lock(&dentry->d_count, &dcache_lock))
		return;

	spin_lock(&dentry->d_lock);
	if (atomic_read(&dentry->d_count)) {
		spin_unlock(&dentry->d_lock);
		spin_unlock(&dcache_lock);
		return;
	}

	/*
	 * AV: ->d_delete() is _NOT_ allowed to block now.
	 */
	if (dentry->d_op && dentry->d_op->d_delete) {
		if (dentry->d_op->d_delete(dentry))
			goto unhash_it;
	}
	/* Unreachable? Get rid of it */
 	if (d_unhashed(dentry))
		goto kill_it;
  	if (list_empty(&dentry->d_lru)) {
  		dentry->d_flags |= DCACHE_REFERENCED;
245
		dentry_lru_add(dentry);
L
Linus Torvalds 已提交
246 247 248 249 250 251 252
  	}
 	spin_unlock(&dentry->d_lock);
	spin_unlock(&dcache_lock);
	return;

unhash_it:
	__d_drop(dentry);
253
kill_it:
254 255
	/* if dentry was on the d_lru list delete it from there */
	dentry_lru_del(dentry);
256 257 258
	dentry = d_kill(dentry);
	if (dentry)
		goto repeat;
L
Linus Torvalds 已提交
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
}

/**
 * d_invalidate - invalidate a dentry
 * @dentry: dentry to invalidate
 *
 * Try to invalidate the dentry if it turns out to be
 * possible. If there are other dentries that can be
 * reached through this one we can't delete it and we
 * return -EBUSY. On success we return 0.
 *
 * no dcache lock.
 */
 
int d_invalidate(struct dentry * dentry)
{
	/*
	 * If it's already been dropped, return OK.
	 */
	spin_lock(&dcache_lock);
	if (d_unhashed(dentry)) {
		spin_unlock(&dcache_lock);
		return 0;
	}
	/*
	 * Check whether to do a partial shrink_dcache
	 * to get rid of unused child entries.
	 */
	if (!list_empty(&dentry->d_subdirs)) {
		spin_unlock(&dcache_lock);
		shrink_dcache_parent(dentry);
		spin_lock(&dcache_lock);
	}

	/*
	 * Somebody else still using it?
	 *
	 * If it's a directory, we can't drop it
	 * for fear of somebody re-populating it
	 * with children (even though dropping it
	 * would make it unreachable from the root,
	 * we might still populate it if it was a
	 * working directory or similar).
	 */
	spin_lock(&dentry->d_lock);
	if (atomic_read(&dentry->d_count) > 1) {
		if (dentry->d_inode && S_ISDIR(dentry->d_inode->i_mode)) {
			spin_unlock(&dentry->d_lock);
			spin_unlock(&dcache_lock);
			return -EBUSY;
		}
	}

	__d_drop(dentry);
	spin_unlock(&dentry->d_lock);
	spin_unlock(&dcache_lock);
	return 0;
}

/* This should be called _only_ with dcache_lock held */

static inline struct dentry * __dget_locked(struct dentry *dentry)
{
	atomic_inc(&dentry->d_count);
323
	dentry_lru_del_init(dentry);
L
Linus Torvalds 已提交
324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343
	return dentry;
}

struct dentry * dget_locked(struct dentry *dentry)
{
	return __dget_locked(dentry);
}

/**
 * d_find_alias - grab a hashed alias of inode
 * @inode: inode in question
 * @want_discon:  flag, used by d_splice_alias, to request
 *          that only a DISCONNECTED alias be returned.
 *
 * If inode has a hashed alias, or is a directory and has any alias,
 * acquire the reference to alias and return it. Otherwise return NULL.
 * Notice that if inode is a directory there can be only one alias and
 * it can be unhashed only if it has no children, or if it is the root
 * of a filesystem.
 *
344
 * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer
L
Linus Torvalds 已提交
345
 * any other hashed alias over that one unless @want_discon is set,
346
 * in which case only return an IS_ROOT, DCACHE_DISCONNECTED alias.
L
Linus Torvalds 已提交
347 348 349 350 351 352 353 354 355 356 357 358 359 360 361
 */

static struct dentry * __d_find_alias(struct inode *inode, int want_discon)
{
	struct list_head *head, *next, *tmp;
	struct dentry *alias, *discon_alias=NULL;

	head = &inode->i_dentry;
	next = inode->i_dentry.next;
	while (next != head) {
		tmp = next;
		next = tmp->next;
		prefetch(next);
		alias = list_entry(tmp, struct dentry, d_alias);
 		if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) {
362 363
			if (IS_ROOT(alias) &&
			    (alias->d_flags & DCACHE_DISCONNECTED))
L
Linus Torvalds 已提交
364 365 366 367 368 369 370 371 372 373 374 375 376 377
				discon_alias = alias;
			else if (!want_discon) {
				__dget_locked(alias);
				return alias;
			}
		}
	}
	if (discon_alias)
		__dget_locked(discon_alias);
	return discon_alias;
}

struct dentry * d_find_alias(struct inode *inode)
{
D
David Howells 已提交
378 379 380 381 382 383 384
	struct dentry *de = NULL;

	if (!list_empty(&inode->i_dentry)) {
		spin_lock(&dcache_lock);
		de = __d_find_alias(inode, 0);
		spin_unlock(&dcache_lock);
	}
L
Linus Torvalds 已提交
385 386 387 388 389 390 391 392 393
	return de;
}

/*
 *	Try to kill dentries associated with this inode.
 * WARNING: you must own a reference to inode.
 */
void d_prune_aliases(struct inode *inode)
{
394
	struct dentry *dentry;
L
Linus Torvalds 已提交
395 396
restart:
	spin_lock(&dcache_lock);
397
	list_for_each_entry(dentry, &inode->i_dentry, d_alias) {
L
Linus Torvalds 已提交
398 399 400 401 402 403 404 405 406 407 408 409 410 411 412
		spin_lock(&dentry->d_lock);
		if (!atomic_read(&dentry->d_count)) {
			__dget_locked(dentry);
			__d_drop(dentry);
			spin_unlock(&dentry->d_lock);
			spin_unlock(&dcache_lock);
			dput(dentry);
			goto restart;
		}
		spin_unlock(&dentry->d_lock);
	}
	spin_unlock(&dcache_lock);
}

/*
413 414 415
 * Throw away a dentry - free the inode, dput the parent.  This requires that
 * the LRU list has already been removed.
 *
416 417 418
 * Try to prune ancestors as well.  This is necessary to prevent
 * quadratic behavior of shrink_dcache_parent(), but is also expected
 * to be beneficial in reducing dentry cache fragmentation.
L
Linus Torvalds 已提交
419
 */
420
static void prune_one_dentry(struct dentry * dentry)
421 422 423
	__releases(dentry->d_lock)
	__releases(dcache_lock)
	__acquires(dcache_lock)
L
Linus Torvalds 已提交
424 425
{
	__d_drop(dentry);
426 427 428 429 430 431
	dentry = d_kill(dentry);

	/*
	 * Prune ancestors.  Locking is simpler than in dput(),
	 * because dcache_lock needs to be taken anyway.
	 */
L
Linus Torvalds 已提交
432
	spin_lock(&dcache_lock);
433 434 435 436 437 438
	while (dentry) {
		if (!atomic_dec_and_lock(&dentry->d_count, &dentry->d_lock))
			return;

		if (dentry->d_op && dentry->d_op->d_delete)
			dentry->d_op->d_delete(dentry);
439
		dentry_lru_del_init(dentry);
440 441 442 443
		__d_drop(dentry);
		dentry = d_kill(dentry);
		spin_lock(&dcache_lock);
	}
L
Linus Torvalds 已提交
444 445
}

446 447 448 449 450 451 452
/*
 * Shrink the dentry LRU on a given superblock.
 * @sb   : superblock to shrink dentry LRU.
 * @count: If count is NULL, we prune all dentries on superblock.
 * @flags: If flags is non-zero, we need to do special processing based on
 * which flags are set. This means we don't need to maintain multiple
 * similar copies of this loop.
L
Linus Torvalds 已提交
453
 */
454
static void __shrink_dcache_sb(struct super_block *sb, int *count, int flags)
L
Linus Torvalds 已提交
455
{
456 457 458 459
	LIST_HEAD(referenced);
	LIST_HEAD(tmp);
	struct dentry *dentry;
	int cnt = 0;
L
Linus Torvalds 已提交
460

461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480
	BUG_ON(!sb);
	BUG_ON((flags & DCACHE_REFERENCED) && count == NULL);
	spin_lock(&dcache_lock);
	if (count != NULL)
		/* called from prune_dcache() and shrink_dcache_parent() */
		cnt = *count;
restart:
	if (count == NULL)
		list_splice_init(&sb->s_dentry_lru, &tmp);
	else {
		while (!list_empty(&sb->s_dentry_lru)) {
			dentry = list_entry(sb->s_dentry_lru.prev,
					struct dentry, d_lru);
			BUG_ON(dentry->d_sb != sb);

			spin_lock(&dentry->d_lock);
			/*
			 * If we are honouring the DCACHE_REFERENCED flag and
			 * the dentry has this flag set, don't free it. Clear
			 * the flag and put it back on the LRU.
481
			 */
482 483 484
			if ((flags & DCACHE_REFERENCED)
				&& (dentry->d_flags & DCACHE_REFERENCED)) {
				dentry->d_flags &= ~DCACHE_REFERENCED;
N
npiggin@suse.de 已提交
485
				list_move(&dentry->d_lru, &referenced);
486 487 488 489 490 491 492
				spin_unlock(&dentry->d_lock);
			} else {
				list_move_tail(&dentry->d_lru, &tmp);
				spin_unlock(&dentry->d_lock);
				cnt--;
				if (!cnt)
					break;
493
			}
494
			cond_resched_lock(&dcache_lock);
495
		}
496 497 498 499 500
	}
	while (!list_empty(&tmp)) {
		dentry = list_entry(tmp.prev, struct dentry, d_lru);
		dentry_lru_del_init(dentry);
		spin_lock(&dentry->d_lock);
L
Linus Torvalds 已提交
501 502
		/*
		 * We found an inuse dentry which was not removed from
503 504
		 * the LRU because of laziness during lookup.  Do not free
		 * it - just keep it off the LRU list.
L
Linus Torvalds 已提交
505
		 */
506 507
		if (atomic_read(&dentry->d_count)) {
			spin_unlock(&dentry->d_lock);
L
Linus Torvalds 已提交
508 509
			continue;
		}
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550
		prune_one_dentry(dentry);
		/* dentry->d_lock was dropped in prune_one_dentry() */
		cond_resched_lock(&dcache_lock);
	}
	if (count == NULL && !list_empty(&sb->s_dentry_lru))
		goto restart;
	if (count != NULL)
		*count = cnt;
	if (!list_empty(&referenced))
		list_splice(&referenced, &sb->s_dentry_lru);
	spin_unlock(&dcache_lock);
}

/**
 * prune_dcache - shrink the dcache
 * @count: number of entries to try to free
 *
 * Shrink the dcache. This is done when we need more memory, or simply when we
 * need to unmount something (at which point we need to unuse all dentries).
 *
 * This function may fail to free any resources if all the dentries are in use.
 */
static void prune_dcache(int count)
{
	struct super_block *sb;
	int w_count;
	int unused = dentry_stat.nr_unused;
	int prune_ratio;
	int pruned;

	if (unused == 0 || count == 0)
		return;
	spin_lock(&dcache_lock);
restart:
	if (count >= unused)
		prune_ratio = 1;
	else
		prune_ratio = unused / count;
	spin_lock(&sb_lock);
	list_for_each_entry(sb, &super_blocks, s_list) {
		if (sb->s_nr_dentry_unused == 0)
L
Linus Torvalds 已提交
551
			continue;
552 553 554 555 556 557 558 559 560
		sb->s_count++;
		/* Now, we reclaim unused dentrins with fairness.
		 * We reclaim them same percentage from each superblock.
		 * We calculate number of dentries to scan on this sb
		 * as follows, but the implementation is arranged to avoid
		 * overflows:
		 * number of dentries to scan on this sb =
		 * count * (number of dentries on this sb /
		 * number of dentries in the machine)
561
		 */
562 563 564 565 566 567
		spin_unlock(&sb_lock);
		if (prune_ratio != 1)
			w_count = (sb->s_nr_dentry_unused / prune_ratio) + 1;
		else
			w_count = sb->s_nr_dentry_unused;
		pruned = w_count;
568
		/*
569 570 571 572 573
		 * We need to be sure this filesystem isn't being unmounted,
		 * otherwise we could race with generic_shutdown_super(), and
		 * end up holding a reference to an inode while the filesystem
		 * is unmounted.  So we try to get s_umount, and make sure
		 * s_root isn't NULL.
574
		 */
575 576 577 578 579 580 581 582
		if (down_read_trylock(&sb->s_umount)) {
			if ((sb->s_root != NULL) &&
			    (!list_empty(&sb->s_dentry_lru))) {
				spin_unlock(&dcache_lock);
				__shrink_dcache_sb(sb, &w_count,
						DCACHE_REFERENCED);
				pruned -= w_count;
				spin_lock(&dcache_lock);
583
			}
584
			up_read(&sb->s_umount);
585
		}
586 587
		spin_lock(&sb_lock);
		count -= pruned;
588
		/*
589 590
		 * restart only when sb is no longer on the list and
		 * we have more work to do.
591
		 */
592 593 594 595
		if (__put_super_and_need_restart(sb) && count > 0) {
			spin_unlock(&sb_lock);
			goto restart;
		}
L
Linus Torvalds 已提交
596
	}
597
	spin_unlock(&sb_lock);
L
Linus Torvalds 已提交
598 599 600 601 602 603 604 605 606 607 608 609 610
	spin_unlock(&dcache_lock);
}

/**
 * shrink_dcache_sb - shrink dcache for a superblock
 * @sb: superblock
 *
 * Shrink the dcache for the specified super block. This
 * is used to free the dcache before unmounting a file
 * system
 */
void shrink_dcache_sb(struct super_block * sb)
{
611
	__shrink_dcache_sb(sb, NULL, 0);
L
Linus Torvalds 已提交
612 613
}

614 615 616 617 618 619 620 621
/*
 * destroy a single subtree of dentries for unmount
 * - see the comments on shrink_dcache_for_umount() for a description of the
 *   locking
 */
static void shrink_dcache_for_umount_subtree(struct dentry *dentry)
{
	struct dentry *parent;
622
	unsigned detached = 0;
623 624 625 626 627

	BUG_ON(!IS_ROOT(dentry));

	/* detach this root from the system */
	spin_lock(&dcache_lock);
628
	dentry_lru_del_init(dentry);
629 630 631 632 633 634 635 636 637 638 639 640 641
	__d_drop(dentry);
	spin_unlock(&dcache_lock);

	for (;;) {
		/* descend to the first leaf in the current subtree */
		while (!list_empty(&dentry->d_subdirs)) {
			struct dentry *loop;

			/* this is a branch with children - detach all of them
			 * from the system in one go */
			spin_lock(&dcache_lock);
			list_for_each_entry(loop, &dentry->d_subdirs,
					    d_u.d_child) {
642
				dentry_lru_del_init(loop);
643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
				__d_drop(loop);
				cond_resched_lock(&dcache_lock);
			}
			spin_unlock(&dcache_lock);

			/* move to the first child */
			dentry = list_entry(dentry->d_subdirs.next,
					    struct dentry, d_u.d_child);
		}

		/* consume the dentries from this leaf up through its parents
		 * until we find one with children or run out altogether */
		do {
			struct inode *inode;

			if (atomic_read(&dentry->d_count) != 0) {
				printk(KERN_ERR
				       "BUG: Dentry %p{i=%lx,n=%s}"
				       " still in use (%d)"
				       " [unmount of %s %s]\n",
				       dentry,
				       dentry->d_inode ?
				       dentry->d_inode->i_ino : 0UL,
				       dentry->d_name.name,
				       atomic_read(&dentry->d_count),
				       dentry->d_sb->s_type->name,
				       dentry->d_sb->s_id);
				BUG();
			}

673
			if (IS_ROOT(dentry))
674
				parent = NULL;
675 676
			else {
				parent = dentry->d_parent;
677
				atomic_dec(&parent->d_count);
678
			}
679 680

			list_del(&dentry->d_u.d_child);
681
			detached++;
682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698

			inode = dentry->d_inode;
			if (inode) {
				dentry->d_inode = NULL;
				list_del_init(&dentry->d_alias);
				if (dentry->d_op && dentry->d_op->d_iput)
					dentry->d_op->d_iput(dentry, inode);
				else
					iput(inode);
			}

			d_free(dentry);

			/* finished when we fall off the top of the tree,
			 * otherwise we ascend to the parent and move to the
			 * next sibling if there is one */
			if (!parent)
699
				goto out;
700 701 702 703 704 705 706 707

			dentry = parent;

		} while (list_empty(&dentry->d_subdirs));

		dentry = list_entry(dentry->d_subdirs.next,
				    struct dentry, d_u.d_child);
	}
708 709 710 711 712
out:
	/* several dentries were freed, need to correct nr_dentry */
	spin_lock(&dcache_lock);
	dentry_stat.nr_dentry -= detached;
	spin_unlock(&dcache_lock);
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743
}

/*
 * destroy the dentries attached to a superblock on unmounting
 * - we don't need to use dentry->d_lock, and only need dcache_lock when
 *   removing the dentry from the system lists and hashes because:
 *   - the superblock is detached from all mountings and open files, so the
 *     dentry trees will not be rearranged by the VFS
 *   - s_umount is write-locked, so the memory pressure shrinker will ignore
 *     any dentries belonging to this superblock that it comes across
 *   - the filesystem itself is no longer permitted to rearrange the dentries
 *     in this superblock
 */
void shrink_dcache_for_umount(struct super_block *sb)
{
	struct dentry *dentry;

	if (down_read_trylock(&sb->s_umount))
		BUG();

	dentry = sb->s_root;
	sb->s_root = NULL;
	atomic_dec(&dentry->d_count);
	shrink_dcache_for_umount_subtree(dentry);

	while (!hlist_empty(&sb->s_anon)) {
		dentry = hlist_entry(sb->s_anon.first, struct dentry, d_hash);
		shrink_dcache_for_umount_subtree(dentry);
	}
}

L
Linus Torvalds 已提交
744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770
/*
 * Search for at least 1 mount point in the dentry's subdirs.
 * We descend to the next level whenever the d_subdirs
 * list is non-empty and continue searching.
 */
 
/**
 * have_submounts - check for mounts over a dentry
 * @parent: dentry to check.
 *
 * Return true if the parent or its subdirectories contain
 * a mount point
 */
 
int have_submounts(struct dentry *parent)
{
	struct dentry *this_parent = parent;
	struct list_head *next;

	spin_lock(&dcache_lock);
	if (d_mountpoint(parent))
		goto positive;
repeat:
	next = this_parent->d_subdirs.next;
resume:
	while (next != &this_parent->d_subdirs) {
		struct list_head *tmp = next;
E
Eric Dumazet 已提交
771
		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
L
Linus Torvalds 已提交
772 773 774 775 776 777 778 779 780 781 782 783 784
		next = tmp->next;
		/* Have we found a mount point ? */
		if (d_mountpoint(dentry))
			goto positive;
		if (!list_empty(&dentry->d_subdirs)) {
			this_parent = dentry;
			goto repeat;
		}
	}
	/*
	 * All done at this level ... ascend and resume the search.
	 */
	if (this_parent != parent) {
E
Eric Dumazet 已提交
785
		next = this_parent->d_u.d_child.next;
L
Linus Torvalds 已提交
786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821
		this_parent = this_parent->d_parent;
		goto resume;
	}
	spin_unlock(&dcache_lock);
	return 0; /* No mount points found in tree */
positive:
	spin_unlock(&dcache_lock);
	return 1;
}

/*
 * Search the dentry child list for the specified parent,
 * and move any unused dentries to the end of the unused
 * list for prune_dcache(). We descend to the next level
 * whenever the d_subdirs list is non-empty and continue
 * searching.
 *
 * It returns zero iff there are no unused children,
 * otherwise  it returns the number of children moved to
 * the end of the unused list. This may not be the total
 * number of unused children, because select_parent can
 * drop the lock and return early due to latency
 * constraints.
 */
static int select_parent(struct dentry * parent)
{
	struct dentry *this_parent = parent;
	struct list_head *next;
	int found = 0;

	spin_lock(&dcache_lock);
repeat:
	next = this_parent->d_subdirs.next;
resume:
	while (next != &this_parent->d_subdirs) {
		struct list_head *tmp = next;
E
Eric Dumazet 已提交
822
		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
L
Linus Torvalds 已提交
823 824
		next = tmp->next;

825
		dentry_lru_del_init(dentry);
L
Linus Torvalds 已提交
826 827 828 829 830
		/* 
		 * move only zero ref count dentries to the end 
		 * of the unused list for prune_dcache
		 */
		if (!atomic_read(&dentry->d_count)) {
831
			dentry_lru_add_tail(dentry);
L
Linus Torvalds 已提交
832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854
			found++;
		}

		/*
		 * We can return to the caller if we have found some (this
		 * ensures forward progress). We'll be coming back to find
		 * the rest.
		 */
		if (found && need_resched())
			goto out;

		/*
		 * Descend a level if the d_subdirs list is non-empty.
		 */
		if (!list_empty(&dentry->d_subdirs)) {
			this_parent = dentry;
			goto repeat;
		}
	}
	/*
	 * All done at this level ... ascend and resume the search.
	 */
	if (this_parent != parent) {
E
Eric Dumazet 已提交
855
		next = this_parent->d_u.d_child.next;
L
Linus Torvalds 已提交
856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872
		this_parent = this_parent->d_parent;
		goto resume;
	}
out:
	spin_unlock(&dcache_lock);
	return found;
}

/**
 * shrink_dcache_parent - prune dcache
 * @parent: parent of entries to prune
 *
 * Prune the dcache to remove unused children of the parent dentry.
 */
 
void shrink_dcache_parent(struct dentry * parent)
{
873
	struct super_block *sb = parent->d_sb;
L
Linus Torvalds 已提交
874 875 876
	int found;

	while ((found = select_parent(parent)) != 0)
877
		__shrink_dcache_sb(sb, &found, 0);
L
Linus Torvalds 已提交
878 879 880 881 882 883 884 885 886 887 888 889 890 891
}

/*
 * Scan `nr' dentries and return the number which remain.
 *
 * We need to avoid reentering the filesystem if the caller is performing a
 * GFP_NOFS allocation attempt.  One example deadlock is:
 *
 * ext2_new_block->getblk->GFP->shrink_dcache_memory->prune_dcache->
 * prune_one_dentry->dput->dentry_iput->iput->inode->i_sb->s_op->put_inode->
 * ext2_discard_prealloc->ext2_free_blocks->lock_super->DEADLOCK.
 *
 * In this case we return -1 to tell the caller that we baled.
 */
A
Al Viro 已提交
892
static int shrink_dcache_memory(int nr, gfp_t gfp_mask)
L
Linus Torvalds 已提交
893 894 895 896
{
	if (nr) {
		if (!(gfp_mask & __GFP_FS))
			return -1;
897
		prune_dcache(nr);
L
Linus Torvalds 已提交
898 899 900 901
	}
	return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
}

902 903 904 905 906
static struct shrinker dcache_shrinker = {
	.shrink = shrink_dcache_memory,
	.seeks = DEFAULT_SEEKS,
};

L
Linus Torvalds 已提交
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921
/**
 * d_alloc	-	allocate a dcache entry
 * @parent: parent of entry to allocate
 * @name: qstr of the name
 *
 * Allocates a dentry. It returns %NULL if there is insufficient memory
 * available. On a success the dentry is returned. The name passed in is
 * copied and the copy passed in may be reused after this call.
 */
 
struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
{
	struct dentry *dentry;
	char *dname;

922
	dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL);
L
Linus Torvalds 已提交
923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959
	if (!dentry)
		return NULL;

	if (name->len > DNAME_INLINE_LEN-1) {
		dname = kmalloc(name->len + 1, GFP_KERNEL);
		if (!dname) {
			kmem_cache_free(dentry_cache, dentry); 
			return NULL;
		}
	} else  {
		dname = dentry->d_iname;
	}	
	dentry->d_name.name = dname;

	dentry->d_name.len = name->len;
	dentry->d_name.hash = name->hash;
	memcpy(dname, name->name, name->len);
	dname[name->len] = 0;

	atomic_set(&dentry->d_count, 1);
	dentry->d_flags = DCACHE_UNHASHED;
	spin_lock_init(&dentry->d_lock);
	dentry->d_inode = NULL;
	dentry->d_parent = NULL;
	dentry->d_sb = NULL;
	dentry->d_op = NULL;
	dentry->d_fsdata = NULL;
	dentry->d_mounted = 0;
	INIT_HLIST_NODE(&dentry->d_hash);
	INIT_LIST_HEAD(&dentry->d_lru);
	INIT_LIST_HEAD(&dentry->d_subdirs);
	INIT_LIST_HEAD(&dentry->d_alias);

	if (parent) {
		dentry->d_parent = dget(parent);
		dentry->d_sb = parent->d_sb;
	} else {
E
Eric Dumazet 已提交
960
		INIT_LIST_HEAD(&dentry->d_u.d_child);
L
Linus Torvalds 已提交
961 962 963 964
	}

	spin_lock(&dcache_lock);
	if (parent)
E
Eric Dumazet 已提交
965
		list_add(&dentry->d_u.d_child, &parent->d_subdirs);
L
Linus Torvalds 已提交
966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981
	dentry_stat.nr_dentry++;
	spin_unlock(&dcache_lock);

	return dentry;
}

struct dentry *d_alloc_name(struct dentry *parent, const char *name)
{
	struct qstr q;

	q.name = name;
	q.len = strlen(name);
	q.hash = full_name_hash(q.name, q.len);
	return d_alloc(parent, &q);
}

982 983 984 985 986 987 988 989 990
/* the caller must hold dcache_lock */
static void __d_instantiate(struct dentry *dentry, struct inode *inode)
{
	if (inode)
		list_add(&dentry->d_alias, &inode->i_dentry);
	dentry->d_inode = inode;
	fsnotify_d_instantiate(dentry, inode);
}

L
Linus Torvalds 已提交
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
/**
 * d_instantiate - fill in inode information for a dentry
 * @entry: dentry to complete
 * @inode: inode to attach to this dentry
 *
 * Fill in inode information in the entry.
 *
 * This turns negative dentries into productive full members
 * of society.
 *
 * NOTE! This assumes that the inode count has been incremented
 * (or otherwise set) by the caller to indicate that it is now
 * in use by the dcache.
 */
 
void d_instantiate(struct dentry *entry, struct inode * inode)
{
1008
	BUG_ON(!list_empty(&entry->d_alias));
L
Linus Torvalds 已提交
1009
	spin_lock(&dcache_lock);
1010
	__d_instantiate(entry, inode);
L
Linus Torvalds 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
	spin_unlock(&dcache_lock);
	security_d_instantiate(entry, inode);
}

/**
 * d_instantiate_unique - instantiate a non-aliased dentry
 * @entry: dentry to instantiate
 * @inode: inode to attach to this dentry
 *
 * Fill in inode information in the entry. On success, it returns NULL.
 * If an unhashed alias of "entry" already exists, then we return the
1022
 * aliased dentry instead and drop one reference to inode.
L
Linus Torvalds 已提交
1023 1024 1025
 *
 * Note that in order to avoid conflicts with rename() etc, the caller
 * had better be holding the parent directory semaphore.
1026 1027 1028 1029
 *
 * This also assumes that the inode count has been incremented
 * (or otherwise set) by the caller to indicate that it is now
 * in use by the dcache.
L
Linus Torvalds 已提交
1030
 */
1031 1032
static struct dentry *__d_instantiate_unique(struct dentry *entry,
					     struct inode *inode)
L
Linus Torvalds 已提交
1033 1034 1035 1036 1037 1038
{
	struct dentry *alias;
	int len = entry->d_name.len;
	const char *name = entry->d_name.name;
	unsigned int hash = entry->d_name.hash;

1039
	if (!inode) {
1040
		__d_instantiate(entry, NULL);
1041 1042 1043
		return NULL;
	}

L
Linus Torvalds 已提交
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
	list_for_each_entry(alias, &inode->i_dentry, d_alias) {
		struct qstr *qstr = &alias->d_name;

		if (qstr->hash != hash)
			continue;
		if (alias->d_parent != entry->d_parent)
			continue;
		if (qstr->len != len)
			continue;
		if (memcmp(qstr->name, name, len))
			continue;
		dget_locked(alias);
		return alias;
	}
1058

1059
	__d_instantiate(entry, inode);
L
Linus Torvalds 已提交
1060 1061
	return NULL;
}
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082

struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
{
	struct dentry *result;

	BUG_ON(!list_empty(&entry->d_alias));

	spin_lock(&dcache_lock);
	result = __d_instantiate_unique(entry, inode);
	spin_unlock(&dcache_lock);

	if (!result) {
		security_d_instantiate(entry, inode);
		return NULL;
	}

	BUG_ON(!d_unhashed(result));
	iput(inode);
	return result;
}

L
Linus Torvalds 已提交
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118
EXPORT_SYMBOL(d_instantiate_unique);

/**
 * d_alloc_root - allocate root dentry
 * @root_inode: inode to allocate the root for
 *
 * Allocate a root ("/") dentry for the inode given. The inode is
 * instantiated and returned. %NULL is returned if there is insufficient
 * memory or the inode passed is %NULL.
 */
 
struct dentry * d_alloc_root(struct inode * root_inode)
{
	struct dentry *res = NULL;

	if (root_inode) {
		static const struct qstr name = { .name = "/", .len = 1 };

		res = d_alloc(NULL, &name);
		if (res) {
			res->d_sb = root_inode->i_sb;
			res->d_parent = res;
			d_instantiate(res, root_inode);
		}
	}
	return res;
}

static inline struct hlist_head *d_hash(struct dentry *parent,
					unsigned long hash)
{
	hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES;
	hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS);
	return dentry_hashtable + (hash & D_HASHMASK);
}

1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
/**
 * d_obtain_alias - find or allocate a dentry for a given inode
 * @inode: inode to allocate the dentry for
 *
 * Obtain a dentry for an inode resulting from NFS filehandle conversion or
 * similar open by handle operations.  The returned dentry may be anonymous,
 * or may have a full name (if the inode was already in the cache).
 *
 * When called on a directory inode, we must ensure that the inode only ever
 * has one dentry.  If a dentry is found, that is returned instead of
 * allocating a new one.
 *
 * On successful return, the reference to the inode has been transferred
1132 1133 1134 1135
 * to the dentry.  In case of an error the reference on the inode is released.
 * To make it easier to use in export operations a %NULL or IS_ERR inode may
 * be passed in and will be the error will be propagate to the return value,
 * with a %NULL @inode replaced by ERR_PTR(-ESTALE).
1136 1137 1138
 */
struct dentry *d_obtain_alias(struct inode *inode)
{
C
Christoph Hellwig 已提交
1139 1140 1141
	static const struct qstr anonstring = { .name = "" };
	struct dentry *tmp;
	struct dentry *res;
1142 1143

	if (!inode)
1144
		return ERR_PTR(-ESTALE);
1145 1146 1147
	if (IS_ERR(inode))
		return ERR_CAST(inode);

C
Christoph Hellwig 已提交
1148 1149 1150 1151 1152 1153 1154 1155
	res = d_find_alias(inode);
	if (res)
		goto out_iput;

	tmp = d_alloc(NULL, &anonstring);
	if (!tmp) {
		res = ERR_PTR(-ENOMEM);
		goto out_iput;
1156
	}
C
Christoph Hellwig 已提交
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182
	tmp->d_parent = tmp; /* make sure dput doesn't croak */

	spin_lock(&dcache_lock);
	res = __d_find_alias(inode, 0);
	if (res) {
		spin_unlock(&dcache_lock);
		dput(tmp);
		goto out_iput;
	}

	/* attach a disconnected dentry */
	spin_lock(&tmp->d_lock);
	tmp->d_sb = inode->i_sb;
	tmp->d_inode = inode;
	tmp->d_flags |= DCACHE_DISCONNECTED;
	tmp->d_flags &= ~DCACHE_UNHASHED;
	list_add(&tmp->d_alias, &inode->i_dentry);
	hlist_add_head(&tmp->d_hash, &inode->i_sb->s_anon);
	spin_unlock(&tmp->d_lock);

	spin_unlock(&dcache_lock);
	return tmp;

 out_iput:
	iput(inode);
	return res;
1183
}
1184
EXPORT_SYMBOL(d_obtain_alias);
L
Linus Torvalds 已提交
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205

/**
 * d_splice_alias - splice a disconnected dentry into the tree if one exists
 * @inode:  the inode which may have a disconnected dentry
 * @dentry: a negative dentry which we want to point to the inode.
 *
 * If inode is a directory and has a 'disconnected' dentry (i.e. IS_ROOT and
 * DCACHE_DISCONNECTED), then d_move that in place of the given dentry
 * and return it, else simply d_add the inode to the dentry and return NULL.
 *
 * This is needed in the lookup routine of any filesystem that is exportable
 * (via knfsd) so that we can build dcache paths to directories effectively.
 *
 * If a dentry was found and moved, then it is returned.  Otherwise NULL
 * is returned.  This matches the expected return value of ->lookup.
 *
 */
struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
{
	struct dentry *new = NULL;

1206
	if (inode && S_ISDIR(inode->i_mode)) {
L
Linus Torvalds 已提交
1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
		spin_lock(&dcache_lock);
		new = __d_find_alias(inode, 1);
		if (new) {
			BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
			spin_unlock(&dcache_lock);
			security_d_instantiate(new, inode);
			d_rehash(dentry);
			d_move(new, dentry);
			iput(inode);
		} else {
1217 1218
			/* already taking dcache_lock, so d_add() by hand */
			__d_instantiate(dentry, inode);
L
Linus Torvalds 已提交
1219 1220 1221 1222 1223 1224 1225 1226 1227
			spin_unlock(&dcache_lock);
			security_d_instantiate(dentry, inode);
			d_rehash(dentry);
		}
	} else
		d_add(dentry, inode);
	return new;
}

1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243
/**
 * d_add_ci - lookup or allocate new dentry with case-exact name
 * @inode:  the inode case-insensitive lookup has found
 * @dentry: the negative dentry that was passed to the parent's lookup func
 * @name:   the case-exact name to be associated with the returned dentry
 *
 * This is to avoid filling the dcache with case-insensitive names to the
 * same inode, only the actual correct case is stored in the dcache for
 * case-insensitive filesystems.
 *
 * For a case-insensitive lookup match and if the the case-exact dentry
 * already exists in in the dcache, use it and return it.
 *
 * If no entry exists with the exact case name, allocate new dentry with
 * the exact case, and return the spliced entry.
 */
1244
struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode,
1245 1246 1247 1248 1249 1250
			struct qstr *name)
{
	int error;
	struct dentry *found;
	struct dentry *new;

C
Christoph Hellwig 已提交
1251 1252 1253 1254
	/*
	 * First check if a dentry matching the name already exists,
	 * if not go ahead and create it now.
	 */
1255 1256 1257 1258 1259 1260 1261
	found = d_hash_and_lookup(dentry->d_parent, name);
	if (!found) {
		new = d_alloc(dentry->d_parent, name);
		if (!new) {
			error = -ENOMEM;
			goto err_out;
		}
C
Christoph Hellwig 已提交
1262

1263 1264 1265 1266 1267 1268 1269
		found = d_splice_alias(inode, new);
		if (found) {
			dput(new);
			return found;
		}
		return new;
	}
C
Christoph Hellwig 已提交
1270 1271 1272 1273 1274 1275 1276

	/*
	 * If a matching dentry exists, and it's not negative use it.
	 *
	 * Decrement the reference count to balance the iget() done
	 * earlier on.
	 */
1277 1278 1279 1280 1281 1282 1283 1284 1285
	if (found->d_inode) {
		if (unlikely(found->d_inode != inode)) {
			/* This can't happen because bad inodes are unhashed. */
			BUG_ON(!is_bad_inode(inode));
			BUG_ON(!is_bad_inode(found->d_inode));
		}
		iput(inode);
		return found;
	}
C
Christoph Hellwig 已提交
1286

1287 1288
	/*
	 * Negative dentry: instantiate it unless the inode is a directory and
C
Christoph Hellwig 已提交
1289
	 * already has a dentry.
1290 1291
	 */
	spin_lock(&dcache_lock);
C
Christoph Hellwig 已提交
1292
	if (!S_ISDIR(inode->i_mode) || list_empty(&inode->i_dentry)) {
1293
		__d_instantiate(found, inode);
1294 1295 1296 1297
		spin_unlock(&dcache_lock);
		security_d_instantiate(found, inode);
		return found;
	}
C
Christoph Hellwig 已提交
1298

1299
	/*
C
Christoph Hellwig 已提交
1300 1301
	 * In case a directory already has a (disconnected) entry grab a
	 * reference to it, move it in place and use it.
1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
	 */
	new = list_entry(inode->i_dentry.next, struct dentry, d_alias);
	dget_locked(new);
	spin_unlock(&dcache_lock);
	security_d_instantiate(found, inode);
	d_move(new, found);
	iput(inode);
	dput(found);
	return new;

err_out:
	iput(inode);
	return ERR_PTR(error);
}
L
Linus Torvalds 已提交
1316 1317 1318 1319 1320 1321 1322 1323

/**
 * d_lookup - search for a dentry
 * @parent: parent dentry
 * @name: qstr of name we wish to find
 *
 * Searches the children of the parent dentry for the name in question. If
 * the dentry is found its reference count is incremented and the dentry
1324
 * is returned. The caller must use dput to free the entry when it has
L
Linus Torvalds 已提交
1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336
 * finished using it. %NULL is returned on failure.
 *
 * __d_lookup is dcache_lock free. The hash list is protected using RCU.
 * Memory barriers are used while updating and doing lockless traversal. 
 * To avoid races with d_move while rename is happening, d_lock is used.
 *
 * Overflows in memcmp(), while d_move, are avoided by keeping the length
 * and name pointer in one structure pointed by d_qstr.
 *
 * rcu_read_lock() and rcu_read_unlock() are used to disable preemption while
 * lookup is going on.
 *
1337
 * The dentry unused LRU is not updated even if lookup finds the required dentry
L
Linus Torvalds 已提交
1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367
 * in there. It is updated in places such as prune_dcache, shrink_dcache_sb,
 * select_parent and __dget_locked. This laziness saves lookup from dcache_lock
 * acquisition.
 *
 * d_lookup() is protected against the concurrent renames in some unrelated
 * directory using the seqlockt_t rename_lock.
 */

struct dentry * d_lookup(struct dentry * parent, struct qstr * name)
{
	struct dentry * dentry = NULL;
	unsigned long seq;

        do {
                seq = read_seqbegin(&rename_lock);
                dentry = __d_lookup(parent, name);
                if (dentry)
			break;
	} while (read_seqretry(&rename_lock, seq));
	return dentry;
}

struct dentry * __d_lookup(struct dentry * parent, struct qstr * name)
{
	unsigned int len = name->len;
	unsigned int hash = name->hash;
	const unsigned char *str = name->name;
	struct hlist_head *head = d_hash(parent,hash);
	struct dentry *found = NULL;
	struct hlist_node *node;
1368
	struct dentry *dentry;
L
Linus Torvalds 已提交
1369 1370 1371

	rcu_read_lock();
	
1372
	hlist_for_each_entry_rcu(dentry, node, head, d_hash) {
L
Linus Torvalds 已提交
1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389
		struct qstr *qstr;

		if (dentry->d_name.hash != hash)
			continue;
		if (dentry->d_parent != parent)
			continue;

		spin_lock(&dentry->d_lock);

		/*
		 * Recheck the dentry after taking the lock - d_move may have
		 * changed things.  Don't bother checking the hash because we're
		 * about to compare the whole name anyway.
		 */
		if (dentry->d_parent != parent)
			goto next;

1390 1391 1392 1393
		/* non-existing due to RCU? */
		if (d_unhashed(dentry))
			goto next;

L
Linus Torvalds 已提交
1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408
		/*
		 * It is safe to compare names since d_move() cannot
		 * change the qstr (protected by d_lock).
		 */
		qstr = &dentry->d_name;
		if (parent->d_op && parent->d_op->d_compare) {
			if (parent->d_op->d_compare(parent, qstr, name))
				goto next;
		} else {
			if (qstr->len != len)
				goto next;
			if (memcmp(qstr->name, str, len))
				goto next;
		}

1409 1410
		atomic_inc(&dentry->d_count);
		found = dentry;
L
Linus Torvalds 已提交
1411 1412 1413 1414 1415 1416 1417 1418 1419 1420
		spin_unlock(&dentry->d_lock);
		break;
next:
		spin_unlock(&dentry->d_lock);
 	}
 	rcu_read_unlock();

 	return found;
}

1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446
/**
 * d_hash_and_lookup - hash the qstr then search for a dentry
 * @dir: Directory to search in
 * @name: qstr of name we wish to find
 *
 * On hash failure or on lookup failure NULL is returned.
 */
struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name)
{
	struct dentry *dentry = NULL;

	/*
	 * Check for a fs-specific hash function. Note that we must
	 * calculate the standard hash first, as the d_op->d_hash()
	 * routine may choose to leave the hash value unchanged.
	 */
	name->hash = full_name_hash(name->name, name->len);
	if (dir->d_op && dir->d_op->d_hash) {
		if (dir->d_op->d_hash(dir, name) < 0)
			goto out;
	}
	dentry = d_lookup(dir, name);
out:
	return dentry;
}

L
Linus Torvalds 已提交
1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471
/**
 * d_validate - verify dentry provided from insecure source
 * @dentry: The dentry alleged to be valid child of @dparent
 * @dparent: The parent dentry (known to be valid)
 *
 * An insecure source has sent us a dentry, here we verify it and dget() it.
 * This is used by ncpfs in its readdir implementation.
 * Zero is returned in the dentry is invalid.
 */
 
int d_validate(struct dentry *dentry, struct dentry *dparent)
{
	struct hlist_head *base;
	struct hlist_node *lhp;

	/* Check whether the ptr might be valid at all.. */
	if (!kmem_ptr_validate(dentry_cache, dentry))
		goto out;

	if (dentry->d_parent != dparent)
		goto out;

	spin_lock(&dcache_lock);
	base = d_hash(dparent, dentry->d_name.hash);
	hlist_for_each(lhp,base) { 
1472
		/* hlist_for_each_entry_rcu() not required for d_hash list
L
Linus Torvalds 已提交
1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
		 * as it is parsed under dcache_lock
		 */
		if (dentry == hlist_entry(lhp, struct dentry, d_hash)) {
			__dget_locked(dentry);
			spin_unlock(&dcache_lock);
			return 1;
		}
	}
	spin_unlock(&dcache_lock);
out:
	return 0;
}

/*
 * When a file is deleted, we have two options:
 * - turn this dentry into a negative dentry
 * - unhash this dentry and free it.
 *
 * Usually, we want to just turn this into
 * a negative dentry, but if anybody else is
 * currently using the dentry or the inode
 * we can't do that and we fall back on removing
 * it from the hash queues and waiting for
 * it to be deleted later when it has no users
 */
 
/**
 * d_delete - delete a dentry
 * @dentry: The dentry to delete
 *
 * Turn the dentry into a negative dentry if possible, otherwise
 * remove it from the hash queues so it can be deleted later
 */
 
void d_delete(struct dentry * dentry)
{
1509
	int isdir = 0;
L
Linus Torvalds 已提交
1510 1511 1512 1513 1514
	/*
	 * Are we the only user?
	 */
	spin_lock(&dcache_lock);
	spin_lock(&dentry->d_lock);
1515
	isdir = S_ISDIR(dentry->d_inode->i_mode);
L
Linus Torvalds 已提交
1516 1517
	if (atomic_read(&dentry->d_count) == 1) {
		dentry_iput(dentry);
1518
		fsnotify_nameremove(dentry, isdir);
L
Linus Torvalds 已提交
1519 1520 1521 1522 1523 1524 1525 1526
		return;
	}

	if (!d_unhashed(dentry))
		__d_drop(dentry);

	spin_unlock(&dentry->d_lock);
	spin_unlock(&dcache_lock);
1527 1528

	fsnotify_nameremove(dentry, isdir);
L
Linus Torvalds 已提交
1529 1530 1531 1532 1533 1534 1535 1536 1537
}

static void __d_rehash(struct dentry * entry, struct hlist_head *list)
{

 	entry->d_flags &= ~DCACHE_UNHASHED;
 	hlist_add_head_rcu(&entry->d_hash, list);
}

1538 1539 1540 1541 1542
static void _d_rehash(struct dentry * entry)
{
	__d_rehash(entry, d_hash(entry->d_parent, entry->d_name.hash));
}

L
Linus Torvalds 已提交
1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553
/**
 * d_rehash	- add an entry back to the hash
 * @entry: dentry to add to the hash
 *
 * Adds a dentry to the hash according to its name.
 */
 
void d_rehash(struct dentry * entry)
{
	spin_lock(&dcache_lock);
	spin_lock(&entry->d_lock);
1554
	_d_rehash(entry);
L
Linus Torvalds 已提交
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576
	spin_unlock(&entry->d_lock);
	spin_unlock(&dcache_lock);
}

/*
 * When switching names, the actual string doesn't strictly have to
 * be preserved in the target - because we're dropping the target
 * anyway. As such, we can just do a simple memcpy() to copy over
 * the new name before we switch.
 *
 * Note that we have to be a lot more careful about getting the hash
 * switched - we have to switch the hash value properly even if it
 * then no longer matches the actual (corrupted) string of the target.
 * The hash value has to match the hash queue that the dentry is on..
 */
static void switch_names(struct dentry *dentry, struct dentry *target)
{
	if (dname_external(target)) {
		if (dname_external(dentry)) {
			/*
			 * Both external: swap the pointers
			 */
1577
			swap(target->d_name.name, dentry->d_name.name);
L
Linus Torvalds 已提交
1578 1579 1580 1581 1582
		} else {
			/*
			 * dentry:internal, target:external.  Steal target's
			 * storage and make target internal.
			 */
1583 1584
			memcpy(target->d_iname, dentry->d_name.name,
					dentry->d_name.len + 1);
L
Linus Torvalds 已提交
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603
			dentry->d_name.name = target->d_name.name;
			target->d_name.name = target->d_iname;
		}
	} else {
		if (dname_external(dentry)) {
			/*
			 * dentry:external, target:internal.  Give dentry's
			 * storage to target and make dentry internal
			 */
			memcpy(dentry->d_iname, target->d_name.name,
					target->d_name.len + 1);
			target->d_name.name = dentry->d_name.name;
			dentry->d_name.name = dentry->d_iname;
		} else {
			/*
			 * Both are internal.  Just copy target to dentry
			 */
			memcpy(dentry->d_iname, target->d_name.name,
					target->d_name.len + 1);
1604 1605
			dentry->d_name.len = target->d_name.len;
			return;
L
Linus Torvalds 已提交
1606 1607
		}
	}
1608
	swap(dentry->d_name.len, target->d_name.len);
L
Linus Torvalds 已提交
1609 1610 1611 1612 1613 1614 1615 1616 1617 1618
}

/*
 * We cannibalize "target" when moving dentry on top of it,
 * because it's going to be thrown away anyway. We could be more
 * polite about it, though.
 *
 * This forceful removal will result in ugly /proc output if
 * somebody holds a file open that got deleted due to a rename.
 * We could be nicer about the deleted file, and let it show
J
J. Bruce Fields 已提交
1619 1620
 * up under the name it had before it was deleted rather than
 * under the original name of the file that was moved on top of it.
L
Linus Torvalds 已提交
1621 1622
 */
 
1623 1624
/*
 * d_move_locked - move a dentry
L
Linus Torvalds 已提交
1625 1626 1627 1628 1629 1630
 * @dentry: entry to move
 * @target: new dentry
 *
 * Update the dcache to reflect the move of a file name. Negative
 * dcache entries should not be moved in this way.
 */
1631
static void d_move_locked(struct dentry * dentry, struct dentry * target)
L
Linus Torvalds 已提交
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643
{
	struct hlist_head *list;

	if (!dentry->d_inode)
		printk(KERN_WARNING "VFS: moving negative dcache entry\n");

	write_seqlock(&rename_lock);
	/*
	 * XXXX: do we really need to take target->d_lock?
	 */
	if (target < dentry) {
		spin_lock(&target->d_lock);
I
Ingo Molnar 已提交
1644
		spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED);
L
Linus Torvalds 已提交
1645 1646
	} else {
		spin_lock(&dentry->d_lock);
I
Ingo Molnar 已提交
1647
		spin_lock_nested(&target->d_lock, DENTRY_D_LOCK_NESTED);
L
Linus Torvalds 已提交
1648 1649 1650
	}

	/* Move the dentry to the target hash queue, if on different bucket */
1651
	if (d_unhashed(dentry))
L
Linus Torvalds 已提交
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662
		goto already_unhashed;

	hlist_del_rcu(&dentry->d_hash);

already_unhashed:
	list = d_hash(target->d_parent, target->d_name.hash);
	__d_rehash(dentry, list);

	/* Unhash the target: dput() will then get rid of it */
	__d_drop(target);

E
Eric Dumazet 已提交
1663 1664
	list_del(&dentry->d_u.d_child);
	list_del(&target->d_u.d_child);
L
Linus Torvalds 已提交
1665 1666 1667

	/* Switch the names.. */
	switch_names(dentry, target);
1668
	swap(dentry->d_name.hash, target->d_name.hash);
L
Linus Torvalds 已提交
1669 1670 1671 1672 1673

	/* ... and switch the parents */
	if (IS_ROOT(dentry)) {
		dentry->d_parent = target->d_parent;
		target->d_parent = target;
E
Eric Dumazet 已提交
1674
		INIT_LIST_HEAD(&target->d_u.d_child);
L
Linus Torvalds 已提交
1675
	} else {
1676
		swap(dentry->d_parent, target->d_parent);
L
Linus Torvalds 已提交
1677 1678

		/* And add them back to the (new) parent lists */
E
Eric Dumazet 已提交
1679
		list_add(&target->d_u.d_child, &target->d_parent->d_subdirs);
L
Linus Torvalds 已提交
1680 1681
	}

E
Eric Dumazet 已提交
1682
	list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs);
L
Linus Torvalds 已提交
1683
	spin_unlock(&target->d_lock);
1684
	fsnotify_d_move(dentry);
L
Linus Torvalds 已提交
1685 1686
	spin_unlock(&dentry->d_lock);
	write_sequnlock(&rename_lock);
1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701
}

/**
 * d_move - move a dentry
 * @dentry: entry to move
 * @target: new dentry
 *
 * Update the dcache to reflect the move of a file name. Negative
 * dcache entries should not be moved in this way.
 */

void d_move(struct dentry * dentry, struct dentry * target)
{
	spin_lock(&dcache_lock);
	d_move_locked(dentry, target);
L
Linus Torvalds 已提交
1702 1703 1704
	spin_unlock(&dcache_lock);
}

1705 1706 1707 1708 1709 1710 1711
/**
 * d_ancestor - search for an ancestor
 * @p1: ancestor dentry
 * @p2: child dentry
 *
 * Returns the ancestor dentry of p2 which is a child of p1, if p1 is
 * an ancestor of p2, else NULL.
1712
 */
1713
struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2)
1714 1715 1716
{
	struct dentry *p;

1717
	for (p = p2; !IS_ROOT(p); p = p->d_parent) {
1718
		if (p->d_parent == p1)
1719
			return p;
1720
	}
1721
	return NULL;
1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733
}

/*
 * This helper attempts to cope with remotely renamed directories
 *
 * It assumes that the caller is already holding
 * dentry->d_parent->d_inode->i_mutex and the dcache_lock
 *
 * Note: If ever the locking in lock_rename() changes, then please
 * remember to update this too...
 */
static struct dentry *__d_unalias(struct dentry *dentry, struct dentry *alias)
1734
	__releases(dcache_lock)
1735 1736 1737 1738 1739 1740 1741 1742 1743 1744
{
	struct mutex *m1 = NULL, *m2 = NULL;
	struct dentry *ret;

	/* If alias and dentry share a parent, then no extra locks required */
	if (alias->d_parent == dentry->d_parent)
		goto out_unalias;

	/* Check for loops */
	ret = ERR_PTR(-ELOOP);
1745
	if (d_ancestor(alias, dentry))
1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767
		goto out_err;

	/* See lock_rename() */
	ret = ERR_PTR(-EBUSY);
	if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex))
		goto out_err;
	m1 = &dentry->d_sb->s_vfs_rename_mutex;
	if (!mutex_trylock(&alias->d_parent->d_inode->i_mutex))
		goto out_err;
	m2 = &alias->d_parent->d_inode->i_mutex;
out_unalias:
	d_move_locked(alias, dentry);
	ret = alias;
out_err:
	spin_unlock(&dcache_lock);
	if (m2)
		mutex_unlock(m2);
	if (m1)
		mutex_unlock(m1);
	return ret;
}

1768 1769 1770 1771 1772 1773 1774 1775 1776
/*
 * Prepare an anonymous dentry for life in the superblock's dentry tree as a
 * named dentry in place of the dentry to be replaced.
 */
static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon)
{
	struct dentry *dparent, *aparent;

	switch_names(dentry, anon);
1777
	swap(dentry->d_name.hash, anon->d_name.hash);
1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808

	dparent = dentry->d_parent;
	aparent = anon->d_parent;

	dentry->d_parent = (aparent == anon) ? dentry : aparent;
	list_del(&dentry->d_u.d_child);
	if (!IS_ROOT(dentry))
		list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs);
	else
		INIT_LIST_HEAD(&dentry->d_u.d_child);

	anon->d_parent = (dparent == dentry) ? anon : dparent;
	list_del(&anon->d_u.d_child);
	if (!IS_ROOT(anon))
		list_add(&anon->d_u.d_child, &anon->d_parent->d_subdirs);
	else
		INIT_LIST_HEAD(&anon->d_u.d_child);

	anon->d_flags &= ~DCACHE_DISCONNECTED;
}

/**
 * d_materialise_unique - introduce an inode into the tree
 * @dentry: candidate dentry
 * @inode: inode to bind to the dentry, to which aliases may be attached
 *
 * Introduces an dentry into the tree, substituting an extant disconnected
 * root directory alias in its place if there is one
 */
struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
{
1809
	struct dentry *actual;
1810 1811 1812 1813 1814 1815 1816

	BUG_ON(!d_unhashed(dentry));

	spin_lock(&dcache_lock);

	if (!inode) {
		actual = dentry;
1817
		__d_instantiate(dentry, NULL);
1818 1819 1820
		goto found_lock;
	}

1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841
	if (S_ISDIR(inode->i_mode)) {
		struct dentry *alias;

		/* Does an aliased dentry already exist? */
		alias = __d_find_alias(inode, 0);
		if (alias) {
			actual = alias;
			/* Is this an anonymous mountpoint that we could splice
			 * into our tree? */
			if (IS_ROOT(alias)) {
				spin_lock(&alias->d_lock);
				__d_materialise_dentry(dentry, alias);
				__d_drop(alias);
				goto found;
			}
			/* Nope, but we must(!) avoid directory aliasing */
			actual = __d_unalias(dentry, alias);
			if (IS_ERR(actual))
				dput(alias);
			goto out_nolock;
		}
1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856
	}

	/* Add a unique reference */
	actual = __d_instantiate_unique(dentry, inode);
	if (!actual)
		actual = dentry;
	else if (unlikely(!d_unhashed(actual)))
		goto shouldnt_be_hashed;

found_lock:
	spin_lock(&actual->d_lock);
found:
	_d_rehash(actual);
	spin_unlock(&actual->d_lock);
	spin_unlock(&dcache_lock);
1857
out_nolock:
1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870
	if (actual == dentry) {
		security_d_instantiate(dentry, inode);
		return NULL;
	}

	iput(inode);
	return actual;

shouldnt_be_hashed:
	spin_unlock(&dcache_lock);
	BUG();
}

1871
static int prepend(char **buffer, int *buflen, const char *str, int namelen)
1872 1873 1874 1875 1876 1877 1878 1879 1880
{
	*buflen -= namelen;
	if (*buflen < 0)
		return -ENAMETOOLONG;
	*buffer -= namelen;
	memcpy(*buffer, str, namelen);
	return 0;
}

1881 1882 1883 1884 1885
static int prepend_name(char **buffer, int *buflen, struct qstr *name)
{
	return prepend(buffer, buflen, name->name, name->len);
}

L
Linus Torvalds 已提交
1886
/**
1887
 * __d_path - return the path of a dentry
1888 1889
 * @path: the dentry/vfsmount to report
 * @root: root vfsmnt/dentry (may be modified by this function)
L
Linus Torvalds 已提交
1890 1891 1892
 * @buffer: buffer to return value in
 * @buflen: buffer length
 *
1893 1894
 * Convert a dentry into an ASCII path name. If the entry has been deleted
 * the string " (deleted)" is appended. Note that this is ambiguous.
L
Linus Torvalds 已提交
1895
 *
1896 1897
 * Returns a pointer into the buffer or an error code if the
 * path was too long.
1898 1899
 *
 * "buflen" should be positive. Caller holds the dcache_lock.
1900 1901 1902
 *
 * If path is not reachable from the supplied root, then the value of
 * root is changed (without modifying refcounts).
L
Linus Torvalds 已提交
1903
 */
1904 1905
char *__d_path(const struct path *path, struct path *root,
	       char *buffer, int buflen)
L
Linus Torvalds 已提交
1906
{
1907 1908
	struct dentry *dentry = path->dentry;
	struct vfsmount *vfsmnt = path->mnt;
1909 1910
	char *end = buffer + buflen;
	char *retval;
1911

1912
	spin_lock(&vfsmount_lock);
1913
	prepend(&end, &buflen, "\0", 1);
1914
	if (d_unlinked(dentry) &&
1915
		(prepend(&end, &buflen, " (deleted)", 10) != 0))
1916 1917 1918 1919 1920 1921 1922 1923 1924
			goto Elong;

	if (buflen < 1)
		goto Elong;
	/* Get '/' right */
	retval = end-1;
	*retval = '/';

	for (;;) {
L
Linus Torvalds 已提交
1925 1926
		struct dentry * parent;

J
Jan Blunck 已提交
1927
		if (dentry == root->dentry && vfsmnt == root->mnt)
1928
			break;
L
Linus Torvalds 已提交
1929
		if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) {
1930
			/* Global root? */
L
Linus Torvalds 已提交
1931 1932 1933 1934 1935 1936 1937 1938 1939
			if (vfsmnt->mnt_parent == vfsmnt) {
				goto global_root;
			}
			dentry = vfsmnt->mnt_mountpoint;
			vfsmnt = vfsmnt->mnt_parent;
			continue;
		}
		parent = dentry->d_parent;
		prefetch(parent);
1940
		if ((prepend_name(&end, &buflen, &dentry->d_name) != 0) ||
1941
		    (prepend(&end, &buflen, "/", 1) != 0))
1942 1943
			goto Elong;
		retval = end;
L
Linus Torvalds 已提交
1944 1945 1946
		dentry = parent;
	}

1947 1948
out:
	spin_unlock(&vfsmount_lock);
1949
	return retval;
L
Linus Torvalds 已提交
1950 1951

global_root:
1952
	retval += 1;	/* hit the slash */
1953
	if (prepend_name(&retval, &buflen, &dentry->d_name) != 0)
L
Linus Torvalds 已提交
1954
		goto Elong;
1955 1956
	root->mnt = vfsmnt;
	root->dentry = dentry;
1957 1958
	goto out;

L
Linus Torvalds 已提交
1959
Elong:
1960 1961
	retval = ERR_PTR(-ENAMETOOLONG);
	goto out;
L
Linus Torvalds 已提交
1962 1963
}

J
Jan Blunck 已提交
1964 1965
/**
 * d_path - return the path of a dentry
1966
 * @path: path to report
J
Jan Blunck 已提交
1967 1968 1969 1970 1971 1972
 * @buf: buffer to return value in
 * @buflen: buffer length
 *
 * Convert a dentry into an ASCII path name. If the entry has been deleted
 * the string " (deleted)" is appended. Note that this is ambiguous.
 *
1973 1974 1975 1976
 * Returns a pointer into the buffer or an error code if the path was
 * too long. Note: Callers should use the returned pointer, not the passed
 * in buffer, to use the name! The implementation often starts at an offset
 * into the buffer, and may leave 0 bytes at the start.
J
Jan Blunck 已提交
1977
 *
1978
 * "buflen" should be positive.
J
Jan Blunck 已提交
1979
 */
1980
char *d_path(const struct path *path, char *buf, int buflen)
L
Linus Torvalds 已提交
1981 1982
{
	char *res;
J
Jan Blunck 已提交
1983
	struct path root;
1984
	struct path tmp;
L
Linus Torvalds 已提交
1985

1986 1987 1988 1989 1990 1991 1992
	/*
	 * We have various synthetic filesystems that never get mounted.  On
	 * these filesystems dentries are never used for lookup purposes, and
	 * thus don't need to be hashed.  They also don't need a name until a
	 * user wants to identify the object in /proc/pid/fd/.  The little hack
	 * below allows us to generate a name for these objects on demand:
	 */
1993 1994
	if (path->dentry->d_op && path->dentry->d_op->d_dname)
		return path->dentry->d_op->d_dname(path->dentry, buf, buflen);
1995

L
Linus Torvalds 已提交
1996
	read_lock(&current->fs->lock);
J
Jan Blunck 已提交
1997
	root = current->fs->root;
1998
	path_get(&root);
L
Linus Torvalds 已提交
1999
	read_unlock(&current->fs->lock);
2000
	spin_lock(&dcache_lock);
2001 2002
	tmp = root;
	res = __d_path(path, &tmp, buf, buflen);
2003
	spin_unlock(&dcache_lock);
J
Jan Blunck 已提交
2004
	path_put(&root);
L
Linus Torvalds 已提交
2005 2006 2007
	return res;
}

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028
/*
 * Helper function for dentry_operations.d_dname() members
 */
char *dynamic_dname(struct dentry *dentry, char *buffer, int buflen,
			const char *fmt, ...)
{
	va_list args;
	char temp[64];
	int sz;

	va_start(args, fmt);
	sz = vsnprintf(temp, sizeof(temp), fmt, args) + 1;
	va_end(args);

	if (sz > sizeof(temp) || sz > buflen)
		return ERR_PTR(-ENAMETOOLONG);

	buffer += buflen - sz;
	return memcpy(buffer, temp, sz);
}

2029 2030 2031 2032 2033 2034 2035 2036 2037 2038
/*
 * Write full pathname from the root of the filesystem into the buffer.
 */
char *dentry_path(struct dentry *dentry, char *buf, int buflen)
{
	char *end = buf + buflen;
	char *retval;

	spin_lock(&dcache_lock);
	prepend(&end, &buflen, "\0", 1);
2039
	if (d_unlinked(dentry) &&
2040 2041 2042 2043 2044 2045 2046 2047
		(prepend(&end, &buflen, "//deleted", 9) != 0))
			goto Elong;
	if (buflen < 1)
		goto Elong;
	/* Get '/' right */
	retval = end-1;
	*retval = '/';

2048 2049
	while (!IS_ROOT(dentry)) {
		struct dentry *parent = dentry->d_parent;
2050 2051

		prefetch(parent);
2052
		if ((prepend_name(&end, &buflen, &dentry->d_name) != 0) ||
2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065
		    (prepend(&end, &buflen, "/", 1) != 0))
			goto Elong;

		retval = end;
		dentry = parent;
	}
	spin_unlock(&dcache_lock);
	return retval;
Elong:
	spin_unlock(&dcache_lock);
	return ERR_PTR(-ENAMETOOLONG);
}

L
Linus Torvalds 已提交
2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083
/*
 * NOTE! The user-level library version returns a
 * character pointer. The kernel system call just
 * returns the length of the buffer filled (which
 * includes the ending '\0' character), or a negative
 * error value. So libc would do something like
 *
 *	char *getcwd(char * buf, size_t size)
 *	{
 *		int retval;
 *
 *		retval = sys_getcwd(buf, size);
 *		if (retval >= 0)
 *			return buf;
 *		errno = -retval;
 *		return NULL;
 *	}
 */
2084
SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size)
L
Linus Torvalds 已提交
2085
{
2086
	int error;
J
Jan Blunck 已提交
2087
	struct path pwd, root;
2088
	char *page = (char *) __get_free_page(GFP_USER);
L
Linus Torvalds 已提交
2089 2090 2091 2092 2093

	if (!page)
		return -ENOMEM;

	read_lock(&current->fs->lock);
J
Jan Blunck 已提交
2094
	pwd = current->fs->pwd;
2095
	path_get(&pwd);
J
Jan Blunck 已提交
2096
	root = current->fs->root;
2097
	path_get(&root);
L
Linus Torvalds 已提交
2098 2099
	read_unlock(&current->fs->lock);

2100 2101
	error = -ENOENT;
	spin_lock(&dcache_lock);
2102
	if (!d_unlinked(pwd.dentry)) {
2103
		unsigned long len;
2104
		struct path tmp = root;
2105
		char * cwd;
L
Linus Torvalds 已提交
2106

2107
		cwd = __d_path(&pwd, &tmp, page, PAGE_SIZE);
2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122
		spin_unlock(&dcache_lock);

		error = PTR_ERR(cwd);
		if (IS_ERR(cwd))
			goto out;

		error = -ERANGE;
		len = PAGE_SIZE + page - cwd;
		if (len <= size) {
			error = len;
			if (copy_to_user(buf, cwd, len))
				error = -EFAULT;
		}
	} else
		spin_unlock(&dcache_lock);
L
Linus Torvalds 已提交
2123 2124

out:
J
Jan Blunck 已提交
2125 2126
	path_put(&pwd);
	path_put(&root);
L
Linus Torvalds 已提交
2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146
	free_page((unsigned long) page);
	return error;
}

/*
 * Test whether new_dentry is a subdirectory of old_dentry.
 *
 * Trivially implemented using the dcache structure
 */

/**
 * is_subdir - is new dentry a subdirectory of old_dentry
 * @new_dentry: new dentry
 * @old_dentry: old dentry
 *
 * Returns 1 if new_dentry is a subdirectory of the parent (at any depth).
 * Returns 0 otherwise.
 * Caller must ensure that "new_dentry" is pinned before calling is_subdir()
 */
  
2147
int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry)
L
Linus Torvalds 已提交
2148 2149 2150 2151
{
	int result;
	unsigned long seq;

2152 2153 2154 2155 2156 2157
	if (new_dentry == old_dentry)
		return 1;

	/*
	 * Need rcu_readlock to protect against the d_parent trashing
	 * due to d_move
L
Linus Torvalds 已提交
2158 2159
	 */
	rcu_read_lock();
2160
	do {
L
Linus Torvalds 已提交
2161 2162
		/* for restarting inner loop in case of seq retry */
		seq = read_seqbegin(&rename_lock);
2163
		if (d_ancestor(old_dentry, new_dentry))
L
Linus Torvalds 已提交
2164
			result = 1;
2165 2166
		else
			result = 0;
L
Linus Torvalds 已提交
2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183
	} while (read_seqretry(&rename_lock, seq));
	rcu_read_unlock();

	return result;
}

void d_genocide(struct dentry *root)
{
	struct dentry *this_parent = root;
	struct list_head *next;

	spin_lock(&dcache_lock);
repeat:
	next = this_parent->d_subdirs.next;
resume:
	while (next != &this_parent->d_subdirs) {
		struct list_head *tmp = next;
E
Eric Dumazet 已提交
2184
		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
L
Linus Torvalds 已提交
2185 2186 2187 2188 2189 2190 2191 2192 2193 2194
		next = tmp->next;
		if (d_unhashed(dentry)||!dentry->d_inode)
			continue;
		if (!list_empty(&dentry->d_subdirs)) {
			this_parent = dentry;
			goto repeat;
		}
		atomic_dec(&dentry->d_count);
	}
	if (this_parent != root) {
E
Eric Dumazet 已提交
2195
		next = this_parent->d_u.d_child.next;
L
Linus Torvalds 已提交
2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221
		atomic_dec(&this_parent->d_count);
		this_parent = this_parent->d_parent;
		goto resume;
	}
	spin_unlock(&dcache_lock);
}

/**
 * find_inode_number - check for dentry with name
 * @dir: directory to check
 * @name: Name to find.
 *
 * Check whether a dentry already exists for the given name,
 * and return the inode number if it has an inode. Otherwise
 * 0 is returned.
 *
 * This routine is used to post-process directory listings for
 * filesystems using synthetic inode numbers, and is necessary
 * to keep getcwd() working.
 */
 
ino_t find_inode_number(struct dentry *dir, struct qstr *name)
{
	struct dentry * dentry;
	ino_t ino = 0;

2222 2223
	dentry = d_hash_and_lookup(dir, name);
	if (dentry) {
L
Linus Torvalds 已提交
2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264
		if (dentry->d_inode)
			ino = dentry->d_inode->i_ino;
		dput(dentry);
	}
	return ino;
}

static __initdata unsigned long dhash_entries;
static int __init set_dhash_entries(char *str)
{
	if (!str)
		return 0;
	dhash_entries = simple_strtoul(str, &str, 0);
	return 1;
}
__setup("dhash_entries=", set_dhash_entries);

static void __init dcache_init_early(void)
{
	int loop;

	/* If hashes are distributed across NUMA nodes, defer
	 * hash allocation until vmalloc space is available.
	 */
	if (hashdist)
		return;

	dentry_hashtable =
		alloc_large_system_hash("Dentry cache",
					sizeof(struct hlist_head),
					dhash_entries,
					13,
					HASH_EARLY,
					&d_hash_shift,
					&d_hash_mask,
					0);

	for (loop = 0; loop < (1 << d_hash_shift); loop++)
		INIT_HLIST_HEAD(&dentry_hashtable[loop]);
}

2265
static void __init dcache_init(void)
L
Linus Torvalds 已提交
2266 2267 2268 2269 2270 2271 2272 2273
{
	int loop;

	/* 
	 * A constructor could be added for stable state like the lists,
	 * but it is probably not worth it because of the cache nature
	 * of the dcache. 
	 */
2274 2275
	dentry_cache = KMEM_CACHE(dentry,
		SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|SLAB_MEM_SPREAD);
L
Linus Torvalds 已提交
2276
	
2277
	register_shrinker(&dcache_shrinker);
L
Linus Torvalds 已提交
2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297

	/* Hash may have been set up in dcache_init_early */
	if (!hashdist)
		return;

	dentry_hashtable =
		alloc_large_system_hash("Dentry cache",
					sizeof(struct hlist_head),
					dhash_entries,
					13,
					0,
					&d_hash_shift,
					&d_hash_mask,
					0);

	for (loop = 0; loop < (1 << d_hash_shift); loop++)
		INIT_HLIST_HEAD(&dentry_hashtable[loop]);
}

/* SLAB cache for __getname() consumers */
2298
struct kmem_cache *names_cachep __read_mostly;
L
Linus Torvalds 已提交
2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318

EXPORT_SYMBOL(d_genocide);

void __init vfs_caches_init_early(void)
{
	dcache_init_early();
	inode_init_early();
}

void __init vfs_caches_init(unsigned long mempages)
{
	unsigned long reserve;

	/* Base hash sizes on available memory, with a reserve equal to
           150% of current kernel size */

	reserve = min((mempages - nr_free_pages()) * 3/2, mempages - 1);
	mempages -= reserve;

	names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0,
2319
			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
L
Linus Torvalds 已提交
2320

2321 2322
	dcache_init();
	inode_init();
L
Linus Torvalds 已提交
2323
	files_init(mempages);
2324
	mnt_init();
L
Linus Torvalds 已提交
2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336
	bdev_cache_init();
	chrdev_init();
}

EXPORT_SYMBOL(d_alloc);
EXPORT_SYMBOL(d_alloc_root);
EXPORT_SYMBOL(d_delete);
EXPORT_SYMBOL(d_find_alias);
EXPORT_SYMBOL(d_instantiate);
EXPORT_SYMBOL(d_invalidate);
EXPORT_SYMBOL(d_lookup);
EXPORT_SYMBOL(d_move);
2337
EXPORT_SYMBOL_GPL(d_materialise_unique);
L
Linus Torvalds 已提交
2338 2339 2340 2341
EXPORT_SYMBOL(d_path);
EXPORT_SYMBOL(d_prune_aliases);
EXPORT_SYMBOL(d_rehash);
EXPORT_SYMBOL(d_splice_alias);
2342
EXPORT_SYMBOL(d_add_ci);
L
Linus Torvalds 已提交
2343 2344 2345 2346 2347 2348 2349 2350
EXPORT_SYMBOL(d_validate);
EXPORT_SYMBOL(dget_locked);
EXPORT_SYMBOL(dput);
EXPORT_SYMBOL(find_inode_number);
EXPORT_SYMBOL(have_submounts);
EXPORT_SYMBOL(names_cachep);
EXPORT_SYMBOL(shrink_dcache_parent);
EXPORT_SYMBOL(shrink_dcache_sb);