inode.c 43.2 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
/*
 * linux/fs/inode.c
 *
 * (C) 1997 Linus Torvalds
 */

#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/dcache.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/writeback.h>
#include <linux/module.h>
#include <linux/backing-dev.h>
#include <linux/wait.h>
Nick Piggin's avatar
Nick Piggin committed
16
#include <linux/rwsem.h>
Linus Torvalds's avatar
Linus Torvalds committed
17 18 19 20 21 22
#include <linux/hash.h>
#include <linux/swap.h>
#include <linux/security.h>
#include <linux/pagemap.h>
#include <linux/cdev.h>
#include <linux/bootmem.h>
23
#include <linux/fsnotify.h>
24
#include <linux/mount.h>
25
#include <linux/async.h>
Al Viro's avatar
Al Viro committed
26
#include <linux/posix_acl.h>
Linus Torvalds's avatar
Linus Torvalds committed
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

/*
 * This is needed for the following functions:
 *  - inode_has_buffers
 *  - invalidate_bdev
 *
 * FIXME: remove all knowledge of the buffer layer from this file
 */
#include <linux/buffer_head.h>

/*
 * New inode.c implementation.
 *
 * This implementation has the basic premise of trying
 * to be extremely low-overhead and SMP-safe, yet be
 * simple enough to be "obviously correct".
 *
 * Famous last words.
 */

/* inode dynamic allocation 1999, Andrea Arcangeli <andrea@suse.de> */

/* #define INODE_PARANOIA 1 */
/* #define INODE_DEBUG 1 */

/*
 * Inode lookup is no longer as critical as it used to be:
 * most of the lookups are going to be through the dcache.
 */
#define I_HASHBITS	i_hash_shift
#define I_HASHMASK	i_hash_mask

59 60
static unsigned int i_hash_mask __read_mostly;
static unsigned int i_hash_shift __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
61 62 63 64 65 66 67 68 69 70 71 72 73

/*
 * Each inode can be on two separate lists. One is
 * the hash list of the inode, used for lookups. The
 * other linked list is the "type" list:
 *  "in_use" - valid inode, i_count > 0, i_nlink > 0
 *  "dirty"  - as "in_use" but also dirty
 *  "unused" - valid inode, i_count = 0
 *
 * A "dirty" list is maintained for each super block,
 * allowing for low-overhead inode sync() operations.
 */

74
static LIST_HEAD(inode_unused);
75
static struct hlist_head *inode_hashtable __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
76 77 78 79 80 81 82 83 84 85

/*
 * A simple spinlock to protect the list manipulations.
 *
 * NOTE! You also have to own the lock if you change
 * the i_state of an inode while it is in use..
 */
DEFINE_SPINLOCK(inode_lock);

/*
Nick Piggin's avatar
Nick Piggin committed
86
 * iprune_sem provides exclusion between the kswapd or try_to_free_pages
Linus Torvalds's avatar
Linus Torvalds committed
87 88 89 90 91
 * icache shrinking path, and the umount path.  Without this exclusion,
 * by the time prune_icache calls iput for the inode whose pages it has
 * been invalidating, or by the time it calls clear_inode & destroy_inode
 * from its final dispose_list, the struct super_block they refer to
 * (for inode->i_sb->s_op) may already have been freed and reused.
Nick Piggin's avatar
Nick Piggin committed
92 93 94 95
 *
 * We make this an rwsem because the fastpath is icache shrinking. In
 * some cases a filesystem may be doing a significant amount of work in
 * its inode reclaim code, so this should improve parallelism.
Linus Torvalds's avatar
Linus Torvalds committed
96
 */
Nick Piggin's avatar
Nick Piggin committed
97
static DECLARE_RWSEM(iprune_sem);
Linus Torvalds's avatar
Linus Torvalds committed
98 99 100 101 102 103

/*
 * Statistics gathering..
 */
struct inodes_stat_t inodes_stat;

104 105 106
static struct percpu_counter nr_inodes __cacheline_aligned_in_smp;
static struct percpu_counter nr_inodes_unused __cacheline_aligned_in_smp;

107
static struct kmem_cache *inode_cachep __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
108

109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138
static inline int get_nr_inodes(void)
{
	return percpu_counter_sum_positive(&nr_inodes);
}

static inline int get_nr_inodes_unused(void)
{
	return percpu_counter_sum_positive(&nr_inodes_unused);
}

int get_nr_dirty_inodes(void)
{
	int nr_dirty = get_nr_inodes() - get_nr_inodes_unused();
	return nr_dirty > 0 ? nr_dirty : 0;

}

/*
 * Handle nr_inode sysctl
 */
#ifdef CONFIG_SYSCTL
int proc_nr_inodes(ctl_table *table, int write,
		   void __user *buffer, size_t *lenp, loff_t *ppos)
{
	inodes_stat.nr_inodes = get_nr_inodes();
	inodes_stat.nr_unused = get_nr_inodes_unused();
	return proc_dointvec(table, write, buffer, lenp, ppos);
}
#endif

Joern Engel's avatar
Joern Engel committed
139 140 141 142 143 144
static void wake_up_inode(struct inode *inode)
{
	/*
	 * Prevent speculative execution through spin_unlock(&inode_lock);
	 */
	smp_mb();
Christoph Hellwig's avatar
Christoph Hellwig committed
145
	wake_up_bit(&inode->i_state, __I_NEW);
Joern Engel's avatar
Joern Engel committed
146 147
}

148 149
/**
 * inode_init_always - perform inode structure intialisation
150 151
 * @sb: superblock inode belongs to
 * @inode: inode to initialise
152 153 154 155
 *
 * These are initializations that need to be done on every inode
 * allocation as the fields are not initialised by slab allocation.
 */
156
int inode_init_always(struct super_block *sb, struct inode *inode)
Linus Torvalds's avatar
Linus Torvalds committed
157
{
158
	static const struct address_space_operations empty_aops;
159
	static const struct inode_operations empty_iops;
160
	static const struct file_operations empty_fops;
161
	struct address_space *const mapping = &inode->i_data;
162 163 164 165 166 167 168 169

	inode->i_sb = sb;
	inode->i_blkbits = sb->s_blocksize_bits;
	inode->i_flags = 0;
	atomic_set(&inode->i_count, 1);
	inode->i_op = &empty_iops;
	inode->i_fop = &empty_fops;
	inode->i_nlink = 1;
Al Viro's avatar
Al Viro committed
170 171
	inode->i_uid = 0;
	inode->i_gid = 0;
172 173 174 175 176
	atomic_set(&inode->i_writecount, 0);
	inode->i_size = 0;
	inode->i_blocks = 0;
	inode->i_bytes = 0;
	inode->i_generation = 0;
Linus Torvalds's avatar
Linus Torvalds committed
177
#ifdef CONFIG_QUOTA
178
	memset(&inode->i_dquot, 0, sizeof(inode->i_dquot));
Linus Torvalds's avatar
Linus Torvalds committed
179
#endif
180 181 182 183 184
	inode->i_pipe = NULL;
	inode->i_bdev = NULL;
	inode->i_cdev = NULL;
	inode->i_rdev = 0;
	inode->dirtied_when = 0;
Mimi Zohar's avatar
Mimi Zohar committed
185 186

	if (security_inode_alloc(inode))
187
		goto out;
188 189 190 191 192 193 194 195 196 197 198 199
	spin_lock_init(&inode->i_lock);
	lockdep_set_class(&inode->i_lock, &sb->s_type->i_lock_key);

	mutex_init(&inode->i_mutex);
	lockdep_set_class(&inode->i_mutex, &sb->s_type->i_mutex_key);

	init_rwsem(&inode->i_alloc_sem);
	lockdep_set_class(&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key);

	mapping->a_ops = &empty_aops;
	mapping->host = inode;
	mapping->flags = 0;
200
	mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE);
201 202 203 204 205 206 207 208 209 210 211 212
	mapping->assoc_mapping = NULL;
	mapping->backing_dev_info = &default_backing_dev_info;
	mapping->writeback_index = 0;

	/*
	 * If the block_device provides a backing_dev_info for client
	 * inodes then use that.  Otherwise the inode share the bdev's
	 * backing_dev_info.
	 */
	if (sb->s_bdev) {
		struct backing_dev_info *bdi;

213
		bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info;
214 215 216 217
		mapping->backing_dev_info = bdi;
	}
	inode->i_private = NULL;
	inode->i_mapping = mapping;
Al Viro's avatar
Al Viro committed
218 219 220
#ifdef CONFIG_FS_POSIX_ACL
	inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED;
#endif
221

222 223 224 225
#ifdef CONFIG_FSNOTIFY
	inode->i_fsnotify_mask = 0;
#endif

226 227
	percpu_counter_inc(&nr_inodes);

228 229 230
	return 0;
out:
	return -ENOMEM;
Linus Torvalds's avatar
Linus Torvalds committed
231
}
232 233 234 235 236 237 238 239 240 241 242
EXPORT_SYMBOL(inode_init_always);

static struct inode *alloc_inode(struct super_block *sb)
{
	struct inode *inode;

	if (sb->s_op->alloc_inode)
		inode = sb->s_op->alloc_inode(sb);
	else
		inode = kmem_cache_alloc(inode_cachep, GFP_KERNEL);

243 244 245 246 247 248 249 250 251 252 253 254
	if (!inode)
		return NULL;

	if (unlikely(inode_init_always(sb, inode))) {
		if (inode->i_sb->s_op->destroy_inode)
			inode->i_sb->s_op->destroy_inode(inode);
		else
			kmem_cache_free(inode_cachep, inode);
		return NULL;
	}

	return inode;
255
}
Linus Torvalds's avatar
Linus Torvalds committed
256

Christoph Hellwig's avatar
Christoph Hellwig committed
257
void __destroy_inode(struct inode *inode)
Linus Torvalds's avatar
Linus Torvalds committed
258
{
259
	BUG_ON(inode_has_buffers(inode));
Linus Torvalds's avatar
Linus Torvalds committed
260
	security_inode_free(inode);
261
	fsnotify_inode_delete(inode);
Al Viro's avatar
Al Viro committed
262 263 264 265 266 267
#ifdef CONFIG_FS_POSIX_ACL
	if (inode->i_acl && inode->i_acl != ACL_NOT_CACHED)
		posix_acl_release(inode->i_acl);
	if (inode->i_default_acl && inode->i_default_acl != ACL_NOT_CACHED)
		posix_acl_release(inode->i_default_acl);
#endif
268
	percpu_counter_dec(&nr_inodes);
Christoph Hellwig's avatar
Christoph Hellwig committed
269 270 271
}
EXPORT_SYMBOL(__destroy_inode);

272
static void destroy_inode(struct inode *inode)
Christoph Hellwig's avatar
Christoph Hellwig committed
273 274
{
	__destroy_inode(inode);
Linus Torvalds's avatar
Linus Torvalds committed
275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291
	if (inode->i_sb->s_op->destroy_inode)
		inode->i_sb->s_op->destroy_inode(inode);
	else
		kmem_cache_free(inode_cachep, (inode));
}

/*
 * These are initializations that only need to be done
 * once, because the fields are idempotent across use
 * of the inode, so let the slab aware of that.
 */
void inode_init_once(struct inode *inode)
{
	memset(inode, 0, sizeof(*inode));
	INIT_HLIST_NODE(&inode->i_hash);
	INIT_LIST_HEAD(&inode->i_dentry);
	INIT_LIST_HEAD(&inode->i_devices);
292
	INIT_LIST_HEAD(&inode->i_list);
Linus Torvalds's avatar
Linus Torvalds committed
293
	INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
Nick Piggin's avatar
Nick Piggin committed
294
	spin_lock_init(&inode->i_data.tree_lock);
Linus Torvalds's avatar
Linus Torvalds committed
295 296 297 298 299 300
	spin_lock_init(&inode->i_data.i_mmap_lock);
	INIT_LIST_HEAD(&inode->i_data.private_list);
	spin_lock_init(&inode->i_data.private_lock);
	INIT_RAW_PRIO_TREE_ROOT(&inode->i_data.i_mmap);
	INIT_LIST_HEAD(&inode->i_data.i_mmap_nonlinear);
	i_size_ordered_init(inode);
301
#ifdef CONFIG_FSNOTIFY
302
	INIT_HLIST_HEAD(&inode->i_fsnotify_marks);
303
#endif
Linus Torvalds's avatar
Linus Torvalds committed
304 305 306
}
EXPORT_SYMBOL(inode_init_once);

307
static void init_once(void *foo)
Linus Torvalds's avatar
Linus Torvalds committed
308
{
309
	struct inode *inode = (struct inode *) foo;
Linus Torvalds's avatar
Linus Torvalds committed
310

311
	inode_init_once(inode);
Linus Torvalds's avatar
Linus Torvalds committed
312 313 314 315 316
}

/*
 * inode_lock must be held
 */
317
void __iget(struct inode *inode)
Linus Torvalds's avatar
Linus Torvalds committed
318
{
319 320
	atomic_inc(&inode->i_count);
}
321

Al Viro's avatar
Al Viro committed
322 323 324 325 326 327 328 329 330
/*
 * get additional reference to inode; caller must already hold one.
 */
void ihold(struct inode *inode)
{
	WARN_ON(atomic_inc_return(&inode->i_count) < 2);
}
EXPORT_SYMBOL(ihold);

331 332 333 334 335 336 337 338 339 340 341 342 343 344
static void inode_lru_list_add(struct inode *inode)
{
	if (list_empty(&inode->i_list)) {
		list_add(&inode->i_list, &inode_unused);
		percpu_counter_inc(&nr_inodes_unused);
	}
}

static void inode_lru_list_del(struct inode *inode)
{
	if (!list_empty(&inode->i_list)) {
		list_del_init(&inode->i_list);
		percpu_counter_dec(&nr_inodes_unused);
	}
Linus Torvalds's avatar
Linus Torvalds committed
345 346
}

347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368
static inline void __inode_sb_list_add(struct inode *inode)
{
	list_add(&inode->i_sb_list, &inode->i_sb->s_inodes);
}

/**
 * inode_sb_list_add - add inode to the superblock list of inodes
 * @inode: inode to add
 */
void inode_sb_list_add(struct inode *inode)
{
	spin_lock(&inode_lock);
	__inode_sb_list_add(inode);
	spin_unlock(&inode_lock);
}
EXPORT_SYMBOL_GPL(inode_sb_list_add);

static inline void __inode_sb_list_del(struct inode *inode)
{
	list_del_init(&inode->i_sb_list);
}

369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
static unsigned long hash(struct super_block *sb, unsigned long hashval)
{
	unsigned long tmp;

	tmp = (hashval * (unsigned long)sb) ^ (GOLDEN_RATIO_PRIME + hashval) /
			L1_CACHE_BYTES;
	tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> I_HASHBITS);
	return tmp & I_HASHMASK;
}

/**
 *	__insert_inode_hash - hash an inode
 *	@inode: unhashed inode
 *	@hashval: unsigned long value used to locate this object in the
 *		inode_hashtable.
 *
 *	Add an inode to the inode hash for this superblock.
 */
void __insert_inode_hash(struct inode *inode, unsigned long hashval)
{
389 390
	struct hlist_head *b = inode_hashtable + hash(inode->i_sb, hashval);

391
	spin_lock(&inode_lock);
392
	hlist_add_head(&inode->i_hash, b);
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
	spin_unlock(&inode_lock);
}
EXPORT_SYMBOL(__insert_inode_hash);

/**
 *	__remove_inode_hash - remove an inode from the hash
 *	@inode: inode to unhash
 *
 *	Remove an inode from the superblock.
 */
static void __remove_inode_hash(struct inode *inode)
{
	hlist_del_init(&inode->i_hash);
}

/**
 *	remove_inode_hash - remove an inode from the hash
 *	@inode: inode to unhash
 *
 *	Remove an inode from the superblock.
 */
void remove_inode_hash(struct inode *inode)
{
	spin_lock(&inode_lock);
	hlist_del_init(&inode->i_hash);
	spin_unlock(&inode_lock);
}
EXPORT_SYMBOL(remove_inode_hash);

Al Viro's avatar
Al Viro committed
422 423 424 425 426 427 428 429 430 431 432 433
void end_writeback(struct inode *inode)
{
	might_sleep();
	BUG_ON(inode->i_data.nrpages);
	BUG_ON(!list_empty(&inode->i_data.private_list));
	BUG_ON(!(inode->i_state & I_FREEING));
	BUG_ON(inode->i_state & I_CLEAR);
	inode_sync_wait(inode);
	inode->i_state = I_FREEING | I_CLEAR;
}
EXPORT_SYMBOL(end_writeback);

434
static void evict(struct inode *inode)
435 436 437
{
	const struct super_operations *op = inode->i_sb->s_op;

Al Viro's avatar
Al Viro committed
438 439
	if (op->evict_inode) {
		op->evict_inode(inode);
440 441 442
	} else {
		if (inode->i_data.nrpages)
			truncate_inode_pages(&inode->i_data, 0);
Al Viro's avatar
Al Viro committed
443
		end_writeback(inode);
444
	}
445 446 447 448
	if (S_ISBLK(inode->i_mode) && inode->i_bdev)
		bd_forget(inode);
	if (S_ISCHR(inode->i_mode) && inode->i_cdev)
		cd_forget(inode);
449 450
}

Linus Torvalds's avatar
Linus Torvalds committed
451 452 453 454 455 456 457 458 459 460 461 462
/*
 * dispose_list - dispose of the contents of a local list
 * @head: the head of the list to free
 *
 * Dispose-list gets a local list with local inodes in it, so it doesn't
 * need to worry about list corruption and SMP locks.
 */
static void dispose_list(struct list_head *head)
{
	while (!list_empty(head)) {
		struct inode *inode;

463
		inode = list_first_entry(head, struct inode, i_list);
464
		list_del_init(&inode->i_list);
Linus Torvalds's avatar
Linus Torvalds committed
465

466
		evict(inode);
467 468

		spin_lock(&inode_lock);
469
		__remove_inode_hash(inode);
470
		__inode_sb_list_del(inode);
471 472 473
		spin_unlock(&inode_lock);

		wake_up_inode(inode);
Linus Torvalds's avatar
Linus Torvalds committed
474 475 476 477 478 479 480 481 482 483
		destroy_inode(inode);
	}
}

/*
 * Invalidate all inodes for a device.
 */
static int invalidate_list(struct list_head *head, struct list_head *dispose)
{
	struct list_head *next;
484
	int busy = 0;
Linus Torvalds's avatar
Linus Torvalds committed
485 486 487

	next = head->next;
	for (;;) {
488 489
		struct list_head *tmp = next;
		struct inode *inode;
Linus Torvalds's avatar
Linus Torvalds committed
490 491 492 493

		/*
		 * We can reschedule here without worrying about the list's
		 * consistency because the per-sb list of inodes must not
Nick Piggin's avatar
Nick Piggin committed
494
		 * change during umount anymore, and because iprune_sem keeps
Linus Torvalds's avatar
Linus Torvalds committed
495 496 497 498 499 500 501 502
		 * shrink_icache_memory() away.
		 */
		cond_resched_lock(&inode_lock);

		next = next->next;
		if (tmp == head)
			break;
		inode = list_entry(tmp, struct inode, i_sb_list);
Nick Piggin's avatar
Nick Piggin committed
503 504
		if (inode->i_state & I_NEW)
			continue;
505 506
		if (atomic_read(&inode->i_count)) {
			busy = 1;
Linus Torvalds's avatar
Linus Torvalds committed
507 508
			continue;
		}
509 510 511 512 513

		list_move(&inode->i_list, dispose);
		inode->i_state |= I_FREEING;
		if (!(inode->i_state & (I_DIRTY | I_SYNC)))
			percpu_counter_dec(&nr_inodes_unused);
Linus Torvalds's avatar
Linus Torvalds committed
514 515 516 517 518 519 520 521 522 523 524 525
	}
	return busy;
}

/**
 *	invalidate_inodes	- discard the inodes on a device
 *	@sb: superblock
 *
 *	Discard all of the inodes for a given superblock. If the discard
 *	fails because there are busy inodes then a non zero value is returned.
 *	If the discard is successful all the inodes have been discarded.
 */
526
int invalidate_inodes(struct super_block *sb)
Linus Torvalds's avatar
Linus Torvalds committed
527 528 529 530
{
	int busy;
	LIST_HEAD(throw_away);

Nick Piggin's avatar
Nick Piggin committed
531
	down_write(&iprune_sem);
Linus Torvalds's avatar
Linus Torvalds committed
532
	spin_lock(&inode_lock);
533
	fsnotify_unmount_inodes(&sb->s_inodes);
Linus Torvalds's avatar
Linus Torvalds committed
534 535 536 537
	busy = invalidate_list(&sb->s_inodes, &throw_away);
	spin_unlock(&inode_lock);

	dispose_list(&throw_away);
Nick Piggin's avatar
Nick Piggin committed
538
	up_write(&iprune_sem);
Linus Torvalds's avatar
Linus Torvalds committed
539 540 541 542 543 544

	return busy;
}

static int can_unuse(struct inode *inode)
{
545
	if (inode->i_state & ~I_REFERENCED)
Linus Torvalds's avatar
Linus Torvalds committed
546 547 548 549 550 551 552 553 554 555 556
		return 0;
	if (inode_has_buffers(inode))
		return 0;
	if (atomic_read(&inode->i_count))
		return 0;
	if (inode->i_data.nrpages)
		return 0;
	return 1;
}

/*
557 558
 * Scan `goal' inodes on the unused list for freeable ones. They are moved to a
 * temporary list and then are freed outside inode_lock by dispose_list().
Linus Torvalds's avatar
Linus Torvalds committed
559 560
 *
 * Any inodes which are pinned purely because of attached pagecache have their
561 562
 * pagecache removed.  If the inode has metadata buffers attached to
 * mapping->private_list then try to remove them.
Linus Torvalds's avatar
Linus Torvalds committed
563
 *
564 565 566 567 568 569 570
 * If the inode has the I_REFERENCED flag set, then it means that it has been
 * used recently - the flag is set in iput_final(). When we encounter such an
 * inode, clear the flag and move it to the back of the LRU so it gets another
 * pass through the LRU before it gets reclaimed. This is necessary because of
 * the fact we are doing lazy LRU updates to minimise lock contention so the
 * LRU does not have strict ordering. Hence we don't want to reclaim inodes
 * with this flag set because they are the inodes that are out of order.
Linus Torvalds's avatar
Linus Torvalds committed
571 572 573 574 575 576 577
 */
static void prune_icache(int nr_to_scan)
{
	LIST_HEAD(freeable);
	int nr_scanned;
	unsigned long reap = 0;

Nick Piggin's avatar
Nick Piggin committed
578
	down_read(&iprune_sem);
Linus Torvalds's avatar
Linus Torvalds committed
579 580 581 582 583 584 585 586 587
	spin_lock(&inode_lock);
	for (nr_scanned = 0; nr_scanned < nr_to_scan; nr_scanned++) {
		struct inode *inode;

		if (list_empty(&inode_unused))
			break;

		inode = list_entry(inode_unused.prev, struct inode, i_list);

588 589 590 591 592 593 594 595 596 597 598 599 600
		/*
		 * Referenced or dirty inodes are still in use. Give them
		 * another pass through the LRU as we canot reclaim them now.
		 */
		if (atomic_read(&inode->i_count) ||
		    (inode->i_state & ~I_REFERENCED)) {
			list_del_init(&inode->i_list);
			percpu_counter_dec(&nr_inodes_unused);
			continue;
		}

		/* recently referenced inodes get one more pass */
		if (inode->i_state & I_REFERENCED) {
Linus Torvalds's avatar
Linus Torvalds committed
601
			list_move(&inode->i_list, &inode_unused);
602
			inode->i_state &= ~I_REFERENCED;
Linus Torvalds's avatar
Linus Torvalds committed
603 604 605 606 607 608
			continue;
		}
		if (inode_has_buffers(inode) || inode->i_data.nrpages) {
			__iget(inode);
			spin_unlock(&inode_lock);
			if (remove_inode_buffers(inode))
609 610
				reap += invalidate_mapping_pages(&inode->i_data,
								0, -1);
Linus Torvalds's avatar
Linus Torvalds committed
611 612 613 614 615 616 617 618 619 620
			iput(inode);
			spin_lock(&inode_lock);

			if (inode != list_entry(inode_unused.next,
						struct inode, i_list))
				continue;	/* wrong inode or list_empty */
			if (!can_unuse(inode))
				continue;
		}
		list_move(&inode->i_list, &freeable);
621
		WARN_ON(inode->i_state & I_NEW);
Linus Torvalds's avatar
Linus Torvalds committed
622
		inode->i_state |= I_FREEING;
623
		percpu_counter_dec(&nr_inodes_unused);
Linus Torvalds's avatar
Linus Torvalds committed
624
	}
625 626 627 628
	if (current_is_kswapd())
		__count_vm_events(KSWAPD_INODESTEAL, reap);
	else
		__count_vm_events(PGINODESTEAL, reap);
Linus Torvalds's avatar
Linus Torvalds committed
629 630 631
	spin_unlock(&inode_lock);

	dispose_list(&freeable);
Nick Piggin's avatar
Nick Piggin committed
632
	up_read(&iprune_sem);
Linus Torvalds's avatar
Linus Torvalds committed
633 634 635 636 637 638 639 640 641 642 643
}

/*
 * shrink_icache_memory() will attempt to reclaim some unused inodes.  Here,
 * "unused" means that no dentries are referring to the inodes: the files are
 * not open and the dcache references to those inodes have already been
 * reclaimed.
 *
 * This function is passed the number of inodes to scan, and it returns the
 * total number of remaining possibly-reclaimable inodes.
 */
644
static int shrink_icache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask)
Linus Torvalds's avatar
Linus Torvalds committed
645 646 647 648 649 650
{
	if (nr) {
		/*
		 * Nasty deadlock avoidance.  We may hold various FS locks,
		 * and we don't want to recurse into the FS that called us
		 * in clear_inode() and friends..
651
		 */
Linus Torvalds's avatar
Linus Torvalds committed
652 653 654 655
		if (!(gfp_mask & __GFP_FS))
			return -1;
		prune_icache(nr);
	}
656
	return (get_nr_inodes_unused() / 100) * sysctl_vfs_cache_pressure;
Linus Torvalds's avatar
Linus Torvalds committed
657 658
}

659 660 661 662 663
static struct shrinker icache_shrinker = {
	.shrink = shrink_icache_memory,
	.seeks = DEFAULT_SEEKS,
};

Linus Torvalds's avatar
Linus Torvalds committed
664 665 666 667
static void __wait_on_freeing_inode(struct inode *inode);
/*
 * Called with the inode lock held.
 */
668 669 670 671
static struct inode *find_inode(struct super_block *sb,
				struct hlist_head *head,
				int (*test)(struct inode *, void *),
				void *data)
Linus Torvalds's avatar
Linus Torvalds committed
672 673
{
	struct hlist_node *node;
674
	struct inode *inode = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
675 676

repeat:
677
	hlist_for_each_entry(inode, node, head, i_hash) {
Linus Torvalds's avatar
Linus Torvalds committed
678 679 680 681
		if (inode->i_sb != sb)
			continue;
		if (!test(inode, data))
			continue;
682
		if (inode->i_state & (I_FREEING|I_WILL_FREE)) {
Linus Torvalds's avatar
Linus Torvalds committed
683 684 685
			__wait_on_freeing_inode(inode);
			goto repeat;
		}
686 687
		__iget(inode);
		return inode;
Linus Torvalds's avatar
Linus Torvalds committed
688
	}
689
	return NULL;
Linus Torvalds's avatar
Linus Torvalds committed
690 691 692 693 694 695
}

/*
 * find_inode_fast is the fast path version of find_inode, see the comment at
 * iget_locked for details.
 */
696 697
static struct inode *find_inode_fast(struct super_block *sb,
				struct hlist_head *head, unsigned long ino)
Linus Torvalds's avatar
Linus Torvalds committed
698 699
{
	struct hlist_node *node;
700
	struct inode *inode = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
701 702

repeat:
703
	hlist_for_each_entry(inode, node, head, i_hash) {
Linus Torvalds's avatar
Linus Torvalds committed
704 705 706 707
		if (inode->i_ino != ino)
			continue;
		if (inode->i_sb != sb)
			continue;
708
		if (inode->i_state & (I_FREEING|I_WILL_FREE)) {
Linus Torvalds's avatar
Linus Torvalds committed
709 710 711
			__wait_on_freeing_inode(inode);
			goto repeat;
		}
712 713
		__iget(inode);
		return inode;
Linus Torvalds's avatar
Linus Torvalds committed
714
	}
715
	return NULL;
Linus Torvalds's avatar
Linus Torvalds committed
716 717
}

718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735
/*
 * Each cpu owns a range of LAST_INO_BATCH numbers.
 * 'shared_last_ino' is dirtied only once out of LAST_INO_BATCH allocations,
 * to renew the exhausted range.
 *
 * This does not significantly increase overflow rate because every CPU can
 * consume at most LAST_INO_BATCH-1 unused inode numbers. So there is
 * NR_CPUS*(LAST_INO_BATCH-1) wastage. At 4096 and 1024, this is ~0.1% of the
 * 2^32 range, and is a worst-case. Even a 50% wastage would only increase
 * overflow rate by 2x, which does not seem too significant.
 *
 * On a 32bit, non LFS stat() call, glibc will generate an EOVERFLOW
 * error if st_ino won't fit in target struct field. Use 32bit counter
 * here to attempt to avoid that.
 */
#define LAST_INO_BATCH 1024
static DEFINE_PER_CPU(unsigned int, last_ino);

736
unsigned int get_next_ino(void)
737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753
{
	unsigned int *p = &get_cpu_var(last_ino);
	unsigned int res = *p;

#ifdef CONFIG_SMP
	if (unlikely((res & (LAST_INO_BATCH-1)) == 0)) {
		static atomic_t shared_last_ino;
		int next = atomic_add_return(LAST_INO_BATCH, &shared_last_ino);

		res = next - LAST_INO_BATCH;
	}
#endif

	*p = ++res;
	put_cpu_var(last_ino);
	return res;
}
754
EXPORT_SYMBOL(get_next_ino);
755

Linus Torvalds's avatar
Linus Torvalds committed
756 757 758 759
/**
 *	new_inode 	- obtain an inode
 *	@sb: superblock
 *
760
 *	Allocates a new inode for given superblock. The default gfp_mask
761
 *	for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE.
762 763 764 765 766
 *	If HIGHMEM pages are unsuitable or it is known that pages allocated
 *	for the page cache are not reclaimable or migratable,
 *	mapping_set_gfp_mask() must be called with suitable flags on the
 *	newly created inode's mapping
 *
Linus Torvalds's avatar
Linus Torvalds committed
767 768 769
 */
struct inode *new_inode(struct super_block *sb)
{
770
	struct inode *inode;
Linus Torvalds's avatar
Linus Torvalds committed
771 772

	spin_lock_prefetch(&inode_lock);
773

Linus Torvalds's avatar
Linus Torvalds committed
774 775 776
	inode = alloc_inode(sb);
	if (inode) {
		spin_lock(&inode_lock);
777
		__inode_sb_list_add(inode);
Linus Torvalds's avatar
Linus Torvalds committed
778 779 780 781 782 783 784 785 786
		inode->i_state = 0;
		spin_unlock(&inode_lock);
	}
	return inode;
}
EXPORT_SYMBOL(new_inode);

void unlock_new_inode(struct inode *inode)
{
787
#ifdef CONFIG_DEBUG_LOCK_ALLOC
788
	if (S_ISDIR(inode->i_mode)) {
789 790
		struct file_system_type *type = inode->i_sb->s_type;

791 792 793 794 795 796 797 798 799 800 801
		/* Set new key only if filesystem hasn't already changed it */
		if (!lockdep_match_class(&inode->i_mutex,
		    &type->i_mutex_key)) {
			/*
			 * ensure nobody is actually holding i_mutex
			 */
			mutex_destroy(&inode->i_mutex);
			mutex_init(&inode->i_mutex);
			lockdep_set_class(&inode->i_mutex,
					  &type->i_mutex_dir_key);
		}
802
	}
803
#endif
Linus Torvalds's avatar
Linus Torvalds committed
804
	/*
Christoph Hellwig's avatar
Christoph Hellwig committed
805
	 * This is special!  We do not need the spinlock when clearing I_NEW,
806 807
	 * because we're guaranteed that nobody else tries to do anything about
	 * the state of the inode when it is locked, as we just created it (so
Christoph Hellwig's avatar
Christoph Hellwig committed
808
	 * there can be no old holders that haven't tested I_NEW).
809
	 * However we must emit the memory barrier so that other CPUs reliably
Christoph Hellwig's avatar
Christoph Hellwig committed
810
	 * see the clearing of I_NEW after the other inode initialisation has
811
	 * completed.
Linus Torvalds's avatar
Linus Torvalds committed
812
	 */
813
	smp_mb();
Christoph Hellwig's avatar
Christoph Hellwig committed
814 815
	WARN_ON(!(inode->i_state & I_NEW));
	inode->i_state &= ~I_NEW;
Linus Torvalds's avatar
Linus Torvalds committed
816 817 818 819 820 821 822 823 824 825
	wake_up_inode(inode);
}
EXPORT_SYMBOL(unlock_new_inode);

/*
 * This is called without the inode lock held.. Be careful.
 *
 * We no longer cache the sb_flags in i_flags - see fs.h
 *	-- rmk@arm.uk.linux.org
 */
826 827 828 829 830
static struct inode *get_new_inode(struct super_block *sb,
				struct hlist_head *head,
				int (*test)(struct inode *, void *),
				int (*set)(struct inode *, void *),
				void *data)
Linus Torvalds's avatar
Linus Torvalds committed
831
{
832
	struct inode *inode;
Linus Torvalds's avatar
Linus Torvalds committed
833 834 835

	inode = alloc_inode(sb);
	if (inode) {
836
		struct inode *old;
Linus Torvalds's avatar
Linus Torvalds committed
837 838 839 840 841 842 843 844

		spin_lock(&inode_lock);
		/* We released the lock, so.. */
		old = find_inode(sb, head, test, data);
		if (!old) {
			if (set(inode, data))
				goto set_failed;

845 846
			hlist_add_head(&inode->i_hash, head);
			__inode_sb_list_add(inode);
Christoph Hellwig's avatar
Christoph Hellwig committed
847
			inode->i_state = I_NEW;
Linus Torvalds's avatar
Linus Torvalds committed
848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877
			spin_unlock(&inode_lock);

			/* Return the locked inode with I_NEW set, the
			 * caller is responsible for filling in the contents
			 */
			return inode;
		}

		/*
		 * Uhhuh, somebody else created the same inode under
		 * us. Use the old inode instead of the one we just
		 * allocated.
		 */
		spin_unlock(&inode_lock);
		destroy_inode(inode);
		inode = old;
		wait_on_inode(inode);
	}
	return inode;

set_failed:
	spin_unlock(&inode_lock);
	destroy_inode(inode);
	return NULL;
}

/*
 * get_new_inode_fast is the fast path version of get_new_inode, see the
 * comment at iget_locked for details.
 */
878 879
static struct inode *get_new_inode_fast(struct super_block *sb,
				struct hlist_head *head, unsigned long ino)
Linus Torvalds's avatar
Linus Torvalds committed
880
{
881
	struct inode *inode;
Linus Torvalds's avatar
Linus Torvalds committed
882 883 884

	inode = alloc_inode(sb);
	if (inode) {
885
		struct inode *old;
Linus Torvalds's avatar
Linus Torvalds committed
886 887 888 889 890 891

		spin_lock(&inode_lock);
		/* We released the lock, so.. */
		old = find_inode_fast(sb, head, ino);
		if (!old) {
			inode->i_ino = ino;
892 893
			hlist_add_head(&inode->i_hash, head);
			__inode_sb_list_add(inode);
Christoph Hellwig's avatar
Christoph Hellwig committed
894
			inode->i_state = I_NEW;
Linus Torvalds's avatar
Linus Torvalds committed
895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915
			spin_unlock(&inode_lock);

			/* Return the locked inode with I_NEW set, the
			 * caller is responsible for filling in the contents
			 */
			return inode;
		}

		/*
		 * Uhhuh, somebody else created the same inode under
		 * us. Use the old inode instead of the one we just
		 * allocated.
		 */
		spin_unlock(&inode_lock);
		destroy_inode(inode);
		inode = old;
		wait_on_inode(inode);
	}
	return inode;
}

916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936
/*
 * search the inode cache for a matching inode number.
 * If we find one, then the inode number we are trying to
 * allocate is not unique and so we should not use it.
 *
 * Returns 1 if the inode number is unique, 0 if it is not.
 */
static int test_inode_iunique(struct super_block *sb, unsigned long ino)
{
	struct hlist_head *b = inode_hashtable + hash(sb, ino);
	struct hlist_node *node;
	struct inode *inode;

	hlist_for_each_entry(inode, node, b, i_hash) {
		if (inode->i_ino == ino && inode->i_sb == sb)
			return 0;
	}

	return 1;
}

Linus Torvalds's avatar
Linus Torvalds committed
937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952
/**
 *	iunique - get a unique inode number
 *	@sb: superblock
 *	@max_reserved: highest reserved inode number
 *
 *	Obtain an inode number that is unique on the system for a given
 *	superblock. This is used by file systems that have no natural
 *	permanent inode numbering system. An inode number is returned that
 *	is higher than the reserved limit but unique.
 *
 *	BUGS:
 *	With a large number of inodes live on the file system this function
 *	currently becomes quite slow.
 */
ino_t iunique(struct super_block *sb, ino_t max_reserved)
{
953 954 955 956 957
	/*
	 * On a 32bit, non LFS stat() call, glibc will generate an EOVERFLOW
	 * error if st_ino won't fit in target struct field. Use 32bit counter
	 * here to attempt to avoid that.
	 */
958
	static DEFINE_SPINLOCK(iunique_lock);
959
	static unsigned int counter;
Linus Torvalds's avatar
Linus Torvalds committed
960
	ino_t res;
961

Linus Torvalds's avatar
Linus Torvalds committed
962
	spin_lock(&inode_lock);
963
	spin_lock(&iunique_lock);
964 965 966
	do {
		if (counter <= max_reserved)
			counter = max_reserved + 1;
Linus Torvalds's avatar
Linus Torvalds committed
967
		res = counter++;
968 969
	} while (!test_inode_iunique(sb, res));
	spin_unlock(&iunique_lock);
970
	spin_unlock(&inode_lock);
Linus Torvalds's avatar
Linus Torvalds committed
971

972 973
	return res;
}
Linus Torvalds's avatar
Linus Torvalds committed
974 975 976 977 978
EXPORT_SYMBOL(iunique);

struct inode *igrab(struct inode *inode)
{
	spin_lock(&inode_lock);
979
	if (!(inode->i_state & (I_FREEING|I_WILL_FREE)))
Linus Torvalds's avatar
Linus Torvalds committed
980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998
		__iget(inode);
	else
		/*
		 * Handle the case where s_op->clear_inode is not been
		 * called yet, and somebody is calling igrab
		 * while the inode is getting freed.
		 */
		inode = NULL;
	spin_unlock(&inode_lock);
	return inode;
}
EXPORT_SYMBOL(igrab);

/**
 * ifind - internal function, you want ilookup5() or iget5().
 * @sb:		super block of file system to search
 * @head:       the head of the list to search
 * @test:	callback used for comparisons between inodes
 * @data:	opaque data pointer to pass to @test
999
 * @wait:	if true wait for the inode to be unlocked, if false do not
Linus Torvalds's avatar
Linus Torvalds committed
1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
 *
 * ifind() searches for the inode specified by @data in the inode
 * cache. This is a generalized version of ifind_fast() for file systems where
 * the inode number is not sufficient for unique identification of an inode.
 *
 * If the inode is in the cache, the inode is returned with an incremented
 * reference count.
 *
 * Otherwise NULL is returned.
 *
 * Note, @test is called with the inode_lock held, so can't sleep.
 */
1012
static struct inode *ifind(struct super_block *sb,
Linus Torvalds's avatar
Linus Torvalds committed
1013
		struct hlist_head *head, int (*test)(struct inode *, void *),
1014
		void *data, const int wait)
Linus Torvalds's avatar
Linus Torvalds committed
1015 1016 1017 1018 1019 1020 1021
{
	struct inode *inode;

	spin_lock(&inode_lock);
	inode = find_inode(sb, head, test, data);
	if (inode) {
		spin_unlock(&inode_lock);
1022 1023
		if (likely(wait))
			wait_on_inode(inode);
Linus Torvalds's avatar
Linus Torvalds committed
1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044
		return inode;
	}
	spin_unlock(&inode_lock);
	return NULL;
}

/**
 * ifind_fast - internal function, you want ilookup() or iget().
 * @sb:		super block of file system to search
 * @head:       head of the list to search
 * @ino:	inode number to search for
 *
 * ifind_fast() searches for the inode @ino in the inode cache. This is for
 * file systems where the inode number is sufficient for unique identification
 * of an inode.
 *
 * If the inode is in the cache, the inode is returned with an incremented
 * reference count.
 *
 * Otherwise NULL is returned.
 */
1045
static struct inode *ifind_fast(struct super_block *sb,
Linus Torvalds's avatar
Linus Torvalds committed
1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061
		struct hlist_head *head, unsigned long ino)
{
	struct inode *inode;

	spin_lock(&inode_lock);
	inode = find_inode_fast(sb, head, ino);
	if (inode) {
		spin_unlock(&inode_lock);
		wait_on_inode(inode);
		return inode;
	}
	spin_unlock(&inode_lock);
	return NULL;
}

/**
1062
 * ilookup5_nowait - search for an inode in the inode cache
Linus Torvalds's avatar
Linus Torvalds committed
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073
 * @sb:		super block of file system to search
 * @hashval:	hash value (usually inode number) to search for
 * @test:	callback used for comparisons between inodes
 * @data:	opaque data pointer to pass to @test
 *
 * ilookup5() uses ifind() to search for the inode specified by @hashval and
 * @data in the inode cache. This is a generalized version of ilookup() for
 * file systems where the inode number is not sufficient for unique
 * identification of an inode.
 *
 * If the inode is in the cache, the inode is returned with an incremented
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104
 * reference count.  Note, the inode lock is not waited upon so you have to be
 * very careful what you do with the returned inode.  You probably should be
 * using ilookup5() instead.
 *
 * Otherwise NULL is returned.
 *
 * Note, @test is called with the inode_lock held, so can't sleep.
 */
struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval,
		int (*test)(struct inode *, void *), void *data)
{
	struct hlist_head *head = inode_hashtable + hash(sb, hashval);

	return ifind(sb, head, test, data, 0);
}
EXPORT_SYMBOL(ilookup5_nowait);

/**
 * ilookup5 - search for an inode in the inode cache
 * @sb:		super block of file system to search
 * @hashval:	hash value (usually inode number) to search for
 * @test:	callback used for comparisons between inodes
 * @data:	opaque data pointer to pass to @test
 *
 * ilookup5() uses ifind() to search for the inode specified by @hashval and
 * @data in the inode cache. This is a generalized version of ilookup() for
 * file systems where the inode number is not sufficient for unique
 * identification of an inode.
 *
 * If the inode is in the cache, the inode lock is waited upon and the inode is
 * returned with an incremented reference count.
Linus Torvalds's avatar
Linus Torvalds committed
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114
 *
 * Otherwise NULL is returned.
 *
 * Note, @test is called with the inode_lock held, so can't sleep.
 */
struct inode *ilookup5(struct super_block *sb, unsigned long hashval,
		int (*test)(struct inode *, void *), void *data)
{
	struct hlist_head *head = inode_hashtable + hash(sb, hashval);

1115
	return ifind(sb, head, test, data, 1);
Linus Torvalds's avatar
Linus Torvalds committed
1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167
}
EXPORT_SYMBOL(ilookup5);

/**
 * ilookup - search for an inode in the inode cache
 * @sb:		super block of file system to search
 * @ino:	inode number to search for
 *
 * ilookup() uses ifind_fast() to search for the inode @ino in the inode cache.
 * This is for file systems where the inode number is sufficient for unique
 * identification of an inode.
 *
 * If the inode is in the cache, the inode is returned with an incremented
 * reference count.
 *
 * Otherwise NULL is returned.
 */
struct inode *ilookup(struct super_block *sb, unsigned long ino)
{
	struct hlist_head *head = inode_hashtable + hash(sb, ino);

	return ifind_fast(sb, head, ino);
}
EXPORT_SYMBOL(ilookup);

/**
 * iget5_locked - obtain an inode from a mounted file system
 * @sb:		super block of file system
 * @hashval:	hash value (usually inode number) to get
 * @test:	callback used for comparisons between inodes
 * @set:	callback used to initialize a new struct inode
 * @data:	opaque data pointer to pass to @test and @set
 *
 * iget5_locked() uses ifind() to search for the inode specified by @hashval
 * and @data in the inode cache and if present it is returned with an increased
 * reference count. This is a generalized version of iget_locked() for file
 * systems where the inode number is not sufficient for unique identification
 * of an inode.
 *
 * If the inode is not in cache, get_new_inode() is called to allocate a new
 * inode and this is returned locked, hashed, and with the I_NEW flag set. The
 * file system gets to fill it in before unlocking it via unlock_new_inode().
 *
 * Note both @test and @set are called with the inode_lock held, so can't sleep.
 */
struct inode *iget5_locked(struct super_block *sb, unsigned long hashval,
		int (*test)(struct inode *, void *),
		int (*set)(struct inode *, void *), void *data)
{
	struct hlist_head *head = inode_hashtable + hash(sb, hashval);
	struct inode *inode;

1168
	inode = ifind(sb, head, test, data, 1);
Linus Torvalds's avatar
Linus Torvalds committed
1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
	if (inode)
		return inode;
	/*
	 * get_new_inode() will do the right thing, re-trying the search
	 * in case it had to block at any point.
	 */
	return get_new_inode(sb, head, test, set, data);
}
EXPORT_SYMBOL(iget5_locked);

/**
 * iget_locked - obtain an inode from a mounted file system
 * @sb:		super block of file system
 * @ino:	inode number to get
 *
 * iget_locked() uses ifind_fast() to search for the inode specified by @ino in
 * the inode cache and if present it is returned with an increased reference
 * count. This is for file systems where the inode number is sufficient for
 * unique identification of an inode.
 *
 * If the inode is not in cache, get_new_inode_fast() is called to allocate a
 * new inode and this is returned locked, hashed, and with the I_NEW flag set.
 * The file system gets to fill it in before unlocking it via
 * unlock_new_inode().
 */
struct inode *iget_locked(struct super_block *sb, unsigned long ino)
{
	struct hlist_head *head = inode_hashtable + hash(sb, ino);
	struct inode *inode;

	inode = ifind_fast(sb, head, ino);
	if (inode)
		return inode;
	/*
	 * get_new_inode_fast() will do the right thing, re-trying the search
	 * in case it had to block at any point.
	 */
	return get_new_inode_fast(sb, head, ino);
}
EXPORT_SYMBOL(iget_locked);

1210 1211 1212 1213 1214 1215
int insert_inode_locked(struct inode *inode)
{
	struct super_block *sb = inode->i_sb;
	ino_t ino = inode->i_ino;
	struct hlist_head *head = inode_hashtable + hash(sb, ino);

Christoph Hellwig's avatar
Christoph Hellwig committed
1216
	inode->i_state |= I_NEW;
1217
	while (1) {
1218 1219
		struct hlist_node *node;
		struct inode *old = NULL;
1220
		spin_lock(&inode_lock);
1221 1222 1223 1224 1225
		hlist_for_each_entry(old, node, head, i_hash) {
			if (old->i_ino != ino)
				continue;
			if (old->i_sb != sb)
				continue;
1226
			if (old->i_state & (I_FREEING|I_WILL_FREE))
1227 1228 1229 1230
				continue;
			break;
		}
		if (likely(!node)) {
1231 1232 1233 1234 1235 1236 1237
			hlist_add_head(&inode->i_hash, head);
			spin_unlock(&inode_lock);
			return 0;
		}
		__iget(old);
		spin_unlock(&inode_lock);
		wait_on_inode(old);
Al Viro's avatar
Al Viro committed
1238
		if (unlikely(!inode_unhashed(old))) {
1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252
			iput(old);
			return -EBUSY;
		}
		iput(old);
	}
}
EXPORT_SYMBOL(insert_inode_locked);

int insert_inode_locked4(struct inode *inode, unsigned long hashval,
		int (*test)(struct inode *, void *), void *data)
{
	struct super_block *sb = inode->i_sb;
	struct hlist_head *head = inode_hashtable + hash(sb, hashval);

Christoph Hellwig's avatar
Christoph Hellwig committed
1253
	inode->i_state |= I_NEW;
1254 1255

	while (1) {
1256 1257 1258
		struct hlist_node *node;
		struct inode *old = NULL;

1259
		spin_lock(&inode_lock);
1260 1261 1262 1263 1264
		hlist_for_each_entry(old, node, head, i_hash) {
			if (old->i_sb != sb)
				continue;
			if (!test(old, data))
				continue;
1265
			if (old->i_state & (I_FREEING|I_WILL_FREE))
1266 1267 1268 1269
				continue;
			break;
		}
		if (likely(!node)) {
1270 1271 1272 1273 1274 1275 1276
			hlist_add_head(&inode->i_hash, head);
			spin_unlock(&inode_lock);
			return 0;
		}
		__iget(old);
		spin_unlock(&inode_lock);
		wait_on_inode(old);
Al Viro's avatar
Al Viro committed
1277
		if (unlikely(!inode_unhashed(old))) {
1278 1279 1280 1281 1282 1283 1284 1285
			iput(old);
			return -EBUSY;
		}
		iput(old);
	}
}
EXPORT_SYMBOL(insert_inode_locked4);

Linus Torvalds's avatar
Linus Torvalds committed
1286

1287 1288 1289 1290 1291 1292
int generic_delete_inode(struct inode *inode)
{
	return 1;
}
EXPORT_SYMBOL(generic_delete_inode);

Linus Torvalds's avatar
Linus Torvalds committed
1293
/*
1294 1295 1296
 * Normal UNIX filesystem behaviour: delete the
 * inode when the usage count drops to zero, and
 * i_nlink is zero.
Linus Torvalds's avatar
Linus Torvalds committed
1297
 */
1298
int generic_drop_inode(struct inode *inode)
Linus Torvalds's avatar
Linus Torvalds committed
1299
{
Al Viro's avatar
Al Viro committed
1300
	return !inode->i_nlink || inode_unhashed(inode);
Linus Torvalds's avatar
Linus Torvalds committed
1301
}
1302
EXPORT_SYMBOL_GPL(generic_drop_inode);
Linus Torvalds's avatar
Linus Torvalds committed
1303