inode.c 42.2 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
/*
 * 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/quotaops.h>
#include <linux/slab.h>
#include <linux/writeback.h>
#include <linux/module.h>
#include <linux/backing-dev.h>
#include <linux/wait.h>
N
Nick Piggin 已提交
17
#include <linux/rwsem.h>
L
Linus Torvalds 已提交
18 19 20
#include <linux/hash.h>
#include <linux/swap.h>
#include <linux/security.h>
M
Mimi Zohar 已提交
21
#include <linux/ima.h>
L
Linus Torvalds 已提交
22 23 24
#include <linux/pagemap.h>
#include <linux/cdev.h>
#include <linux/bootmem.h>
R
Robert Love 已提交
25
#include <linux/inotify.h>
26
#include <linux/fsnotify.h>
27
#include <linux/mount.h>
28
#include <linux/async.h>
A
Al Viro 已提交
29
#include <linux/posix_acl.h>
L
Linus Torvalds 已提交
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 59 60 61 62

/*
 * This is needed for the following functions:
 *  - inode_has_buffers
 *  - invalidate_inode_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

63 64
static unsigned int i_hash_mask __read_mostly;
static unsigned int i_hash_shift __read_mostly;
L
Linus Torvalds 已提交
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79

/*
 * 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.
 */

LIST_HEAD(inode_in_use);
LIST_HEAD(inode_unused);
80
static struct hlist_head *inode_hashtable __read_mostly;
L
Linus Torvalds 已提交
81 82 83 84 85 86 87 88 89 90

/*
 * 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);

/*
N
Nick Piggin 已提交
91
 * iprune_sem provides exclusion between the kswapd or try_to_free_pages
L
Linus Torvalds 已提交
92 93 94 95 96
 * 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.
N
Nick Piggin 已提交
97 98 99 100
 *
 * 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.
L
Linus Torvalds 已提交
101
 */
N
Nick Piggin 已提交
102
static DECLARE_RWSEM(iprune_sem);
L
Linus Torvalds 已提交
103 104 105 106 107 108

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

109
static struct kmem_cache *inode_cachep __read_mostly;
L
Linus Torvalds 已提交
110

J
Joern Engel 已提交
111 112 113 114 115 116 117 118 119
static void wake_up_inode(struct inode *inode)
{
	/*
	 * Prevent speculative execution through spin_unlock(&inode_lock);
	 */
	smp_mb();
	wake_up_bit(&inode->i_state, __I_LOCK);
}

D
David Chinner 已提交
120 121
/**
 * inode_init_always - perform inode structure intialisation
R
Randy Dunlap 已提交
122 123
 * @sb: superblock inode belongs to
 * @inode: inode to initialise
D
David Chinner 已提交
124 125 126 127
 *
 * These are initializations that need to be done on every inode
 * allocation as the fields are not initialised by slab allocation.
 */
128
int inode_init_always(struct super_block *sb, struct inode *inode)
L
Linus Torvalds 已提交
129
{
130
	static const struct address_space_operations empty_aops;
131
	static const struct inode_operations empty_iops;
132
	static const struct file_operations empty_fops;
133
	struct address_space *const mapping = &inode->i_data;
D
David Chinner 已提交
134 135 136 137 138 139 140 141

	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;
A
Al Viro 已提交
142 143
	inode->i_uid = 0;
	inode->i_gid = 0;
D
David Chinner 已提交
144 145 146 147 148
	atomic_set(&inode->i_writecount, 0);
	inode->i_size = 0;
	inode->i_blocks = 0;
	inode->i_bytes = 0;
	inode->i_generation = 0;
L
Linus Torvalds 已提交
149
#ifdef CONFIG_QUOTA
D
David Chinner 已提交
150
	memset(&inode->i_dquot, 0, sizeof(inode->i_dquot));
L
Linus Torvalds 已提交
151
#endif
D
David Chinner 已提交
152 153 154 155 156
	inode->i_pipe = NULL;
	inode->i_bdev = NULL;
	inode->i_cdev = NULL;
	inode->i_rdev = 0;
	inode->dirtied_when = 0;
M
Mimi Zohar 已提交
157 158

	if (security_inode_alloc(inode))
159
		goto out;
M
Mimi Zohar 已提交
160 161 162 163

	/* allocate and initialize an i_integrity */
	if (ima_inode_alloc(inode))
		goto out_free_security;
D
David Chinner 已提交
164 165 166 167 168 169 170 171 172 173 174 175 176

	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;
H
Hugh Dickins 已提交
177
	mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE);
D
David Chinner 已提交
178 179 180 181 182 183 184 185 186 187 188 189
	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;

190
		bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info;
D
David Chinner 已提交
191 192 193 194
		mapping->backing_dev_info = bdi;
	}
	inode->i_private = NULL;
	inode->i_mapping = mapping;
A
Al Viro 已提交
195 196 197
#ifdef CONFIG_FS_POSIX_ACL
	inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED;
#endif
D
David Chinner 已提交
198

199 200 201 202
#ifdef CONFIG_FSNOTIFY
	inode->i_fsnotify_mask = 0;
#endif

203
	return 0;
M
Mimi Zohar 已提交
204 205 206

out_free_security:
	security_inode_free(inode);
207 208
out:
	return -ENOMEM;
L
Linus Torvalds 已提交
209
}
D
David Chinner 已提交
210 211 212 213 214 215 216 217 218 219 220
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);

221 222 223 224 225 226 227 228 229 230 231 232
	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;
D
David Chinner 已提交
233
}
L
Linus Torvalds 已提交
234

C
Christoph Hellwig 已提交
235
void __destroy_inode(struct inode *inode)
L
Linus Torvalds 已提交
236
{
237
	BUG_ON(inode_has_buffers(inode));
H
Hugh Dickins 已提交
238
	ima_inode_free(inode);
L
Linus Torvalds 已提交
239
	security_inode_free(inode);
240
	fsnotify_inode_delete(inode);
A
Al Viro 已提交
241 242 243 244 245 246
#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
C
Christoph Hellwig 已提交
247 248 249 250 251 252
}
EXPORT_SYMBOL(__destroy_inode);

void destroy_inode(struct inode *inode)
{
	__destroy_inode(inode);
L
Linus Torvalds 已提交
253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270
	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);
	INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
N
Nick Piggin 已提交
271
	spin_lock_init(&inode->i_data.tree_lock);
L
Linus Torvalds 已提交
272 273 274 275 276 277
	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);
R
Robert Love 已提交
278 279
#ifdef CONFIG_INOTIFY
	INIT_LIST_HEAD(&inode->inotify_watches);
I
Ingo Molnar 已提交
280
	mutex_init(&inode->inotify_mutex);
R
Robert Love 已提交
281
#endif
282 283 284
#ifdef CONFIG_FSNOTIFY
	INIT_HLIST_HEAD(&inode->i_fsnotify_mark_entries);
#endif
L
Linus Torvalds 已提交
285 286 287
}
EXPORT_SYMBOL(inode_init_once);

288
static void init_once(void *foo)
L
Linus Torvalds 已提交
289
{
290
	struct inode *inode = (struct inode *) foo;
L
Linus Torvalds 已提交
291

C
Christoph Lameter 已提交
292
	inode_init_once(inode);
L
Linus Torvalds 已提交
293 294 295 296 297
}

/*
 * inode_lock must be held
 */
298
void __iget(struct inode *inode)
L
Linus Torvalds 已提交
299 300 301 302 303 304
{
	if (atomic_read(&inode->i_count)) {
		atomic_inc(&inode->i_count);
		return;
	}
	atomic_inc(&inode->i_count);
J
Joern Engel 已提交
305
	if (!(inode->i_state & (I_DIRTY|I_SYNC)))
L
Linus Torvalds 已提交
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321
		list_move(&inode->i_list, &inode_in_use);
	inodes_stat.nr_unused--;
}

/**
 * clear_inode - clear an inode
 * @inode: inode to clear
 *
 * This is called by the filesystem to tell us
 * that the inode is no longer useful. We just
 * terminate it with extreme prejudice.
 */
void clear_inode(struct inode *inode)
{
	might_sleep();
	invalidate_inode_buffers(inode);
322

323 324 325
	BUG_ON(inode->i_data.nrpages);
	BUG_ON(!(inode->i_state & I_FREEING));
	BUG_ON(inode->i_state & I_CLEAR);
J
Joern Engel 已提交
326
	inode_sync_wait(inode);
327
	vfs_dq_drop(inode);
328
	if (inode->i_sb->s_op->clear_inode)
L
Linus Torvalds 已提交
329
		inode->i_sb->s_op->clear_inode(inode);
330
	if (S_ISBLK(inode->i_mode) && inode->i_bdev)
L
Linus Torvalds 已提交
331
		bd_forget(inode);
332
	if (S_ISCHR(inode->i_mode) && inode->i_cdev)
L
Linus Torvalds 已提交
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
		cd_forget(inode);
	inode->i_state = I_CLEAR;
}
EXPORT_SYMBOL(clear_inode);

/*
 * 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)
{
	int nr_disposed = 0;

	while (!list_empty(head)) {
		struct inode *inode;

352
		inode = list_first_entry(head, struct inode, i_list);
L
Linus Torvalds 已提交
353 354 355 356 357
		list_del(&inode->i_list);

		if (inode->i_data.nrpages)
			truncate_inode_pages(&inode->i_data, 0);
		clear_inode(inode);
358 359 360 361 362 363 364

		spin_lock(&inode_lock);
		hlist_del_init(&inode->i_hash);
		list_del_init(&inode->i_sb_list);
		spin_unlock(&inode_lock);

		wake_up_inode(inode);
L
Linus Torvalds 已提交
365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
		destroy_inode(inode);
		nr_disposed++;
	}
	spin_lock(&inode_lock);
	inodes_stat.nr_inodes -= nr_disposed;
	spin_unlock(&inode_lock);
}

/*
 * Invalidate all inodes for a device.
 */
static int invalidate_list(struct list_head *head, struct list_head *dispose)
{
	struct list_head *next;
	int busy = 0, count = 0;

	next = head->next;
	for (;;) {
383 384
		struct list_head *tmp = next;
		struct inode *inode;
L
Linus Torvalds 已提交
385 386 387 388

		/*
		 * We can reschedule here without worrying about the list's
		 * consistency because the per-sb list of inodes must not
N
Nick Piggin 已提交
389
		 * change during umount anymore, and because iprune_sem keeps
L
Linus Torvalds 已提交
390 391 392 393 394 395 396 397
		 * 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);
N
Nick Piggin 已提交
398 399
		if (inode->i_state & I_NEW)
			continue;
L
Linus Torvalds 已提交
400 401 402
		invalidate_inode_buffers(inode);
		if (!atomic_read(&inode->i_count)) {
			list_move(&inode->i_list, dispose);
403
			WARN_ON(inode->i_state & I_NEW);
L
Linus Torvalds 已提交
404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
			inode->i_state |= I_FREEING;
			count++;
			continue;
		}
		busy = 1;
	}
	/* only unused inodes may be cached with i_count zero */
	inodes_stat.nr_unused -= count;
	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.
 */
423
int invalidate_inodes(struct super_block *sb)
L
Linus Torvalds 已提交
424 425 426 427
{
	int busy;
	LIST_HEAD(throw_away);

N
Nick Piggin 已提交
428
	down_write(&iprune_sem);
L
Linus Torvalds 已提交
429
	spin_lock(&inode_lock);
R
Robert Love 已提交
430
	inotify_unmount_inodes(&sb->s_inodes);
431
	fsnotify_unmount_inodes(&sb->s_inodes);
L
Linus Torvalds 已提交
432 433 434 435
	busy = invalidate_list(&sb->s_inodes, &throw_away);
	spin_unlock(&inode_lock);

	dispose_list(&throw_away);
N
Nick Piggin 已提交
436
	up_write(&iprune_sem);
L
Linus Torvalds 已提交
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474

	return busy;
}
EXPORT_SYMBOL(invalidate_inodes);

static int can_unuse(struct inode *inode)
{
	if (inode->i_state)
		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;
}

/*
 * 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().
 *
 * Any inodes which are pinned purely because of attached pagecache have their
 * pagecache removed.  We expect the final iput() on that inode to add it to
 * the front of the inode_unused list.  So look for it there and if the
 * inode is still freeable, proceed.  The right inode is found 99.9% of the
 * time in testing on a 4-way.
 *
 * If the inode has metadata buffers attached to mapping->private_list then
 * try to remove them.
 */
static void prune_icache(int nr_to_scan)
{
	LIST_HEAD(freeable);
	int nr_pruned = 0;
	int nr_scanned;
	unsigned long reap = 0;

N
Nick Piggin 已提交
475
	down_read(&iprune_sem);
L
Linus Torvalds 已提交
476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492
	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);

		if (inode->i_state || atomic_read(&inode->i_count)) {
			list_move(&inode->i_list, &inode_unused);
			continue;
		}
		if (inode_has_buffers(inode) || inode->i_data.nrpages) {
			__iget(inode);
			spin_unlock(&inode_lock);
			if (remove_inode_buffers(inode))
493 494
				reap += invalidate_mapping_pages(&inode->i_data,
								0, -1);
L
Linus Torvalds 已提交
495 496 497 498 499 500 501 502 503 504
			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);
505
		WARN_ON(inode->i_state & I_NEW);
L
Linus Torvalds 已提交
506 507 508 509
		inode->i_state |= I_FREEING;
		nr_pruned++;
	}
	inodes_stat.nr_unused -= nr_pruned;
510 511 512 513
	if (current_is_kswapd())
		__count_vm_events(KSWAPD_INODESTEAL, reap);
	else
		__count_vm_events(PGINODESTEAL, reap);
L
Linus Torvalds 已提交
514 515 516
	spin_unlock(&inode_lock);

	dispose_list(&freeable);
N
Nick Piggin 已提交
517
	up_read(&iprune_sem);
L
Linus Torvalds 已提交
518 519 520 521 522 523 524 525 526 527 528
}

/*
 * 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.
 */
A
Al Viro 已提交
529
static int shrink_icache_memory(int nr, gfp_t gfp_mask)
L
Linus Torvalds 已提交
530 531 532 533 534 535
{
	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..
536
		 */
L
Linus Torvalds 已提交
537 538 539 540 541 542 543
		if (!(gfp_mask & __GFP_FS))
			return -1;
		prune_icache(nr);
	}
	return (inodes_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
}

544 545 546 547 548
static struct shrinker icache_shrinker = {
	.shrink = shrink_icache_memory,
	.seeks = DEFAULT_SEEKS,
};

L
Linus Torvalds 已提交
549 550 551 552 553 554 555
static void __wait_on_freeing_inode(struct inode *inode);
/*
 * Called with the inode lock held.
 * NOTE: we are not increasing the inode-refcount, you must call __iget()
 * by hand after calling find_inode now! This simplifies iunique and won't
 * add any additional branch in the common code.
 */
556 557 558 559
static struct inode *find_inode(struct super_block *sb,
				struct hlist_head *head,
				int (*test)(struct inode *, void *),
				void *data)
L
Linus Torvalds 已提交
560 561
{
	struct hlist_node *node;
562
	struct inode *inode = NULL;
L
Linus Torvalds 已提交
563 564

repeat:
565
	hlist_for_each_entry(inode, node, head, i_hash) {
L
Linus Torvalds 已提交
566 567 568 569
		if (inode->i_sb != sb)
			continue;
		if (!test(inode, data))
			continue;
570
		if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
L
Linus Torvalds 已提交
571 572 573 574 575 576 577 578 579 580 581 582
			__wait_on_freeing_inode(inode);
			goto repeat;
		}
		break;
	}
	return node ? inode : NULL;
}

/*
 * find_inode_fast is the fast path version of find_inode, see the comment at
 * iget_locked for details.
 */
583 584
static struct inode *find_inode_fast(struct super_block *sb,
				struct hlist_head *head, unsigned long ino)
L
Linus Torvalds 已提交
585 586
{
	struct hlist_node *node;
587
	struct inode *inode = NULL;
L
Linus Torvalds 已提交
588 589

repeat:
590
	hlist_for_each_entry(inode, node, head, i_hash) {
L
Linus Torvalds 已提交
591 592 593 594
		if (inode->i_ino != ino)
			continue;
		if (inode->i_sb != sb)
			continue;
595
		if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
L
Linus Torvalds 已提交
596 597 598 599 600 601 602 603
			__wait_on_freeing_inode(inode);
			goto repeat;
		}
		break;
	}
	return node ? inode : NULL;
}

604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626
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;
}

static inline void
__inode_add_to_lists(struct super_block *sb, struct hlist_head *head,
			struct inode *inode)
{
	inodes_stat.nr_inodes++;
	list_add(&inode->i_list, &inode_in_use);
	list_add(&inode->i_sb_list, &sb->s_inodes);
	if (head)
		hlist_add_head(&inode->i_hash, head);
}

/**
 * inode_add_to_lists - add a new inode to relevant lists
R
Randy Dunlap 已提交
627 628
 * @sb: superblock inode belongs to
 * @inode: inode to mark in use
629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
 *
 * When an inode is allocated it needs to be accounted for, added to the in use
 * list, the owning superblock and the inode hash. This needs to be done under
 * the inode_lock, so export a function to do this rather than the inode lock
 * itself. We calculate the hash list to add to here so it is all internal
 * which requires the caller to have already set up the inode number in the
 * inode to add.
 */
void inode_add_to_lists(struct super_block *sb, struct inode *inode)
{
	struct hlist_head *head = inode_hashtable + hash(sb, inode->i_ino);

	spin_lock(&inode_lock);
	__inode_add_to_lists(sb, head, inode);
	spin_unlock(&inode_lock);
}
EXPORT_SYMBOL_GPL(inode_add_to_lists);

L
Linus Torvalds 已提交
647 648 649 650
/**
 *	new_inode 	- obtain an inode
 *	@sb: superblock
 *
651
 *	Allocates a new inode for given superblock. The default gfp_mask
H
Hugh Dickins 已提交
652
 *	for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE.
653 654 655 656 657
 *	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
 *
L
Linus Torvalds 已提交
658 659 660
 */
struct inode *new_inode(struct super_block *sb)
{
661 662 663 664 665 666
	/*
	 * 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.
	 */
	static unsigned int last_ino;
667
	struct inode *inode;
L
Linus Torvalds 已提交
668 669

	spin_lock_prefetch(&inode_lock);
670

L
Linus Torvalds 已提交
671 672 673
	inode = alloc_inode(sb);
	if (inode) {
		spin_lock(&inode_lock);
674
		__inode_add_to_lists(sb, NULL, inode);
L
Linus Torvalds 已提交
675 676 677 678 679 680 681 682 683 684
		inode->i_ino = ++last_ino;
		inode->i_state = 0;
		spin_unlock(&inode_lock);
	}
	return inode;
}
EXPORT_SYMBOL(new_inode);

void unlock_new_inode(struct inode *inode)
{
685
#ifdef CONFIG_DEBUG_LOCK_ALLOC
686 687 688
	if (inode->i_mode & S_IFDIR) {
		struct file_system_type *type = inode->i_sb->s_type;

689 690 691 692 693 694 695 696 697 698 699
		/* 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);
		}
700
	}
701
#endif
L
Linus Torvalds 已提交
702
	/*
703 704 705 706 707 708 709
	 * This is special!  We do not need the spinlock when clearing I_LOCK,
	 * 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
	 * there can be no old holders that haven't tested I_LOCK).
	 * However we must emit the memory barrier so that other CPUs reliably
	 * see the clearing of I_LOCK after the other inode initialisation has
	 * completed.
L
Linus Torvalds 已提交
710
	 */
711
	smp_mb();
712
	WARN_ON((inode->i_state & (I_LOCK|I_NEW)) != (I_LOCK|I_NEW));
L
Linus Torvalds 已提交
713 714 715 716 717 718 719 720 721 722 723
	inode->i_state &= ~(I_LOCK|I_NEW);
	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
 */
724 725 726 727 728
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)
L
Linus Torvalds 已提交
729
{
730
	struct inode *inode;
L
Linus Torvalds 已提交
731 732 733

	inode = alloc_inode(sb);
	if (inode) {
734
		struct inode *old;
L
Linus Torvalds 已提交
735 736 737 738 739 740 741 742

		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;

743
			__inode_add_to_lists(sb, head, inode);
L
Linus Torvalds 已提交
744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775
			inode->i_state = I_LOCK|I_NEW;
			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.
		 */
		__iget(old);
		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.
 */
776 777
static struct inode *get_new_inode_fast(struct super_block *sb,
				struct hlist_head *head, unsigned long ino)
L
Linus Torvalds 已提交
778
{
779
	struct inode *inode;
L
Linus Torvalds 已提交
780 781 782

	inode = alloc_inode(sb);
	if (inode) {
783
		struct inode *old;
L
Linus Torvalds 已提交
784 785 786 787 788 789

		spin_lock(&inode_lock);
		/* We released the lock, so.. */
		old = find_inode_fast(sb, head, ino);
		if (!old) {
			inode->i_ino = ino;
790
			__inode_add_to_lists(sb, head, inode);
L
Linus Torvalds 已提交
791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829
			inode->i_state = I_LOCK|I_NEW;
			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.
		 */
		__iget(old);
		spin_unlock(&inode_lock);
		destroy_inode(inode);
		inode = old;
		wait_on_inode(inode);
	}
	return inode;
}

/**
 *	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)
{
830 831 832 833 834 835
	/*
	 * 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.
	 */
	static unsigned int counter;
L
Linus Torvalds 已提交
836
	struct inode *inode;
837
	struct hlist_head *head;
L
Linus Torvalds 已提交
838
	ino_t res;
839

L
Linus Torvalds 已提交
840
	spin_lock(&inode_lock);
841 842 843
	do {
		if (counter <= max_reserved)
			counter = max_reserved + 1;
L
Linus Torvalds 已提交
844
		res = counter++;
845
		head = inode_hashtable + hash(sb, res);
L
Linus Torvalds 已提交
846
		inode = find_inode_fast(sb, head, res);
847 848
	} while (inode != NULL);
	spin_unlock(&inode_lock);
L
Linus Torvalds 已提交
849

850 851
	return res;
}
L
Linus Torvalds 已提交
852 853 854 855 856
EXPORT_SYMBOL(iunique);

struct inode *igrab(struct inode *inode)
{
	spin_lock(&inode_lock);
857
	if (!(inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)))
L
Linus Torvalds 已提交
858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
		__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
877
 * @wait:	if true wait for the inode to be unlocked, if false do not
L
Linus Torvalds 已提交
878 879 880 881 882 883 884 885 886 887 888 889
 *
 * 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.
 */
890
static struct inode *ifind(struct super_block *sb,
L
Linus Torvalds 已提交
891
		struct hlist_head *head, int (*test)(struct inode *, void *),
892
		void *data, const int wait)
L
Linus Torvalds 已提交
893 894 895 896 897 898 899 900
{
	struct inode *inode;

	spin_lock(&inode_lock);
	inode = find_inode(sb, head, test, data);
	if (inode) {
		__iget(inode);
		spin_unlock(&inode_lock);
901 902
		if (likely(wait))
			wait_on_inode(inode);
L
Linus Torvalds 已提交
903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923
		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.
 */
924
static struct inode *ifind_fast(struct super_block *sb,
L
Linus Torvalds 已提交
925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941
		struct hlist_head *head, unsigned long ino)
{
	struct inode *inode;

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

/**
942
 * ilookup5_nowait - search for an inode in the inode cache
L
Linus Torvalds 已提交
943 944 945 946 947 948 949 950 951 952 953
 * @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
954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984
 * 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.
L
Linus Torvalds 已提交
985 986 987 988 989 990 991 992 993 994
 *
 * 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);

995
	return ifind(sb, head, test, data, 1);
L
Linus Torvalds 已提交
996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
}
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;

1048
	inode = ifind(sb, head, test, data, 1);
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
	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);

A
Al Viro 已提交
1090 1091 1092 1093 1094 1095 1096 1097
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);

	inode->i_state |= I_LOCK|I_NEW;
	while (1) {
1098 1099
		struct hlist_node *node;
		struct inode *old = NULL;
A
Al Viro 已提交
1100
		spin_lock(&inode_lock);
1101 1102 1103 1104 1105 1106 1107 1108 1109 1110
		hlist_for_each_entry(old, node, head, i_hash) {
			if (old->i_ino != ino)
				continue;
			if (old->i_sb != sb)
				continue;
			if (old->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE))
				continue;
			break;
		}
		if (likely(!node)) {
A
Al Viro 已提交
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135
			hlist_add_head(&inode->i_hash, head);
			spin_unlock(&inode_lock);
			return 0;
		}
		__iget(old);
		spin_unlock(&inode_lock);
		wait_on_inode(old);
		if (unlikely(!hlist_unhashed(&old->i_hash))) {
			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);

	inode->i_state |= I_LOCK|I_NEW;

	while (1) {
1136 1137 1138
		struct hlist_node *node;
		struct inode *old = NULL;

A
Al Viro 已提交
1139
		spin_lock(&inode_lock);
1140 1141 1142 1143 1144 1145 1146 1147 1148 1149
		hlist_for_each_entry(old, node, head, i_hash) {
			if (old->i_sb != sb)
				continue;
			if (!test(old, data))
				continue;
			if (old->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE))
				continue;
			break;
		}
		if (likely(!node)) {
A
Al Viro 已提交
1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
			hlist_add_head(&inode->i_hash, head);
			spin_unlock(&inode_lock);
			return 0;
		}
		__iget(old);
		spin_unlock(&inode_lock);
		wait_on_inode(old);
		if (unlikely(!hlist_unhashed(&old->i_hash))) {
			iput(old);
			return -EBUSY;
		}
		iput(old);
	}
}
EXPORT_SYMBOL(insert_inode_locked4);

L
Linus Torvalds 已提交
1166 1167 1168 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
/**
 *	__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)
{
	struct hlist_head *head = inode_hashtable + hash(inode->i_sb, hashval);
	spin_lock(&inode_lock);
	hlist_add_head(&inode->i_hash, head);
	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.
 */
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);

/*
 * Tell the filesystem that this inode is no longer of any interest and should
 * be completely destroyed.
 *
 * We leave the inode in the inode hash table until *after* the filesystem's
 * ->delete_inode completes.  This ensures that an iget (such as nfsd might
 * instigate) will always find up-to-date information either in the hash or on
 * disk.
 *
 * I_FREEING is set so that no-one will take a new reference to the inode while
 * it is being deleted.
 */
1209
void generic_delete_inode(struct inode *inode)
L
Linus Torvalds 已提交
1210
{
1211
	const struct super_operations *op = inode->i_sb->s_op;
L
Linus Torvalds 已提交
1212

1213 1214
	list_del_init(&inode->i_list);
	list_del_init(&inode->i_sb_list);
1215
	WARN_ON(inode->i_state & I_NEW);
1216 1217 1218 1219
	inode->i_state |= I_FREEING;
	inodes_stat.nr_inodes--;
	spin_unlock(&inode_lock);

L
Linus Torvalds 已提交
1220 1221 1222 1223 1224
	security_inode_delete(inode);

	if (op->delete_inode) {
		void (*delete)(struct inode *) = op->delete_inode;
		if (!is_bad_inode(inode))
1225
			vfs_dq_init(inode);
1226 1227 1228 1229
		/* Filesystems implementing their own
		 * s_op->delete_inode are required to call
		 * truncate_inode_pages and clear_inode()
		 * internally */
L
Linus Torvalds 已提交
1230
		delete(inode);
1231 1232
	} else {
		truncate_inode_pages(&inode->i_data, 0);
L
Linus Torvalds 已提交
1233
		clear_inode(inode);
1234
	}
L
Linus Torvalds 已提交
1235 1236 1237 1238
	spin_lock(&inode_lock);
	hlist_del_init(&inode->i_hash);
	spin_unlock(&inode_lock);
	wake_up_inode(inode);
1239
	BUG_ON(inode->i_state != I_CLEAR);
L
Linus Torvalds 已提交
1240 1241 1242 1243
	destroy_inode(inode);
}
EXPORT_SYMBOL(generic_delete_inode);

1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
/**
 *	generic_detach_inode - remove inode from inode lists
 *	@inode: inode to remove
 *
 *	Remove inode from inode lists, write it if it's dirty. This is just an
 *	internal VFS helper exported for hugetlbfs. Do not use!
 *
 *	Returns 1 if inode should be completely destroyed.
 */
int generic_detach_inode(struct inode *inode)
L
Linus Torvalds 已提交
1254 1255 1256 1257
{
	struct super_block *sb = inode->i_sb;

	if (!hlist_unhashed(&inode->i_hash)) {
J
Joern Engel 已提交
1258
		if (!(inode->i_state & (I_DIRTY|I_SYNC)))
L
Linus Torvalds 已提交
1259 1260
			list_move(&inode->i_list, &inode_unused);
		inodes_stat.nr_unused++;
1261
		if (sb->s_flags & MS_ACTIVE) {
1262
			spin_unlock(&inode_lock);
1263
			return 0;
1264
		}
1265
		WARN_ON(inode->i_state & I_NEW);
1266 1267
		inode->i_state |= I_WILL_FREE;
		spin_unlock(&inode_lock);
L
Linus Torvalds 已提交
1268 1269
		write_inode_now(inode, 1);
		spin_lock(&inode_lock);
1270
		WARN_ON(inode->i_state & I_NEW);
1271
		inode->i_state &= ~I_WILL_FREE;
L
Linus Torvalds 已提交
1272 1273 1274 1275 1276
		inodes_stat.nr_unused--;
		hlist_del_init(&inode->i_hash);
	}
	list_del_init(&inode->i_list);
	list_del_init(&inode->i_sb_list);
1277
	WARN_ON(inode->i_state & I_NEW);
1278
	inode->i_state |= I_FREEING;
L
Linus Torvalds 已提交
1279 1280
	inodes_stat.nr_inodes--;
	spin_unlock(&inode_lock);
1281 1282 1283 1284 1285 1286 1287 1288
	return 1;
}
EXPORT_SYMBOL_GPL(generic_detach_inode);

static void generic_forget_inode(struct inode *inode)
{
	if (!generic_detach_inode(inode))
		return;
L
Linus Torvalds 已提交
1289 1290 1291
	if (inode->i_data.nrpages)
		truncate_inode_pages(&inode->i_data, 0);
	clear_inode(inode);
1292
	wake_up_inode(inode);
L
Linus Torvalds 已提交
1293 1294 1295 1296 1297 1298 1299 1300
	destroy_inode(inode);
}

/*
 * Normal UNIX filesystem behaviour: delete the
 * inode when the usage count drops to zero, and
 * i_nlink is zero.
 */
1301
void generic_drop_inode(struct inode *inode)
L
Linus Torvalds 已提交
1302 1303 1304 1305 1306 1307
{
	if (!inode->i_nlink)
		generic_delete_inode(inode);
	else
		generic_forget_inode(inode);
}
1308 1309
EXPORT_SYMBOL_GPL(generic_drop_inode);

L
Linus Torvalds 已提交
1310 1311
/*
 * Called when we're dropping the last reference
1312
 * to an inode.
L
Linus Torvalds 已提交
1313 1314 1315 1316 1317 1318 1319 1320 1321 1322
 *
 * Call the FS "drop()" function, defaulting to
 * the legacy UNIX filesystem behaviour..
 *
 * NOTE! NOTE! NOTE! We're called with the inode lock
 * held, and the drop function is supposed to release
 * the lock!
 */
static inline void iput_final(struct inode *inode)
{
1323
	const struct super_operations *op = inode->i_sb->s_op;
L
Linus Torvalds 已提交
1324 1325 1326 1327 1328 1329 1330 1331
	void (*drop)(struct inode *) = generic_drop_inode;

	if (op && op->drop_inode)
		drop = op->drop_inode;
	drop(inode);
}

/**
1332
 *	iput	- put an inode
L
Linus Torvalds 已提交
1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358
 *	@inode: inode to put
 *
 *	Puts an inode, dropping its usage count. If the inode use count hits
 *	zero, the inode is then freed and may also be destroyed.
 *
 *	Consequently, iput() can sleep.
 */
void iput(struct inode *inode)
{
	if (inode) {
		BUG_ON(inode->i_state == I_CLEAR);

		if (atomic_dec_and_lock(&inode->i_count, &inode_lock))
			iput_final(inode);
	}
}
EXPORT_SYMBOL(iput);

/**
 *	bmap	- find a block number in a file
 *	@inode: inode of file
 *	@block: block to find
 *
 *	Returns the block number on the device holding the inode that
 *	is the disk block number for the block of the file requested.
 *	That is, asked for block 4 of inode 1 the function will return the
1359
 *	disk block relative to the disk start that holds that block of the
L
Linus Torvalds 已提交
1360 1361
 *	file.
 */
1362
sector_t bmap(struct inode *inode, sector_t block)
L
Linus Torvalds 已提交
1363 1364 1365 1366 1367 1368 1369 1370
{
	sector_t res = 0;
	if (inode->i_mapping->a_ops->bmap)
		res = inode->i_mapping->a_ops->bmap(inode->i_mapping, block);
	return res;
}
EXPORT_SYMBOL(bmap);

1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404
/*
 * With relative atime, only update atime if the previous atime is
 * earlier than either the ctime or mtime or if at least a day has
 * passed since the last atime update.
 */
static int relatime_need_update(struct vfsmount *mnt, struct inode *inode,
			     struct timespec now)
{

	if (!(mnt->mnt_flags & MNT_RELATIME))
		return 1;
	/*
	 * Is mtime younger than atime? If yes, update atime:
	 */
	if (timespec_compare(&inode->i_mtime, &inode->i_atime) >= 0)
		return 1;
	/*
	 * Is ctime younger than atime? If yes, update atime:
	 */
	if (timespec_compare(&inode->i_ctime, &inode->i_atime) >= 0)
		return 1;

	/*
	 * Is the previous atime value older than a day? If yes,
	 * update atime:
	 */
	if ((long)(now.tv_sec - inode->i_atime.tv_sec) >= 24*60*60)
		return 1;
	/*
	 * Good, we can skip the atime update:
	 */
	return 0;
}

L
Linus Torvalds 已提交
1405
/**
C
Christoph Hellwig 已提交
1406 1407
 *	touch_atime	-	update the access time
 *	@mnt: mount the inode is accessed on
1408
 *	@dentry: dentry accessed
L
Linus Torvalds 已提交
1409 1410 1411 1412 1413
 *
 *	Update the accessed time on an inode and mark it for writeback.
 *	This function automatically handles read only file systems and media,
 *	as well as the "noatime" flag and inode specific "noatime" markers.
 */
C
Christoph Hellwig 已提交
1414
void touch_atime(struct vfsmount *mnt, struct dentry *dentry)
L
Linus Torvalds 已提交
1415
{
C
Christoph Hellwig 已提交
1416
	struct inode *inode = dentry->d_inode;
L
Linus Torvalds 已提交
1417 1418
	struct timespec now;

1419
	if (inode->i_flags & S_NOATIME)
A
Andi Kleen 已提交
1420
		return;
1421
	if (IS_NOATIME(inode))
A
Andi Kleen 已提交
1422
		return;
A
Andrew Morton 已提交
1423
	if ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode))
A
Andi Kleen 已提交
1424
		return;
V
Valerie Henson 已提交
1425

1426
	if (mnt->mnt_flags & MNT_NOATIME)
A
Andi Kleen 已提交
1427
		return;
1428
	if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode))
A
Andi Kleen 已提交
1429
		return;
L
Linus Torvalds 已提交
1430 1431

	now = current_fs_time(inode->i_sb);
1432 1433

	if (!relatime_need_update(mnt, inode, now))
A
Andi Kleen 已提交
1434
		return;
1435

V
Valerie Henson 已提交
1436
	if (timespec_equal(&inode->i_atime, &now))
A
Andi Kleen 已提交
1437 1438 1439 1440
		return;

	if (mnt_want_write(mnt))
		return;
V
Valerie Henson 已提交
1441 1442 1443

	inode->i_atime = now;
	mark_inode_dirty_sync(inode);
1444
	mnt_drop_write(mnt);
L
Linus Torvalds 已提交
1445
}
C
Christoph Hellwig 已提交
1446
EXPORT_SYMBOL(touch_atime);
L
Linus Torvalds 已提交
1447 1448

/**
1449 1450
 *	file_update_time	-	update mtime and ctime time
 *	@file: file accessed
L
Linus Torvalds 已提交
1451
 *
1452 1453 1454 1455
 *	Update the mtime and ctime members of an inode and mark the inode
 *	for writeback.  Note that this function is meant exclusively for
 *	usage in the file write path of filesystems, and filesystems may
 *	choose to explicitly ignore update via this function with the
1456
 *	S_NOCMTIME inode flag, e.g. for network filesystem where these
1457
 *	timestamps are handled by the server.
L
Linus Torvalds 已提交
1458 1459
 */

1460
void file_update_time(struct file *file)
L
Linus Torvalds 已提交
1461
{
1462
	struct inode *inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
1463 1464
	struct timespec now;
	int sync_it = 0;
1465
	int err;
L
Linus Torvalds 已提交
1466 1467 1468

	if (IS_NOCMTIME(inode))
		return;
1469

N
npiggin@suse.de 已提交
1470
	err = mnt_want_write_file(file);
1471
	if (err)
L
Linus Torvalds 已提交
1472 1473 1474
		return;

	now = current_fs_time(inode->i_sb);
A
Andreas Mohr 已提交
1475 1476
	if (!timespec_equal(&inode->i_mtime, &now)) {
		inode->i_mtime = now;
L
Linus Torvalds 已提交
1477
		sync_it = 1;
A
Andreas Mohr 已提交
1478
	}
L
Linus Torvalds 已提交
1479

A
Andreas Mohr 已提交
1480 1481
	if (!timespec_equal(&inode->i_ctime, &now)) {
		inode->i_ctime = now;
1482
		sync_it = 1;
A
Andreas Mohr 已提交
1483
	}
1484

1485 1486 1487 1488 1489
	if (IS_I_VERSION(inode)) {
		inode_inc_iversion(inode);
		sync_it = 1;
	}

L
Linus Torvalds 已提交
1490 1491
	if (sync_it)
		mark_inode_dirty_sync(inode);
1492
	mnt_drop_write(file->f_path.mnt);
L
Linus Torvalds 已提交
1493
}
1494
EXPORT_SYMBOL(file_update_time);
L
Linus Torvalds 已提交
1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510

int inode_needs_sync(struct inode *inode)
{
	if (IS_SYNC(inode))
		return 1;
	if (S_ISDIR(inode->i_mode) && IS_DIRSYNC(inode))
		return 1;
	return 0;
}
EXPORT_SYMBOL(inode_needs_sync);

int inode_wait(void *word)
{
	schedule();
	return 0;
}
1511
EXPORT_SYMBOL(inode_wait);
L
Linus Torvalds 已提交
1512 1513

/*
1514 1515 1516 1517 1518 1519 1520 1521 1522
 * If we try to find an inode in the inode hash while it is being
 * deleted, we have to wait until the filesystem completes its
 * deletion before reporting that it isn't found.  This function waits
 * until the deletion _might_ have completed.  Callers are responsible
 * to recheck inode state.
 *
 * It doesn't matter if I_LOCK is not set initially, a call to
 * wake_up_inode() after removing from the hash list will DTRT.
 *
L
Linus Torvalds 已提交
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
 * This is called with inode_lock held.
 */
static void __wait_on_freeing_inode(struct inode *inode)
{
	wait_queue_head_t *wq;
	DEFINE_WAIT_BIT(wait, &inode->i_state, __I_LOCK);
	wq = bit_waitqueue(&inode->i_state, __I_LOCK);
	prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
	spin_unlock(&inode_lock);
	schedule();
	finish_wait(wq, &wait.wait);
	spin_lock(&inode_lock);
}

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

/*
 * Initialize the waitqueues and inode hash table.
 */
void __init inode_init_early(void)
{
	int loop;

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

	inode_hashtable =
		alloc_large_system_hash("Inode-cache",
					sizeof(struct hlist_head),
					ihash_entries,
					14,
					HASH_EARLY,
					&i_hash_shift,
					&i_hash_mask,
					0);

	for (loop = 0; loop < (1 << i_hash_shift); loop++)
		INIT_HLIST_HEAD(&inode_hashtable[loop]);
}

1574
void __init inode_init(void)
L
Linus Torvalds 已提交
1575 1576 1577 1578
{
	int loop;

	/* inode slab cache */
1579 1580 1581 1582 1583
	inode_cachep = kmem_cache_create("inode_cache",
					 sizeof(struct inode),
					 0,
					 (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|
					 SLAB_MEM_SPREAD),
1584
					 init_once);
1585
	register_shrinker(&icache_shrinker);
L
Linus Torvalds 已提交
1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618

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

	inode_hashtable =
		alloc_large_system_hash("Inode-cache",
					sizeof(struct hlist_head),
					ihash_entries,
					14,
					0,
					&i_hash_shift,
					&i_hash_mask,
					0);

	for (loop = 0; loop < (1 << i_hash_shift); loop++)
		INIT_HLIST_HEAD(&inode_hashtable[loop]);
}

void init_special_inode(struct inode *inode, umode_t mode, dev_t rdev)
{
	inode->i_mode = mode;
	if (S_ISCHR(mode)) {
		inode->i_fop = &def_chr_fops;
		inode->i_rdev = rdev;
	} else if (S_ISBLK(mode)) {
		inode->i_fop = &def_blk_fops;
		inode->i_rdev = rdev;
	} else if (S_ISFIFO(mode))
		inode->i_fop = &def_fifo_fops;
	else if (S_ISSOCK(mode))
		inode->i_fop = &bad_sock_fops;
	else
1619 1620 1621
		printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o) for"
				  " inode %s:%lu\n", mode, inode->i_sb->s_id,
				  inode->i_ino);
L
Linus Torvalds 已提交
1622 1623
}
EXPORT_SYMBOL(init_special_inode);