locks.c 65.9 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 16 17 18 19 20 21 22 23 24 25 26 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 59 60 61 62
/*
 *  linux/fs/locks.c
 *
 *  Provide support for fcntl()'s F_GETLK, F_SETLK, and F_SETLKW calls.
 *  Doug Evans (dje@spiff.uucp), August 07, 1992
 *
 *  Deadlock detection added.
 *  FIXME: one thing isn't handled yet:
 *	- mandatory locks (requires lots of changes elsewhere)
 *  Kelly Carmichael (kelly@[142.24.8.65]), September 17, 1994.
 *
 *  Miscellaneous edits, and a total rewrite of posix_lock_file() code.
 *  Kai Petzke (wpp@marie.physik.tu-berlin.de), 1994
 *  
 *  Converted file_lock_table to a linked list from an array, which eliminates
 *  the limits on how many active file locks are open.
 *  Chad Page (pageone@netcom.com), November 27, 1994
 * 
 *  Removed dependency on file descriptors. dup()'ed file descriptors now
 *  get the same locks as the original file descriptors, and a close() on
 *  any file descriptor removes ALL the locks on the file for the current
 *  process. Since locks still depend on the process id, locks are inherited
 *  after an exec() but not after a fork(). This agrees with POSIX, and both
 *  BSD and SVR4 practice.
 *  Andy Walker (andy@lysaker.kvaerner.no), February 14, 1995
 *
 *  Scrapped free list which is redundant now that we allocate locks
 *  dynamically with kmalloc()/kfree().
 *  Andy Walker (andy@lysaker.kvaerner.no), February 21, 1995
 *
 *  Implemented two lock personalities - FL_FLOCK and FL_POSIX.
 *
 *  FL_POSIX locks are created with calls to fcntl() and lockf() through the
 *  fcntl() system call. They have the semantics described above.
 *
 *  FL_FLOCK locks are created with calls to flock(), through the flock()
 *  system call, which is new. Old C libraries implement flock() via fcntl()
 *  and will continue to use the old, broken implementation.
 *
 *  FL_FLOCK locks follow the 4.4 BSD flock() semantics. They are associated
 *  with a file pointer (filp). As a result they can be shared by a parent
 *  process and its children after a fork(). They are removed when the last
 *  file descriptor referring to the file pointer is closed (unless explicitly
 *  unlocked). 
 *
 *  FL_FLOCK locks never deadlock, an existing lock is always removed before
 *  upgrading from shared to exclusive (or vice versa). When this happens
 *  any processes blocked by the current lock are woken up and allowed to
 *  run before the new lock is applied.
 *  Andy Walker (andy@lysaker.kvaerner.no), June 09, 1995
 *
 *  Removed some race conditions in flock_lock_file(), marked other possible
 *  races. Just grep for FIXME to see them. 
 *  Dmitry Gorodchanin (pgmdsg@ibi.com), February 09, 1996.
 *
 *  Addressed Dmitry's concerns. Deadlock checking no longer recursive.
 *  Lock allocation changed to GFP_ATOMIC as we can't afford to sleep
 *  once we've checked for blocking and deadlocking.
 *  Andy Walker (andy@lysaker.kvaerner.no), April 03, 1996.
 *
 *  Initial implementation of mandatory locks. SunOS turned out to be
 *  a rotten model, so I implemented the "obvious" semantics.
Paul Bolle's avatar
Paul Bolle committed
63
 *  See 'Documentation/filesystems/mandatory-locking.txt' for details.
Linus Torvalds's avatar
Linus Torvalds committed
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
 *  Andy Walker (andy@lysaker.kvaerner.no), April 06, 1996.
 *
 *  Don't allow mandatory locks on mmap()'ed files. Added simple functions to
 *  check if a file has mandatory locks, used by mmap(), open() and creat() to
 *  see if system call should be rejected. Ref. HP-UX/SunOS/Solaris Reference
 *  Manual, Section 2.
 *  Andy Walker (andy@lysaker.kvaerner.no), April 09, 1996.
 *
 *  Tidied up block list handling. Added '/proc/locks' interface.
 *  Andy Walker (andy@lysaker.kvaerner.no), April 24, 1996.
 *
 *  Fixed deadlock condition for pathological code that mixes calls to
 *  flock() and fcntl().
 *  Andy Walker (andy@lysaker.kvaerner.no), April 29, 1996.
 *
 *  Allow only one type of locking scheme (FL_POSIX or FL_FLOCK) to be in use
 *  for a given file at a time. Changed the CONFIG_LOCK_MANDATORY scheme to
 *  guarantee sensible behaviour in the case where file system modules might
 *  be compiled with different options than the kernel itself.
 *  Andy Walker (andy@lysaker.kvaerner.no), May 15, 1996.
 *
 *  Added a couple of missing wake_up() calls. Thanks to Thomas Meckel
 *  (Thomas.Meckel@mni.fh-giessen.de) for spotting this.
 *  Andy Walker (andy@lysaker.kvaerner.no), May 15, 1996.
 *
 *  Changed FL_POSIX locks to use the block list in the same way as FL_FLOCK
 *  locks. Changed process synchronisation to avoid dereferencing locks that
 *  have already been freed.
 *  Andy Walker (andy@lysaker.kvaerner.no), Sep 21, 1996.
 *
 *  Made the block list a circular list to minimise searching in the list.
 *  Andy Walker (andy@lysaker.kvaerner.no), Sep 25, 1996.
 *
 *  Made mandatory locking a mount option. Default is not to allow mandatory
 *  locking.
 *  Andy Walker (andy@lysaker.kvaerner.no), Oct 04, 1996.
 *
 *  Some adaptations for NFS support.
 *  Olaf Kirch (okir@monad.swb.de), Dec 1996,
 *
 *  Fixed /proc/locks interface so that we can't overrun the buffer we are handed.
 *  Andy Walker (andy@lysaker.kvaerner.no), May 12, 1997.
 *
 *  Use slab allocator instead of kmalloc/kfree.
 *  Use generic list implementation from <linux/list.h>.
 *  Sped up posix_locks_deadlock by only considering blocked locks.
 *  Matthew Wilcox <willy@debian.org>, March, 2000.
 *
 *  Leases and LOCK_MAND
 *  Matthew Wilcox <willy@debian.org>, June, 2000.
 *  Stephen Rothwell <sfr@canb.auug.org.au>, June, 2000.
 */

#include <linux/capability.h>
#include <linux/file.h>
Al Viro's avatar
Al Viro committed
119
#include <linux/fdtable.h>
Linus Torvalds's avatar
Linus Torvalds committed
120 121 122 123 124 125 126
#include <linux/fs.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/security.h>
#include <linux/slab.h>
#include <linux/syscalls.h>
#include <linux/time.h>
127
#include <linux/rcupdate.h>
128
#include <linux/pid_namespace.h>
129
#include <linux/hashtable.h>
130 131
#include <linux/percpu.h>
#include <linux/lglock.h>
Linus Torvalds's avatar
Linus Torvalds committed
132 133 134 135 136

#include <asm/uaccess.h>

#define IS_POSIX(fl)	(fl->fl_flags & FL_POSIX)
#define IS_FLOCK(fl)	(fl->fl_flags & FL_FLOCK)
137
#define IS_LEASE(fl)	(fl->fl_flags & (FL_LEASE|FL_DELEG))
Linus Torvalds's avatar
Linus Torvalds committed
138

J. Bruce Fields's avatar
J. Bruce Fields committed
139 140
static bool lease_breaking(struct file_lock *fl)
{
141 142 143 144 145 146 147 148 149 150
	return fl->fl_flags & (FL_UNLOCK_PENDING | FL_DOWNGRADE_PENDING);
}

static int target_leasetype(struct file_lock *fl)
{
	if (fl->fl_flags & FL_UNLOCK_PENDING)
		return F_UNLCK;
	if (fl->fl_flags & FL_DOWNGRADE_PENDING)
		return F_RDLCK;
	return fl->fl_type;
J. Bruce Fields's avatar
J. Bruce Fields committed
151 152
}

Linus Torvalds's avatar
Linus Torvalds committed
153 154 155 156 157 158
int leases_enable = 1;
int lease_break_time = 45;

#define for_each_lock(inode, lockp) \
	for (lockp = &inode->i_flock; *lockp != NULL; lockp = &(*lockp)->fl_next)

159
/*
160 161 162 163
 * The global file_lock_list is only used for displaying /proc/locks, so we
 * keep a list on each CPU, with each list protected by its own spinlock via
 * the file_lock_lglock. Note that alterations to the list also require that
 * the relevant i_lock is held.
164
 */
165 166
DEFINE_STATIC_LGLOCK(file_lock_lglock);
static DEFINE_PER_CPU(struct hlist_head, file_lock_list);
167

168
/*
169
 * The blocked_hash is used to find POSIX lock loops for deadlock detection.
170
 * It is protected by blocked_lock_lock.
171 172 173 174 175 176 177
 *
 * We hash locks by lockowner in order to optimize searching for the lock a
 * particular lockowner is waiting on.
 *
 * FIXME: make this value scale via some heuristic? We generally will want more
 * buckets when we have more lockowners holding locks, but that's a little
 * difficult to determine without knowing what the workload will look like.
178
 */
179 180
#define BLOCKED_HASH_BITS	7
static DEFINE_HASHTABLE(blocked_hash, BLOCKED_HASH_BITS);
181

182
/*
183 184
 * This lock protects the blocked_hash. Generally, if you're accessing it, you
 * want to be holding this lock.
185 186 187 188 189 190 191 192 193 194 195 196 197 198
 *
 * In addition, it also protects the fl->fl_block list, and the fl->fl_next
 * pointer for file_lock structures that are acting as lock requests (in
 * contrast to those that are acting as records of acquired locks).
 *
 * Note that when we acquire this lock in order to change the above fields,
 * we often hold the i_lock as well. In certain cases, when reading the fields
 * protected by this lock, we can skip acquiring it iff we already hold the
 * i_lock.
 *
 * In particular, adding an entry to the fl_block list requires that you hold
 * both the i_lock and the blocked_lock_lock (acquired in that order). Deleting
 * an entry from the list however only requires the file_lock_lock.
 */
199
static DEFINE_SPINLOCK(blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
200

201
static struct kmem_cache *filelock_cache __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
202

Miklos Szeredi's avatar
Miklos Szeredi committed
203
static void locks_init_lock_heads(struct file_lock *fl)
Miklos Szeredi's avatar
Miklos Szeredi committed
204
{
205
	INIT_HLIST_NODE(&fl->fl_link);
Miklos Szeredi's avatar
Miklos Szeredi committed
206 207
	INIT_LIST_HEAD(&fl->fl_block);
	init_waitqueue_head(&fl->fl_wait);
Miklos Szeredi's avatar
Miklos Szeredi committed
208 209
}

Linus Torvalds's avatar
Linus Torvalds committed
210
/* Allocate an empty lock structure. */
211
struct file_lock *locks_alloc_lock(void)
Linus Torvalds's avatar
Linus Torvalds committed
212
{
Miklos Szeredi's avatar
Miklos Szeredi committed
213
	struct file_lock *fl = kmem_cache_zalloc(filelock_cache, GFP_KERNEL);
Miklos Szeredi's avatar
Miklos Szeredi committed
214 215

	if (fl)
Miklos Szeredi's avatar
Miklos Szeredi committed
216
		locks_init_lock_heads(fl);
Miklos Szeredi's avatar
Miklos Szeredi committed
217 218

	return fl;
Linus Torvalds's avatar
Linus Torvalds committed
219
}
220
EXPORT_SYMBOL_GPL(locks_alloc_lock);
Linus Torvalds's avatar
Linus Torvalds committed
221

222
void locks_release_private(struct file_lock *fl)
223 224 225 226 227 228
{
	if (fl->fl_ops) {
		if (fl->fl_ops->fl_release_private)
			fl->fl_ops->fl_release_private(fl);
		fl->fl_ops = NULL;
	}
229
	fl->fl_lmops = NULL;
230 231

}
232
EXPORT_SYMBOL_GPL(locks_release_private);
233

Linus Torvalds's avatar
Linus Torvalds committed
234
/* Free a lock which is not in use. */
235
void locks_free_lock(struct file_lock *fl)
Linus Torvalds's avatar
Linus Torvalds committed
236
{
Miklos Szeredi's avatar
Miklos Szeredi committed
237 238
	BUG_ON(waitqueue_active(&fl->fl_wait));
	BUG_ON(!list_empty(&fl->fl_block));
239
	BUG_ON(!hlist_unhashed(&fl->fl_link));
Linus Torvalds's avatar
Linus Torvalds committed
240

241
	locks_release_private(fl);
Linus Torvalds's avatar
Linus Torvalds committed
242 243
	kmem_cache_free(filelock_cache, fl);
}
244
EXPORT_SYMBOL(locks_free_lock);
Linus Torvalds's avatar
Linus Torvalds committed
245 246 247

void locks_init_lock(struct file_lock *fl)
{
Miklos Szeredi's avatar
Miklos Szeredi committed
248 249
	memset(fl, 0, sizeof(struct file_lock));
	locks_init_lock_heads(fl);
Linus Torvalds's avatar
Linus Torvalds committed
250 251 252 253
}

EXPORT_SYMBOL(locks_init_lock);

254 255 256 257 258 259 260
static void locks_copy_private(struct file_lock *new, struct file_lock *fl)
{
	if (fl->fl_ops) {
		if (fl->fl_ops->fl_copy_lock)
			fl->fl_ops->fl_copy_lock(new, fl);
		new->fl_ops = fl->fl_ops;
	}
261
	if (fl->fl_lmops)
262 263 264
		new->fl_lmops = fl->fl_lmops;
}

Linus Torvalds's avatar
Linus Torvalds committed
265 266 267
/*
 * Initialize a new lock from an existing file_lock structure.
 */
268
void __locks_copy_lock(struct file_lock *new, const struct file_lock *fl)
Linus Torvalds's avatar
Linus Torvalds committed
269 270 271
{
	new->fl_owner = fl->fl_owner;
	new->fl_pid = fl->fl_pid;
272
	new->fl_file = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
273 274 275 276
	new->fl_flags = fl->fl_flags;
	new->fl_type = fl->fl_type;
	new->fl_start = fl->fl_start;
	new->fl_end = fl->fl_end;
277 278 279
	new->fl_ops = NULL;
	new->fl_lmops = NULL;
}
280
EXPORT_SYMBOL(__locks_copy_lock);
281 282 283 284 285 286 287

void locks_copy_lock(struct file_lock *new, struct file_lock *fl)
{
	locks_release_private(new);

	__locks_copy_lock(new, fl);
	new->fl_file = fl->fl_file;
Linus Torvalds's avatar
Linus Torvalds committed
288 289
	new->fl_ops = fl->fl_ops;
	new->fl_lmops = fl->fl_lmops;
290 291

	locks_copy_private(new, fl);
Linus Torvalds's avatar
Linus Torvalds committed
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
}

EXPORT_SYMBOL(locks_copy_lock);

static inline int flock_translate_cmd(int cmd) {
	if (cmd & LOCK_MAND)
		return cmd & (LOCK_MAND | LOCK_RW);
	switch (cmd) {
	case LOCK_SH:
		return F_RDLCK;
	case LOCK_EX:
		return F_WRLCK;
	case LOCK_UN:
		return F_UNLCK;
	}
	return -EINVAL;
}

/* Fill in a file_lock structure with an appropriate FLOCK lock. */
static int flock_make_lock(struct file *filp, struct file_lock **lock,
		unsigned int cmd)
{
	struct file_lock *fl;
	int type = flock_translate_cmd(cmd);
	if (type < 0)
		return type;
	
	fl = locks_alloc_lock();
	if (fl == NULL)
		return -ENOMEM;

	fl->fl_file = filp;
	fl->fl_pid = current->tgid;
	fl->fl_flags = FL_FLOCK;
	fl->fl_type = type;
	fl->fl_end = OFFSET_MAX;
	
	*lock = fl;
	return 0;
}

333
static int assign_type(struct file_lock *fl, long type)
Linus Torvalds's avatar
Linus Torvalds committed
334 335 336 337 338 339 340 341 342 343 344 345 346
{
	switch (type) {
	case F_RDLCK:
	case F_WRLCK:
	case F_UNLCK:
		fl->fl_type = type;
		break;
	default:
		return -EINVAL;
	}
	return 0;
}

347 348
static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl,
				 struct flock64 *l)
Linus Torvalds's avatar
Linus Torvalds committed
349 350
{
	switch (l->l_whence) {
351
	case SEEK_SET:
352
		fl->fl_start = 0;
Linus Torvalds's avatar
Linus Torvalds committed
353
		break;
354
	case SEEK_CUR:
355
		fl->fl_start = filp->f_pos;
Linus Torvalds's avatar
Linus Torvalds committed
356
		break;
357
	case SEEK_END:
358
		fl->fl_start = i_size_read(file_inode(filp));
Linus Torvalds's avatar
Linus Torvalds committed
359 360 361 362
		break;
	default:
		return -EINVAL;
	}
363 364 365 366 367
	if (l->l_start > OFFSET_MAX - fl->fl_start)
		return -EOVERFLOW;
	fl->fl_start += l->l_start;
	if (fl->fl_start < 0)
		return -EINVAL;
Linus Torvalds's avatar
Linus Torvalds committed
368 369 370

	/* POSIX-1996 leaves the case l->l_len < 0 undefined;
	   POSIX-2001 defines it. */
371
	if (l->l_len > 0) {
372 373 374 375
		if (l->l_len - 1 > OFFSET_MAX - fl->fl_start)
			return -EOVERFLOW;
		fl->fl_end = fl->fl_start + l->l_len - 1;

376
	} else if (l->l_len < 0) {
377
		if (fl->fl_start + l->l_len < 0)
378
			return -EINVAL;
379 380 381 382 383
		fl->fl_end = fl->fl_start - 1;
		fl->fl_start += l->l_len;
	} else
		fl->fl_end = OFFSET_MAX;

Linus Torvalds's avatar
Linus Torvalds committed
384 385 386 387 388 389 390
	fl->fl_owner = current->files;
	fl->fl_pid = current->tgid;
	fl->fl_file = filp;
	fl->fl_flags = FL_POSIX;
	fl->fl_ops = NULL;
	fl->fl_lmops = NULL;

391 392 393 394 395 396 397 398 399 400 401 402
	/* Ensure that fl->fl_filp has compatible f_mode */
	switch (l->l_type) {
	case F_RDLCK:
		if (!(filp->f_mode & FMODE_READ))
			return -EBADF;
		break;
	case F_WRLCK:
		if (!(filp->f_mode & FMODE_WRITE))
			return -EBADF;
		break;
	}

Linus Torvalds's avatar
Linus Torvalds committed
403 404 405
	return assign_type(fl, l->l_type);
}

406 407 408 409 410
/* Verify a "struct flock" and copy it to a "struct file_lock" as a POSIX
 * style lock.
 */
static int flock_to_posix_lock(struct file *filp, struct file_lock *fl,
			       struct flock *l)
Linus Torvalds's avatar
Linus Torvalds committed
411
{
412 413 414 415 416 417 418 419
	struct flock64 ll = {
		.l_type = l->l_type,
		.l_whence = l->l_whence,
		.l_start = l->l_start,
		.l_len = l->l_len,
	};

	return flock64_to_posix_lock(filp, fl, &ll);
Linus Torvalds's avatar
Linus Torvalds committed
420 421 422 423 424 425 426 427
}

/* default lease lock manager operations */
static void lease_break_callback(struct file_lock *fl)
{
	kill_fasync(&fl->fl_fasync, SIGIO, POLL_MSG);
}

428
static const struct lock_manager_operations lease_manager_ops = {
429 430
	.lm_break = lease_break_callback,
	.lm_change = lease_modify,
Linus Torvalds's avatar
Linus Torvalds committed
431 432 433 434 435
};

/*
 * Initialize a lease, use the default lock manager operations
 */
436
static int lease_init(struct file *filp, long type, struct file_lock *fl)
Linus Torvalds's avatar
Linus Torvalds committed
437
 {
438 439 440
	if (assign_type(fl, type) != 0)
		return -EINVAL;

Linus Torvalds's avatar
Linus Torvalds committed
441 442 443 444 445 446 447 448 449 450 451 452 453
	fl->fl_owner = current->files;
	fl->fl_pid = current->tgid;

	fl->fl_file = filp;
	fl->fl_flags = FL_LEASE;
	fl->fl_start = 0;
	fl->fl_end = OFFSET_MAX;
	fl->fl_ops = NULL;
	fl->fl_lmops = &lease_manager_ops;
	return 0;
}

/* Allocate a file_lock initialised to this type of lease */
454
static struct file_lock *lease_alloc(struct file *filp, long type)
Linus Torvalds's avatar
Linus Torvalds committed
455 456
{
	struct file_lock *fl = locks_alloc_lock();
457
	int error = -ENOMEM;
Linus Torvalds's avatar
Linus Torvalds committed
458 459

	if (fl == NULL)
460
		return ERR_PTR(error);
Linus Torvalds's avatar
Linus Torvalds committed
461 462

	error = lease_init(filp, type, fl);
463 464
	if (error) {
		locks_free_lock(fl);
465
		return ERR_PTR(error);
466
	}
467
	return fl;
Linus Torvalds's avatar
Linus Torvalds committed
468 469 470 471 472 473 474 475 476 477 478 479 480
}

/* Check if two locks overlap each other.
 */
static inline int locks_overlap(struct file_lock *fl1, struct file_lock *fl2)
{
	return ((fl1->fl_end >= fl2->fl_start) &&
		(fl2->fl_end >= fl1->fl_start));
}

/*
 * Check whether two locks have the same owner.
 */
481
static int posix_same_owner(struct file_lock *fl1, struct file_lock *fl2)
Linus Torvalds's avatar
Linus Torvalds committed
482
{
483
	if (fl1->fl_lmops && fl1->fl_lmops->lm_compare_owner)
Linus Torvalds's avatar
Linus Torvalds committed
484
		return fl2->fl_lmops == fl1->fl_lmops &&
485
			fl1->fl_lmops->lm_compare_owner(fl1, fl2);
Linus Torvalds's avatar
Linus Torvalds committed
486 487 488
	return fl1->fl_owner == fl2->fl_owner;
}

489
/* Must be called with the i_lock held! */
490
static void locks_insert_global_locks(struct file_lock *fl)
491
{
492 493 494 495
	lg_local_lock(&file_lock_lglock);
	fl->fl_link_cpu = smp_processor_id();
	hlist_add_head(&fl->fl_link, this_cpu_ptr(&file_lock_list));
	lg_local_unlock(&file_lock_lglock);
496 497
}

498
/* Must be called with the i_lock held! */
499
static void locks_delete_global_locks(struct file_lock *fl)
500
{
501 502 503 504 505 506 507 508
	/*
	 * Avoid taking lock if already unhashed. This is safe since this check
	 * is done while holding the i_lock, and new insertions into the list
	 * also require that it be held.
	 */
	if (hlist_unhashed(&fl->fl_link))
		return;
	lg_local_lock_cpu(&file_lock_lglock, fl->fl_link_cpu);
509
	hlist_del_init(&fl->fl_link);
510
	lg_local_unlock_cpu(&file_lock_lglock, fl->fl_link_cpu);
511 512
}

513 514 515 516 517 518 519 520
static unsigned long
posix_owner_key(struct file_lock *fl)
{
	if (fl->fl_lmops && fl->fl_lmops->lm_owner_key)
		return fl->fl_lmops->lm_owner_key(fl);
	return (unsigned long)fl->fl_owner;
}

521
static void locks_insert_global_blocked(struct file_lock *waiter)
522
{
523
	hash_add(blocked_hash, &waiter->fl_link, posix_owner_key(waiter));
524 525
}

526
static void locks_delete_global_blocked(struct file_lock *waiter)
527
{
528
	hash_del(&waiter->fl_link);
529 530
}

Linus Torvalds's avatar
Linus Torvalds committed
531 532
/* Remove waiter from blocker's block list.
 * When blocker ends up pointing to itself then the list is empty.
533
 *
534
 * Must be called with blocked_lock_lock held.
Linus Torvalds's avatar
Linus Torvalds committed
535
 */
536
static void __locks_delete_block(struct file_lock *waiter)
Linus Torvalds's avatar
Linus Torvalds committed
537
{
538
	locks_delete_global_blocked(waiter);
Linus Torvalds's avatar
Linus Torvalds committed
539 540 541 542
	list_del_init(&waiter->fl_block);
	waiter->fl_next = NULL;
}

543
static void locks_delete_block(struct file_lock *waiter)
Linus Torvalds's avatar
Linus Torvalds committed
544
{
545
	spin_lock(&blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
546
	__locks_delete_block(waiter);
547
	spin_unlock(&blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
548 549 550 551 552 553
}

/* Insert waiter into blocker's block list.
 * We use a circular list so that processes can be easily woken up in
 * the order they blocked. The documentation doesn't require this but
 * it seems like the reasonable thing to do.
554
 *
555
 * Must be called with both the i_lock and blocked_lock_lock held. The fl_block
Jeff Layton's avatar
Jeff Layton committed
556
 * list itself is protected by the blocked_lock_lock, but by ensuring that the
557
 * i_lock is also held on insertions we can avoid taking the blocked_lock_lock
558
 * in some cases when we see that the fl_block list is empty.
Linus Torvalds's avatar
Linus Torvalds committed
559
 */
560 561
static void __locks_insert_block(struct file_lock *blocker,
					struct file_lock *waiter)
Linus Torvalds's avatar
Linus Torvalds committed
562
{
563
	BUG_ON(!list_empty(&waiter->fl_block));
Linus Torvalds's avatar
Linus Torvalds committed
564
	waiter->fl_next = blocker;
565
	list_add_tail(&waiter->fl_block, &blocker->fl_block);
Linus Torvalds's avatar
Linus Torvalds committed
566
	if (IS_POSIX(blocker))
567 568 569 570 571 572 573
		locks_insert_global_blocked(waiter);
}

/* Must be called with i_lock held. */
static void locks_insert_block(struct file_lock *blocker,
					struct file_lock *waiter)
{
574
	spin_lock(&blocked_lock_lock);
575
	__locks_insert_block(blocker, waiter);
576
	spin_unlock(&blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
577 578
}

579 580 581
/*
 * Wake up processes blocked waiting for blocker.
 *
582
 * Must be called with the inode->i_lock held!
Linus Torvalds's avatar
Linus Torvalds committed
583 584 585
 */
static void locks_wake_up_blocks(struct file_lock *blocker)
{
586 587 588 589 590
	/*
	 * Avoid taking global lock if list is empty. This is safe since new
	 * blocked requests are only added to the list under the i_lock, and
	 * the i_lock is always held here. Note that removal from the fl_block
	 * list does not require the i_lock, so we must recheck list_empty()
591
	 * after acquiring the blocked_lock_lock.
592 593 594 595
	 */
	if (list_empty(&blocker->fl_block))
		return;

596
	spin_lock(&blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
597
	while (!list_empty(&blocker->fl_block)) {
598 599 600
		struct file_lock *waiter;

		waiter = list_first_entry(&blocker->fl_block,
Linus Torvalds's avatar
Linus Torvalds committed
601 602
				struct file_lock, fl_block);
		__locks_delete_block(waiter);
603 604
		if (waiter->fl_lmops && waiter->fl_lmops->lm_notify)
			waiter->fl_lmops->lm_notify(waiter);
Linus Torvalds's avatar
Linus Torvalds committed
605 606 607
		else
			wake_up(&waiter->fl_wait);
	}
608
	spin_unlock(&blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
609 610 611 612
}

/* Insert file lock fl into an inode's lock list at the position indicated
 * by pos. At the same time add the lock to the global file lock list.
613 614
 *
 * Must be called with the i_lock held!
Linus Torvalds's avatar
Linus Torvalds committed
615 616 617
 */
static void locks_insert_lock(struct file_lock **pos, struct file_lock *fl)
{
618 619
	fl->fl_nspid = get_pid(task_tgid(current));

Linus Torvalds's avatar
Linus Torvalds committed
620 621 622
	/* insert into file's list */
	fl->fl_next = *pos;
	*pos = fl;
623 624

	locks_insert_global_locks(fl);
Linus Torvalds's avatar
Linus Torvalds committed
625 626
}

627 628 629 630 631 632 633 634
/**
 * locks_delete_lock - Delete a lock and then free it.
 * @thisfl_p: pointer that points to the fl_next field of the previous
 * 	      inode->i_flock list entry
 *
 * Unlink a lock from all lists and free the namespace reference, but don't
 * free it yet. Wake up processes that are blocked waiting for this lock and
 * notify the FS that the lock has been cleared.
635 636
 *
 * Must be called with the i_lock held!
Linus Torvalds's avatar
Linus Torvalds committed
637
 */
638
static void locks_unlink_lock(struct file_lock **thisfl_p)
Linus Torvalds's avatar
Linus Torvalds committed
639 640 641
{
	struct file_lock *fl = *thisfl_p;

642 643
	locks_delete_global_locks(fl);

Linus Torvalds's avatar
Linus Torvalds committed
644 645 646
	*thisfl_p = fl->fl_next;
	fl->fl_next = NULL;

647 648 649 650 651
	if (fl->fl_nspid) {
		put_pid(fl->fl_nspid);
		fl->fl_nspid = NULL;
	}

Linus Torvalds's avatar
Linus Torvalds committed
652
	locks_wake_up_blocks(fl);
653 654 655 656 657 658 659 660 661 662 663 664
}

/*
 * Unlink a lock from all lists and free it.
 *
 * Must be called with i_lock held!
 */
static void locks_delete_lock(struct file_lock **thisfl_p)
{
	struct file_lock *fl = *thisfl_p;

	locks_unlink_lock(thisfl_p);
Linus Torvalds's avatar
Linus Torvalds committed
665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713
	locks_free_lock(fl);
}

/* Determine if lock sys_fl blocks lock caller_fl. Common functionality
 * checks for shared/exclusive status of overlapping locks.
 */
static int locks_conflict(struct file_lock *caller_fl, struct file_lock *sys_fl)
{
	if (sys_fl->fl_type == F_WRLCK)
		return 1;
	if (caller_fl->fl_type == F_WRLCK)
		return 1;
	return 0;
}

/* Determine if lock sys_fl blocks lock caller_fl. POSIX specific
 * checking before calling the locks_conflict().
 */
static int posix_locks_conflict(struct file_lock *caller_fl, struct file_lock *sys_fl)
{
	/* POSIX locks owned by the same process do not conflict with
	 * each other.
	 */
	if (!IS_POSIX(sys_fl) || posix_same_owner(caller_fl, sys_fl))
		return (0);

	/* Check whether they overlap */
	if (!locks_overlap(caller_fl, sys_fl))
		return 0;

	return (locks_conflict(caller_fl, sys_fl));
}

/* Determine if lock sys_fl blocks lock caller_fl. FLOCK specific
 * checking before calling the locks_conflict().
 */
static int flock_locks_conflict(struct file_lock *caller_fl, struct file_lock *sys_fl)
{
	/* FLOCK locks referring to the same filp do not conflict with
	 * each other.
	 */
	if (!IS_FLOCK(sys_fl) || (caller_fl->fl_file == sys_fl->fl_file))
		return (0);
	if ((caller_fl->fl_type & LOCK_MAND) || (sys_fl->fl_type & LOCK_MAND))
		return 0;

	return (locks_conflict(caller_fl, sys_fl));
}

714
void
715
posix_test_lock(struct file *filp, struct file_lock *fl)
Linus Torvalds's avatar
Linus Torvalds committed
716 717
{
	struct file_lock *cfl;
718
	struct inode *inode = file_inode(filp);
Linus Torvalds's avatar
Linus Torvalds committed
719

720
	spin_lock(&inode->i_lock);
Al Viro's avatar
Al Viro committed
721
	for (cfl = file_inode(filp)->i_flock; cfl; cfl = cfl->fl_next) {
Linus Torvalds's avatar
Linus Torvalds committed
722 723
		if (!IS_POSIX(cfl))
			continue;
724
		if (posix_locks_conflict(fl, cfl))
Linus Torvalds's avatar
Linus Torvalds committed
725 726
			break;
	}
727
	if (cfl) {
728
		__locks_copy_lock(fl, cfl);
729
		if (cfl->fl_nspid)
730
			fl->fl_pid = pid_vnr(cfl->fl_nspid);
731
	} else
732
		fl->fl_type = F_UNLCK;
733
	spin_unlock(&inode->i_lock);
734
	return;
Linus Torvalds's avatar
Linus Torvalds committed
735 736 737
}
EXPORT_SYMBOL(posix_test_lock);

738 739 740 741 742
/*
 * Deadlock detection:
 *
 * We attempt to detect deadlocks that are due purely to posix file
 * locks.
Linus Torvalds's avatar
Linus Torvalds committed
743
 *
744 745 746 747 748 749 750
 * We assume that a task can be waiting for at most one lock at a time.
 * So for any acquired lock, the process holding that lock may be
 * waiting on at most one other lock.  That lock in turns may be held by
 * someone waiting for at most one other lock.  Given a requested lock
 * caller_fl which is about to wait for a conflicting lock block_fl, we
 * follow this chain of waiters to ensure we are not about to create a
 * cycle.
Linus Torvalds's avatar
Linus Torvalds committed
751
 *
752 753 754
 * Since we do this before we ever put a process to sleep on a lock, we
 * are ensured that there is never a cycle; that is what guarantees that
 * the while() loop in posix_locks_deadlock() eventually completes.
755
 *
756 757 758 759 760
 * Note: the above assumption may not be true when handling lock
 * requests from a broken NFS client. It may also fail in the presence
 * of tasks (such as posix threads) sharing the same open file table.
 *
 * To handle those cases, we just bail out after a few iterations.
Linus Torvalds's avatar
Linus Torvalds committed
761
 */
762 763 764

#define MAX_DEADLK_ITERATIONS 10

765 766 767 768 769
/* Find a lock that the owner of the given block_fl is blocking on. */
static struct file_lock *what_owner_is_waiting_for(struct file_lock *block_fl)
{
	struct file_lock *fl;

770
	hash_for_each_possible(blocked_hash, fl, fl_link, posix_owner_key(block_fl)) {
771 772 773 774 775 776
		if (posix_same_owner(fl, block_fl))
			return fl->fl_next;
	}
	return NULL;
}

777
/* Must be called with the blocked_lock_lock held! */
778
static int posix_locks_deadlock(struct file_lock *caller_fl,
Linus Torvalds's avatar
Linus Torvalds committed
779 780
				struct file_lock *block_fl)
{
781
	int i = 0;
Linus Torvalds's avatar
Linus Torvalds committed
782

783 784 785 786 787
	while ((block_fl = what_owner_is_waiting_for(block_fl))) {
		if (i++ > MAX_DEADLK_ITERATIONS)
			return 0;
		if (posix_same_owner(caller_fl, block_fl))
			return 1;
Linus Torvalds's avatar
Linus Torvalds committed
788 789 790 791 792
	}
	return 0;
}

/* Try to create a FLOCK lock on filp. We always insert new FLOCK locks
793
 * after any leases, but before any posix locks.
794 795 796 797
 *
 * Note that if called with an FL_EXISTS argument, the caller may determine
 * whether or not a lock was successfully freed by testing the return
 * value for -ENOENT.
Linus Torvalds's avatar
Linus Torvalds committed
798
 */
799
static int flock_lock_file(struct file *filp, struct file_lock *request)
Linus Torvalds's avatar
Linus Torvalds committed
800
{
801
	struct file_lock *new_fl = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
802
	struct file_lock **before;
Al Viro's avatar
Al Viro committed
803
	struct inode * inode = file_inode(filp);
Linus Torvalds's avatar
Linus Torvalds committed
804 805 806
	int error = 0;
	int found = 0;

807
	if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) {
808
		new_fl = locks_alloc_lock();
809 810
		if (!new_fl)
			return -ENOMEM;
811 812
	}

813
	spin_lock(&inode->i_lock);
814 815 816
	if (request->fl_flags & FL_ACCESS)
		goto find_conflict;

Linus Torvalds's avatar
Linus Torvalds committed
817 818 819 820 821 822 823 824
	for_each_lock(inode, before) {
		struct file_lock *fl = *before;
		if (IS_POSIX(fl))
			break;
		if (IS_LEASE(fl))
			continue;
		if (filp != fl->fl_file)
			continue;
825
		if (request->fl_type == fl->fl_type)
Linus Torvalds's avatar
Linus Torvalds committed
826 827 828 829 830 831
			goto out;
		found = 1;
		locks_delete_lock(before);
		break;
	}

832 833 834
	if (request->fl_type == F_UNLCK) {
		if ((request->fl_flags & FL_EXISTS) && !found)
			error = -ENOENT;
835
		goto out;
836
	}
Linus Torvalds's avatar
Linus Torvalds committed
837 838 839 840 841

	/*
	 * If a higher-priority process was blocked on the old file lock,
	 * give it the opportunity to lock the file.
	 */
842
	if (found) {
843
		spin_unlock(&inode->i_lock);
844
		cond_resched();
845
		spin_lock(&inode->i_lock);
846
	}
Linus Torvalds's avatar
Linus Torvalds committed
847

848
find_conflict:
Linus Torvalds's avatar
Linus Torvalds committed
849 850 851 852 853 854
	for_each_lock(inode, before) {
		struct file_lock *fl = *before;
		if (IS_POSIX(fl))
			break;
		if (IS_LEASE(fl))
			continue;
855
		if (!flock_locks_conflict(request, fl))
Linus Torvalds's avatar
Linus Torvalds committed
856 857
			continue;
		error = -EAGAIN;
858 859 860 861
		if (!(request->fl_flags & FL_SLEEP))
			goto out;
		error = FILE_LOCK_DEFERRED;
		locks_insert_block(fl, request);
Linus Torvalds's avatar
Linus Torvalds committed
862 863
		goto out;
	}
864 865
	if (request->fl_flags & FL_ACCESS)
		goto out;
866
	locks_copy_lock(new_fl, request);
867
	locks_insert_lock(before, new_fl);
868
	new_fl = NULL;
869
	error = 0;
Linus Torvalds's avatar
Linus Torvalds committed
870 871

out:
872
	spin_unlock(&inode->i_lock);
873 874
	if (new_fl)
		locks_free_lock(new_fl);
Linus Torvalds's avatar
Linus Torvalds committed
875 876 877
	return error;
}

878
static int __posix_lock_file(struct inode *inode, struct file_lock *request, struct file_lock *conflock)
Linus Torvalds's avatar
Linus Torvalds committed
879 880
{
	struct file_lock *fl;
881 882
	struct file_lock *new_fl = NULL;
	struct file_lock *new_fl2 = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
883 884 885
	struct file_lock *left = NULL;
	struct file_lock *right = NULL;
	struct file_lock **before;
886 887
	int error;
	bool added = false;
Linus Torvalds's avatar
Linus Torvalds committed
888 889 890 891

	/*
	 * We may need two file_lock structures for this operation,
	 * so we get them in advance to avoid races.
892 893
	 *
	 * In some cases we can be sure, that no new locks will be needed
Linus Torvalds's avatar
Linus Torvalds committed
894
	 */
895 896 897 898 899 900
	if (!(request->fl_flags & FL_ACCESS) &&
	    (request->fl_type != F_UNLCK ||
	     request->fl_start != 0 || request->fl_end != OFFSET_MAX)) {
		new_fl = locks_alloc_lock();
		new_fl2 = locks_alloc_lock();
	}
Linus Torvalds's avatar
Linus Torvalds committed
901

902
	spin_lock(&inode->i_lock);
903 904 905
	/*
	 * New lock request. Walk all POSIX locks and look for conflicts. If
	 * there are any, either return error or put the request on the
906
	 * blocker's list of waiters and the global blocked_hash.
907
	 */
Linus Torvalds's avatar
Linus Torvalds committed
908 909
	if (request->fl_type != F_UNLCK) {
		for_each_lock(inode, before) {
910
			fl = *before;
Linus Torvalds's avatar
Linus Torvalds committed
911 912 913 914
			if (!IS_POSIX(fl))
				continue;
			if (!posix_locks_conflict(request, fl))
				continue;
915
			if (conflock)
916
				__locks_copy_lock(conflock, fl);
Linus Torvalds's avatar
Linus Torvalds committed
917 918 919
			error = -EAGAIN;
			if (!(request->fl_flags & FL_SLEEP))
				goto out;
920 921 922 923
			/*
			 * Deadlock detection and insertion into the blocked
			 * locks list must be done while holding the same lock!
			 */
Linus Torvalds's avatar
Linus Torvalds committed
924
			error = -EDEADLK;
925
			spin_lock(&blocked_lock_lock);
926 927 928 929
			if (likely(!posix_locks_deadlock(request, fl))) {
				error = FILE_LOCK_DEFERRED;
				__locks_insert_block(fl, request);
			}
930
			spin_unlock(&blocked_lock_lock);
Linus Torvalds's avatar
Linus Torvalds committed
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951
			goto out;
  		}
  	}

	/* If we're just looking for a conflict, we're done. */
	error = 0;
	if (request->fl_flags & FL_ACCESS)
		goto out;

	/*
	 * Find the first old lock with the same owner as the new lock.
	 */
	
	before = &inode->i_flock;

	/* First skip locks owned by other processes.  */
	while ((fl = *before) && (!IS_POSIX(fl) ||
				  !posix_same_owner(request, fl))) {
		before = &fl->fl_next;
	}

952
	/* Process locks with this owner. */
Linus Torvalds's avatar
Linus Torvalds committed
953 954 955 956
	while ((fl = *before) && posix_same_owner(request, fl)) {
		/* Detect adjacent or overlapping regions (if same lock type)
		 */
		if (request->fl_type == fl->fl_type) {
Olaf Kirch's avatar
Olaf Kirch committed
957 958 959 960
			/* In all comparisons of start vs end, use
			 * "start - 1" rather than "end + 1". If end
			 * is OFFSET_MAX, end + 1 will become negative.
			 */
Linus Torvalds's avatar
Linus Torvalds committed
961 962 963 964 965
			if (fl->fl_end < request->fl_start - 1)
				goto next_lock;
			/* If the next lock in the list has entirely bigger
			 * addresses than the new one, insert the lock here.
			 */
Olaf Kirch's avatar
Olaf Kirch committed
966
			if (fl->fl_start - 1 > request->fl_end)
Linus Torvalds's avatar
Linus Torvalds committed
967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986
				break;

			/* If we come here, the new and old lock are of the
			 * same type and adjacent or overlapping. Make one
			 * lock yielding from the lower start address of both
			 * locks to the higher end address.
			 */
			if (fl->fl_start > request->fl_start)
				fl->fl_start = request->fl_start;
			else
				request->fl_start = fl->fl_start;
			if (fl->fl_end < request->fl_end)
				fl->fl_end = request->fl_end;
			else
				request->fl_end = fl->fl_end;
			if (added) {
				locks_delete_lock(before);
				continue;
			}
			request = fl;
987
			added = true;
Linus Torvalds's avatar
Linus Torvalds committed
988 989 990 991 992 993 994 995 996 997
		}
		else {
			/* Processing for different lock types is a bit
			 * more complex.
			 */
			if (fl->fl_end < request->fl_start)
				goto next_lock;
			if (fl->fl_start > request->fl_end)
				break;
			if (request->fl_type == F_UNLCK)
998
				added = true;
Linus Torvalds's avatar
Linus Torvalds committed
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
			if (fl->fl_start < request->fl_start)
				left = fl;
			/* If the next lock in the list has a higher end
			 * address than the new one, insert the new one here.
			 */
			if (fl->fl_end > request->fl_end) {
				right = fl;
				break;
			}
			if (fl->fl_start >= request->fl_start) {
				/* The new lock completely replaces an old
				 * one (This may happen several times).
				 */
				if (added) {
					locks_delete_lock(before);
					continue;
				}
				/* Replace the old lock with the new one.
				 * Wake up anybody waiting for the old one,
				 * as the change in lock type might satisfy
				 * their needs.
				 */
				locks_wake_up_blocks(fl);
				fl->fl_start = request->fl_start;
				fl->fl_end = request->fl_end;
				fl->fl_type = request->fl_type;
1025 1026
				locks_release_private(fl);
				locks_copy_private(fl, request);
Linus Torvalds's avatar
Linus Torvalds committed
1027
				request = fl;
1028
				added = true;
Linus Torvalds's avatar
Linus Torvalds committed
1029 1030 1031 1032 1033 1034 1035 1036
			}
		}
		/* Go on to next lock.
		 */
	next_lock:
		before = &fl->fl_next;
	}

1037
	/*
1038 1039 1040
	 * The above code only modifies existing locks in case of merging or
	 * replacing. If new lock(s) need to be inserted all modifications are
	 * done below this, so it's safe yet to bail out.
1041 1042 1043 1044 1045
	 */
	error = -ENOLCK; /* "no luck" */
	if (right && left == right && !new_fl2)
		goto out;

Linus Torvalds's avatar
Linus Torvalds committed
1046 1047
	error = 0;
	if (!added) {
1048 1049 1050
		if (request->fl_type == F_UNLCK) {
			if (request->fl_flags & FL_EXISTS)
				error = -ENOENT;
Linus Torvalds's avatar
Linus Torvalds committed
1051
			goto out;
1052
		}
1053 1054 1055 1056 1057

		if (!new_fl) {
			error = -ENOLCK;
			goto out;
		}
Linus Torvalds's avatar
Linus Torvalds committed
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
		locks_copy_lock(new_fl, request);
		locks_insert_lock(before, new_fl);
		new_fl = NULL;
	}
	if (right) {
		if (left == right) {
			/* The new lock breaks the old one in two pieces,
			 * so we have to use the second new lock.
			 */
			left = new_fl2;
			new_fl2 = NULL;
			locks_copy_lock(left, right);
			locks_insert_lock(before, left);
		}
		right->fl_start = request->fl_end + 1;
		locks_wake_up_blocks(right);
	}
	if (left) {
		left->fl_end = request->fl_start - 1;
		locks_wake_up_blocks(left);
	}
 out:
1080
	spin_unlock(&inode->i_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094
	/*
	 * Free any unused locks.
	 */
	if (new_fl)
		locks_free_lock(new_fl);
	if (new_fl2)
		locks_free_lock(new_fl2);
	return error;
}

/**
 * posix_lock_file - Apply a POSIX-style lock to a file
 * @filp: The file to apply the lock to
 * @fl: The lock to be applied
1095
 * @conflock: Place to return a copy of the conflicting lock, if found.
Linus Torvalds's avatar
Linus Torvalds committed
1096 1097 1098 1099
 *
 * Add a POSIX style lock to a file.
 * We merge adjacent & overlapping locks whenever possible.
 * POSIX locks are sorted by owner task, then by starting address
1100 1101 1102 1103
 *
 * Note that if called with an FL_EXISTS argument, the caller may determine
 * whether or not a lock was successfully freed by testing the return
 * value for -ENOENT.
Linus Torvalds's avatar
Linus Torvalds committed
1104
 */
1105
int posix_lock_file(struct file *filp, struct file_lock *fl,
1106 1107
			struct file_lock *conflock)
{
Al Viro's avatar
Al Viro committed
1108
	return __posix_lock_file(file_inode(filp), fl, conflock);
Linus Torvalds's avatar
Linus Torvalds committed
1109
}
1110
EXPORT_SYMBOL(posix_lock_file);
Linus Torvalds's avatar
Linus Torvalds committed
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125

/**
 * posix_lock_file_wait - Apply a POSIX-style lock to a file
 * @filp: The file to apply the lock to
 * @fl: The lock to be applied
 *
 * Add a POSIX style lock to a file.
 * We merge adjacent & overlapping locks whenever possible.
 * POSIX locks are sorted by owner task, then by starting address
 */
int posix_lock_file_wait(struct file *filp, struct file_lock *fl)
{
	int error;
	might_sleep ();
	for (;;) {
1126
		error = posix_lock_file(filp, fl, NULL);
1127
		if (error != FILE_LOCK_DEFERRED)
Linus Torvalds's avatar
Linus Torvalds committed
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
			break;
		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
		if (!error)
			continue;

		locks_delete_block(fl);
		break;
	}
	return error;
}
EXPORT_SYMBOL(posix_lock_file_wait);

/**
 * locks_mandatory_locked - Check for an active lock
 * @inode: the file to check
 *
 * Searches the inode's list of locks to find any POSIX locks which conflict.
 * This function is called from locks_verify_locked() only.
 */
int locks_mandatory_locked(struct inode *inode)
{
	fl_owner_t owner = current->files;
	struct file_lock *fl;

	/*
	 * Search the lock list for this inode for any POSIX locks.
	 */
1155
	spin_lock(&inode->i_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1156 1157 1158 1159 1160 1161
	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (!IS_POSIX(fl))
			continue;
		if (fl->fl_owner != owner)
			break;
	}
1162
	spin_unlock(&inode->i_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1163 1164 1165 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
	return fl ? -EAGAIN : 0;
}

/**
 * locks_mandatory_area - Check for a conflicting lock
 * @read_write: %FLOCK_VERIFY_WRITE for exclusive access, %FLOCK_VERIFY_READ
 *		for shared
 * @inode:      the file to check
 * @filp:       how the file was opened (if it was)
 * @offset:     start of area to check
 * @count:      length of area to check
 *
 * Searches the inode's list of locks to find any POSIX locks which conflict.
 * This function is called from rw_verify_area() and
 * locks_verify_truncate().
 */
int locks_mandatory_area(int read_write, struct inode *inode,
			 struct file *filp, loff_t offset,
			 size_t count)
{
	struct file_lock fl;
	int error;

	locks_init_lock(&fl);
	fl.fl_owner = current->files;
	fl.fl_pid = current->tgid;
	fl.fl_file = filp;
	fl.fl_flags = FL_POSIX | FL_ACCESS;
	if (filp && !(filp->f_flags & O_NONBLOCK))
		fl.fl_flags |= FL_SLEEP;
	fl.fl_type = (read_write == FLOCK_VERIFY_WRITE) ? F_WRLCK : F_RDLCK;
	fl.fl_start = offset;
	fl.fl_end = offset + count - 1;

	for (;;) {
1198
		error = __posix_lock_file(inode, &fl, NULL);
1199
		if (error != FILE_LOCK_DEFERRED)
Linus Torvalds's avatar
Linus Torvalds committed
1200 1201 1202 1203 1204 1205 1206
			break;
		error = wait_event_interruptible(fl.fl_wait, !fl.fl_next);
		if (!error) {
			/*
			 * If we've been sleeping someone might have
			 * changed the permissions behind our back.
			 */
1207
			if (__mandatory_lock(inode))
Linus Torvalds's avatar
Linus Torvalds committed
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219
				continue;
		}

		locks_delete_block(&fl);
		break;
	}

	return error;
}

EXPORT_SYMBOL(locks_mandatory_area);

1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230
static void lease_clear_pending(struct file_lock *fl, int arg)
{
	switch (arg) {
	case F_UNLCK:
		fl->fl_flags &= ~FL_UNLOCK_PENDING;
		/* fall through: */
	case F_RDLCK:
		fl->fl_flags &= ~FL_DOWNGRADE_PENDING;
	}
}

Linus Torvalds's avatar
Linus Torvalds committed
1231 1232 1233 1234 1235 1236 1237 1238
/* We already had a lease on this file; just change its type */
int lease_modify(struct file_lock **before, int arg)
{
	struct file_lock *fl = *before;
	int error = assign_type(fl, arg);

	if (error)
		return error;
1239
	lease_clear_pending(fl, arg);
Linus Torvalds's avatar
Linus Torvalds committed
1240
	locks_wake_up_blocks(fl);
1241 1242 1243 1244 1245
	if (arg == F_UNLCK) {
		struct file *filp = fl->fl_file;

		f_delown(filp);
		filp->f_owner.signum = 0;
1246 1247 1248 1249 1250
		fasync_helper(0, fl->fl_file, 0, &fl->fl_fasync);
		if (fl->fl_fasync != NULL) {
			printk(KERN_ERR "locks_delete_lock: fasync == %p\n", fl->fl_fasync);
			fl->fl_fasync = NULL;
		}
Linus Torvalds's avatar
Linus Torvalds committed
1251
		locks_delete_lock(before);
1252
	}
Linus Torvalds's avatar
Linus Torvalds committed
1253 1254 1255 1256 1257
	return 0;
}

EXPORT_SYMBOL(lease_modify);

1258 1259 1260 1261 1262 1263 1264 1265
static bool past_time(unsigned long then)
{
	if (!then)
		/* 0 is a special value meaning "this never expires": */
		return false;
	return time_after(jiffies, then);
}

Linus Torvalds's avatar
Linus Torvalds committed
1266 1267 1268 1269 1270 1271
static void time_out_leases(struct inode *inode)
{
	struct file_lock **before;
	struct file_lock *fl;

	before = &inode->i_flock;
J. Bruce Fields's avatar
J. Bruce Fields committed
1272
	while ((fl = *before) && IS_LEASE(fl) && lease_breaking(fl)) {
1273 1274 1275 1276
		if (past_time(fl->fl_downgrade_time))
			lease_modify(before, F_RDLCK);
		if (past_time(fl->fl_break_time))
			lease_modify(before, F_UNLCK);
Linus Torvalds's avatar
Linus Torvalds committed
1277 1278 1279 1280 1281
		if (fl == *before)	/* lease_modify may have freed fl */
			before = &fl->fl_next;
	}
}

1282 1283 1284 1285 1286 1287 1288
static bool leases_conflict(struct file_lock *lease, struct file_lock *breaker)
{
	if ((breaker->fl_flags & FL_DELEG) && (lease->fl_flags & FL_LEASE))
		return false;
	return locks_conflict(breaker, lease);
}

Linus Torvalds's avatar
Linus Torvalds committed
1289 1290 1291
/**
 *	__break_lease	-	revoke all outstanding leases on file
 *	@inode: the inode of the file to return
1292 1293 1294 1295
 *	@mode: O_RDONLY: break only write leases; O_WRONLY or O_RDWR:
 *	    break all leases
 *	@type: FL_LEASE: break leases and delegations; FL_DELEG: break
 *	    only delegations
Linus Torvalds's avatar
Linus Torvalds committed
1296
 *
1297 1298 1299
 *	break_lease (inlined for speed) has checked there already is at least
 *	some kind of lock (maybe a lease) on this file.  Leases are broken on
 *	a call to open() or truncate().  This function can sleep unless you
Linus Torvalds's avatar
Linus Torvalds committed
1300 1301
 *	specified %O_NONBLOCK to your open().
 */
1302
int __break_lease(struct inode *inode, unsigned int mode, unsigned int type)
Linus Torvalds's avatar
Linus Torvalds committed
1303
{
1304
	int error = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1305 1306 1307 1308
	struct file_lock *new_fl, *flock;
	struct file_lock *fl;
	unsigned long break_time;
	int i_have_this_lease = 0;
1309
	bool lease_conflict = false;
1310
	int want_write = (mode & O_ACCMODE) != O_RDONLY;
Linus Torvalds's avatar
Linus Torvalds committed
1311

1312
	new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK);
1313 1314
	if (IS_ERR(new_fl))
		return PTR_ERR(new_fl);
1315
	new_fl->fl_flags = type;
Linus Torvalds's avatar
Linus Torvalds committed
1316

1317
	spin_lock(&inode->i_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1318 1319 1320 1321 1322 1323 1324

	time_out_leases(inode);

	flock = inode->i_flock;
	if ((flock == NULL) || !IS_LEASE(flock))
		goto out;

1325 1326 1327 1328 1329 1330 1331 1332
	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next) {
		if (leases_conflict(fl, new_fl)) {
			lease_conflict = true;
			if (fl->fl_owner == current->files)
				i_have_this_lease = 1;
		}
	}
	if (!lease_conflict)
1333 1334
		goto out;

Linus Torvalds's avatar
Linus Torvalds committed
1335 1336 1337 1338 1339 1340 1341 1342
	break_time = 0;
	if (lease_break_time > 0) {
		break_time = jiffies + lease_break_time * HZ;
		if (break_time == 0)
			break_time++;	/* so that 0 means no break time */
	}

	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next) {
1343 1344
		if (!leases_conflict(fl, new_fl))
			continue;
1345 1346 1347 1348
		if (want_write) {
			if (fl->fl_flags & FL_UNLOCK_PENDING)
				continue;
			fl->fl_flags |= FL_UNLOCK_PENDING;
Linus Torvalds's avatar
Linus Torvalds committed
1349
			fl->fl_break_time = break_time;
1350 1351 1352 1353 1354
		} else {
			if (lease_breaking(flock))
				continue;
			fl->fl_flags |= FL_DOWNGRADE_PENDING;
			fl->fl_downgrade_time = break_time;
Linus Torvalds's avatar
Linus Torvalds committed
1355
		}
1356
		fl->fl_lmops->lm_break(fl);
Linus Torvalds's avatar
Linus Torvalds committed
1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370
	}

	if (i_have_this_lease || (mode & O_NONBLOCK)) {
		error = -EWOULDBLOCK;
		goto out;
	}

restart:
	break_time = flock->fl_break_time;
	if (break_time != 0) {
		break_time -= jiffies;
		if (break_time == 0)
			break_time++;
	}
1371
	locks_insert_block(flock, new_fl);
1372
	spin_unlock(&inode->i_lock);
1373 1374
	error = wait_event_interruptible_timeout(new_fl->fl_wait,
						!new_fl->fl_next, break_time);
1375 1376
	spin_lock(&inode->i_lock);
	locks_delete_block(new_fl);
Linus Torvalds's avatar
Linus Torvalds committed
1377 1378 1379
	if (error >= 0) {
		if (error == 0)
			time_out_leases(inode);
1380 1381 1382 1383
		/*
		 * Wait for the next conflicting lease that has not been
		 * broken yet
		 */
Linus Torvalds's avatar
Linus Torvalds committed
1384 1385
		for (flock = inode->i_flock; flock && IS_LEASE(flock);
				flock = flock->fl_next) {
1386
			if (leases_conflict(new_fl, flock))
Linus Torvalds's avatar
Linus Torvalds committed
1387 1388 1389 1390 1391 1392
				goto restart;
		}
		error = 0;
	}

out:
1393
	spin_unlock(&inode->i_lock);
1394
	locks_free_lock(new_fl);
Linus Torvalds's avatar
Linus Torvalds committed
1395 1396 1397 1398 1399 1400
	return error;
}

EXPORT_SYMBOL(__break_lease);

/**
1401
 *	lease_get_mtime - get the last modified time of an inode
Linus Torvalds's avatar
Linus Torvalds committed
1402 1403 1404 1405 1406
 *	@inode: the inode
 *      @time:  pointer to a timespec which will contain the last modified time
 *
 * This is to force NFS clients to flush their caches for files with
 * exclusive leases.  The justification is that if someone has an
1407
 * exclusive lease, then they could be modifying it.
Linus Torvalds's avatar
Linus Torvalds committed
1408 1409 1410 1411
 */
void lease_get_mtime(struct inode *inode, struct timespec *time)
{
	struct file_lock *flock = inode->i_flock;
1412
	if (flock && IS_LEASE(flock) && (flock->fl_type == F_WRLCK))
Linus Torvalds's avatar
Linus Torvalds committed
1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445
		*time = current_fs_time(inode->i_sb);
	else
		*time = inode->i_mtime;
}

EXPORT_SYMBOL(lease_get_mtime);

/**
 *	fcntl_getlease - Enquire what lease is currently active
 *	@filp: the file
 *
 *	The value returned by this function will be one of
 *	(if no lease break is pending):
 *
 *	%F_RDLCK to indicate a shared lease is held.
 *
 *	%F_WRLCK to indicate an exclusive lease is held.
 *
 *	%F_UNLCK to indicate no lease is held.
 *
 *	(if a lease break is pending):
 *
 *	%F_RDLCK to indicate an exclusive lease needs to be
 *		changed to a shared lease (or removed).
 *
 *	%F_UNLCK to indicate the lease needs to be removed.
 *
 *	XXX: sfr & willy disagree over whether F_INPROGRESS
 *	should be returned to userspace.
 */
int fcntl_getlease(struct file *filp)
{
	struct file_lock *fl;
1446
	struct inode *inode = file_inode(filp);
Linus Torvalds's avatar
Linus Torvalds committed
1447 1448
	int type = F_UNLCK;

1449
	spin_lock(&inode->i_lock);
Al Viro's avatar
Al Viro committed
1450 1451
	time_out_leases(file_inode(filp));
	for (fl = file_inode(filp)->i_flock; fl && IS_LEASE(fl);
Linus Torvalds's avatar
Linus Torvalds committed
1452 1453
			fl = fl->fl_next) {
		if (fl->fl_file == filp) {
1454
			type = target_leasetype(fl);
Linus Torvalds's avatar
Linus Torvalds committed
1455 1456 1457
			break;
		}
	}
1458
	spin_unlock(&inode->i_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1459 1460 1461
	return type;
}