locks.c 58.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 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.
P
Paul Bolle 已提交
63
 *  See 'Documentation/filesystems/mandatory-locking.txt' for details.
L
Linus Torvalds 已提交
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>
A
Al Viro 已提交
119
#include <linux/fdtable.h>
L
Linus Torvalds 已提交
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>
L
Linus Torvalds 已提交
129 130 131 132 133 134 135

#include <asm/uaccess.h>

#define IS_POSIX(fl)	(fl->fl_flags & FL_POSIX)
#define IS_FLOCK(fl)	(fl->fl_flags & FL_FLOCK)
#define IS_LEASE(fl)	(fl->fl_flags & FL_LEASE)

J
J. Bruce Fields 已提交
136 137
static bool lease_breaking(struct file_lock *fl)
{
138 139 140 141 142 143 144 145 146 147
	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
J. Bruce Fields 已提交
148 149
}

L
Linus Torvalds 已提交
150 151 152 153 154 155
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)

156
static LIST_HEAD(file_lock_list);
L
Linus Torvalds 已提交
157
static LIST_HEAD(blocked_list);
158
static DEFINE_SPINLOCK(file_lock_lock);
L
Linus Torvalds 已提交
159

160 161 162 163 164
/*
 * Protects the two list heads above, plus the inode->i_flock list
 */
void lock_flocks(void)
{
165
	spin_lock(&file_lock_lock);
166 167 168 169 170
}
EXPORT_SYMBOL_GPL(lock_flocks);

void unlock_flocks(void)
{
171
	spin_unlock(&file_lock_lock);
172 173 174
}
EXPORT_SYMBOL_GPL(unlock_flocks);

175
static struct kmem_cache *filelock_cache __read_mostly;
L
Linus Torvalds 已提交
176

M
Miklos Szeredi 已提交
177
static void locks_init_lock_heads(struct file_lock *fl)
M
Miklos Szeredi 已提交
178
{
M
Miklos Szeredi 已提交
179 180 181
	INIT_LIST_HEAD(&fl->fl_link);
	INIT_LIST_HEAD(&fl->fl_block);
	init_waitqueue_head(&fl->fl_wait);
M
Miklos Szeredi 已提交
182 183
}

L
Linus Torvalds 已提交
184
/* Allocate an empty lock structure. */
185
struct file_lock *locks_alloc_lock(void)
L
Linus Torvalds 已提交
186
{
M
Miklos Szeredi 已提交
187
	struct file_lock *fl = kmem_cache_zalloc(filelock_cache, GFP_KERNEL);
M
Miklos Szeredi 已提交
188 189

	if (fl)
M
Miklos Szeredi 已提交
190
		locks_init_lock_heads(fl);
M
Miklos Szeredi 已提交
191 192

	return fl;
L
Linus Torvalds 已提交
193
}
194
EXPORT_SYMBOL_GPL(locks_alloc_lock);
L
Linus Torvalds 已提交
195

196
void locks_release_private(struct file_lock *fl)
197 198 199 200 201 202
{
	if (fl->fl_ops) {
		if (fl->fl_ops->fl_release_private)
			fl->fl_ops->fl_release_private(fl);
		fl->fl_ops = NULL;
	}
203
	fl->fl_lmops = NULL;
204 205

}
206
EXPORT_SYMBOL_GPL(locks_release_private);
207

L
Linus Torvalds 已提交
208
/* Free a lock which is not in use. */
209
void locks_free_lock(struct file_lock *fl)
L
Linus Torvalds 已提交
210
{
M
Miklos Szeredi 已提交
211 212 213
	BUG_ON(waitqueue_active(&fl->fl_wait));
	BUG_ON(!list_empty(&fl->fl_block));
	BUG_ON(!list_empty(&fl->fl_link));
L
Linus Torvalds 已提交
214

215
	locks_release_private(fl);
L
Linus Torvalds 已提交
216 217
	kmem_cache_free(filelock_cache, fl);
}
218
EXPORT_SYMBOL(locks_free_lock);
L
Linus Torvalds 已提交
219 220 221

void locks_init_lock(struct file_lock *fl)
{
M
Miklos Szeredi 已提交
222 223
	memset(fl, 0, sizeof(struct file_lock));
	locks_init_lock_heads(fl);
L
Linus Torvalds 已提交
224 225 226 227
}

EXPORT_SYMBOL(locks_init_lock);

228 229 230 231 232 233 234
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;
	}
235
	if (fl->fl_lmops)
236 237 238
		new->fl_lmops = fl->fl_lmops;
}

L
Linus Torvalds 已提交
239 240 241
/*
 * Initialize a new lock from an existing file_lock structure.
 */
242
void __locks_copy_lock(struct file_lock *new, const struct file_lock *fl)
L
Linus Torvalds 已提交
243 244 245
{
	new->fl_owner = fl->fl_owner;
	new->fl_pid = fl->fl_pid;
246
	new->fl_file = NULL;
L
Linus Torvalds 已提交
247 248 249 250
	new->fl_flags = fl->fl_flags;
	new->fl_type = fl->fl_type;
	new->fl_start = fl->fl_start;
	new->fl_end = fl->fl_end;
251 252 253
	new->fl_ops = NULL;
	new->fl_lmops = NULL;
}
254
EXPORT_SYMBOL(__locks_copy_lock);
255 256 257 258 259 260 261

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;
L
Linus Torvalds 已提交
262 263
	new->fl_ops = fl->fl_ops;
	new->fl_lmops = fl->fl_lmops;
264 265

	locks_copy_private(new, fl);
L
Linus Torvalds 已提交
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306
}

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;
}

307
static int assign_type(struct file_lock *fl, long type)
L
Linus Torvalds 已提交
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
{
	switch (type) {
	case F_RDLCK:
	case F_WRLCK:
	case F_UNLCK:
		fl->fl_type = type;
		break;
	default:
		return -EINVAL;
	}
	return 0;
}

/* 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)
{
	off_t start, end;

	switch (l->l_whence) {
330
	case SEEK_SET:
L
Linus Torvalds 已提交
331 332
		start = 0;
		break;
333
	case SEEK_CUR:
L
Linus Torvalds 已提交
334 335
		start = filp->f_pos;
		break;
336
	case SEEK_END:
337
		start = i_size_read(filp->f_path.dentry->d_inode);
L
Linus Torvalds 已提交
338 339 340 341 342 343 344 345
		break;
	default:
		return -EINVAL;
	}

	/* POSIX-1996 leaves the case l->l_len < 0 undefined;
	   POSIX-2001 defines it. */
	start += l->l_start;
346 347 348 349 350 351 352
	if (start < 0)
		return -EINVAL;
	fl->fl_end = OFFSET_MAX;
	if (l->l_len > 0) {
		end = start + l->l_len - 1;
		fl->fl_end = end;
	} else if (l->l_len < 0) {
L
Linus Torvalds 已提交
353
		end = start - 1;
354
		fl->fl_end = end;
L
Linus Torvalds 已提交
355
		start += l->l_len;
356 357
		if (start < 0)
			return -EINVAL;
L
Linus Torvalds 已提交
358 359
	}
	fl->fl_start = start;	/* we record the absolute position */
360 361
	if (fl->fl_end < fl->fl_start)
		return -EOVERFLOW;
L
Linus Torvalds 已提交
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379
	
	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;

	return assign_type(fl, l->l_type);
}

#if BITS_PER_LONG == 32
static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl,
				 struct flock64 *l)
{
	loff_t start;

	switch (l->l_whence) {
380
	case SEEK_SET:
L
Linus Torvalds 已提交
381 382
		start = 0;
		break;
383
	case SEEK_CUR:
L
Linus Torvalds 已提交
384 385
		start = filp->f_pos;
		break;
386
	case SEEK_END:
387
		start = i_size_read(filp->f_path.dentry->d_inode);
L
Linus Torvalds 已提交
388 389 390 391 392
		break;
	default:
		return -EINVAL;
	}

393 394
	start += l->l_start;
	if (start < 0)
L
Linus Torvalds 已提交
395
		return -EINVAL;
396 397 398 399 400 401 402 403 404
	fl->fl_end = OFFSET_MAX;
	if (l->l_len > 0) {
		fl->fl_end = start + l->l_len - 1;
	} else if (l->l_len < 0) {
		fl->fl_end = start - 1;
		start += l->l_len;
		if (start < 0)
			return -EINVAL;
	}
L
Linus Torvalds 已提交
405
	fl->fl_start = start;	/* we record the absolute position */
406 407
	if (fl->fl_end < fl->fl_start)
		return -EOVERFLOW;
L
Linus Torvalds 已提交
408 409 410 411 412 413 414 415
	
	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;

N
Namhyung Kim 已提交
416
	return assign_type(fl, l->l_type);
L
Linus Torvalds 已提交
417 418 419 420 421 422 423 424 425
}
#endif

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

426
static const struct lock_manager_operations lease_manager_ops = {
J
J. Bruce Fields 已提交
427 428
	.lm_break = lease_break_callback,
	.lm_change = lease_modify,
L
Linus Torvalds 已提交
429 430 431 432 433
};

/*
 * Initialize a lease, use the default lock manager operations
 */
434
static int lease_init(struct file *filp, long type, struct file_lock *fl)
L
Linus Torvalds 已提交
435
 {
436 437 438
	if (assign_type(fl, type) != 0)
		return -EINVAL;

L
Linus Torvalds 已提交
439 440 441 442 443 444 445 446 447 448 449 450 451
	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 */
452
static struct file_lock *lease_alloc(struct file *filp, long type)
L
Linus Torvalds 已提交
453 454
{
	struct file_lock *fl = locks_alloc_lock();
455
	int error = -ENOMEM;
L
Linus Torvalds 已提交
456 457

	if (fl == NULL)
J
J. Bruce Fields 已提交
458
		return ERR_PTR(error);
L
Linus Torvalds 已提交
459 460

	error = lease_init(filp, type, fl);
461 462
	if (error) {
		locks_free_lock(fl);
J
J. Bruce Fields 已提交
463
		return ERR_PTR(error);
464
	}
J
J. Bruce Fields 已提交
465
	return fl;
L
Linus Torvalds 已提交
466 467 468 469 470 471 472 473 474 475 476 477 478
}

/* 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.
 */
479
static int posix_same_owner(struct file_lock *fl1, struct file_lock *fl2)
L
Linus Torvalds 已提交
480
{
J
J. Bruce Fields 已提交
481
	if (fl1->fl_lmops && fl1->fl_lmops->lm_compare_owner)
L
Linus Torvalds 已提交
482
		return fl2->fl_lmops == fl1->fl_lmops &&
J
J. Bruce Fields 已提交
483
			fl1->fl_lmops->lm_compare_owner(fl1, fl2);
L
Linus Torvalds 已提交
484 485 486 487 488 489
	return fl1->fl_owner == fl2->fl_owner;
}

/* Remove waiter from blocker's block list.
 * When blocker ends up pointing to itself then the list is empty.
 */
490
static void __locks_delete_block(struct file_lock *waiter)
L
Linus Torvalds 已提交
491 492 493 494 495 496 497 498
{
	list_del_init(&waiter->fl_block);
	list_del_init(&waiter->fl_link);
	waiter->fl_next = NULL;
}

/*
 */
499
void locks_delete_block(struct file_lock *waiter)
L
Linus Torvalds 已提交
500
{
501
	lock_flocks();
L
Linus Torvalds 已提交
502
	__locks_delete_block(waiter);
503
	unlock_flocks();
L
Linus Torvalds 已提交
504
}
505
EXPORT_SYMBOL(locks_delete_block);
L
Linus Torvalds 已提交
506 507 508 509 510 511 512 513 514

/* 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.
 */
static void locks_insert_block(struct file_lock *blocker, 
			       struct file_lock *waiter)
{
515
	BUG_ON(!list_empty(&waiter->fl_block));
L
Linus Torvalds 已提交
516 517 518 519 520 521 522 523 524 525 526 527 528
	list_add_tail(&waiter->fl_block, &blocker->fl_block);
	waiter->fl_next = blocker;
	if (IS_POSIX(blocker))
		list_add(&waiter->fl_link, &blocked_list);
}

/* Wake up processes blocked waiting for blocker.
 * If told to wait then schedule the processes until the block list
 * is empty, otherwise empty the block list ourselves.
 */
static void locks_wake_up_blocks(struct file_lock *blocker)
{
	while (!list_empty(&blocker->fl_block)) {
529 530 531
		struct file_lock *waiter;

		waiter = list_first_entry(&blocker->fl_block,
L
Linus Torvalds 已提交
532 533
				struct file_lock, fl_block);
		__locks_delete_block(waiter);
J
J. Bruce Fields 已提交
534 535
		if (waiter->fl_lmops && waiter->fl_lmops->lm_notify)
			waiter->fl_lmops->lm_notify(waiter);
L
Linus Torvalds 已提交
536 537 538 539 540 541 542 543 544 545 546 547
		else
			wake_up(&waiter->fl_wait);
	}
}

/* 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.
 */
static void locks_insert_lock(struct file_lock **pos, struct file_lock *fl)
{
	list_add(&fl->fl_link, &file_lock_list);

548 549
	fl->fl_nspid = get_pid(task_tgid(current));

L
Linus Torvalds 已提交
550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568
	/* insert into file's list */
	fl->fl_next = *pos;
	*pos = fl;
}

/*
 * Delete a lock and then free it.
 * Wake up processes that are blocked waiting for this lock,
 * notify the FS that the lock has been cleared and
 * finally free the lock.
 */
static void locks_delete_lock(struct file_lock **thisfl_p)
{
	struct file_lock *fl = *thisfl_p;

	*thisfl_p = fl->fl_next;
	fl->fl_next = NULL;
	list_del_init(&fl->fl_link);

569 570 571 572 573
	if (fl->fl_nspid) {
		put_pid(fl->fl_nspid);
		fl->fl_nspid = NULL;
	}

L
Linus Torvalds 已提交
574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623
	locks_wake_up_blocks(fl);
	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));
}

624
void
625
posix_test_lock(struct file *filp, struct file_lock *fl)
L
Linus Torvalds 已提交
626 627 628
{
	struct file_lock *cfl;

629
	lock_flocks();
630
	for (cfl = filp->f_path.dentry->d_inode->i_flock; cfl; cfl = cfl->fl_next) {
L
Linus Torvalds 已提交
631 632
		if (!IS_POSIX(cfl))
			continue;
633
		if (posix_locks_conflict(fl, cfl))
L
Linus Torvalds 已提交
634 635
			break;
	}
636
	if (cfl) {
637
		__locks_copy_lock(fl, cfl);
638
		if (cfl->fl_nspid)
639
			fl->fl_pid = pid_vnr(cfl->fl_nspid);
640
	} else
641
		fl->fl_type = F_UNLCK;
642
	unlock_flocks();
643
	return;
L
Linus Torvalds 已提交
644 645 646
}
EXPORT_SYMBOL(posix_test_lock);

647 648 649 650 651
/*
 * Deadlock detection:
 *
 * We attempt to detect deadlocks that are due purely to posix file
 * locks.
L
Linus Torvalds 已提交
652
 *
653 654 655 656 657 658 659
 * 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.
L
Linus Torvalds 已提交
660
 *
661 662 663
 * 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.
664
 *
665 666 667 668 669
 * 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.
L
Linus Torvalds 已提交
670
 */
671 672 673

#define MAX_DEADLK_ITERATIONS 10

674 675 676 677 678 679 680 681 682 683 684 685
/* 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;

	list_for_each_entry(fl, &blocked_list, fl_link) {
		if (posix_same_owner(fl, block_fl))
			return fl->fl_next;
	}
	return NULL;
}

686
static int posix_locks_deadlock(struct file_lock *caller_fl,
L
Linus Torvalds 已提交
687 688
				struct file_lock *block_fl)
{
689
	int i = 0;
L
Linus Torvalds 已提交
690

691 692 693 694 695
	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;
L
Linus Torvalds 已提交
696 697 698 699 700
	}
	return 0;
}

/* Try to create a FLOCK lock on filp. We always insert new FLOCK locks
701
 * after any leases, but before any posix locks.
702 703 704 705
 *
 * 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.
L
Linus Torvalds 已提交
706
 */
707
static int flock_lock_file(struct file *filp, struct file_lock *request)
L
Linus Torvalds 已提交
708
{
709
	struct file_lock *new_fl = NULL;
L
Linus Torvalds 已提交
710
	struct file_lock **before;
711
	struct inode * inode = filp->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
712 713 714
	int error = 0;
	int found = 0;

715
	if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) {
716
		new_fl = locks_alloc_lock();
717 718
		if (!new_fl)
			return -ENOMEM;
719 720
	}

721 722 723 724
	lock_flocks();
	if (request->fl_flags & FL_ACCESS)
		goto find_conflict;

L
Linus Torvalds 已提交
725 726 727 728 729 730 731 732
	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;
733
		if (request->fl_type == fl->fl_type)
L
Linus Torvalds 已提交
734 735 736 737 738 739
			goto out;
		found = 1;
		locks_delete_lock(before);
		break;
	}

740 741 742
	if (request->fl_type == F_UNLCK) {
		if ((request->fl_flags & FL_EXISTS) && !found)
			error = -ENOENT;
743
		goto out;
744
	}
L
Linus Torvalds 已提交
745 746 747 748 749

	/*
	 * If a higher-priority process was blocked on the old file lock,
	 * give it the opportunity to lock the file.
	 */
750 751
	if (found) {
		unlock_flocks();
752
		cond_resched();
753 754
		lock_flocks();
	}
L
Linus Torvalds 已提交
755

756
find_conflict:
L
Linus Torvalds 已提交
757 758 759 760 761 762
	for_each_lock(inode, before) {
		struct file_lock *fl = *before;
		if (IS_POSIX(fl))
			break;
		if (IS_LEASE(fl))
			continue;
763
		if (!flock_locks_conflict(request, fl))
L
Linus Torvalds 已提交
764 765
			continue;
		error = -EAGAIN;
766 767 768 769
		if (!(request->fl_flags & FL_SLEEP))
			goto out;
		error = FILE_LOCK_DEFERRED;
		locks_insert_block(fl, request);
L
Linus Torvalds 已提交
770 771
		goto out;
	}
772 773
	if (request->fl_flags & FL_ACCESS)
		goto out;
774
	locks_copy_lock(new_fl, request);
P
Pavel Emelyanov 已提交
775
	locks_insert_lock(before, new_fl);
776
	new_fl = NULL;
777
	error = 0;
L
Linus Torvalds 已提交
778 779

out:
780
	unlock_flocks();
781 782
	if (new_fl)
		locks_free_lock(new_fl);
L
Linus Torvalds 已提交
783 784 785
	return error;
}

786
static int __posix_lock_file(struct inode *inode, struct file_lock *request, struct file_lock *conflock)
L
Linus Torvalds 已提交
787 788
{
	struct file_lock *fl;
789 790
	struct file_lock *new_fl = NULL;
	struct file_lock *new_fl2 = NULL;
L
Linus Torvalds 已提交
791 792 793 794 795 796 797 798
	struct file_lock *left = NULL;
	struct file_lock *right = NULL;
	struct file_lock **before;
	int error, added = 0;

	/*
	 * We may need two file_lock structures for this operation,
	 * so we get them in advance to avoid races.
799 800
	 *
	 * In some cases we can be sure, that no new locks will be needed
L
Linus Torvalds 已提交
801
	 */
802 803 804 805 806 807
	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();
	}
L
Linus Torvalds 已提交
808

809
	lock_flocks();
L
Linus Torvalds 已提交
810 811
	if (request->fl_type != F_UNLCK) {
		for_each_lock(inode, before) {
812
			fl = *before;
L
Linus Torvalds 已提交
813 814 815 816
			if (!IS_POSIX(fl))
				continue;
			if (!posix_locks_conflict(request, fl))
				continue;
817
			if (conflock)
818
				__locks_copy_lock(conflock, fl);
L
Linus Torvalds 已提交
819 820 821 822 823 824
			error = -EAGAIN;
			if (!(request->fl_flags & FL_SLEEP))
				goto out;
			error = -EDEADLK;
			if (posix_locks_deadlock(request, fl))
				goto out;
825
			error = FILE_LOCK_DEFERRED;
L
Linus Torvalds 已提交
826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852
			locks_insert_block(fl, request);
			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;
	}

	/* Process locks with this owner.  */
	while ((fl = *before) && posix_same_owner(request, fl)) {
		/* Detect adjacent or overlapping regions (if same lock type)
		 */
		if (request->fl_type == fl->fl_type) {
O
Olaf Kirch 已提交
853 854 855 856
			/* In all comparisons of start vs end, use
			 * "start - 1" rather than "end + 1". If end
			 * is OFFSET_MAX, end + 1 will become negative.
			 */
L
Linus Torvalds 已提交
857 858 859 860 861
			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.
			 */
O
Olaf Kirch 已提交
862
			if (fl->fl_start - 1 > request->fl_end)
L
Linus Torvalds 已提交
863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920
				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;
			added = 1;
		}
		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)
				added = 1;
			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;
921 922
				locks_release_private(fl);
				locks_copy_private(fl, request);
L
Linus Torvalds 已提交
923 924 925 926 927 928 929 930 931 932
				request = fl;
				added = 1;
			}
		}
		/* Go on to next lock.
		 */
	next_lock:
		before = &fl->fl_next;
	}

933 934 935 936 937 938 939 940 941 942
	/*
	 * 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 bellow this, so it's safe yet to
	 * bail out.
	 */
	error = -ENOLCK; /* "no luck" */
	if (right && left == right && !new_fl2)
		goto out;

L
Linus Torvalds 已提交
943 944
	error = 0;
	if (!added) {
945 946 947
		if (request->fl_type == F_UNLCK) {
			if (request->fl_flags & FL_EXISTS)
				error = -ENOENT;
L
Linus Torvalds 已提交
948
			goto out;
949
		}
950 951 952 953 954

		if (!new_fl) {
			error = -ENOLCK;
			goto out;
		}
L
Linus Torvalds 已提交
955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976
		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:
977
	unlock_flocks();
L
Linus Torvalds 已提交
978 979 980 981 982 983 984 985 986 987 988 989 990 991
	/*
	 * 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
992
 * @conflock: Place to return a copy of the conflicting lock, if found.
L
Linus Torvalds 已提交
993 994 995 996
 *
 * 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
997 998 999 1000
 *
 * 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.
L
Linus Torvalds 已提交
1001
 */
1002
int posix_lock_file(struct file *filp, struct file_lock *fl,
1003 1004
			struct file_lock *conflock)
{
1005
	return __posix_lock_file(filp->f_path.dentry->d_inode, fl, conflock);
L
Linus Torvalds 已提交
1006
}
1007
EXPORT_SYMBOL(posix_lock_file);
L
Linus Torvalds 已提交
1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022

/**
 * 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 (;;) {
1023
		error = posix_lock_file(filp, fl, NULL);
1024
		if (error != FILE_LOCK_DEFERRED)
L
Linus Torvalds 已提交
1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051
			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.
	 */
1052
	lock_flocks();
L
Linus Torvalds 已提交
1053 1054 1055 1056 1057 1058
	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (!IS_POSIX(fl))
			continue;
		if (fl->fl_owner != owner)
			break;
	}
1059
	unlock_flocks();
L
Linus Torvalds 已提交
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 1090 1091 1092 1093 1094
	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 (;;) {
1095
		error = __posix_lock_file(inode, &fl, NULL);
1096
		if (error != FILE_LOCK_DEFERRED)
L
Linus Torvalds 已提交
1097 1098 1099 1100 1101 1102 1103
			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.
			 */
1104
			if (__mandatory_lock(inode))
L
Linus Torvalds 已提交
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
				continue;
		}

		locks_delete_block(&fl);
		break;
	}

	return error;
}

EXPORT_SYMBOL(locks_mandatory_area);

1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127
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;
	}
}

L
Linus Torvalds 已提交
1128 1129 1130 1131 1132 1133 1134 1135
/* 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;
1136
	lease_clear_pending(fl, arg);
L
Linus Torvalds 已提交
1137
	locks_wake_up_blocks(fl);
1138 1139 1140 1141 1142
	if (arg == F_UNLCK) {
		struct file *filp = fl->fl_file;

		f_delown(filp);
		filp->f_owner.signum = 0;
1143 1144 1145 1146 1147
		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;
		}
L
Linus Torvalds 已提交
1148
		locks_delete_lock(before);
1149
	}
L
Linus Torvalds 已提交
1150 1151 1152 1153 1154
	return 0;
}

EXPORT_SYMBOL(lease_modify);

1155 1156 1157 1158 1159 1160 1161 1162
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);
}

L
Linus Torvalds 已提交
1163 1164 1165 1166 1167 1168
static void time_out_leases(struct inode *inode)
{
	struct file_lock **before;
	struct file_lock *fl;

	before = &inode->i_flock;
J
J. Bruce Fields 已提交
1169
	while ((fl = *before) && IS_LEASE(fl) && lease_breaking(fl)) {
1170 1171 1172 1173
		if (past_time(fl->fl_downgrade_time))
			lease_modify(before, F_RDLCK);
		if (past_time(fl->fl_break_time))
			lease_modify(before, F_UNLCK);
L
Linus Torvalds 已提交
1174 1175 1176 1177 1178 1179 1180 1181 1182 1183
		if (fl == *before)	/* lease_modify may have freed fl */
			before = &fl->fl_next;
	}
}

/**
 *	__break_lease	-	revoke all outstanding leases on file
 *	@inode: the inode of the file to return
 *	@mode: the open mode (read or write)
 *
1184 1185 1186
 *	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
L
Linus Torvalds 已提交
1187 1188 1189 1190
 *	specified %O_NONBLOCK to your open().
 */
int __break_lease(struct inode *inode, unsigned int mode)
{
1191
	int error = 0;
L
Linus Torvalds 已提交
1192 1193 1194 1195
	struct file_lock *new_fl, *flock;
	struct file_lock *fl;
	unsigned long break_time;
	int i_have_this_lease = 0;
1196
	int want_write = (mode & O_ACCMODE) != O_RDONLY;
L
Linus Torvalds 已提交
1197

1198
	new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK);
1199 1200
	if (IS_ERR(new_fl))
		return PTR_ERR(new_fl);
L
Linus Torvalds 已提交
1201

1202
	lock_flocks();
L
Linus Torvalds 已提交
1203 1204 1205 1206 1207 1208 1209

	time_out_leases(inode);

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

1210 1211 1212
	if (!locks_conflict(flock, new_fl))
		goto out;

L
Linus Torvalds 已提交
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next)
		if (fl->fl_owner == current->files)
			i_have_this_lease = 1;

	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) {
1225 1226 1227 1228
		if (want_write) {
			if (fl->fl_flags & FL_UNLOCK_PENDING)
				continue;
			fl->fl_flags |= FL_UNLOCK_PENDING;
L
Linus Torvalds 已提交
1229
			fl->fl_break_time = break_time;
1230 1231 1232 1233 1234
		} else {
			if (lease_breaking(flock))
				continue;
			fl->fl_flags |= FL_DOWNGRADE_PENDING;
			fl->fl_downgrade_time = break_time;
L
Linus Torvalds 已提交
1235
		}
1236
		fl->fl_lmops->lm_break(fl);
L
Linus Torvalds 已提交
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250
	}

	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++;
	}
1251
	locks_insert_block(flock, new_fl);
1252
	unlock_flocks();
1253 1254
	error = wait_event_interruptible_timeout(new_fl->fl_wait,
						!new_fl->fl_next, break_time);
1255
	lock_flocks();
1256
	__locks_delete_block(new_fl);
L
Linus Torvalds 已提交
1257 1258 1259
	if (error >= 0) {
		if (error == 0)
			time_out_leases(inode);
1260 1261 1262 1263
		/*
		 * Wait for the next conflicting lease that has not been
		 * broken yet
		 */
L
Linus Torvalds 已提交
1264 1265
		for (flock = inode->i_flock; flock && IS_LEASE(flock);
				flock = flock->fl_next) {
1266
			if (locks_conflict(new_fl, flock))
L
Linus Torvalds 已提交
1267 1268 1269 1270 1271 1272
				goto restart;
		}
		error = 0;
	}

out:
1273
	unlock_flocks();
1274
	locks_free_lock(new_fl);
L
Linus Torvalds 已提交
1275 1276 1277 1278 1279 1280
	return error;
}

EXPORT_SYMBOL(__break_lease);

/**
1281
 *	lease_get_mtime - get the last modified time of an inode
L
Linus Torvalds 已提交
1282 1283 1284 1285 1286
 *	@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
1287
 * exclusive lease, then they could be modifying it.
L
Linus Torvalds 已提交
1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327
 */
void lease_get_mtime(struct inode *inode, struct timespec *time)
{
	struct file_lock *flock = inode->i_flock;
	if (flock && IS_LEASE(flock) && (flock->fl_type & F_WRLCK))
		*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;
	int type = F_UNLCK;

1328
	lock_flocks();
1329 1330
	time_out_leases(filp->f_path.dentry->d_inode);
	for (fl = filp->f_path.dentry->d_inode->i_flock; fl && IS_LEASE(fl);
L
Linus Torvalds 已提交
1331 1332
			fl = fl->fl_next) {
		if (fl->fl_file == filp) {
1333
			type = target_leasetype(fl);
L
Linus Torvalds 已提交
1334 1335 1336
			break;
		}
	}
1337
	unlock_flocks();
L
Linus Torvalds 已提交
1338 1339 1340
	return type;
}

1341
int generic_add_lease(struct file *filp, long arg, struct file_lock **flp)
L
Linus Torvalds 已提交
1342
{
1343
	struct file_lock *fl, **before, **my_before = NULL, *lease;
1344
	struct dentry *dentry = filp->f_path.dentry;
L
Linus Torvalds 已提交
1345
	struct inode *inode = dentry->d_inode;
J
J. Bruce Fields 已提交
1346
	int error;
L
Linus Torvalds 已提交
1347

1348 1349
	lease = *flp;

1350 1351
	error = -EAGAIN;
	if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0))
1352
		goto out;
1353 1354 1355
	if ((arg == F_WRLCK)
	    && ((dentry->d_count > 1)
		|| (atomic_read(&inode->i_count) > 1)))
1356
		goto out;
1357

L
Linus Torvalds 已提交
1358 1359 1360 1361 1362 1363 1364 1365
	/*
	 * At this point, we know that if there is an exclusive
	 * lease on this file, then we hold it on this filp
	 * (otherwise our open of this file would have blocked).
	 * And if we are trying to acquire an exclusive lease,
	 * then the file is not open by anyone (including us)
	 * except for this filp.
	 */
J
J. Bruce Fields 已提交
1366
	error = -EAGAIN;
L
Linus Torvalds 已提交
1367 1368 1369
	for (before = &inode->i_flock;
			((fl = *before) != NULL) && IS_LEASE(fl);
			before = &fl->fl_next) {
J
J. Bruce Fields 已提交
1370
		if (fl->fl_file == filp) {
L
Linus Torvalds 已提交
1371
			my_before = before;
J
J. Bruce Fields 已提交
1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385
			continue;
		}
		/*
		 * No exclusive leases if someone else has a lease on
		 * this file:
		 */
		if (arg == F_WRLCK)
			goto out;
		/*
		 * Modifying our existing lease is OK, but no getting a
		 * new lease if someone else is opening for write:
		 */
		if (fl->fl_flags & FL_UNLOCK_PENDING)
			goto out;
L
Linus Torvalds 已提交
1386 1387 1388
	}

	if (my_before != NULL) {
J
J. Bruce Fields 已提交
1389
		error = lease->fl_lmops->lm_change(my_before, arg);
1390 1391
		if (!error)
			*flp = *my_before;
L
Linus Torvalds 已提交
1392 1393 1394 1395 1396 1397 1398
		goto out;
	}

	error = -EINVAL;
	if (!leases_enable)
		goto out;

1399
	locks_insert_lock(before, lease);
1400
	return 0;
L
Linus Torvalds 已提交
1401 1402 1403 1404

out:
	return error;
}
1405 1406 1407 1408 1409 1410 1411 1412 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

int generic_delete_lease(struct file *filp, struct file_lock **flp)
{
	struct file_lock *fl, **before;
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;

	for (before = &inode->i_flock;
			((fl = *before) != NULL) && IS_LEASE(fl);
			before = &fl->fl_next) {
		if (fl->fl_file != filp)
			continue;
		return (*flp)->fl_lmops->lm_change(before, F_UNLCK);
	}
	return -EAGAIN;
}

/**
 *	generic_setlease	-	sets a lease on an open file
 *	@filp: file pointer
 *	@arg: type of lease to obtain
 *	@flp: input - file_lock to use, output - file_lock inserted
 *
 *	The (input) flp->fl_lmops->lm_break function is required
 *	by break_lease().
 *
 *	Called with file_lock_lock held.
 */
int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
{
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;
	int error;

1439
	if ((!uid_eq(current_fsuid(), inode->i_uid)) && !capable(CAP_LEASE))
1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457
		return -EACCES;
	if (!S_ISREG(inode->i_mode))
		return -EINVAL;
	error = security_file_lock(filp, arg);
	if (error)
		return error;

	time_out_leases(inode);

	BUG_ON(!(*flp)->fl_lmops->lm_break);

	switch (arg) {
	case F_UNLCK:
		return generic_delete_lease(filp, flp);
	case F_RDLCK:
	case F_WRLCK:
		return generic_add_lease(filp, arg, flp);
	default:
1458
		return -EINVAL;
1459 1460
	}
}
1461
EXPORT_SYMBOL(generic_setlease);
L
Linus Torvalds 已提交
1462

1463 1464 1465 1466 1467 1468 1469 1470 1471
static int __vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
{
	if (filp->f_op && filp->f_op->setlease)
		return filp->f_op->setlease(filp, arg, lease);
	else
		return generic_setlease(filp, arg, lease);
}

/**
1472
 *	vfs_setlease        -       sets a lease on an open file
L
Linus Torvalds 已提交
1473 1474 1475 1476 1477
 *	@filp: file pointer
 *	@arg: type of lease to obtain
 *	@lease: file_lock to use
 *
 *	Call this to establish a lease on the file.
J
J. Bruce Fields 已提交
1478
 *	The (*lease)->fl_lmops->lm_break operation must be set; if not,
1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489
 *	break_lease will oops!
 *
 *	This will call the filesystem's setlease file method, if
 *	defined.  Note that there is no getlease method; instead, the
 *	filesystem setlease method should call back to setlease() to
 *	add a lease to the inode's lease list, where fcntl_getlease() can
 *	find it.  Since fcntl_getlease() only reports whether the current
 *	task holds a lease, a cluster filesystem need only do this for
 *	leases held by processes on this node.
 *
 *	There is also no break_lease method; filesystems that
1490
 *	handle their own leases should break leases themselves from the
1491 1492 1493 1494 1495
 *	filesystem's open, create, and (on truncate) setattr methods.
 *
 *	Warning: the only current setlease methods exist only to disable
 *	leases in certain cases.  More vfs changes may be required to
 *	allow a full filesystem lease implementation.
L
Linus Torvalds 已提交
1496 1497
 */

1498
int vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
L
Linus Torvalds 已提交
1499 1500 1501
{
	int error;

1502 1503 1504
	lock_flocks();
	error = __vfs_setlease(filp, arg, lease);
	unlock_flocks();
L
Linus Torvalds 已提交
1505 1506 1507

	return error;
}
1508
EXPORT_SYMBOL_GPL(vfs_setlease);
L
Linus Torvalds 已提交
1509

1510 1511 1512 1513 1514 1515 1516 1517 1518 1519
static int do_fcntl_delete_lease(struct file *filp)
{
	struct file_lock fl, *flp = &fl;

	lease_init(filp, F_UNLCK, flp);

	return vfs_setlease(filp, F_UNLCK, &flp);
}

static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg)
L
Linus Torvalds 已提交
1520
{
1521
	struct file_lock *fl, *ret;
1522
	struct fasync_struct *new;
L
Linus Torvalds 已提交
1523 1524
	int error;

1525 1526 1527
	fl = lease_alloc(filp, arg);
	if (IS_ERR(fl))
		return PTR_ERR(fl);
L
Linus Torvalds 已提交
1528

1529 1530 1531 1532 1533
	new = fasync_alloc();
	if (!new) {
		locks_free_lock(fl);
		return -ENOMEM;
	}
1534
	ret = fl;
1535
	lock_flocks();
1536
	error = __vfs_setlease(filp, arg, &ret);
1537 1538 1539 1540 1541
	if (error) {
		unlock_flocks();
		locks_free_lock(fl);
		goto out_free_fasync;
	}
1542 1543
	if (ret != fl)
		locks_free_lock(fl);
L
Linus Torvalds 已提交
1544

1545 1546 1547 1548 1549 1550
	/*
	 * fasync_insert_entry() returns the old entry if any.
	 * If there was no old entry, then it used 'new' and
	 * inserted it into the fasync list. Clear new so that
	 * we don't release it here.
	 */
1551
	if (!fasync_insert_entry(fd, filp, &ret->fl_fasync, new))
1552 1553
		new = NULL;

1554
	error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
1555
	unlock_flocks();
1556 1557

out_free_fasync:
1558 1559
	if (new)
		fasync_free(new);
L
Linus Torvalds 已提交
1560 1561 1562
	return error;
}

1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579
/**
 *	fcntl_setlease	-	sets a lease on an open file
 *	@fd: open file descriptor
 *	@filp: file pointer
 *	@arg: type of lease to obtain
 *
 *	Call this fcntl to establish a lease on the file.
 *	Note that you also need to call %F_SETSIG to
 *	receive a signal when the lease is broken.
 */
int fcntl_setlease(unsigned int fd, struct file *filp, long arg)
{
	if (arg == F_UNLCK)
		return do_fcntl_delete_lease(filp);
	return do_fcntl_add_lease(fd, filp, arg);
}

L
Linus Torvalds 已提交
1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592
/**
 * flock_lock_file_wait - Apply a FLOCK-style lock to a file
 * @filp: The file to apply the lock to
 * @fl: The lock to be applied
 *
 * Add a FLOCK style lock to a file.
 */
int flock_lock_file_wait(struct file *filp, struct file_lock *fl)
{
	int error;
	might_sleep();
	for (;;) {
		error = flock_lock_file(filp, fl);
1593
		if (error != FILE_LOCK_DEFERRED)
L
Linus Torvalds 已提交
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 1619 1620 1621 1622 1623 1624 1625
			break;
		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
		if (!error)
			continue;

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

EXPORT_SYMBOL(flock_lock_file_wait);

/**
 *	sys_flock: - flock() system call.
 *	@fd: the file descriptor to lock.
 *	@cmd: the type of lock to apply.
 *
 *	Apply a %FL_FLOCK style lock to an open file descriptor.
 *	The @cmd can be one of
 *
 *	%LOCK_SH -- a shared lock.
 *
 *	%LOCK_EX -- an exclusive lock.
 *
 *	%LOCK_UN -- remove an existing lock.
 *
 *	%LOCK_MAND -- a `mandatory' flock.  This exists to emulate Windows Share Modes.
 *
 *	%LOCK_MAND can be combined with %LOCK_READ or %LOCK_WRITE to allow other
 *	processes read and write access respectively.
 */
1626
SYSCALL_DEFINE2(flock, unsigned int, fd, unsigned int, cmd)
L
Linus Torvalds 已提交
1627 1628
{
	struct file *filp;
A
Al Viro 已提交
1629
	int fput_needed;
L
Linus Torvalds 已提交
1630 1631 1632 1633 1634
	struct file_lock *lock;
	int can_sleep, unlock;
	int error;

	error = -EBADF;
A
Al Viro 已提交
1635
	filp = fget_light(fd, &fput_needed);
L
Linus Torvalds 已提交
1636 1637 1638 1639 1640 1641 1642
	if (!filp)
		goto out;

	can_sleep = !(cmd & LOCK_NB);
	cmd &= ~LOCK_NB;
	unlock = (cmd == LOCK_UN);

1643 1644
	if (!unlock && !(cmd & LOCK_MAND) &&
	    !(filp->f_mode & (FMODE_READ|FMODE_WRITE)))
L
Linus Torvalds 已提交
1645 1646 1647 1648 1649 1650 1651 1652
		goto out_putf;

	error = flock_make_lock(filp, &lock, cmd);
	if (error)
		goto out_putf;
	if (can_sleep)
		lock->fl_flags |= FL_SLEEP;

1653
	error = security_file_lock(filp, lock->fl_type);
L
Linus Torvalds 已提交
1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664
	if (error)
		goto out_free;

	if (filp->f_op && filp->f_op->flock)
		error = filp->f_op->flock(filp,
					  (can_sleep) ? F_SETLKW : F_SETLK,
					  lock);
	else
		error = flock_lock_file_wait(filp, lock);

 out_free:
1665
	locks_free_lock(lock);
L
Linus Torvalds 已提交
1666 1667

 out_putf:
A
Al Viro 已提交
1668
	fput_light(filp, fput_needed);
L
Linus Torvalds 已提交
1669 1670 1671 1672
 out:
	return error;
}

1673 1674 1675
/**
 * vfs_test_lock - test file byte range lock
 * @filp: The file to test lock for
1676
 * @fl: The lock to test; also used to hold result
1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689
 *
 * Returns -ERRNO on failure.  Indicates presence of conflicting lock by
 * setting conf->fl_type to something other than F_UNLCK.
 */
int vfs_test_lock(struct file *filp, struct file_lock *fl)
{
	if (filp->f_op && filp->f_op->lock)
		return filp->f_op->lock(filp, F_GETLK, fl);
	posix_test_lock(filp, fl);
	return 0;
}
EXPORT_SYMBOL_GPL(vfs_test_lock);

1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706
static int posix_lock_to_flock(struct flock *flock, struct file_lock *fl)
{
	flock->l_pid = fl->fl_pid;
#if BITS_PER_LONG == 32
	/*
	 * Make sure we can represent the posix lock via
	 * legacy 32bit flock.
	 */
	if (fl->fl_start > OFFT_OFFSET_MAX)
		return -EOVERFLOW;
	if (fl->fl_end != OFFSET_MAX && fl->fl_end > OFFT_OFFSET_MAX)
		return -EOVERFLOW;
#endif
	flock->l_start = fl->fl_start;
	flock->l_len = fl->fl_end == OFFSET_MAX ? 0 :
		fl->fl_end - fl->fl_start + 1;
	flock->l_whence = 0;
1707
	flock->l_type = fl->fl_type;
1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722
	return 0;
}

#if BITS_PER_LONG == 32
static void posix_lock_to_flock64(struct flock64 *flock, struct file_lock *fl)
{
	flock->l_pid = fl->fl_pid;
	flock->l_start = fl->fl_start;
	flock->l_len = fl->fl_end == OFFSET_MAX ? 0 :
		fl->fl_end - fl->fl_start + 1;
	flock->l_whence = 0;
	flock->l_type = fl->fl_type;
}
#endif

L
Linus Torvalds 已提交
1723 1724 1725 1726 1727
/* Report the first existing lock that would conflict with l.
 * This implements the F_GETLK command of fcntl().
 */
int fcntl_getlk(struct file *filp, struct flock __user *l)
{
1728
	struct file_lock file_lock;
L
Linus Torvalds 已提交
1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742
	struct flock flock;
	int error;

	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;
	error = -EINVAL;
	if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
		goto out;

	error = flock_to_posix_lock(filp, &file_lock, &flock);
	if (error)
		goto out;

1743 1744 1745
	error = vfs_test_lock(filp, &file_lock);
	if (error)
		goto out;
L
Linus Torvalds 已提交
1746
 
1747 1748 1749
	flock.l_type = file_lock.fl_type;
	if (file_lock.fl_type != F_UNLCK) {
		error = posix_lock_to_flock(&flock, &file_lock);
1750
		if (error)
L
Linus Torvalds 已提交
1751 1752 1753 1754 1755 1756 1757 1758 1759
			goto out;
	}
	error = -EFAULT;
	if (!copy_to_user(l, &flock, sizeof(flock)))
		error = 0;
out:
	return error;
}

1760 1761 1762 1763 1764
/**
 * vfs_lock_file - file byte range lock
 * @filp: The file to apply the lock to
 * @cmd: type of locking operation (F_SETLK, F_GETLK, etc.)
 * @fl: The lock to be applied
1765 1766 1767 1768 1769 1770 1771 1772
 * @conf: Place to return a copy of the conflicting lock, if found.
 *
 * A caller that doesn't care about the conflicting lock may pass NULL
 * as the final argument.
 *
 * If the filesystem defines a private ->lock() method, then @conf will
 * be left unchanged; so a caller that cares should initialize it to
 * some acceptable default.
1773 1774 1775 1776
 *
 * To avoid blocking kernel daemons, such as lockd, that need to acquire POSIX
 * locks, the ->lock() interface may return asynchronously, before the lock has
 * been granted or denied by the underlying filesystem, if (and only if)
J
J. Bruce Fields 已提交
1777
 * lm_grant is set. Callers expecting ->lock() to return asynchronously
1778 1779
 * will only use F_SETLK, not F_SETLKW; they will set FL_SLEEP if (and only if)
 * the request is for a blocking lock. When ->lock() does return asynchronously,
J
J. Bruce Fields 已提交
1780
 * it must return FILE_LOCK_DEFERRED, and call ->lm_grant() when the lock
1781 1782
 * request completes.
 * If the request is for non-blocking lock the file system should return
1783 1784
 * FILE_LOCK_DEFERRED then try to get the lock and call the callback routine
 * with the result. If the request timed out the callback routine will return a
1785 1786 1787 1788 1789
 * nonzero return code and the file system should release the lock. The file
 * system is also responsible to keep a corresponding posix lock when it
 * grants a lock so the VFS can find out which locks are locally held and do
 * the correct lock cleanup when required.
 * The underlying filesystem must not drop the kernel lock or call
J
J. Bruce Fields 已提交
1790
 * ->lm_grant() before returning to the caller with a FILE_LOCK_DEFERRED
1791
 * return code.
1792
 */
1793
int vfs_lock_file(struct file *filp, unsigned int cmd, struct file_lock *fl, struct file_lock *conf)
1794 1795 1796 1797
{
	if (filp->f_op && filp->f_op->lock)
		return filp->f_op->lock(filp, cmd, fl);
	else
1798
		return posix_lock_file(filp, fl, conf);
1799 1800 1801
}
EXPORT_SYMBOL_GPL(vfs_lock_file);

M
Miklos Szeredi 已提交
1802 1803 1804 1805 1806 1807 1808 1809 1810
static int do_lock_file_wait(struct file *filp, unsigned int cmd,
			     struct file_lock *fl)
{
	int error;

	error = security_file_lock(filp, fl->fl_type);
	if (error)
		return error;

1811 1812 1813
	for (;;) {
		error = vfs_lock_file(filp, cmd, fl, NULL);
		if (error != FILE_LOCK_DEFERRED)
M
Miklos Szeredi 已提交
1814
			break;
1815 1816 1817 1818 1819 1820
		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
		if (!error)
			continue;

		locks_delete_block(fl);
		break;
M
Miklos Szeredi 已提交
1821 1822 1823 1824 1825
	}

	return error;
}

L
Linus Torvalds 已提交
1826 1827 1828
/* Apply the lock described by l to an open file descriptor.
 * This implements both the F_SETLK and F_SETLKW commands of fcntl().
 */
1829 1830
int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
		struct flock __user *l)
L
Linus Torvalds 已提交
1831 1832 1833 1834
{
	struct file_lock *file_lock = locks_alloc_lock();
	struct flock flock;
	struct inode *inode;
1835
	struct file *f;
L
Linus Torvalds 已提交
1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847
	int error;

	if (file_lock == NULL)
		return -ENOLCK;

	/*
	 * This might block, so we do it before checking the inode.
	 */
	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;

1848
	inode = filp->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
1849 1850 1851 1852

	/* Don't allow mandatory locks on files that may be memory mapped
	 * and shared.
	 */
1853
	if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) {
L
Linus Torvalds 已提交
1854 1855 1856 1857
		error = -EAGAIN;
		goto out;
	}

1858
again:
L
Linus Torvalds 已提交
1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882
	error = flock_to_posix_lock(filp, file_lock, &flock);
	if (error)
		goto out;
	if (cmd == F_SETLKW) {
		file_lock->fl_flags |= FL_SLEEP;
	}
	
	error = -EBADF;
	switch (flock.l_type) {
	case F_RDLCK:
		if (!(filp->f_mode & FMODE_READ))
			goto out;
		break;
	case F_WRLCK:
		if (!(filp->f_mode & FMODE_WRITE))
			goto out;
		break;
	case F_UNLCK:
		break;
	default:
		error = -EINVAL;
		goto out;
	}

M
Miklos Szeredi 已提交
1883
	error = do_lock_file_wait(filp, cmd, file_lock);
L
Linus Torvalds 已提交
1884

1885 1886 1887 1888
	/*
	 * Attempt to detect a close/fcntl race and recover by
	 * releasing the lock that was just acquired.
	 */
1889 1890 1891 1892 1893 1894 1895 1896 1897
	/*
	 * we need that spin_lock here - it prevents reordering between
	 * update of inode->i_flock and check for it done in close().
	 * rcu_read_lock() wouldn't do.
	 */
	spin_lock(&current->files->file_lock);
	f = fcheck(fd);
	spin_unlock(&current->files->file_lock);
	if (!error && f != filp && flock.l_type != F_UNLCK) {
1898 1899
		flock.l_type = F_UNLCK;
		goto again;
L
Linus Torvalds 已提交
1900 1901
	}

1902
out:
L
Linus Torvalds 已提交
1903 1904 1905 1906 1907 1908 1909 1910 1911 1912
	locks_free_lock(file_lock);
	return error;
}

#if BITS_PER_LONG == 32
/* Report the first existing lock that would conflict with l.
 * This implements the F_GETLK command of fcntl().
 */
int fcntl_getlk64(struct file *filp, struct flock64 __user *l)
{
1913
	struct file_lock file_lock;
L
Linus Torvalds 已提交
1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927
	struct flock64 flock;
	int error;

	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;
	error = -EINVAL;
	if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
		goto out;

	error = flock64_to_posix_lock(filp, &file_lock, &flock);
	if (error)
		goto out;

1928 1929 1930 1931
	error = vfs_test_lock(filp, &file_lock);
	if (error)
		goto out;

1932 1933 1934 1935
	flock.l_type = file_lock.fl_type;
	if (file_lock.fl_type != F_UNLCK)
		posix_lock_to_flock64(&flock, &file_lock);

L
Linus Torvalds 已提交
1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946
	error = -EFAULT;
	if (!copy_to_user(l, &flock, sizeof(flock)))
		error = 0;
  
out:
	return error;
}

/* Apply the lock described by l to an open file descriptor.
 * This implements both the F_SETLK and F_SETLKW commands of fcntl().
 */
1947 1948
int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
		struct flock64 __user *l)
L
Linus Torvalds 已提交
1949 1950 1951 1952
{
	struct file_lock *file_lock = locks_alloc_lock();
	struct flock64 flock;
	struct inode *inode;
1953
	struct file *f;
L
Linus Torvalds 已提交
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965
	int error;

	if (file_lock == NULL)
		return -ENOLCK;

	/*
	 * This might block, so we do it before checking the inode.
	 */
	error = -EFAULT;
	if (copy_from_user(&flock, l, sizeof(flock)))
		goto out;

1966
	inode = filp->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
1967 1968 1969 1970

	/* Don't allow mandatory locks on files that may be memory mapped
	 * and shared.
	 */
1971
	if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) {
L
Linus Torvalds 已提交
1972 1973 1974 1975
		error = -EAGAIN;
		goto out;
	}

1976
again:
L
Linus Torvalds 已提交
1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
	error = flock64_to_posix_lock(filp, file_lock, &flock);
	if (error)
		goto out;
	if (cmd == F_SETLKW64) {
		file_lock->fl_flags |= FL_SLEEP;
	}
	
	error = -EBADF;
	switch (flock.l_type) {
	case F_RDLCK:
		if (!(filp->f_mode & FMODE_READ))
			goto out;
		break;
	case F_WRLCK:
		if (!(filp->f_mode & FMODE_WRITE))
			goto out;
		break;
	case F_UNLCK:
		break;
	default:
		error = -EINVAL;
		goto out;
	}

M
Miklos Szeredi 已提交
2001
	error = do_lock_file_wait(filp, cmd, file_lock);
L
Linus Torvalds 已提交
2002

2003 2004 2005 2006
	/*
	 * Attempt to detect a close/fcntl race and recover by
	 * releasing the lock that was just acquired.
	 */
2007 2008 2009 2010
	spin_lock(&current->files->file_lock);
	f = fcheck(fd);
	spin_unlock(&current->files->file_lock);
	if (!error && f != filp && flock.l_type != F_UNLCK) {
2011 2012
		flock.l_type = F_UNLCK;
		goto again;
L
Linus Torvalds 已提交
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027
	}

out:
	locks_free_lock(file_lock);
	return error;
}
#endif /* BITS_PER_LONG == 32 */

/*
 * This function is called when the file is being removed
 * from the task's fd array.  POSIX locks belonging to this task
 * are deleted at this time.
 */
void locks_remove_posix(struct file *filp, fl_owner_t owner)
{
2028
	struct file_lock lock;
L
Linus Torvalds 已提交
2029 2030 2031 2032 2033 2034

	/*
	 * If there are no locks held on this file, we don't need to call
	 * posix_lock_file().  Another process could be setting a lock on this
	 * file at the same time, but we wouldn't remove that lock anyway.
	 */
2035
	if (!filp->f_path.dentry->d_inode->i_flock)
L
Linus Torvalds 已提交
2036 2037 2038
		return;

	lock.fl_type = F_UNLCK;
2039
	lock.fl_flags = FL_POSIX | FL_CLOSE;
L
Linus Torvalds 已提交
2040 2041 2042 2043 2044 2045 2046 2047
	lock.fl_start = 0;
	lock.fl_end = OFFSET_MAX;
	lock.fl_owner = owner;
	lock.fl_pid = current->tgid;
	lock.fl_file = filp;
	lock.fl_ops = NULL;
	lock.fl_lmops = NULL;

2048
	vfs_lock_file(filp, F_SETLK, &lock, NULL);
L
Linus Torvalds 已提交
2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060

	if (lock.fl_ops && lock.fl_ops->fl_release_private)
		lock.fl_ops->fl_release_private(&lock);
}

EXPORT_SYMBOL(locks_remove_posix);

/*
 * This function is called on the last close of an open file.
 */
void locks_remove_flock(struct file *filp)
{
2061
	struct inode * inode = filp->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076
	struct file_lock *fl;
	struct file_lock **before;

	if (!inode->i_flock)
		return;

	if (filp->f_op && filp->f_op->flock) {
		struct file_lock fl = {
			.fl_pid = current->tgid,
			.fl_file = filp,
			.fl_flags = FL_FLOCK,
			.fl_type = F_UNLCK,
			.fl_end = OFFSET_MAX,
		};
		filp->f_op->flock(filp, F_SETLKW, &fl);
2077 2078
		if (fl.fl_ops && fl.fl_ops->fl_release_private)
			fl.fl_ops->fl_release_private(&fl);
L
Linus Torvalds 已提交
2079 2080
	}

2081
	lock_flocks();
L
Linus Torvalds 已提交
2082 2083 2084 2085
	before = &inode->i_flock;

	while ((fl = *before) != NULL) {
		if (fl->fl_file == filp) {
2086
			if (IS_FLOCK(fl)) {
L
Linus Torvalds 已提交
2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098
				locks_delete_lock(before);
				continue;
			}
			if (IS_LEASE(fl)) {
				lease_modify(before, F_UNLCK);
				continue;
			}
			/* What? */
			BUG();
 		}
		before = &fl->fl_next;
	}
2099
	unlock_flocks();
L
Linus Torvalds 已提交
2100 2101 2102 2103 2104 2105 2106 2107 2108
}

/**
 *	posix_unblock_lock - stop waiting for a file lock
 *      @filp:   how the file was opened
 *	@waiter: the lock which was waiting
 *
 *	lockd needs to block waiting for locks.
 */
J
J. Bruce Fields 已提交
2109
int
L
Linus Torvalds 已提交
2110 2111
posix_unblock_lock(struct file *filp, struct file_lock *waiter)
{
J
J. Bruce Fields 已提交
2112 2113
	int status = 0;

2114
	lock_flocks();
2115
	if (waiter->fl_next)
L
Linus Torvalds 已提交
2116
		__locks_delete_block(waiter);
J
J. Bruce Fields 已提交
2117 2118
	else
		status = -ENOENT;
2119
	unlock_flocks();
J
J. Bruce Fields 已提交
2120
	return status;
L
Linus Torvalds 已提交
2121 2122 2123 2124
}

EXPORT_SYMBOL(posix_unblock_lock);

M
Marc Eshel 已提交
2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140
/**
 * vfs_cancel_lock - file byte range unblock lock
 * @filp: The file to apply the unblock to
 * @fl: The lock to be unblocked
 *
 * Used by lock managers to cancel blocked requests
 */
int vfs_cancel_lock(struct file *filp, struct file_lock *fl)
{
	if (filp->f_op && filp->f_op->lock)
		return filp->f_op->lock(filp, F_CANCELLK, fl);
	return 0;
}

EXPORT_SYMBOL_GPL(vfs_cancel_lock);

2141
#ifdef CONFIG_PROC_FS
2142
#include <linux/proc_fs.h>
2143 2144 2145
#include <linux/seq_file.h>

static void lock_get_status(struct seq_file *f, struct file_lock *fl,
2146
			    loff_t id, char *pfx)
L
Linus Torvalds 已提交
2147 2148
{
	struct inode *inode = NULL;
2149 2150 2151
	unsigned int fl_pid;

	if (fl->fl_nspid)
2152
		fl_pid = pid_vnr(fl->fl_nspid);
2153 2154
	else
		fl_pid = fl->fl_pid;
L
Linus Torvalds 已提交
2155 2156

	if (fl->fl_file != NULL)
2157
		inode = fl->fl_file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
2158

2159
	seq_printf(f, "%lld:%s ", id, pfx);
L
Linus Torvalds 已提交
2160
	if (IS_POSIX(fl)) {
2161
		seq_printf(f, "%6s %s ",
L
Linus Torvalds 已提交
2162 2163
			     (fl->fl_flags & FL_ACCESS) ? "ACCESS" : "POSIX ",
			     (inode == NULL) ? "*NOINODE*" :
2164
			     mandatory_lock(inode) ? "MANDATORY" : "ADVISORY ");
L
Linus Torvalds 已提交
2165 2166
	} else if (IS_FLOCK(fl)) {
		if (fl->fl_type & LOCK_MAND) {
2167
			seq_printf(f, "FLOCK  MSNFS     ");
L
Linus Torvalds 已提交
2168
		} else {
2169
			seq_printf(f, "FLOCK  ADVISORY  ");
L
Linus Torvalds 已提交
2170 2171
		}
	} else if (IS_LEASE(fl)) {
2172
		seq_printf(f, "LEASE  ");
J
J. Bruce Fields 已提交
2173
		if (lease_breaking(fl))
2174
			seq_printf(f, "BREAKING  ");
L
Linus Torvalds 已提交
2175
		else if (fl->fl_file)
2176
			seq_printf(f, "ACTIVE    ");
L
Linus Torvalds 已提交
2177
		else
2178
			seq_printf(f, "BREAKER   ");
L
Linus Torvalds 已提交
2179
	} else {
2180
		seq_printf(f, "UNKNOWN UNKNOWN  ");
L
Linus Torvalds 已提交
2181 2182
	}
	if (fl->fl_type & LOCK_MAND) {
2183
		seq_printf(f, "%s ",
L
Linus Torvalds 已提交
2184 2185 2186 2187
			       (fl->fl_type & LOCK_READ)
			       ? (fl->fl_type & LOCK_WRITE) ? "RW   " : "READ "
			       : (fl->fl_type & LOCK_WRITE) ? "WRITE" : "NONE ");
	} else {
2188
		seq_printf(f, "%s ",
J
J. Bruce Fields 已提交
2189
			       (lease_breaking(fl))
L
Linus Torvalds 已提交
2190 2191 2192 2193 2194
			       ? (fl->fl_type & F_UNLCK) ? "UNLCK" : "READ "
			       : (fl->fl_type & F_WRLCK) ? "WRITE" : "READ ");
	}
	if (inode) {
#ifdef WE_CAN_BREAK_LSLK_NOW
2195
		seq_printf(f, "%d %s:%ld ", fl_pid,
L
Linus Torvalds 已提交
2196 2197 2198
				inode->i_sb->s_id, inode->i_ino);
#else
		/* userspace relies on this representation of dev_t ;-( */
2199
		seq_printf(f, "%d %02x:%02x:%ld ", fl_pid,
L
Linus Torvalds 已提交
2200 2201 2202 2203
				MAJOR(inode->i_sb->s_dev),
				MINOR(inode->i_sb->s_dev), inode->i_ino);
#endif
	} else {
2204
		seq_printf(f, "%d <none>:0 ", fl_pid);
L
Linus Torvalds 已提交
2205 2206 2207
	}
	if (IS_POSIX(fl)) {
		if (fl->fl_end == OFFSET_MAX)
2208
			seq_printf(f, "%Ld EOF\n", fl->fl_start);
L
Linus Torvalds 已提交
2209
		else
2210
			seq_printf(f, "%Ld %Ld\n", fl->fl_start, fl->fl_end);
L
Linus Torvalds 已提交
2211
	} else {
2212
		seq_printf(f, "0 EOF\n");
L
Linus Torvalds 已提交
2213 2214 2215
	}
}

2216
static int locks_show(struct seq_file *f, void *v)
L
Linus Torvalds 已提交
2217
{
2218
	struct file_lock *fl, *bfl;
L
Linus Torvalds 已提交
2219

2220
	fl = list_entry(v, struct file_lock, fl_link);
L
Linus Torvalds 已提交
2221

2222
	lock_get_status(f, fl, *((loff_t *)f->private), "");
L
Linus Torvalds 已提交
2223

2224
	list_for_each_entry(bfl, &fl->fl_block, fl_block)
2225
		lock_get_status(f, bfl, *((loff_t *)f->private), " ->");
2226

2227 2228
	return 0;
}
L
Linus Torvalds 已提交
2229

2230 2231
static void *locks_start(struct seq_file *f, loff_t *pos)
{
2232 2233
	loff_t *p = f->private;

2234
	lock_flocks();
2235
	*p = (*pos + 1);
2236 2237
	return seq_list_start(&file_lock_list, *pos);
}
L
Linus Torvalds 已提交
2238

2239 2240
static void *locks_next(struct seq_file *f, void *v, loff_t *pos)
{
2241 2242
	loff_t *p = f->private;
	++*p;
2243 2244
	return seq_list_next(v, &file_lock_list, pos);
}
L
Linus Torvalds 已提交
2245

2246 2247
static void locks_stop(struct seq_file *f, void *v)
{
2248
	unlock_flocks();
L
Linus Torvalds 已提交
2249 2250
}

2251
static const struct seq_operations locks_seq_operations = {
2252 2253 2254 2255 2256
	.start	= locks_start,
	.next	= locks_next,
	.stop	= locks_stop,
	.show	= locks_show,
};
2257 2258 2259

static int locks_open(struct inode *inode, struct file *filp)
{
2260
	return seq_open_private(filp, &locks_seq_operations, sizeof(loff_t));
2261 2262 2263 2264 2265 2266
}

static const struct file_operations proc_locks_operations = {
	.open		= locks_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
2267
	.release	= seq_release_private,
2268 2269 2270 2271 2272 2273 2274 2275
};

static int __init proc_locks_init(void)
{
	proc_create("locks", 0, NULL, &proc_locks_operations);
	return 0;
}
module_init(proc_locks_init);
2276 2277
#endif

L
Linus Torvalds 已提交
2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294
/**
 *	lock_may_read - checks that the region is free of locks
 *	@inode: the inode that is being read
 *	@start: the first byte to read
 *	@len: the number of bytes to read
 *
 *	Emulates Windows locking requirements.  Whole-file
 *	mandatory locks (share modes) can prohibit a read and
 *	byte-range POSIX locks can prohibit a read if they overlap.
 *
 *	N.B. this function is only ever called
 *	from knfsd and ownership of locks is never checked.
 */
int lock_may_read(struct inode *inode, loff_t start, unsigned long len)
{
	struct file_lock *fl;
	int result = 1;
2295
	lock_flocks();
L
Linus Torvalds 已提交
2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311
	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (IS_POSIX(fl)) {
			if (fl->fl_type == F_RDLCK)
				continue;
			if ((fl->fl_end < start) || (fl->fl_start > (start + len)))
				continue;
		} else if (IS_FLOCK(fl)) {
			if (!(fl->fl_type & LOCK_MAND))
				continue;
			if (fl->fl_type & LOCK_READ)
				continue;
		} else
			continue;
		result = 0;
		break;
	}
2312
	unlock_flocks();
L
Linus Torvalds 已提交
2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334
	return result;
}

EXPORT_SYMBOL(lock_may_read);

/**
 *	lock_may_write - checks that the region is free of locks
 *	@inode: the inode that is being written
 *	@start: the first byte to write
 *	@len: the number of bytes to write
 *
 *	Emulates Windows locking requirements.  Whole-file
 *	mandatory locks (share modes) can prohibit a write and
 *	byte-range POSIX locks can prohibit a write if they overlap.
 *
 *	N.B. this function is only ever called
 *	from knfsd and ownership of locks is never checked.
 */
int lock_may_write(struct inode *inode, loff_t start, unsigned long len)
{
	struct file_lock *fl;
	int result = 1;
2335
	lock_flocks();
L
Linus Torvalds 已提交
2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349
	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (IS_POSIX(fl)) {
			if ((fl->fl_end < start) || (fl->fl_start > (start + len)))
				continue;
		} else if (IS_FLOCK(fl)) {
			if (!(fl->fl_type & LOCK_MAND))
				continue;
			if (fl->fl_type & LOCK_WRITE)
				continue;
		} else
			continue;
		result = 0;
		break;
	}
2350
	unlock_flocks();
L
Linus Torvalds 已提交
2351 2352 2353 2354 2355 2356 2357 2358
	return result;
}

EXPORT_SYMBOL(lock_may_write);

static int __init filelock_init(void)
{
	filelock_cache = kmem_cache_create("file_lock_cache",
M
Miklos Szeredi 已提交
2359 2360
			sizeof(struct file_lock), 0, SLAB_PANIC, NULL);

L
Linus Torvalds 已提交
2361 2362 2363 2364
	return 0;
}

core_initcall(filelock_init);