Blame view
fs/locks.c
67.8 KB
1da177e4c
|
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. |
395cf9691
|
63 |
* See 'Documentation/filesystems/mandatory-locking.txt' for details. |
1da177e4c
|
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> |
9f3acc314
|
119 |
#include <linux/fdtable.h> |
1da177e4c
|
120 121 122 123 124 |
#include <linux/fs.h> #include <linux/init.h> #include <linux/module.h> #include <linux/security.h> #include <linux/slab.h> |
1da177e4c
|
125 126 |
#include <linux/syscalls.h> #include <linux/time.h> |
4fb3a5386
|
127 |
#include <linux/rcupdate.h> |
ab1f16116
|
128 |
#include <linux/pid_namespace.h> |
48f741865
|
129 |
#include <linux/hashtable.h> |
7012b02a2
|
130 131 |
#include <linux/percpu.h> #include <linux/lglock.h> |
1da177e4c
|
132 |
|
62af4f1f7
|
133 134 |
#define CREATE_TRACE_POINTS #include <trace/events/filelock.h> |
1da177e4c
|
135 136 137 138 |
#include <asm/uaccess.h> #define IS_POSIX(fl) (fl->fl_flags & FL_POSIX) #define IS_FLOCK(fl) (fl->fl_flags & FL_FLOCK) |
617588d51
|
139 |
#define IS_LEASE(fl) (fl->fl_flags & (FL_LEASE|FL_DELEG)) |
cff2fce58
|
140 |
#define IS_OFDLCK(fl) (fl->fl_flags & FL_OFDLCK) |
1da177e4c
|
141 |
|
ab83fa4b4
|
142 143 |
static bool lease_breaking(struct file_lock *fl) { |
778fc546f
|
144 145 146 147 148 149 150 151 152 153 |
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; |
ab83fa4b4
|
154 |
} |
1da177e4c
|
155 156 157 158 159 |
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) |
1c8c601a8
|
160 |
/* |
7012b02a2
|
161 162 163 164 |
* 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. |
1c8c601a8
|
165 |
*/ |
7012b02a2
|
166 167 |
DEFINE_STATIC_LGLOCK(file_lock_lglock); static DEFINE_PER_CPU(struct hlist_head, file_lock_list); |
889746917
|
168 |
|
1c8c601a8
|
169 |
/* |
48f741865
|
170 |
* The blocked_hash is used to find POSIX lock loops for deadlock detection. |
7b2296afb
|
171 |
* It is protected by blocked_lock_lock. |
48f741865
|
172 173 174 175 176 177 178 |
* * 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. |
1c8c601a8
|
179 |
*/ |
48f741865
|
180 181 |
#define BLOCKED_HASH_BITS 7 static DEFINE_HASHTABLE(blocked_hash, BLOCKED_HASH_BITS); |
889746917
|
182 |
|
1c8c601a8
|
183 |
/* |
7b2296afb
|
184 185 |
* This lock protects the blocked_hash. Generally, if you're accessing it, you * want to be holding this lock. |
1c8c601a8
|
186 187 188 189 190 191 192 193 194 195 196 197 198 199 |
* * 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. */ |
7b2296afb
|
200 |
static DEFINE_SPINLOCK(blocked_lock_lock); |
1da177e4c
|
201 |
|
e18b890bb
|
202 |
static struct kmem_cache *filelock_cache __read_mostly; |
1da177e4c
|
203 |
|
ee19cc406
|
204 |
static void locks_init_lock_heads(struct file_lock *fl) |
a51cb91d8
|
205 |
{ |
139ca04ee
|
206 |
INIT_HLIST_NODE(&fl->fl_link); |
ee19cc406
|
207 208 |
INIT_LIST_HEAD(&fl->fl_block); init_waitqueue_head(&fl->fl_wait); |
a51cb91d8
|
209 |
} |
1da177e4c
|
210 |
/* Allocate an empty lock structure. */ |
c5b1f0d92
|
211 |
struct file_lock *locks_alloc_lock(void) |
1da177e4c
|
212 |
{ |
ee19cc406
|
213 |
struct file_lock *fl = kmem_cache_zalloc(filelock_cache, GFP_KERNEL); |
a51cb91d8
|
214 215 |
if (fl) |
ee19cc406
|
216 |
locks_init_lock_heads(fl); |
a51cb91d8
|
217 218 |
return fl; |
1da177e4c
|
219 |
} |
c5b1f0d92
|
220 |
EXPORT_SYMBOL_GPL(locks_alloc_lock); |
1da177e4c
|
221 |
|
a9e61e25f
|
222 |
void locks_release_private(struct file_lock *fl) |
47831f35b
|
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; } |
47831f35b
|
229 |
|
5c97d7b14
|
230 231 232 233 234 |
if (fl->fl_lmops) { if (fl->fl_lmops->lm_put_owner) fl->fl_lmops->lm_put_owner(fl); fl->fl_lmops = NULL; } |
47831f35b
|
235 |
} |
a9e61e25f
|
236 |
EXPORT_SYMBOL_GPL(locks_release_private); |
47831f35b
|
237 |
|
1da177e4c
|
238 |
/* Free a lock which is not in use. */ |
05fa3135f
|
239 |
void locks_free_lock(struct file_lock *fl) |
1da177e4c
|
240 |
{ |
5ce29646e
|
241 242 |
BUG_ON(waitqueue_active(&fl->fl_wait)); BUG_ON(!list_empty(&fl->fl_block)); |
139ca04ee
|
243 |
BUG_ON(!hlist_unhashed(&fl->fl_link)); |
1da177e4c
|
244 |
|
47831f35b
|
245 |
locks_release_private(fl); |
1da177e4c
|
246 247 |
kmem_cache_free(filelock_cache, fl); } |
05fa3135f
|
248 |
EXPORT_SYMBOL(locks_free_lock); |
1da177e4c
|
249 |
|
ed9814d85
|
250 251 252 253 254 255 256 257 258 259 260 |
static void locks_dispose_list(struct list_head *dispose) { struct file_lock *fl; while (!list_empty(dispose)) { fl = list_first_entry(dispose, struct file_lock, fl_block); list_del_init(&fl->fl_block); locks_free_lock(fl); } } |
1da177e4c
|
261 262 |
void locks_init_lock(struct file_lock *fl) { |
ee19cc406
|
263 264 |
memset(fl, 0, sizeof(struct file_lock)); locks_init_lock_heads(fl); |
1da177e4c
|
265 266 267 |
} EXPORT_SYMBOL(locks_init_lock); |
1da177e4c
|
268 269 270 |
/* * Initialize a new lock from an existing file_lock structure. */ |
3fe0fff18
|
271 |
void locks_copy_conflock(struct file_lock *new, struct file_lock *fl) |
1da177e4c
|
272 273 274 |
{ new->fl_owner = fl->fl_owner; new->fl_pid = fl->fl_pid; |
0996905f9
|
275 |
new->fl_file = NULL; |
1da177e4c
|
276 277 278 279 |
new->fl_flags = fl->fl_flags; new->fl_type = fl->fl_type; new->fl_start = fl->fl_start; new->fl_end = fl->fl_end; |
f328296e2
|
280 |
new->fl_lmops = fl->fl_lmops; |
0996905f9
|
281 |
new->fl_ops = NULL; |
f328296e2
|
282 283 284 285 286 |
if (fl->fl_lmops) { if (fl->fl_lmops->lm_get_owner) fl->fl_lmops->lm_get_owner(new, fl); } |
0996905f9
|
287 |
} |
3fe0fff18
|
288 |
EXPORT_SYMBOL(locks_copy_conflock); |
0996905f9
|
289 290 291 |
void locks_copy_lock(struct file_lock *new, struct file_lock *fl) { |
566709bd6
|
292 293 |
/* "new" must be a freshly-initialized lock */ WARN_ON_ONCE(new->fl_ops); |
0996905f9
|
294 |
|
3fe0fff18
|
295 |
locks_copy_conflock(new, fl); |
f328296e2
|
296 |
|
0996905f9
|
297 |
new->fl_file = fl->fl_file; |
1da177e4c
|
298 |
new->fl_ops = fl->fl_ops; |
47831f35b
|
299 |
|
f328296e2
|
300 301 302 303 |
if (fl->fl_ops) { if (fl->fl_ops->fl_copy_lock) fl->fl_ops->fl_copy_lock(new, fl); } |
1da177e4c
|
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 |
} 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. */ |
6e129d006
|
323 324 |
static struct file_lock * flock_make_lock(struct file *filp, unsigned int cmd) |
1da177e4c
|
325 326 327 |
{ struct file_lock *fl; int type = flock_translate_cmd(cmd); |
6e129d006
|
328 |
|
1da177e4c
|
329 |
if (type < 0) |
6e129d006
|
330 |
return ERR_PTR(type); |
1da177e4c
|
331 332 333 |
fl = locks_alloc_lock(); if (fl == NULL) |
6e129d006
|
334 |
return ERR_PTR(-ENOMEM); |
1da177e4c
|
335 336 |
fl->fl_file = filp; |
73a8f5f7e
|
337 |
fl->fl_owner = filp; |
1da177e4c
|
338 339 340 341 342 |
fl->fl_pid = current->tgid; fl->fl_flags = FL_FLOCK; fl->fl_type = type; fl->fl_end = OFFSET_MAX; |
6e129d006
|
343 |
return fl; |
1da177e4c
|
344 |
} |
0ec4f431e
|
345 |
static int assign_type(struct file_lock *fl, long type) |
1da177e4c
|
346 347 348 349 350 351 352 353 354 355 356 357 |
{ switch (type) { case F_RDLCK: case F_WRLCK: case F_UNLCK: fl->fl_type = type; break; default: return -EINVAL; } return 0; } |
ef12e72a0
|
358 359 |
static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl, struct flock64 *l) |
1da177e4c
|
360 |
{ |
1da177e4c
|
361 |
switch (l->l_whence) { |
f5579f8c7
|
362 |
case SEEK_SET: |
ef12e72a0
|
363 |
fl->fl_start = 0; |
1da177e4c
|
364 |
break; |
f5579f8c7
|
365 |
case SEEK_CUR: |
ef12e72a0
|
366 |
fl->fl_start = filp->f_pos; |
1da177e4c
|
367 |
break; |
f5579f8c7
|
368 |
case SEEK_END: |
ef12e72a0
|
369 |
fl->fl_start = i_size_read(file_inode(filp)); |
1da177e4c
|
370 371 372 373 |
break; default: return -EINVAL; } |
ef12e72a0
|
374 375 376 377 378 |
if (l->l_start > OFFSET_MAX - fl->fl_start) return -EOVERFLOW; fl->fl_start += l->l_start; if (fl->fl_start < 0) return -EINVAL; |
1da177e4c
|
379 380 381 |
/* POSIX-1996 leaves the case l->l_len < 0 undefined; POSIX-2001 defines it. */ |
4c780a468
|
382 |
if (l->l_len > 0) { |
ef12e72a0
|
383 384 385 |
if (l->l_len - 1 > OFFSET_MAX - fl->fl_start) return -EOVERFLOW; fl->fl_end = fl->fl_start + l->l_len - 1; |
4c780a468
|
386 |
} else if (l->l_len < 0) { |
ef12e72a0
|
387 |
if (fl->fl_start + l->l_len < 0) |
4c780a468
|
388 |
return -EINVAL; |
ef12e72a0
|
389 390 391 392 |
fl->fl_end = fl->fl_start - 1; fl->fl_start += l->l_len; } else fl->fl_end = OFFSET_MAX; |
1da177e4c
|
393 394 395 396 397 398 399 400 401 |
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); } |
ef12e72a0
|
402 403 404 405 406 |
/* 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) |
1da177e4c
|
407 |
{ |
ef12e72a0
|
408 409 410 411 412 413 414 415 |
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); |
1da177e4c
|
416 |
} |
1da177e4c
|
417 418 |
/* default lease lock manager operations */ |
4d01b7f5e
|
419 420 |
static bool lease_break_callback(struct file_lock *fl) |
1da177e4c
|
421 422 |
{ kill_fasync(&fl->fl_fasync, SIGIO, POLL_MSG); |
4d01b7f5e
|
423 |
return false; |
1da177e4c
|
424 |
} |
1c7dd2ff4
|
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 |
static void lease_setup(struct file_lock *fl, void **priv) { struct file *filp = fl->fl_file; struct fasync_struct *fa = *priv; /* * fasync_insert_entry() returns the old entry if any. If there was no * old entry, then it used "priv" and inserted it into the fasync list. * Clear the pointer to indicate that it shouldn't be freed. */ if (!fasync_insert_entry(fa->fa_fd, filp, &fl->fl_fasync, fa)) *priv = NULL; __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); } |
7b021967c
|
441 |
static const struct lock_manager_operations lease_manager_ops = { |
8fb47a4fb
|
442 |
.lm_break = lease_break_callback, |
8fb47a4fb
|
443 |
.lm_change = lease_modify, |
1c7dd2ff4
|
444 |
.lm_setup = lease_setup, |
1da177e4c
|
445 446 447 448 449 |
}; /* * Initialize a lease, use the default lock manager operations */ |
0ec4f431e
|
450 |
static int lease_init(struct file *filp, long type, struct file_lock *fl) |
1da177e4c
|
451 |
{ |
75dff55af
|
452 453 |
if (assign_type(fl, type) != 0) return -EINVAL; |
7ca76311f
|
454 |
fl->fl_owner = filp; |
1da177e4c
|
455 456 457 458 |
fl->fl_pid = current->tgid; fl->fl_file = filp; fl->fl_flags = FL_LEASE; |
1da177e4c
|
459 460 461 462 463 464 465 466 |
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 */ |
0ec4f431e
|
467 |
static struct file_lock *lease_alloc(struct file *filp, long type) |
1da177e4c
|
468 469 |
{ struct file_lock *fl = locks_alloc_lock(); |
75dff55af
|
470 |
int error = -ENOMEM; |
1da177e4c
|
471 472 |
if (fl == NULL) |
e32b8ee27
|
473 |
return ERR_PTR(error); |
1da177e4c
|
474 475 |
error = lease_init(filp, type, fl); |
75dff55af
|
476 477 |
if (error) { locks_free_lock(fl); |
e32b8ee27
|
478 |
return ERR_PTR(error); |
75dff55af
|
479 |
} |
e32b8ee27
|
480 |
return fl; |
1da177e4c
|
481 482 483 484 485 486 487 488 489 490 491 492 493 |
} /* 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. */ |
33443c42f
|
494 |
static int posix_same_owner(struct file_lock *fl1, struct file_lock *fl2) |
1da177e4c
|
495 |
{ |
8fb47a4fb
|
496 |
if (fl1->fl_lmops && fl1->fl_lmops->lm_compare_owner) |
1da177e4c
|
497 |
return fl2->fl_lmops == fl1->fl_lmops && |
8fb47a4fb
|
498 |
fl1->fl_lmops->lm_compare_owner(fl1, fl2); |
1da177e4c
|
499 500 |
return fl1->fl_owner == fl2->fl_owner; } |
7012b02a2
|
501 |
/* Must be called with the i_lock held! */ |
6ca10ed8e
|
502 |
static void locks_insert_global_locks(struct file_lock *fl) |
889746917
|
503 |
{ |
7012b02a2
|
504 505 506 507 |
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); |
889746917
|
508 |
} |
7012b02a2
|
509 |
/* Must be called with the i_lock held! */ |
6ca10ed8e
|
510 |
static void locks_delete_global_locks(struct file_lock *fl) |
889746917
|
511 |
{ |
7012b02a2
|
512 513 514 515 516 517 518 519 |
/* * 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); |
139ca04ee
|
520 |
hlist_del_init(&fl->fl_link); |
7012b02a2
|
521 |
lg_local_unlock_cpu(&file_lock_lglock, fl->fl_link_cpu); |
889746917
|
522 |
} |
3999e4936
|
523 524 525 526 527 528 529 |
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; } |
6ca10ed8e
|
530 |
static void locks_insert_global_blocked(struct file_lock *waiter) |
889746917
|
531 |
{ |
3999e4936
|
532 |
hash_add(blocked_hash, &waiter->fl_link, posix_owner_key(waiter)); |
889746917
|
533 |
} |
6ca10ed8e
|
534 |
static void locks_delete_global_blocked(struct file_lock *waiter) |
889746917
|
535 |
{ |
48f741865
|
536 |
hash_del(&waiter->fl_link); |
889746917
|
537 |
} |
1da177e4c
|
538 539 |
/* Remove waiter from blocker's block list. * When blocker ends up pointing to itself then the list is empty. |
1c8c601a8
|
540 |
* |
7b2296afb
|
541 |
* Must be called with blocked_lock_lock held. |
1da177e4c
|
542 |
*/ |
33443c42f
|
543 |
static void __locks_delete_block(struct file_lock *waiter) |
1da177e4c
|
544 |
{ |
889746917
|
545 |
locks_delete_global_blocked(waiter); |
1da177e4c
|
546 |
list_del_init(&waiter->fl_block); |
1da177e4c
|
547 548 |
waiter->fl_next = NULL; } |
1a9e64a71
|
549 |
static void locks_delete_block(struct file_lock *waiter) |
1da177e4c
|
550 |
{ |
7b2296afb
|
551 |
spin_lock(&blocked_lock_lock); |
1da177e4c
|
552 |
__locks_delete_block(waiter); |
7b2296afb
|
553 |
spin_unlock(&blocked_lock_lock); |
1da177e4c
|
554 555 556 557 558 559 |
} /* 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. |
1c8c601a8
|
560 |
* |
7b2296afb
|
561 |
* Must be called with both the i_lock and blocked_lock_lock held. The fl_block |
46dad7603
|
562 |
* list itself is protected by the blocked_lock_lock, but by ensuring that the |
7b2296afb
|
563 |
* i_lock is also held on insertions we can avoid taking the blocked_lock_lock |
4e8c765d3
|
564 |
* in some cases when we see that the fl_block list is empty. |
1da177e4c
|
565 |
*/ |
1c8c601a8
|
566 567 |
static void __locks_insert_block(struct file_lock *blocker, struct file_lock *waiter) |
1da177e4c
|
568 |
{ |
6dc0fe8f8
|
569 |
BUG_ON(!list_empty(&waiter->fl_block)); |
1da177e4c
|
570 |
waiter->fl_next = blocker; |
889746917
|
571 |
list_add_tail(&waiter->fl_block, &blocker->fl_block); |
cff2fce58
|
572 |
if (IS_POSIX(blocker) && !IS_OFDLCK(blocker)) |
1c8c601a8
|
573 574 575 576 577 578 579 |
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) { |
7b2296afb
|
580 |
spin_lock(&blocked_lock_lock); |
1c8c601a8
|
581 |
__locks_insert_block(blocker, waiter); |
7b2296afb
|
582 |
spin_unlock(&blocked_lock_lock); |
1da177e4c
|
583 |
} |
1cb360125
|
584 585 586 |
/* * Wake up processes blocked waiting for blocker. * |
1c8c601a8
|
587 |
* Must be called with the inode->i_lock held! |
1da177e4c
|
588 589 590 |
*/ static void locks_wake_up_blocks(struct file_lock *blocker) { |
4e8c765d3
|
591 592 593 594 595 |
/* * 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() |
7b2296afb
|
596 |
* after acquiring the blocked_lock_lock. |
4e8c765d3
|
597 598 599 |
*/ if (list_empty(&blocker->fl_block)) return; |
7b2296afb
|
600 |
spin_lock(&blocked_lock_lock); |
1da177e4c
|
601 |
while (!list_empty(&blocker->fl_block)) { |
f0c1cd0ea
|
602 603 604 |
struct file_lock *waiter; waiter = list_first_entry(&blocker->fl_block, |
1da177e4c
|
605 606 |
struct file_lock, fl_block); __locks_delete_block(waiter); |
8fb47a4fb
|
607 608 |
if (waiter->fl_lmops && waiter->fl_lmops->lm_notify) waiter->fl_lmops->lm_notify(waiter); |
1da177e4c
|
609 610 611 |
else wake_up(&waiter->fl_wait); } |
7b2296afb
|
612 |
spin_unlock(&blocked_lock_lock); |
1da177e4c
|
613 614 615 616 |
} /* 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. |
1c8c601a8
|
617 618 |
* * Must be called with the i_lock held! |
1da177e4c
|
619 620 621 |
*/ static void locks_insert_lock(struct file_lock **pos, struct file_lock *fl) { |
ab1f16116
|
622 |
fl->fl_nspid = get_pid(task_tgid(current)); |
1da177e4c
|
623 624 625 |
/* insert into file's list */ fl->fl_next = *pos; *pos = fl; |
889746917
|
626 627 |
locks_insert_global_locks(fl); |
1da177e4c
|
628 |
} |
24cbe7845
|
629 630 631 632 633 634 635 636 |
/** * 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. |
1c8c601a8
|
637 638 |
* * Must be called with the i_lock held! |
1da177e4c
|
639 |
*/ |
24cbe7845
|
640 |
static void locks_unlink_lock(struct file_lock **thisfl_p) |
1da177e4c
|
641 642 |
{ struct file_lock *fl = *thisfl_p; |
889746917
|
643 |
locks_delete_global_locks(fl); |
1da177e4c
|
644 645 |
*thisfl_p = fl->fl_next; fl->fl_next = NULL; |
1da177e4c
|
646 |
|
ab1f16116
|
647 648 649 650 |
if (fl->fl_nspid) { put_pid(fl->fl_nspid); fl->fl_nspid = NULL; } |
1da177e4c
|
651 |
locks_wake_up_blocks(fl); |
24cbe7845
|
652 653 654 655 656 657 658 |
} /* * Unlink a lock from all lists and free it. * * Must be called with i_lock held! */ |
ed9814d85
|
659 660 |
static void locks_delete_lock(struct file_lock **thisfl_p, struct list_head *dispose) |
24cbe7845
|
661 662 663 664 |
{ struct file_lock *fl = *thisfl_p; locks_unlink_lock(thisfl_p); |
ed9814d85
|
665 666 667 668 |
if (dispose) list_add(&fl->fl_block, dispose); else locks_free_lock(fl); |
1da177e4c
|
669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 |
} /* 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)); } |
6d34ac199
|
716 |
void |
9d6a8c5c2
|
717 |
posix_test_lock(struct file *filp, struct file_lock *fl) |
1da177e4c
|
718 719 |
{ struct file_lock *cfl; |
1c8c601a8
|
720 |
struct inode *inode = file_inode(filp); |
1da177e4c
|
721 |
|
1c8c601a8
|
722 |
spin_lock(&inode->i_lock); |
496ad9aa8
|
723 |
for (cfl = file_inode(filp)->i_flock; cfl; cfl = cfl->fl_next) { |
1da177e4c
|
724 725 |
if (!IS_POSIX(cfl)) continue; |
b842e240f
|
726 |
if (posix_locks_conflict(fl, cfl)) |
1da177e4c
|
727 728 |
break; } |
ab1f16116
|
729 |
if (cfl) { |
3fe0fff18
|
730 |
locks_copy_conflock(fl, cfl); |
ab1f16116
|
731 |
if (cfl->fl_nspid) |
6c5f3e7b4
|
732 |
fl->fl_pid = pid_vnr(cfl->fl_nspid); |
ab1f16116
|
733 |
} else |
129a84de2
|
734 |
fl->fl_type = F_UNLCK; |
1c8c601a8
|
735 |
spin_unlock(&inode->i_lock); |
6d34ac199
|
736 |
return; |
1da177e4c
|
737 |
} |
1da177e4c
|
738 |
EXPORT_SYMBOL(posix_test_lock); |
b533184fc
|
739 740 741 742 743 |
/* * Deadlock detection: * * We attempt to detect deadlocks that are due purely to posix file * locks. |
1da177e4c
|
744 |
* |
b533184fc
|
745 746 747 748 749 750 751 |
* 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. |
1da177e4c
|
752 |
* |
b533184fc
|
753 754 755 |
* 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. |
97855b49b
|
756 |
* |
b533184fc
|
757 758 759 |
* 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. |
b533184fc
|
760 |
* To handle those cases, we just bail out after a few iterations. |
57b65325f
|
761 |
* |
cff2fce58
|
762 |
* For FL_OFDLCK locks, the owner is the filp, not the files_struct. |
57b65325f
|
763 764 765 766 |
* Because the owner is not even nominally tied to a thread of * execution, the deadlock detection below can't reasonably work well. Just * skip it for those. * |
cff2fce58
|
767 |
* In principle, we could do a more limited deadlock detection on FL_OFDLCK |
57b65325f
|
768 769 |
* locks that just checks for the case where two tasks are attempting to * upgrade from read to write locks on the same inode. |
1da177e4c
|
770 |
*/ |
97855b49b
|
771 772 |
#define MAX_DEADLK_ITERATIONS 10 |
b533184fc
|
773 774 775 776 |
/* 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; |
3999e4936
|
777 |
hash_for_each_possible(blocked_hash, fl, fl_link, posix_owner_key(block_fl)) { |
b533184fc
|
778 779 780 781 782 |
if (posix_same_owner(fl, block_fl)) return fl->fl_next; } return NULL; } |
7b2296afb
|
783 |
/* Must be called with the blocked_lock_lock held! */ |
b0904e147
|
784 |
static int posix_locks_deadlock(struct file_lock *caller_fl, |
1da177e4c
|
785 786 |
struct file_lock *block_fl) { |
97855b49b
|
787 |
int i = 0; |
1da177e4c
|
788 |
|
57b65325f
|
789 790 |
/* * This deadlock detector can't reasonably detect deadlocks with |
cff2fce58
|
791 |
* FL_OFDLCK locks, since they aren't owned by a process, per-se. |
57b65325f
|
792 |
*/ |
cff2fce58
|
793 |
if (IS_OFDLCK(caller_fl)) |
57b65325f
|
794 |
return 0; |
b533184fc
|
795 796 797 798 799 |
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; |
1da177e4c
|
800 801 802 |
} return 0; } |
1da177e4c
|
803 |
/* Try to create a FLOCK lock on filp. We always insert new FLOCK locks |
02888f41e
|
804 |
* after any leases, but before any posix locks. |
f475ae957
|
805 806 807 808 |
* * 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. |
1da177e4c
|
809 |
*/ |
993dfa877
|
810 |
static int flock_lock_file(struct file *filp, struct file_lock *request) |
1da177e4c
|
811 |
{ |
993dfa877
|
812 |
struct file_lock *new_fl = NULL; |
1da177e4c
|
813 |
struct file_lock **before; |
496ad9aa8
|
814 |
struct inode * inode = file_inode(filp); |
1da177e4c
|
815 816 |
int error = 0; int found = 0; |
ed9814d85
|
817 |
LIST_HEAD(dispose); |
1da177e4c
|
818 |
|
b89f43213
|
819 |
if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) { |
84d535ade
|
820 |
new_fl = locks_alloc_lock(); |
b89f43213
|
821 822 |
if (!new_fl) return -ENOMEM; |
84d535ade
|
823 |
} |
1c8c601a8
|
824 |
spin_lock(&inode->i_lock); |
b89f43213
|
825 826 |
if (request->fl_flags & FL_ACCESS) goto find_conflict; |
1da177e4c
|
827 828 829 830 831 832 833 834 |
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; |
993dfa877
|
835 |
if (request->fl_type == fl->fl_type) |
1da177e4c
|
836 837 |
goto out; found = 1; |
ed9814d85
|
838 |
locks_delete_lock(before, &dispose); |
1da177e4c
|
839 840 |
break; } |
1da177e4c
|
841 |
|
f475ae957
|
842 843 844 |
if (request->fl_type == F_UNLCK) { if ((request->fl_flags & FL_EXISTS) && !found) error = -ENOENT; |
993dfa877
|
845 |
goto out; |
f475ae957
|
846 |
} |
1da177e4c
|
847 848 849 850 851 |
/* * If a higher-priority process was blocked on the old file lock, * give it the opportunity to lock the file. */ |
b89f43213
|
852 |
if (found) { |
1c8c601a8
|
853 |
spin_unlock(&inode->i_lock); |
def01bc53
|
854 |
cond_resched(); |
1c8c601a8
|
855 |
spin_lock(&inode->i_lock); |
b89f43213
|
856 |
} |
1da177e4c
|
857 |
|
f07f18dd6
|
858 |
find_conflict: |
1da177e4c
|
859 860 861 862 863 864 |
for_each_lock(inode, before) { struct file_lock *fl = *before; if (IS_POSIX(fl)) break; if (IS_LEASE(fl)) continue; |
993dfa877
|
865 |
if (!flock_locks_conflict(request, fl)) |
1da177e4c
|
866 867 |
continue; error = -EAGAIN; |
bde74e4bc
|
868 869 870 871 |
if (!(request->fl_flags & FL_SLEEP)) goto out; error = FILE_LOCK_DEFERRED; locks_insert_block(fl, request); |
1da177e4c
|
872 873 |
goto out; } |
f07f18dd6
|
874 875 |
if (request->fl_flags & FL_ACCESS) goto out; |
993dfa877
|
876 |
locks_copy_lock(new_fl, request); |
0e2f6db88
|
877 |
locks_insert_lock(before, new_fl); |
993dfa877
|
878 |
new_fl = NULL; |
9cedc194a
|
879 |
error = 0; |
1da177e4c
|
880 881 |
out: |
1c8c601a8
|
882 |
spin_unlock(&inode->i_lock); |
993dfa877
|
883 884 |
if (new_fl) locks_free_lock(new_fl); |
ed9814d85
|
885 |
locks_dispose_list(&dispose); |
1da177e4c
|
886 887 |
return error; } |
150b39345
|
888 |
static int __posix_lock_file(struct inode *inode, struct file_lock *request, struct file_lock *conflock) |
1da177e4c
|
889 890 |
{ struct file_lock *fl; |
39005d022
|
891 892 |
struct file_lock *new_fl = NULL; struct file_lock *new_fl2 = NULL; |
1da177e4c
|
893 894 895 |
struct file_lock *left = NULL; struct file_lock *right = NULL; struct file_lock **before; |
b9746ef80
|
896 897 |
int error; bool added = false; |
ed9814d85
|
898 |
LIST_HEAD(dispose); |
1da177e4c
|
899 900 901 902 |
/* * We may need two file_lock structures for this operation, * so we get them in advance to avoid races. |
39005d022
|
903 904 |
* * In some cases we can be sure, that no new locks will be needed |
1da177e4c
|
905 |
*/ |
39005d022
|
906 907 908 909 910 911 |
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(); } |
1da177e4c
|
912 |
|
1c8c601a8
|
913 |
spin_lock(&inode->i_lock); |
1cb360125
|
914 915 916 |
/* * New lock request. Walk all POSIX locks and look for conflicts. If * there are any, either return error or put the request on the |
48f741865
|
917 |
* blocker's list of waiters and the global blocked_hash. |
1cb360125
|
918 |
*/ |
1da177e4c
|
919 920 |
if (request->fl_type != F_UNLCK) { for_each_lock(inode, before) { |
526985b9d
|
921 |
fl = *before; |
1da177e4c
|
922 923 924 925 |
if (!IS_POSIX(fl)) continue; if (!posix_locks_conflict(request, fl)) continue; |
5842add2f
|
926 |
if (conflock) |
3fe0fff18
|
927 |
locks_copy_conflock(conflock, fl); |
1da177e4c
|
928 929 930 |
error = -EAGAIN; if (!(request->fl_flags & FL_SLEEP)) goto out; |
1c8c601a8
|
931 932 933 934 |
/* * Deadlock detection and insertion into the blocked * locks list must be done while holding the same lock! */ |
1da177e4c
|
935 |
error = -EDEADLK; |
7b2296afb
|
936 |
spin_lock(&blocked_lock_lock); |
1c8c601a8
|
937 938 939 940 |
if (likely(!posix_locks_deadlock(request, fl))) { error = FILE_LOCK_DEFERRED; __locks_insert_block(fl, request); } |
7b2296afb
|
941 |
spin_unlock(&blocked_lock_lock); |
1da177e4c
|
942 943 944 945 946 947 948 949 |
goto out; } } /* If we're just looking for a conflict, we're done. */ error = 0; if (request->fl_flags & FL_ACCESS) goto out; |
1da177e4c
|
950 |
/* |
1da177e4c
|
951 952 953 954 955 956 957 958 959 960 |
* 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; } |
1cb360125
|
961 |
/* Process locks with this owner. */ |
1da177e4c
|
962 963 964 965 |
while ((fl = *before) && posix_same_owner(request, fl)) { /* Detect adjacent or overlapping regions (if same lock type) */ if (request->fl_type == fl->fl_type) { |
449231d6d
|
966 967 968 969 |
/* In all comparisons of start vs end, use * "start - 1" rather than "end + 1". If end * is OFFSET_MAX, end + 1 will become negative. */ |
1da177e4c
|
970 971 972 973 974 |
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. */ |
449231d6d
|
975 |
if (fl->fl_start - 1 > request->fl_end) |
1da177e4c
|
976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 |
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) { |
ed9814d85
|
992 |
locks_delete_lock(before, &dispose); |
1da177e4c
|
993 994 995 |
continue; } request = fl; |
b9746ef80
|
996 |
added = true; |
1da177e4c
|
997 998 999 1000 1001 1002 1003 1004 1005 1006 |
} 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) |
b9746ef80
|
1007 |
added = true; |
1da177e4c
|
1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 |
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) { |
ed9814d85
|
1022 |
locks_delete_lock(before, &dispose); |
1da177e4c
|
1023 1024 |
continue; } |
b84d49f94
|
1025 1026 1027 1028 1029 1030 |
/* * Replace the old lock with new_fl, and * remove the old one. It's safe to do the * insert here since we know that we won't be * using new_fl later, and that the lock is * just replacing an existing lock. |
1da177e4c
|
1031 |
*/ |
b84d49f94
|
1032 1033 1034 1035 1036 1037 |
error = -ENOLCK; if (!new_fl) goto out; locks_copy_lock(new_fl, request); request = new_fl; new_fl = NULL; |
ed9814d85
|
1038 |
locks_delete_lock(before, &dispose); |
b84d49f94
|
1039 |
locks_insert_lock(before, request); |
b9746ef80
|
1040 |
added = true; |
1da177e4c
|
1041 1042 1043 1044 1045 1046 1047 |
} } /* Go on to next lock. */ next_lock: before = &fl->fl_next; } |
0d9a490ab
|
1048 |
/* |
1cb360125
|
1049 1050 1051 |
* 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. |
0d9a490ab
|
1052 1053 1054 1055 |
*/ error = -ENOLCK; /* "no luck" */ if (right && left == right && !new_fl2) goto out; |
1da177e4c
|
1056 1057 |
error = 0; if (!added) { |
f475ae957
|
1058 1059 1060 |
if (request->fl_type == F_UNLCK) { if (request->fl_flags & FL_EXISTS) error = -ENOENT; |
1da177e4c
|
1061 |
goto out; |
f475ae957
|
1062 |
} |
0d9a490ab
|
1063 1064 1065 1066 1067 |
if (!new_fl) { error = -ENOLCK; goto out; } |
1da177e4c
|
1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 |
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: |
1c8c601a8
|
1090 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1091 1092 1093 1094 1095 1096 1097 |
/* * Free any unused locks. */ if (new_fl) locks_free_lock(new_fl); if (new_fl2) locks_free_lock(new_fl2); |
ed9814d85
|
1098 |
locks_dispose_list(&dispose); |
1da177e4c
|
1099 1100 1101 1102 1103 1104 1105 |
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 |
150b39345
|
1106 |
* @conflock: Place to return a copy of the conflicting lock, if found. |
1da177e4c
|
1107 1108 1109 1110 |
* * 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 |
f475ae957
|
1111 1112 1113 1114 |
* * 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. |
1da177e4c
|
1115 |
*/ |
150b39345
|
1116 |
int posix_lock_file(struct file *filp, struct file_lock *fl, |
5842add2f
|
1117 1118 |
struct file_lock *conflock) { |
496ad9aa8
|
1119 |
return __posix_lock_file(file_inode(filp), fl, conflock); |
1da177e4c
|
1120 |
} |
150b39345
|
1121 |
EXPORT_SYMBOL(posix_lock_file); |
1da177e4c
|
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 |
/** * 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 (;;) { |
150b39345
|
1137 |
error = posix_lock_file(filp, fl, NULL); |
bde74e4bc
|
1138 |
if (error != FILE_LOCK_DEFERRED) |
1da177e4c
|
1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 |
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 |
d7a06983a
|
1153 |
* @file: the file to check |
1da177e4c
|
1154 1155 1156 1157 |
* * Searches the inode's list of locks to find any POSIX locks which conflict. * This function is called from locks_verify_locked() only. */ |
d7a06983a
|
1158 |
int locks_mandatory_locked(struct file *file) |
1da177e4c
|
1159 |
{ |
d7a06983a
|
1160 |
struct inode *inode = file_inode(file); |
1da177e4c
|
1161 1162 1163 1164 1165 |
struct file_lock *fl; /* * Search the lock list for this inode for any POSIX locks. */ |
1c8c601a8
|
1166 |
spin_lock(&inode->i_lock); |
1da177e4c
|
1167 1168 1169 |
for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (!IS_POSIX(fl)) continue; |
73a8f5f7e
|
1170 1171 |
if (fl->fl_owner != current->files && fl->fl_owner != file) |
1da177e4c
|
1172 1173 |
break; } |
1c8c601a8
|
1174 |
spin_unlock(&inode->i_lock); |
1da177e4c
|
1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 |
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; |
29723adee
|
1197 |
bool sleep = false; |
1da177e4c
|
1198 1199 |
locks_init_lock(&fl); |
1da177e4c
|
1200 1201 1202 1203 |
fl.fl_pid = current->tgid; fl.fl_file = filp; fl.fl_flags = FL_POSIX | FL_ACCESS; if (filp && !(filp->f_flags & O_NONBLOCK)) |
29723adee
|
1204 |
sleep = true; |
1da177e4c
|
1205 1206 1207 1208 1209 |
fl.fl_type = (read_write == FLOCK_VERIFY_WRITE) ? F_WRLCK : F_RDLCK; fl.fl_start = offset; fl.fl_end = offset + count - 1; for (;;) { |
29723adee
|
1210 |
if (filp) { |
73a8f5f7e
|
1211 |
fl.fl_owner = filp; |
29723adee
|
1212 1213 1214 1215 1216 1217 1218 1219 1220 |
fl.fl_flags &= ~FL_SLEEP; error = __posix_lock_file(inode, &fl, NULL); if (!error) break; } if (sleep) fl.fl_flags |= FL_SLEEP; fl.fl_owner = current->files; |
150b39345
|
1221 |
error = __posix_lock_file(inode, &fl, NULL); |
bde74e4bc
|
1222 |
if (error != FILE_LOCK_DEFERRED) |
1da177e4c
|
1223 1224 1225 1226 1227 1228 1229 |
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. */ |
a16877ca9
|
1230 |
if (__mandatory_lock(inode)) |
1da177e4c
|
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 |
continue; } locks_delete_block(&fl); break; } return error; } EXPORT_SYMBOL(locks_mandatory_area); |
778fc546f
|
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 |
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; } } |
1da177e4c
|
1252 |
/* We already had a lease on this file; just change its type */ |
c45198eda
|
1253 |
int lease_modify(struct file_lock **before, int arg, struct list_head *dispose) |
1da177e4c
|
1254 1255 1256 1257 1258 1259 |
{ struct file_lock *fl = *before; int error = assign_type(fl, arg); if (error) return error; |
778fc546f
|
1260 |
lease_clear_pending(fl, arg); |
1da177e4c
|
1261 |
locks_wake_up_blocks(fl); |
3b6e2723f
|
1262 1263 1264 1265 1266 |
if (arg == F_UNLCK) { struct file *filp = fl->fl_file; f_delown(filp); filp->f_owner.signum = 0; |
96d6d59ce
|
1267 1268 1269 1270 1271 1272 |
fasync_helper(0, fl->fl_file, 0, &fl->fl_fasync); if (fl->fl_fasync != NULL) { printk(KERN_ERR "locks_delete_lock: fasync == %p ", fl->fl_fasync); fl->fl_fasync = NULL; } |
c45198eda
|
1273 |
locks_delete_lock(before, dispose); |
3b6e2723f
|
1274 |
} |
1da177e4c
|
1275 1276 |
return 0; } |
1da177e4c
|
1277 |
EXPORT_SYMBOL(lease_modify); |
778fc546f
|
1278 1279 1280 1281 1282 1283 1284 |
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); } |
c45198eda
|
1285 |
static void time_out_leases(struct inode *inode, struct list_head *dispose) |
1da177e4c
|
1286 1287 1288 |
{ struct file_lock **before; struct file_lock *fl; |
f82b4b678
|
1289 |
lockdep_assert_held(&inode->i_lock); |
1da177e4c
|
1290 |
before = &inode->i_flock; |
ab83fa4b4
|
1291 |
while ((fl = *before) && IS_LEASE(fl) && lease_breaking(fl)) { |
62af4f1f7
|
1292 |
trace_time_out_leases(inode, fl); |
778fc546f
|
1293 |
if (past_time(fl->fl_downgrade_time)) |
c45198eda
|
1294 |
lease_modify(before, F_RDLCK, dispose); |
778fc546f
|
1295 |
if (past_time(fl->fl_break_time)) |
c45198eda
|
1296 |
lease_modify(before, F_UNLCK, dispose); |
1da177e4c
|
1297 1298 1299 1300 |
if (fl == *before) /* lease_modify may have freed fl */ before = &fl->fl_next; } } |
df4e8d2c1
|
1301 1302 1303 1304 1305 1306 |
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); } |
03d12ddf8
|
1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 |
static bool any_leases_conflict(struct inode *inode, struct file_lock *breaker) { struct file_lock *fl; lockdep_assert_held(&inode->i_lock); for (fl = inode->i_flock ; fl && IS_LEASE(fl); fl = fl->fl_next) { if (leases_conflict(fl, breaker)) return true; } return false; } |
1da177e4c
|
1320 1321 1322 |
/** * __break_lease - revoke all outstanding leases on file * @inode: the inode of the file to return |
df4e8d2c1
|
1323 1324 1325 1326 |
* @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 |
1da177e4c
|
1327 |
* |
87250dd26
|
1328 1329 1330 |
* 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 |
1da177e4c
|
1331 1332 |
* specified %O_NONBLOCK to your open(). */ |
df4e8d2c1
|
1333 |
int __break_lease(struct inode *inode, unsigned int mode, unsigned int type) |
1da177e4c
|
1334 |
{ |
778fc546f
|
1335 |
int error = 0; |
03d12ddf8
|
1336 |
struct file_lock *new_fl; |
4d01b7f5e
|
1337 |
struct file_lock *fl, **before; |
1da177e4c
|
1338 |
unsigned long break_time; |
8737c9305
|
1339 |
int want_write = (mode & O_ACCMODE) != O_RDONLY; |
c45198eda
|
1340 |
LIST_HEAD(dispose); |
1da177e4c
|
1341 |
|
8737c9305
|
1342 |
new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK); |
6d4b9e38d
|
1343 1344 |
if (IS_ERR(new_fl)) return PTR_ERR(new_fl); |
df4e8d2c1
|
1345 |
new_fl->fl_flags = type; |
1da177e4c
|
1346 |
|
1c8c601a8
|
1347 |
spin_lock(&inode->i_lock); |
1da177e4c
|
1348 |
|
c45198eda
|
1349 |
time_out_leases(inode, &dispose); |
1da177e4c
|
1350 |
|
03d12ddf8
|
1351 |
if (!any_leases_conflict(inode, new_fl)) |
778fc546f
|
1352 |
goto out; |
1da177e4c
|
1353 1354 1355 1356 1357 1358 |
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 */ } |
4d01b7f5e
|
1359 1360 1361 |
for (before = &inode->i_flock; ((fl = *before) != NULL) && IS_LEASE(fl); before = &fl->fl_next) { |
df4e8d2c1
|
1362 1363 |
if (!leases_conflict(fl, new_fl)) continue; |
778fc546f
|
1364 1365 1366 1367 |
if (want_write) { if (fl->fl_flags & FL_UNLOCK_PENDING) continue; fl->fl_flags |= FL_UNLOCK_PENDING; |
1da177e4c
|
1368 |
fl->fl_break_time = break_time; |
778fc546f
|
1369 |
} else { |
03d12ddf8
|
1370 |
if (lease_breaking(inode->i_flock)) |
778fc546f
|
1371 1372 1373 |
continue; fl->fl_flags |= FL_DOWNGRADE_PENDING; fl->fl_downgrade_time = break_time; |
1da177e4c
|
1374 |
} |
4d01b7f5e
|
1375 1376 |
if (fl->fl_lmops->lm_break(fl)) locks_delete_lock(before, &dispose); |
1da177e4c
|
1377 |
} |
4d01b7f5e
|
1378 1379 1380 |
fl = inode->i_flock; if (!fl || !IS_LEASE(fl)) goto out; |
843c6b2f4
|
1381 |
if (mode & O_NONBLOCK) { |
62af4f1f7
|
1382 |
trace_break_lease_noblock(inode, new_fl); |
1da177e4c
|
1383 1384 1385 1386 1387 |
error = -EWOULDBLOCK; goto out; } restart: |
03d12ddf8
|
1388 |
break_time = inode->i_flock->fl_break_time; |
f1c6bb2cb
|
1389 |
if (break_time != 0) |
1da177e4c
|
1390 |
break_time -= jiffies; |
f1c6bb2cb
|
1391 1392 |
if (break_time == 0) break_time++; |
03d12ddf8
|
1393 |
locks_insert_block(inode->i_flock, new_fl); |
62af4f1f7
|
1394 |
trace_break_lease_block(inode, new_fl); |
1c8c601a8
|
1395 |
spin_unlock(&inode->i_lock); |
c45198eda
|
1396 |
locks_dispose_list(&dispose); |
4321e01e7
|
1397 1398 |
error = wait_event_interruptible_timeout(new_fl->fl_wait, !new_fl->fl_next, break_time); |
1c8c601a8
|
1399 |
spin_lock(&inode->i_lock); |
62af4f1f7
|
1400 |
trace_break_lease_unblock(inode, new_fl); |
1c8c601a8
|
1401 |
locks_delete_block(new_fl); |
1da177e4c
|
1402 |
if (error >= 0) { |
778fc546f
|
1403 1404 1405 1406 |
/* * Wait for the next conflicting lease that has not been * broken yet */ |
03d12ddf8
|
1407 1408 1409 1410 |
if (error == 0) time_out_leases(inode, &dispose); if (any_leases_conflict(inode, new_fl)) goto restart; |
1da177e4c
|
1411 1412 1413 1414 |
error = 0; } out: |
1c8c601a8
|
1415 |
spin_unlock(&inode->i_lock); |
c45198eda
|
1416 |
locks_dispose_list(&dispose); |
6d4b9e38d
|
1417 |
locks_free_lock(new_fl); |
1da177e4c
|
1418 1419 1420 1421 1422 1423 |
return error; } EXPORT_SYMBOL(__break_lease); /** |
a6b91919e
|
1424 |
* lease_get_mtime - get the last modified time of an inode |
1da177e4c
|
1425 1426 1427 1428 1429 |
* @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 |
a6b91919e
|
1430 |
* exclusive lease, then they could be modifying it. |
1da177e4c
|
1431 1432 1433 |
*/ void lease_get_mtime(struct inode *inode, struct timespec *time) { |
bfe860243
|
1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 |
bool has_lease = false; struct file_lock *flock; if (inode->i_flock) { spin_lock(&inode->i_lock); flock = inode->i_flock; if (flock && IS_LEASE(flock) && (flock->fl_type == F_WRLCK)) has_lease = true; spin_unlock(&inode->i_lock); } if (has_lease) |
1da177e4c
|
1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 |
*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; |
1c8c601a8
|
1479 |
struct inode *inode = file_inode(filp); |
1da177e4c
|
1480 |
int type = F_UNLCK; |
c45198eda
|
1481 |
LIST_HEAD(dispose); |
1da177e4c
|
1482 |
|
1c8c601a8
|
1483 |
spin_lock(&inode->i_lock); |
c45198eda
|
1484 |
time_out_leases(file_inode(filp), &dispose); |
496ad9aa8
|
1485 |
for (fl = file_inode(filp)->i_flock; fl && IS_LEASE(fl); |
1da177e4c
|
1486 1487 |
fl = fl->fl_next) { if (fl->fl_file == filp) { |
778fc546f
|
1488 |
type = target_leasetype(fl); |
1da177e4c
|
1489 1490 1491 |
break; } } |
1c8c601a8
|
1492 |
spin_unlock(&inode->i_lock); |
c45198eda
|
1493 |
locks_dispose_list(&dispose); |
1da177e4c
|
1494 1495 |
return type; } |
24cbe7845
|
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 |
/** * check_conflicting_open - see if the given dentry points to a file that has * an existing open that would conflict with the * desired lease. * @dentry: dentry to check * @arg: type of lease that we're trying to acquire * * Check to see if there's an existing open fd on this file that would * conflict with the lease we're trying to set. */ static int check_conflicting_open(const struct dentry *dentry, const long arg) { int ret = 0; struct inode *inode = dentry->d_inode; if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0)) return -EAGAIN; if ((arg == F_WRLCK) && ((d_count(dentry) > 1) || (atomic_read(&inode->i_count) > 1))) ret = -EAGAIN; return ret; } |
e6f5c7893
|
1521 1522 |
static int generic_add_lease(struct file *filp, long arg, struct file_lock **flp, void **priv) |
1da177e4c
|
1523 |
{ |
7eaae2828
|
1524 |
struct file_lock *fl, **before, **my_before = NULL, *lease; |
0f7fc9e4d
|
1525 |
struct dentry *dentry = filp->f_path.dentry; |
1da177e4c
|
1526 |
struct inode *inode = dentry->d_inode; |
df4e8d2c1
|
1527 |
bool is_deleg = (*flp)->fl_flags & FL_DELEG; |
c1f24ef4e
|
1528 |
int error; |
c45198eda
|
1529 |
LIST_HEAD(dispose); |
1da177e4c
|
1530 |
|
096657b65
|
1531 |
lease = *flp; |
62af4f1f7
|
1532 |
trace_generic_add_lease(inode, lease); |
df4e8d2c1
|
1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 |
/* * In the delegation case we need mutual exclusion with * a number of operations that take the i_mutex. We trylock * because delegations are an optional optimization, and if * there's some chance of a conflict--we'd rather not * bother, maybe that's a sign this just isn't a good file to * hand out a delegation on. */ if (is_deleg && !mutex_trylock(&inode->i_mutex)) return -EAGAIN; if (is_deleg && arg == F_WRLCK) { /* Write delegations are not currently supported: */ |
4fdb793ff
|
1546 |
mutex_unlock(&inode->i_mutex); |
df4e8d2c1
|
1547 1548 1549 |
WARN_ON_ONCE(1); return -EINVAL; } |
096657b65
|
1550 |
|
f82b4b678
|
1551 |
spin_lock(&inode->i_lock); |
c45198eda
|
1552 |
time_out_leases(inode, &dispose); |
24cbe7845
|
1553 1554 |
error = check_conflicting_open(dentry, arg); if (error) |
096657b65
|
1555 |
goto out; |
6d5e8b05c
|
1556 |
|
1da177e4c
|
1557 1558 1559 1560 1561 1562 1563 1564 |
/* * 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. */ |
c1f24ef4e
|
1565 |
error = -EAGAIN; |
1da177e4c
|
1566 1567 1568 |
for (before = &inode->i_flock; ((fl = *before) != NULL) && IS_LEASE(fl); before = &fl->fl_next) { |
c1f24ef4e
|
1569 |
if (fl->fl_file == filp) { |
1da177e4c
|
1570 |
my_before = before; |
c1f24ef4e
|
1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 |
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; |
1da177e4c
|
1585 |
} |
1da177e4c
|
1586 |
if (my_before != NULL) { |
1c7dd2ff4
|
1587 |
lease = *my_before; |
c45198eda
|
1588 |
error = lease->fl_lmops->lm_change(my_before, arg, &dispose); |
1c7dd2ff4
|
1589 1590 1591 |
if (error) goto out; goto out_setup; |
1da177e4c
|
1592 |
} |
1da177e4c
|
1593 1594 1595 |
error = -EINVAL; if (!leases_enable) goto out; |
c5b1f0d92
|
1596 |
locks_insert_lock(before, lease); |
24cbe7845
|
1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 |
/* * The check in break_lease() is lockless. It's possible for another * open to race in after we did the earlier check for a conflicting * open but before the lease was inserted. Check again for a * conflicting open and cancel the lease if there is one. * * We also add a barrier here to ensure that the insertion of the lock * precedes these checks. */ smp_mb(); error = check_conflicting_open(dentry, arg); if (error) |
e6f5c7893
|
1609 |
goto out_unlink; |
1c7dd2ff4
|
1610 1611 1612 1613 |
out_setup: if (lease->fl_lmops->lm_setup) lease->fl_lmops->lm_setup(lease, priv); |
1da177e4c
|
1614 |
out: |
f82b4b678
|
1615 |
spin_unlock(&inode->i_lock); |
c45198eda
|
1616 |
locks_dispose_list(&dispose); |
df4e8d2c1
|
1617 1618 |
if (is_deleg) mutex_unlock(&inode->i_mutex); |
1c7dd2ff4
|
1619 1620 |
if (!error && !my_before) *flp = NULL; |
1da177e4c
|
1621 |
return error; |
e6f5c7893
|
1622 1623 1624 |
out_unlink: locks_unlink_lock(before); goto out; |
1da177e4c
|
1625 |
} |
8335ebd94
|
1626 |
|
0efaa7e82
|
1627 |
static int generic_delete_lease(struct file *filp) |
8335ebd94
|
1628 |
{ |
0efaa7e82
|
1629 |
int error = -EAGAIN; |
8335ebd94
|
1630 1631 1632 |
struct file_lock *fl, **before; struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; |
c45198eda
|
1633 |
LIST_HEAD(dispose); |
8335ebd94
|
1634 |
|
f82b4b678
|
1635 |
spin_lock(&inode->i_lock); |
c45198eda
|
1636 |
time_out_leases(inode, &dispose); |
8335ebd94
|
1637 1638 1639 |
for (before = &inode->i_flock; ((fl = *before) != NULL) && IS_LEASE(fl); before = &fl->fl_next) { |
0efaa7e82
|
1640 1641 |
if (fl->fl_file == filp) break; |
8335ebd94
|
1642 |
} |
0efaa7e82
|
1643 1644 |
trace_generic_delete_lease(inode, fl); if (fl) |
c45198eda
|
1645 |
error = fl->fl_lmops->lm_change(before, F_UNLCK, &dispose); |
f82b4b678
|
1646 |
spin_unlock(&inode->i_lock); |
c45198eda
|
1647 |
locks_dispose_list(&dispose); |
0efaa7e82
|
1648 |
return error; |
8335ebd94
|
1649 1650 1651 1652 |
} /** * generic_setlease - sets a lease on an open file |
1c7dd2ff4
|
1653 1654 1655 1656 1657 |
* @filp: file pointer * @arg: type of lease to obtain * @flp: input - file_lock to use, output - file_lock inserted * @priv: private data for lm_setup (may be NULL if lm_setup * doesn't require it) |
8335ebd94
|
1658 1659 1660 |
* * The (input) flp->fl_lmops->lm_break function is required * by break_lease(). |
8335ebd94
|
1661 |
*/ |
e6f5c7893
|
1662 1663 |
int generic_setlease(struct file *filp, long arg, struct file_lock **flp, void **priv) |
8335ebd94
|
1664 1665 1666 1667 |
{ struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; int error; |
8e96e3b7b
|
1668 |
if ((!uid_eq(current_fsuid(), inode->i_uid)) && !capable(CAP_LEASE)) |
8335ebd94
|
1669 1670 1671 1672 1673 1674 |
return -EACCES; if (!S_ISREG(inode->i_mode)) return -EINVAL; error = security_file_lock(filp, arg); if (error) return error; |
8335ebd94
|
1675 1676 |
switch (arg) { case F_UNLCK: |
0efaa7e82
|
1677 |
return generic_delete_lease(filp); |
8335ebd94
|
1678 1679 |
case F_RDLCK: case F_WRLCK: |
0efaa7e82
|
1680 1681 1682 1683 |
if (!(*flp)->fl_lmops->lm_break) { WARN_ON_ONCE(1); return -ENOLCK; } |
e6f5c7893
|
1684 |
return generic_add_lease(filp, arg, flp, priv); |
8335ebd94
|
1685 |
default: |
8d657eb3b
|
1686 |
return -EINVAL; |
8335ebd94
|
1687 1688 |
} } |
0af1a4504
|
1689 |
EXPORT_SYMBOL(generic_setlease); |
1da177e4c
|
1690 |
|
b89f43213
|
1691 |
/** |
e51673aa5
|
1692 |
* vfs_setlease - sets a lease on an open file |
1c7dd2ff4
|
1693 1694 1695 1696 1697 |
* @filp: file pointer * @arg: type of lease to obtain * @lease: file_lock to use when adding a lease * @priv: private info for lm_setup when adding a lease (may be * NULL if lm_setup doesn't require it) |
e51673aa5
|
1698 1699 1700 1701 1702 1703 |
* * Call this to establish a lease on the file. The "lease" argument is not * used for F_UNLCK requests and may be NULL. For commands that set or alter * an existing lease, the (*lease)->fl_lmops->lm_break operation must be set; * if not, this function will return -ENOLCK (and generate a scary-looking * stack trace). |
1c7dd2ff4
|
1704 1705 1706 |
* * The "priv" pointer is passed directly to the lm_setup function as-is. It * may be NULL if the lm_setup operation doesn't require it. |
1da177e4c
|
1707 |
*/ |
e6f5c7893
|
1708 1709 |
int vfs_setlease(struct file *filp, long arg, struct file_lock **lease, void **priv) |
1da177e4c
|
1710 |
{ |
1c7dd2ff4
|
1711 |
if (filp->f_op->setlease) |
f82b4b678
|
1712 |
return filp->f_op->setlease(filp, arg, lease, priv); |
1c7dd2ff4
|
1713 |
else |
f82b4b678
|
1714 |
return generic_setlease(filp, arg, lease, priv); |
1da177e4c
|
1715 |
} |
a9933cea7
|
1716 |
EXPORT_SYMBOL_GPL(vfs_setlease); |
1da177e4c
|
1717 |
|
0ceaf6c70
|
1718 |
static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) |
1da177e4c
|
1719 |
{ |
1c7dd2ff4
|
1720 |
struct file_lock *fl; |
f7347ce4e
|
1721 |
struct fasync_struct *new; |
1da177e4c
|
1722 |
int error; |
c5b1f0d92
|
1723 1724 1725 |
fl = lease_alloc(filp, arg); if (IS_ERR(fl)) return PTR_ERR(fl); |
1da177e4c
|
1726 |
|
f7347ce4e
|
1727 1728 1729 1730 1731 |
new = fasync_alloc(); if (!new) { locks_free_lock(fl); return -ENOMEM; } |
1c7dd2ff4
|
1732 |
new->fa_fd = fd; |
f7347ce4e
|
1733 |
|
1c7dd2ff4
|
1734 |
error = vfs_setlease(filp, arg, &fl, (void **)&new); |
2dfb928f7
|
1735 1736 |
if (fl) locks_free_lock(fl); |
f7347ce4e
|
1737 1738 |
if (new) fasync_free(new); |
1da177e4c
|
1739 1740 1741 1742 |
return error; } /** |
0ceaf6c70
|
1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 |
* 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) |
e6f5c7893
|
1755 |
return vfs_setlease(filp, F_UNLCK, NULL, NULL); |
0ceaf6c70
|
1756 1757 1758 1759 |
return do_fcntl_add_lease(fd, filp, arg); } /** |
1da177e4c
|
1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 |
* 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); |
bde74e4bc
|
1772 |
if (error != FILE_LOCK_DEFERRED) |
1da177e4c
|
1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 |
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. */ |
002c8976e
|
1805 |
SYSCALL_DEFINE2(flock, unsigned int, fd, unsigned int, cmd) |
1da177e4c
|
1806 |
{ |
2903ff019
|
1807 |
struct fd f = fdget(fd); |
1da177e4c
|
1808 1809 1810 1811 1812 |
struct file_lock *lock; int can_sleep, unlock; int error; error = -EBADF; |
2903ff019
|
1813 |
if (!f.file) |
1da177e4c
|
1814 1815 1816 1817 1818 |
goto out; can_sleep = !(cmd & LOCK_NB); cmd &= ~LOCK_NB; unlock = (cmd == LOCK_UN); |
aeb5d7270
|
1819 |
if (!unlock && !(cmd & LOCK_MAND) && |
2903ff019
|
1820 |
!(f.file->f_mode & (FMODE_READ|FMODE_WRITE))) |
1da177e4c
|
1821 |
goto out_putf; |
6e129d006
|
1822 1823 1824 |
lock = flock_make_lock(f.file, cmd); if (IS_ERR(lock)) { error = PTR_ERR(lock); |
1da177e4c
|
1825 |
goto out_putf; |
6e129d006
|
1826 |
} |
1da177e4c
|
1827 1828 |
if (can_sleep) lock->fl_flags |= FL_SLEEP; |
2903ff019
|
1829 |
error = security_file_lock(f.file, lock->fl_type); |
1da177e4c
|
1830 1831 |
if (error) goto out_free; |
72c2d5319
|
1832 |
if (f.file->f_op->flock) |
2903ff019
|
1833 |
error = f.file->f_op->flock(f.file, |
1da177e4c
|
1834 1835 1836 |
(can_sleep) ? F_SETLKW : F_SETLK, lock); else |
2903ff019
|
1837 |
error = flock_lock_file_wait(f.file, lock); |
1da177e4c
|
1838 1839 |
out_free: |
993dfa877
|
1840 |
locks_free_lock(lock); |
1da177e4c
|
1841 1842 |
out_putf: |
2903ff019
|
1843 |
fdput(f); |
1da177e4c
|
1844 1845 1846 |
out: return error; } |
3ee17abd1
|
1847 1848 1849 |
/** * vfs_test_lock - test file byte range lock * @filp: The file to test lock for |
6924c5549
|
1850 |
* @fl: The lock to test; also used to hold result |
3ee17abd1
|
1851 1852 1853 1854 1855 1856 |
* * 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) { |
72c2d5319
|
1857 |
if (filp->f_op->lock) |
3ee17abd1
|
1858 1859 1860 1861 1862 |
return filp->f_op->lock(filp, F_GETLK, fl); posix_test_lock(filp, fl); return 0; } EXPORT_SYMBOL_GPL(vfs_test_lock); |
c2fa1b8a6
|
1863 1864 |
static int posix_lock_to_flock(struct flock *flock, struct file_lock *fl) { |
cff2fce58
|
1865 |
flock->l_pid = IS_OFDLCK(fl) ? -1 : fl->fl_pid; |
c2fa1b8a6
|
1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 |
#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; |
129a84de2
|
1880 |
flock->l_type = fl->fl_type; |
c2fa1b8a6
|
1881 1882 1883 1884 1885 1886 |
return 0; } #if BITS_PER_LONG == 32 static void posix_lock_to_flock64(struct flock64 *flock, struct file_lock *fl) { |
cff2fce58
|
1887 |
flock->l_pid = IS_OFDLCK(fl) ? -1 : fl->fl_pid; |
c2fa1b8a6
|
1888 1889 1890 1891 1892 1893 1894 |
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 |
1da177e4c
|
1895 1896 1897 |
/* Report the first existing lock that would conflict with l. * This implements the F_GETLK command of fcntl(). */ |
c1e62b8fc
|
1898 |
int fcntl_getlk(struct file *filp, unsigned int cmd, struct flock __user *l) |
1da177e4c
|
1899 |
{ |
9d6a8c5c2
|
1900 |
struct file_lock file_lock; |
1da177e4c
|
1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 |
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; |
0d3f7a2dd
|
1914 |
if (cmd == F_OFD_GETLK) { |
90478939d
|
1915 1916 1917 |
error = -EINVAL; if (flock.l_pid != 0) goto out; |
5d50ffd7c
|
1918 |
cmd = F_GETLK; |
cff2fce58
|
1919 |
file_lock.fl_flags |= FL_OFDLCK; |
73a8f5f7e
|
1920 |
file_lock.fl_owner = filp; |
5d50ffd7c
|
1921 |
} |
3ee17abd1
|
1922 1923 1924 |
error = vfs_test_lock(filp, &file_lock); if (error) goto out; |
1da177e4c
|
1925 |
|
9d6a8c5c2
|
1926 1927 1928 |
flock.l_type = file_lock.fl_type; if (file_lock.fl_type != F_UNLCK) { error = posix_lock_to_flock(&flock, &file_lock); |
c2fa1b8a6
|
1929 |
if (error) |
f328296e2
|
1930 |
goto rel_priv; |
1da177e4c
|
1931 1932 1933 1934 |
} error = -EFAULT; if (!copy_to_user(l, &flock, sizeof(flock))) error = 0; |
f328296e2
|
1935 1936 |
rel_priv: locks_release_private(&file_lock); |
1da177e4c
|
1937 1938 1939 |
out: return error; } |
7723ec977
|
1940 1941 1942 1943 1944 |
/** * 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 |
150b39345
|
1945 1946 1947 1948 1949 1950 1951 1952 |
* @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. |
2beb6614f
|
1953 1954 1955 1956 |
* * 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) |
8fb47a4fb
|
1957 |
* lm_grant is set. Callers expecting ->lock() to return asynchronously |
2beb6614f
|
1958 1959 |
* 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, |
8fb47a4fb
|
1960 |
* it must return FILE_LOCK_DEFERRED, and call ->lm_grant() when the lock |
2beb6614f
|
1961 1962 |
* request completes. * If the request is for non-blocking lock the file system should return |
bde74e4bc
|
1963 1964 |
* 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 |
2beb6614f
|
1965 1966 1967 1968 1969 |
* 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 |
8fb47a4fb
|
1970 |
* ->lm_grant() before returning to the caller with a FILE_LOCK_DEFERRED |
2beb6614f
|
1971 |
* return code. |
7723ec977
|
1972 |
*/ |
150b39345
|
1973 |
int vfs_lock_file(struct file *filp, unsigned int cmd, struct file_lock *fl, struct file_lock *conf) |
7723ec977
|
1974 |
{ |
72c2d5319
|
1975 |
if (filp->f_op->lock) |
7723ec977
|
1976 1977 |
return filp->f_op->lock(filp, cmd, fl); else |
150b39345
|
1978 |
return posix_lock_file(filp, fl, conf); |
7723ec977
|
1979 1980 |
} EXPORT_SYMBOL_GPL(vfs_lock_file); |
b648a6de0
|
1981 1982 1983 1984 1985 1986 1987 1988 |
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; |
764c76b37
|
1989 1990 1991 |
for (;;) { error = vfs_lock_file(filp, cmd, fl, NULL); if (error != FILE_LOCK_DEFERRED) |
b648a6de0
|
1992 |
break; |
764c76b37
|
1993 1994 1995 1996 1997 1998 |
error = wait_event_interruptible(fl->fl_wait, !fl->fl_next); if (!error) continue; locks_delete_block(fl); break; |
b648a6de0
|
1999 2000 2001 2002 |
} return error; } |
cf01f4eef
|
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 |
/* Ensure that fl->fl_filp has compatible f_mode for F_SETLK calls */ static int check_fmode_for_setlk(struct file_lock *fl) { switch (fl->fl_type) { case F_RDLCK: if (!(fl->fl_file->f_mode & FMODE_READ)) return -EBADF; break; case F_WRLCK: if (!(fl->fl_file->f_mode & FMODE_WRITE)) return -EBADF; } return 0; } |
1da177e4c
|
2018 2019 2020 |
/* Apply the lock described by l to an open file descriptor. * This implements both the F_SETLK and F_SETLKW commands of fcntl(). */ |
c293621bb
|
2021 2022 |
int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd, struct flock __user *l) |
1da177e4c
|
2023 2024 2025 2026 |
{ struct file_lock *file_lock = locks_alloc_lock(); struct flock flock; struct inode *inode; |
0b2bac2f1
|
2027 |
struct file *f; |
1da177e4c
|
2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 |
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; |
496ad9aa8
|
2039 |
inode = file_inode(filp); |
1da177e4c
|
2040 2041 2042 2043 |
/* Don't allow mandatory locks on files that may be memory mapped * and shared. */ |
a16877ca9
|
2044 |
if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) { |
1da177e4c
|
2045 2046 2047 |
error = -EAGAIN; goto out; } |
c293621bb
|
2048 |
again: |
1da177e4c
|
2049 2050 2051 |
error = flock_to_posix_lock(filp, file_lock, &flock); if (error) goto out; |
5d50ffd7c
|
2052 |
|
cf01f4eef
|
2053 2054 2055 |
error = check_fmode_for_setlk(file_lock); if (error) goto out; |
5d50ffd7c
|
2056 2057 |
/* * If the cmd is requesting file-private locks, then set the |
cff2fce58
|
2058 |
* FL_OFDLCK flag and override the owner. |
5d50ffd7c
|
2059 2060 |
*/ switch (cmd) { |
0d3f7a2dd
|
2061 |
case F_OFD_SETLK: |
90478939d
|
2062 2063 2064 |
error = -EINVAL; if (flock.l_pid != 0) goto out; |
5d50ffd7c
|
2065 |
cmd = F_SETLK; |
cff2fce58
|
2066 |
file_lock->fl_flags |= FL_OFDLCK; |
73a8f5f7e
|
2067 |
file_lock->fl_owner = filp; |
5d50ffd7c
|
2068 |
break; |
0d3f7a2dd
|
2069 |
case F_OFD_SETLKW: |
90478939d
|
2070 2071 2072 |
error = -EINVAL; if (flock.l_pid != 0) goto out; |
5d50ffd7c
|
2073 |
cmd = F_SETLKW; |
cff2fce58
|
2074 |
file_lock->fl_flags |= FL_OFDLCK; |
73a8f5f7e
|
2075 |
file_lock->fl_owner = filp; |
5d50ffd7c
|
2076 2077 |
/* Fallthrough */ case F_SETLKW: |
1da177e4c
|
2078 2079 |
file_lock->fl_flags |= FL_SLEEP; } |
5d50ffd7c
|
2080 |
|
b648a6de0
|
2081 |
error = do_lock_file_wait(filp, cmd, file_lock); |
1da177e4c
|
2082 |
|
c293621bb
|
2083 2084 2085 2086 |
/* * Attempt to detect a close/fcntl race and recover by * releasing the lock that was just acquired. */ |
0b2bac2f1
|
2087 2088 2089 2090 2091 2092 2093 2094 2095 |
/* * 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(¤t->files->file_lock); f = fcheck(fd); spin_unlock(¤t->files->file_lock); if (!error && f != filp && flock.l_type != F_UNLCK) { |
c293621bb
|
2096 2097 |
flock.l_type = F_UNLCK; goto again; |
1da177e4c
|
2098 |
} |
c293621bb
|
2099 |
out: |
1da177e4c
|
2100 2101 2102 2103 2104 2105 2106 2107 |
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(). */ |
c1e62b8fc
|
2108 |
int fcntl_getlk64(struct file *filp, unsigned int cmd, struct flock64 __user *l) |
1da177e4c
|
2109 |
{ |
9d6a8c5c2
|
2110 |
struct file_lock file_lock; |
1da177e4c
|
2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 |
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; |
0d3f7a2dd
|
2124 |
if (cmd == F_OFD_GETLK) { |
90478939d
|
2125 2126 2127 |
error = -EINVAL; if (flock.l_pid != 0) goto out; |
5d50ffd7c
|
2128 |
cmd = F_GETLK64; |
cff2fce58
|
2129 |
file_lock.fl_flags |= FL_OFDLCK; |
73a8f5f7e
|
2130 |
file_lock.fl_owner = filp; |
5d50ffd7c
|
2131 |
} |
3ee17abd1
|
2132 2133 2134 |
error = vfs_test_lock(filp, &file_lock); if (error) goto out; |
9d6a8c5c2
|
2135 2136 2137 |
flock.l_type = file_lock.fl_type; if (file_lock.fl_type != F_UNLCK) posix_lock_to_flock64(&flock, &file_lock); |
1da177e4c
|
2138 2139 2140 |
error = -EFAULT; if (!copy_to_user(l, &flock, sizeof(flock))) error = 0; |
f328296e2
|
2141 2142 |
locks_release_private(&file_lock); |
1da177e4c
|
2143 2144 2145 2146 2147 2148 2149 |
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(). */ |
c293621bb
|
2150 2151 |
int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd, struct flock64 __user *l) |
1da177e4c
|
2152 2153 2154 2155 |
{ struct file_lock *file_lock = locks_alloc_lock(); struct flock64 flock; struct inode *inode; |
0b2bac2f1
|
2156 |
struct file *f; |
1da177e4c
|
2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 |
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; |
496ad9aa8
|
2168 |
inode = file_inode(filp); |
1da177e4c
|
2169 2170 2171 2172 |
/* Don't allow mandatory locks on files that may be memory mapped * and shared. */ |
a16877ca9
|
2173 |
if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) { |
1da177e4c
|
2174 2175 2176 |
error = -EAGAIN; goto out; } |
c293621bb
|
2177 |
again: |
1da177e4c
|
2178 2179 2180 |
error = flock64_to_posix_lock(filp, file_lock, &flock); if (error) goto out; |
5d50ffd7c
|
2181 |
|
cf01f4eef
|
2182 2183 2184 |
error = check_fmode_for_setlk(file_lock); if (error) goto out; |
5d50ffd7c
|
2185 2186 |
/* * If the cmd is requesting file-private locks, then set the |
cff2fce58
|
2187 |
* FL_OFDLCK flag and override the owner. |
5d50ffd7c
|
2188 2189 |
*/ switch (cmd) { |
0d3f7a2dd
|
2190 |
case F_OFD_SETLK: |
90478939d
|
2191 2192 2193 |
error = -EINVAL; if (flock.l_pid != 0) goto out; |
5d50ffd7c
|
2194 |
cmd = F_SETLK64; |
cff2fce58
|
2195 |
file_lock->fl_flags |= FL_OFDLCK; |
73a8f5f7e
|
2196 |
file_lock->fl_owner = filp; |
5d50ffd7c
|
2197 |
break; |
0d3f7a2dd
|
2198 |
case F_OFD_SETLKW: |
90478939d
|
2199 2200 2201 |
error = -EINVAL; if (flock.l_pid != 0) goto out; |
5d50ffd7c
|
2202 |
cmd = F_SETLKW64; |
cff2fce58
|
2203 |
file_lock->fl_flags |= FL_OFDLCK; |
73a8f5f7e
|
2204 |
file_lock->fl_owner = filp; |
5d50ffd7c
|
2205 2206 |
/* Fallthrough */ case F_SETLKW64: |
1da177e4c
|
2207 2208 |
file_lock->fl_flags |= FL_SLEEP; } |
5d50ffd7c
|
2209 |
|
b648a6de0
|
2210 |
error = do_lock_file_wait(filp, cmd, file_lock); |
1da177e4c
|
2211 |
|
c293621bb
|
2212 2213 2214 2215 |
/* * Attempt to detect a close/fcntl race and recover by * releasing the lock that was just acquired. */ |
0b2bac2f1
|
2216 2217 2218 2219 |
spin_lock(¤t->files->file_lock); f = fcheck(fd); spin_unlock(¤t->files->file_lock); if (!error && f != filp && flock.l_type != F_UNLCK) { |
c293621bb
|
2220 2221 |
flock.l_type = F_UNLCK; goto again; |
1da177e4c
|
2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 |
} 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) { |
ff7b86b82
|
2237 |
struct file_lock lock; |
1da177e4c
|
2238 2239 2240 2241 2242 2243 |
/* * 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. */ |
496ad9aa8
|
2244 |
if (!file_inode(filp)->i_flock) |
1da177e4c
|
2245 2246 2247 |
return; lock.fl_type = F_UNLCK; |
75e1fcc0b
|
2248 |
lock.fl_flags = FL_POSIX | FL_CLOSE; |
1da177e4c
|
2249 2250 2251 2252 2253 2254 2255 |
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; |
150b39345
|
2256 |
vfs_lock_file(filp, F_SETLK, &lock, NULL); |
1da177e4c
|
2257 |
|
1da177e4c
|
2258 2259 2260 2261 2262 2263 2264 2265 2266 |
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. */ |
78ed8a133
|
2267 |
void locks_remove_file(struct file *filp) |
1da177e4c
|
2268 |
{ |
496ad9aa8
|
2269 |
struct inode * inode = file_inode(filp); |
1da177e4c
|
2270 2271 |
struct file_lock *fl; struct file_lock **before; |
ed9814d85
|
2272 |
LIST_HEAD(dispose); |
1da177e4c
|
2273 2274 2275 |
if (!inode->i_flock) return; |
73a8f5f7e
|
2276 |
locks_remove_posix(filp, filp); |
5d50ffd7c
|
2277 |
|
72c2d5319
|
2278 |
if (filp->f_op->flock) { |
1da177e4c
|
2279 |
struct file_lock fl = { |
73a8f5f7e
|
2280 |
.fl_owner = filp, |
1da177e4c
|
2281 2282 2283 2284 2285 2286 2287 |
.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); |
80fec4c62
|
2288 2289 |
if (fl.fl_ops && fl.fl_ops->fl_release_private) fl.fl_ops->fl_release_private(&fl); |
1da177e4c
|
2290 |
} |
1c8c601a8
|
2291 |
spin_lock(&inode->i_lock); |
1da177e4c
|
2292 2293 2294 2295 |
before = &inode->i_flock; while ((fl = *before) != NULL) { if (fl->fl_file == filp) { |
1da177e4c
|
2296 |
if (IS_LEASE(fl)) { |
c45198eda
|
2297 |
lease_modify(before, F_UNLCK, &dispose); |
1da177e4c
|
2298 2299 |
continue; } |
8c3cac5e6
|
2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 |
/* * There's a leftover lock on the list of a type that * we didn't expect to see. Most likely a classic * POSIX lock that ended up not getting released * properly, or that raced onto the list somehow. Log * some info about it and then just remove it from * the list. */ WARN(!IS_FLOCK(fl), "leftover lock: dev=%u:%u ino=%lu type=%hhd flags=0x%x start=%lld end=%lld ", MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev), inode->i_ino, fl->fl_type, fl->fl_flags, fl->fl_start, fl->fl_end); |
ed9814d85
|
2316 |
locks_delete_lock(before, &dispose); |
8c3cac5e6
|
2317 |
continue; |
1da177e4c
|
2318 2319 2320 |
} before = &fl->fl_next; } |
1c8c601a8
|
2321 |
spin_unlock(&inode->i_lock); |
ed9814d85
|
2322 |
locks_dispose_list(&dispose); |
1da177e4c
|
2323 2324 2325 |
} /** |
1da177e4c
|
2326 |
* posix_unblock_lock - stop waiting for a file lock |
1da177e4c
|
2327 2328 2329 2330 |
* @waiter: the lock which was waiting * * lockd needs to block waiting for locks. */ |
64a318ee2
|
2331 |
int |
f891a29f4
|
2332 |
posix_unblock_lock(struct file_lock *waiter) |
1da177e4c
|
2333 |
{ |
64a318ee2
|
2334 |
int status = 0; |
7b2296afb
|
2335 |
spin_lock(&blocked_lock_lock); |
5996a298d
|
2336 |
if (waiter->fl_next) |
1da177e4c
|
2337 |
__locks_delete_block(waiter); |
64a318ee2
|
2338 2339 |
else status = -ENOENT; |
7b2296afb
|
2340 |
spin_unlock(&blocked_lock_lock); |
64a318ee2
|
2341 |
return status; |
1da177e4c
|
2342 |
} |
1da177e4c
|
2343 |
EXPORT_SYMBOL(posix_unblock_lock); |
9b9d2ab41
|
2344 2345 2346 2347 2348 2349 2350 2351 2352 |
/** * 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) { |
72c2d5319
|
2353 |
if (filp->f_op->lock) |
9b9d2ab41
|
2354 2355 2356 2357 2358 |
return filp->f_op->lock(filp, F_CANCELLK, fl); return 0; } EXPORT_SYMBOL_GPL(vfs_cancel_lock); |
7f8ada98d
|
2359 |
#ifdef CONFIG_PROC_FS |
d8ba7a363
|
2360 |
#include <linux/proc_fs.h> |
7f8ada98d
|
2361 |
#include <linux/seq_file.h> |
7012b02a2
|
2362 2363 2364 2365 |
struct locks_iterator { int li_cpu; loff_t li_pos; }; |
7f8ada98d
|
2366 |
static void lock_get_status(struct seq_file *f, struct file_lock *fl, |
99dc82925
|
2367 |
loff_t id, char *pfx) |
1da177e4c
|
2368 2369 |
{ struct inode *inode = NULL; |
ab1f16116
|
2370 2371 2372 |
unsigned int fl_pid; if (fl->fl_nspid) |
6c5f3e7b4
|
2373 |
fl_pid = pid_vnr(fl->fl_nspid); |
ab1f16116
|
2374 2375 |
else fl_pid = fl->fl_pid; |
1da177e4c
|
2376 2377 |
if (fl->fl_file != NULL) |
496ad9aa8
|
2378 |
inode = file_inode(fl->fl_file); |
1da177e4c
|
2379 |
|
99dc82925
|
2380 |
seq_printf(f, "%lld:%s ", id, pfx); |
1da177e4c
|
2381 |
if (IS_POSIX(fl)) { |
c918d42a2
|
2382 |
if (fl->fl_flags & FL_ACCESS) |
5315c26a6
|
2383 |
seq_puts(f, "ACCESS"); |
cff2fce58
|
2384 |
else if (IS_OFDLCK(fl)) |
5315c26a6
|
2385 |
seq_puts(f, "OFDLCK"); |
c918d42a2
|
2386 |
else |
5315c26a6
|
2387 |
seq_puts(f, "POSIX "); |
c918d42a2
|
2388 2389 |
seq_printf(f, " %s ", |
1da177e4c
|
2390 |
(inode == NULL) ? "*NOINODE*" : |
a16877ca9
|
2391 |
mandatory_lock(inode) ? "MANDATORY" : "ADVISORY "); |
1da177e4c
|
2392 2393 |
} else if (IS_FLOCK(fl)) { if (fl->fl_type & LOCK_MAND) { |
5315c26a6
|
2394 |
seq_puts(f, "FLOCK MSNFS "); |
1da177e4c
|
2395 |
} else { |
5315c26a6
|
2396 |
seq_puts(f, "FLOCK ADVISORY "); |
1da177e4c
|
2397 2398 |
} } else if (IS_LEASE(fl)) { |
8144f1f69
|
2399 2400 2401 2402 |
if (fl->fl_flags & FL_DELEG) seq_puts(f, "DELEG "); else seq_puts(f, "LEASE "); |
ab83fa4b4
|
2403 |
if (lease_breaking(fl)) |
5315c26a6
|
2404 |
seq_puts(f, "BREAKING "); |
1da177e4c
|
2405 |
else if (fl->fl_file) |
5315c26a6
|
2406 |
seq_puts(f, "ACTIVE "); |
1da177e4c
|
2407 |
else |
5315c26a6
|
2408 |
seq_puts(f, "BREAKER "); |
1da177e4c
|
2409 |
} else { |
5315c26a6
|
2410 |
seq_puts(f, "UNKNOWN UNKNOWN "); |
1da177e4c
|
2411 2412 |
} if (fl->fl_type & LOCK_MAND) { |
7f8ada98d
|
2413 |
seq_printf(f, "%s ", |
1da177e4c
|
2414 2415 2416 2417 |
(fl->fl_type & LOCK_READ) ? (fl->fl_type & LOCK_WRITE) ? "RW " : "READ " : (fl->fl_type & LOCK_WRITE) ? "WRITE" : "NONE "); } else { |
7f8ada98d
|
2418 |
seq_printf(f, "%s ", |
ab83fa4b4
|
2419 |
(lease_breaking(fl)) |
0ee5c6d63
|
2420 2421 |
? (fl->fl_type == F_UNLCK) ? "UNLCK" : "READ " : (fl->fl_type == F_WRLCK) ? "WRITE" : "READ "); |
1da177e4c
|
2422 2423 2424 |
} if (inode) { #ifdef WE_CAN_BREAK_LSLK_NOW |
ab1f16116
|
2425 |
seq_printf(f, "%d %s:%ld ", fl_pid, |
1da177e4c
|
2426 2427 2428 |
inode->i_sb->s_id, inode->i_ino); #else /* userspace relies on this representation of dev_t ;-( */ |
ab1f16116
|
2429 |
seq_printf(f, "%d %02x:%02x:%ld ", fl_pid, |
1da177e4c
|
2430 2431 2432 2433 |
MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev), inode->i_ino); #endif } else { |
ab1f16116
|
2434 |
seq_printf(f, "%d <none>:0 ", fl_pid); |
1da177e4c
|
2435 2436 2437 |
} if (IS_POSIX(fl)) { if (fl->fl_end == OFFSET_MAX) |
7f8ada98d
|
2438 2439 |
seq_printf(f, "%Ld EOF ", fl->fl_start); |
1da177e4c
|
2440 |
else |
7f8ada98d
|
2441 2442 |
seq_printf(f, "%Ld %Ld ", fl->fl_start, fl->fl_end); |
1da177e4c
|
2443 |
} else { |
5315c26a6
|
2444 2445 |
seq_puts(f, "0 EOF "); |
1da177e4c
|
2446 2447 |
} } |
7f8ada98d
|
2448 |
static int locks_show(struct seq_file *f, void *v) |
1da177e4c
|
2449 |
{ |
7012b02a2
|
2450 |
struct locks_iterator *iter = f->private; |
7f8ada98d
|
2451 |
struct file_lock *fl, *bfl; |
1da177e4c
|
2452 |
|
139ca04ee
|
2453 |
fl = hlist_entry(v, struct file_lock, fl_link); |
1da177e4c
|
2454 |
|
7012b02a2
|
2455 |
lock_get_status(f, fl, iter->li_pos, ""); |
1da177e4c
|
2456 |
|
7f8ada98d
|
2457 |
list_for_each_entry(bfl, &fl->fl_block, fl_block) |
7012b02a2
|
2458 |
lock_get_status(f, bfl, iter->li_pos, " ->"); |
094f28252
|
2459 |
|
7f8ada98d
|
2460 2461 |
return 0; } |
1da177e4c
|
2462 |
|
7f8ada98d
|
2463 |
static void *locks_start(struct seq_file *f, loff_t *pos) |
b03dfdec0
|
2464 |
__acquires(&blocked_lock_lock) |
7f8ada98d
|
2465 |
{ |
7012b02a2
|
2466 |
struct locks_iterator *iter = f->private; |
99dc82925
|
2467 |
|
7012b02a2
|
2468 2469 |
iter->li_pos = *pos + 1; lg_global_lock(&file_lock_lglock); |
7b2296afb
|
2470 |
spin_lock(&blocked_lock_lock); |
7012b02a2
|
2471 |
return seq_hlist_start_percpu(&file_lock_list, &iter->li_cpu, *pos); |
7f8ada98d
|
2472 |
} |
1da177e4c
|
2473 |
|
7f8ada98d
|
2474 2475 |
static void *locks_next(struct seq_file *f, void *v, loff_t *pos) { |
7012b02a2
|
2476 2477 2478 2479 |
struct locks_iterator *iter = f->private; ++iter->li_pos; return seq_hlist_next_percpu(v, &file_lock_list, &iter->li_cpu, pos); |
7f8ada98d
|
2480 |
} |
1da177e4c
|
2481 |
|
7f8ada98d
|
2482 |
static void locks_stop(struct seq_file *f, void *v) |
b03dfdec0
|
2483 |
__releases(&blocked_lock_lock) |
7f8ada98d
|
2484 |
{ |
7b2296afb
|
2485 |
spin_unlock(&blocked_lock_lock); |
7012b02a2
|
2486 |
lg_global_unlock(&file_lock_lglock); |
1da177e4c
|
2487 |
} |
d8ba7a363
|
2488 |
static const struct seq_operations locks_seq_operations = { |
7f8ada98d
|
2489 2490 2491 2492 2493 |
.start = locks_start, .next = locks_next, .stop = locks_stop, .show = locks_show, }; |
d8ba7a363
|
2494 2495 2496 |
static int locks_open(struct inode *inode, struct file *filp) { |
7012b02a2
|
2497 2498 |
return seq_open_private(filp, &locks_seq_operations, sizeof(struct locks_iterator)); |
d8ba7a363
|
2499 2500 2501 2502 2503 2504 |
} static const struct file_operations proc_locks_operations = { .open = locks_open, .read = seq_read, .llseek = seq_lseek, |
99dc82925
|
2505 |
.release = seq_release_private, |
d8ba7a363
|
2506 2507 2508 2509 2510 2511 2512 2513 |
}; static int __init proc_locks_init(void) { proc_create("locks", 0, NULL, &proc_locks_operations); return 0; } module_init(proc_locks_init); |
7f8ada98d
|
2514 |
#endif |
1da177e4c
|
2515 2516 |
static int __init filelock_init(void) { |
7012b02a2
|
2517 |
int i; |
1da177e4c
|
2518 |
filelock_cache = kmem_cache_create("file_lock_cache", |
ee19cc406
|
2519 |
sizeof(struct file_lock), 0, SLAB_PANIC, NULL); |
7012b02a2
|
2520 2521 2522 2523 |
lg_lock_init(&file_lock_lglock, "file_lock_lglock"); for_each_possible_cpu(i) INIT_HLIST_HEAD(per_cpu_ptr(&file_lock_list, i)); |
1da177e4c
|
2524 2525 2526 2527 |
return 0; } core_initcall(filelock_init); |