aboutsummaryrefslogtreecommitdiffstats
path: root/kernel/futex.c
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 18:20:36 -0400
committerLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 18:20:36 -0400
commit1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 (patch)
tree0bba044c4ce775e45a88a51686b5d9f90697ea9d /kernel/futex.c
Linux-2.6.12-rc2v2.6.12-rc2
Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
Diffstat (limited to 'kernel/futex.c')
-rw-r--r--kernel/futex.c798
1 files changed, 798 insertions, 0 deletions
diff --git a/kernel/futex.c b/kernel/futex.c
new file mode 100644
index 000000000000..7b54a672d0ad
--- /dev/null
+++ b/kernel/futex.c
@@ -0,0 +1,798 @@
1/*
2 * Fast Userspace Mutexes (which I call "Futexes!").
3 * (C) Rusty Russell, IBM 2002
4 *
5 * Generalized futexes, futex requeueing, misc fixes by Ingo Molnar
6 * (C) Copyright 2003 Red Hat Inc, All Rights Reserved
7 *
8 * Removed page pinning, fix privately mapped COW pages and other cleanups
9 * (C) Copyright 2003, 2004 Jamie Lokier
10 *
11 * Thanks to Ben LaHaise for yelling "hashed waitqueues" loudly
12 * enough at me, Linus for the original (flawed) idea, Matthew
13 * Kirkwood for proof-of-concept implementation.
14 *
15 * "The futexes are also cursed."
16 * "But they come in a choice of three flavours!"
17 *
18 * This program is free software; you can redistribute it and/or modify
19 * it under the terms of the GNU General Public License as published by
20 * the Free Software Foundation; either version 2 of the License, or
21 * (at your option) any later version.
22 *
23 * This program is distributed in the hope that it will be useful,
24 * but WITHOUT ANY WARRANTY; without even the implied warranty of
25 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
26 * GNU General Public License for more details.
27 *
28 * You should have received a copy of the GNU General Public License
29 * along with this program; if not, write to the Free Software
30 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
31 */
32#include <linux/slab.h>
33#include <linux/poll.h>
34#include <linux/fs.h>
35#include <linux/file.h>
36#include <linux/jhash.h>
37#include <linux/init.h>
38#include <linux/futex.h>
39#include <linux/mount.h>
40#include <linux/pagemap.h>
41#include <linux/syscalls.h>
42
43#define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8)
44
45/*
46 * Futexes are matched on equal values of this key.
47 * The key type depends on whether it's a shared or private mapping.
48 * Don't rearrange members without looking at hash_futex().
49 *
50 * offset is aligned to a multiple of sizeof(u32) (== 4) by definition.
51 * We set bit 0 to indicate if it's an inode-based key.
52 */
53union futex_key {
54 struct {
55 unsigned long pgoff;
56 struct inode *inode;
57 int offset;
58 } shared;
59 struct {
60 unsigned long uaddr;
61 struct mm_struct *mm;
62 int offset;
63 } private;
64 struct {
65 unsigned long word;
66 void *ptr;
67 int offset;
68 } both;
69};
70
71/*
72 * We use this hashed waitqueue instead of a normal wait_queue_t, so
73 * we can wake only the relevant ones (hashed queues may be shared).
74 *
75 * A futex_q has a woken state, just like tasks have TASK_RUNNING.
76 * It is considered woken when list_empty(&q->list) || q->lock_ptr == 0.
77 * The order of wakup is always to make the first condition true, then
78 * wake up q->waiters, then make the second condition true.
79 */
80struct futex_q {
81 struct list_head list;
82 wait_queue_head_t waiters;
83
84 /* Which hash list lock to use. */
85 spinlock_t *lock_ptr;
86
87 /* Key which the futex is hashed on. */
88 union futex_key key;
89
90 /* For fd, sigio sent using these. */
91 int fd;
92 struct file *filp;
93};
94
95/*
96 * Split the global futex_lock into every hash list lock.
97 */
98struct futex_hash_bucket {
99 spinlock_t lock;
100 struct list_head chain;
101};
102
103static struct futex_hash_bucket futex_queues[1<<FUTEX_HASHBITS];
104
105/* Futex-fs vfsmount entry: */
106static struct vfsmount *futex_mnt;
107
108/*
109 * We hash on the keys returned from get_futex_key (see below).
110 */
111static struct futex_hash_bucket *hash_futex(union futex_key *key)
112{
113 u32 hash = jhash2((u32*)&key->both.word,
114 (sizeof(key->both.word)+sizeof(key->both.ptr))/4,
115 key->both.offset);
116 return &futex_queues[hash & ((1 << FUTEX_HASHBITS)-1)];
117}
118
119/*
120 * Return 1 if two futex_keys are equal, 0 otherwise.
121 */
122static inline int match_futex(union futex_key *key1, union futex_key *key2)
123{
124 return (key1->both.word == key2->both.word
125 && key1->both.ptr == key2->both.ptr
126 && key1->both.offset == key2->both.offset);
127}
128
129/*
130 * Get parameters which are the keys for a futex.
131 *
132 * For shared mappings, it's (page->index, vma->vm_file->f_dentry->d_inode,
133 * offset_within_page). For private mappings, it's (uaddr, current->mm).
134 * We can usually work out the index without swapping in the page.
135 *
136 * Returns: 0, or negative error code.
137 * The key words are stored in *key on success.
138 *
139 * Should be called with &current->mm->mmap_sem but NOT any spinlocks.
140 */
141static int get_futex_key(unsigned long uaddr, union futex_key *key)
142{
143 struct mm_struct *mm = current->mm;
144 struct vm_area_struct *vma;
145 struct page *page;
146 int err;
147
148 /*
149 * The futex address must be "naturally" aligned.
150 */
151 key->both.offset = uaddr % PAGE_SIZE;
152 if (unlikely((key->both.offset % sizeof(u32)) != 0))
153 return -EINVAL;
154 uaddr -= key->both.offset;
155
156 /*
157 * The futex is hashed differently depending on whether
158 * it's in a shared or private mapping. So check vma first.
159 */
160 vma = find_extend_vma(mm, uaddr);
161 if (unlikely(!vma))
162 return -EFAULT;
163
164 /*
165 * Permissions.
166 */
167 if (unlikely((vma->vm_flags & (VM_IO|VM_READ)) != VM_READ))
168 return (vma->vm_flags & VM_IO) ? -EPERM : -EACCES;
169
170 /*
171 * Private mappings are handled in a simple way.
172 *
173 * NOTE: When userspace waits on a MAP_SHARED mapping, even if
174 * it's a read-only handle, it's expected that futexes attach to
175 * the object not the particular process. Therefore we use
176 * VM_MAYSHARE here, not VM_SHARED which is restricted to shared
177 * mappings of _writable_ handles.
178 */
179 if (likely(!(vma->vm_flags & VM_MAYSHARE))) {
180 key->private.mm = mm;
181 key->private.uaddr = uaddr;
182 return 0;
183 }
184
185 /*
186 * Linear file mappings are also simple.
187 */
188 key->shared.inode = vma->vm_file->f_dentry->d_inode;
189 key->both.offset++; /* Bit 0 of offset indicates inode-based key. */
190 if (likely(!(vma->vm_flags & VM_NONLINEAR))) {
191 key->shared.pgoff = (((uaddr - vma->vm_start) >> PAGE_SHIFT)
192 + vma->vm_pgoff);
193 return 0;
194 }
195
196 /*
197 * We could walk the page table to read the non-linear
198 * pte, and get the page index without fetching the page
199 * from swap. But that's a lot of code to duplicate here
200 * for a rare case, so we simply fetch the page.
201 */
202
203 /*
204 * Do a quick atomic lookup first - this is the fastpath.
205 */
206 spin_lock(&current->mm->page_table_lock);
207 page = follow_page(mm, uaddr, 0);
208 if (likely(page != NULL)) {
209 key->shared.pgoff =
210 page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
211 spin_unlock(&current->mm->page_table_lock);
212 return 0;
213 }
214 spin_unlock(&current->mm->page_table_lock);
215
216 /*
217 * Do it the general way.
218 */
219 err = get_user_pages(current, mm, uaddr, 1, 0, 0, &page, NULL);
220 if (err >= 0) {
221 key->shared.pgoff =
222 page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
223 put_page(page);
224 return 0;
225 }
226 return err;
227}
228
229/*
230 * Take a reference to the resource addressed by a key.
231 * Can be called while holding spinlocks.
232 *
233 * NOTE: mmap_sem MUST be held between get_futex_key() and calling this
234 * function, if it is called at all. mmap_sem keeps key->shared.inode valid.
235 */
236static inline void get_key_refs(union futex_key *key)
237{
238 if (key->both.ptr != 0) {
239 if (key->both.offset & 1)
240 atomic_inc(&key->shared.inode->i_count);
241 else
242 atomic_inc(&key->private.mm->mm_count);
243 }
244}
245
246/*
247 * Drop a reference to the resource addressed by a key.
248 * The hash bucket spinlock must not be held.
249 */
250static void drop_key_refs(union futex_key *key)
251{
252 if (key->both.ptr != 0) {
253 if (key->both.offset & 1)
254 iput(key->shared.inode);
255 else
256 mmdrop(key->private.mm);
257 }
258}
259
260static inline int get_futex_value_locked(int *dest, int __user *from)
261{
262 int ret;
263
264 inc_preempt_count();
265 ret = __copy_from_user_inatomic(dest, from, sizeof(int));
266 dec_preempt_count();
267
268 return ret ? -EFAULT : 0;
269}
270
271/*
272 * The hash bucket lock must be held when this is called.
273 * Afterwards, the futex_q must not be accessed.
274 */
275static void wake_futex(struct futex_q *q)
276{
277 list_del_init(&q->list);
278 if (q->filp)
279 send_sigio(&q->filp->f_owner, q->fd, POLL_IN);
280 /*
281 * The lock in wake_up_all() is a crucial memory barrier after the
282 * list_del_init() and also before assigning to q->lock_ptr.
283 */
284 wake_up_all(&q->waiters);
285 /*
286 * The waiting task can free the futex_q as soon as this is written,
287 * without taking any locks. This must come last.
288 */
289 q->lock_ptr = NULL;
290}
291
292/*
293 * Wake up all waiters hashed on the physical page that is mapped
294 * to this virtual address:
295 */
296static int futex_wake(unsigned long uaddr, int nr_wake)
297{
298 union futex_key key;
299 struct futex_hash_bucket *bh;
300 struct list_head *head;
301 struct futex_q *this, *next;
302 int ret;
303
304 down_read(&current->mm->mmap_sem);
305
306 ret = get_futex_key(uaddr, &key);
307 if (unlikely(ret != 0))
308 goto out;
309
310 bh = hash_futex(&key);
311 spin_lock(&bh->lock);
312 head = &bh->chain;
313
314 list_for_each_entry_safe(this, next, head, list) {
315 if (match_futex (&this->key, &key)) {
316 wake_futex(this);
317 if (++ret >= nr_wake)
318 break;
319 }
320 }
321
322 spin_unlock(&bh->lock);
323out:
324 up_read(&current->mm->mmap_sem);
325 return ret;
326}
327
328/*
329 * Requeue all waiters hashed on one physical page to another
330 * physical page.
331 */
332static int futex_requeue(unsigned long uaddr1, unsigned long uaddr2,
333 int nr_wake, int nr_requeue, int *valp)
334{
335 union futex_key key1, key2;
336 struct futex_hash_bucket *bh1, *bh2;
337 struct list_head *head1;
338 struct futex_q *this, *next;
339 int ret, drop_count = 0;
340
341 retry:
342 down_read(&current->mm->mmap_sem);
343
344 ret = get_futex_key(uaddr1, &key1);
345 if (unlikely(ret != 0))
346 goto out;
347 ret = get_futex_key(uaddr2, &key2);
348 if (unlikely(ret != 0))
349 goto out;
350
351 bh1 = hash_futex(&key1);
352 bh2 = hash_futex(&key2);
353
354 if (bh1 < bh2)
355 spin_lock(&bh1->lock);
356 spin_lock(&bh2->lock);
357 if (bh1 > bh2)
358 spin_lock(&bh1->lock);
359
360 if (likely(valp != NULL)) {
361 int curval;
362
363 ret = get_futex_value_locked(&curval, (int __user *)uaddr1);
364
365 if (unlikely(ret)) {
366 spin_unlock(&bh1->lock);
367 if (bh1 != bh2)
368 spin_unlock(&bh2->lock);
369
370 /* If we would have faulted, release mmap_sem, fault
371 * it in and start all over again.
372 */
373 up_read(&current->mm->mmap_sem);
374
375 ret = get_user(curval, (int __user *)uaddr1);
376
377 if (!ret)
378 goto retry;
379
380 return ret;
381 }
382 if (curval != *valp) {
383 ret = -EAGAIN;
384 goto out_unlock;
385 }
386 }
387
388 head1 = &bh1->chain;
389 list_for_each_entry_safe(this, next, head1, list) {
390 if (!match_futex (&this->key, &key1))
391 continue;
392 if (++ret <= nr_wake) {
393 wake_futex(this);
394 } else {
395 list_move_tail(&this->list, &bh2->chain);
396 this->lock_ptr = &bh2->lock;
397 this->key = key2;
398 get_key_refs(&key2);
399 drop_count++;
400
401 if (ret - nr_wake >= nr_requeue)
402 break;
403 /* Make sure to stop if key1 == key2 */
404 if (head1 == &bh2->chain && head1 != &next->list)
405 head1 = &this->list;
406 }
407 }
408
409out_unlock:
410 spin_unlock(&bh1->lock);
411 if (bh1 != bh2)
412 spin_unlock(&bh2->lock);
413
414 /* drop_key_refs() must be called outside the spinlocks. */
415 while (--drop_count >= 0)
416 drop_key_refs(&key1);
417
418out:
419 up_read(&current->mm->mmap_sem);
420 return ret;
421}
422
423/* The key must be already stored in q->key. */
424static inline struct futex_hash_bucket *
425queue_lock(struct futex_q *q, int fd, struct file *filp)
426{
427 struct futex_hash_bucket *bh;
428
429 q->fd = fd;
430 q->filp = filp;
431
432 init_waitqueue_head(&q->waiters);
433
434 get_key_refs(&q->key);
435 bh = hash_futex(&q->key);
436 q->lock_ptr = &bh->lock;
437
438 spin_lock(&bh->lock);
439 return bh;
440}
441
442static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *bh)
443{
444 list_add_tail(&q->list, &bh->chain);
445 spin_unlock(&bh->lock);
446}
447
448static inline void
449queue_unlock(struct futex_q *q, struct futex_hash_bucket *bh)
450{
451 spin_unlock(&bh->lock);
452 drop_key_refs(&q->key);
453}
454
455/*
456 * queue_me and unqueue_me must be called as a pair, each
457 * exactly once. They are called with the hashed spinlock held.
458 */
459
460/* The key must be already stored in q->key. */
461static void queue_me(struct futex_q *q, int fd, struct file *filp)
462{
463 struct futex_hash_bucket *bh;
464 bh = queue_lock(q, fd, filp);
465 __queue_me(q, bh);
466}
467
468/* Return 1 if we were still queued (ie. 0 means we were woken) */
469static int unqueue_me(struct futex_q *q)
470{
471 int ret = 0;
472 spinlock_t *lock_ptr;
473
474 /* In the common case we don't take the spinlock, which is nice. */
475 retry:
476 lock_ptr = q->lock_ptr;
477 if (lock_ptr != 0) {
478 spin_lock(lock_ptr);
479 /*
480 * q->lock_ptr can change between reading it and
481 * spin_lock(), causing us to take the wrong lock. This
482 * corrects the race condition.
483 *
484 * Reasoning goes like this: if we have the wrong lock,
485 * q->lock_ptr must have changed (maybe several times)
486 * between reading it and the spin_lock(). It can
487 * change again after the spin_lock() but only if it was
488 * already changed before the spin_lock(). It cannot,
489 * however, change back to the original value. Therefore
490 * we can detect whether we acquired the correct lock.
491 */
492 if (unlikely(lock_ptr != q->lock_ptr)) {
493 spin_unlock(lock_ptr);
494 goto retry;
495 }
496 WARN_ON(list_empty(&q->list));
497 list_del(&q->list);
498 spin_unlock(lock_ptr);
499 ret = 1;
500 }
501
502 drop_key_refs(&q->key);
503 return ret;
504}
505
506static int futex_wait(unsigned long uaddr, int val, unsigned long time)
507{
508 DECLARE_WAITQUEUE(wait, current);
509 int ret, curval;
510 struct futex_q q;
511 struct futex_hash_bucket *bh;
512
513 retry:
514 down_read(&current->mm->mmap_sem);
515
516 ret = get_futex_key(uaddr, &q.key);
517 if (unlikely(ret != 0))
518 goto out_release_sem;
519
520 bh = queue_lock(&q, -1, NULL);
521
522 /*
523 * Access the page AFTER the futex is queued.
524 * Order is important:
525 *
526 * Userspace waiter: val = var; if (cond(val)) futex_wait(&var, val);
527 * Userspace waker: if (cond(var)) { var = new; futex_wake(&var); }
528 *
529 * The basic logical guarantee of a futex is that it blocks ONLY
530 * if cond(var) is known to be true at the time of blocking, for
531 * any cond. If we queued after testing *uaddr, that would open
532 * a race condition where we could block indefinitely with
533 * cond(var) false, which would violate the guarantee.
534 *
535 * A consequence is that futex_wait() can return zero and absorb
536 * a wakeup when *uaddr != val on entry to the syscall. This is
537 * rare, but normal.
538 *
539 * We hold the mmap semaphore, so the mapping cannot have changed
540 * since we looked it up in get_futex_key.
541 */
542
543 ret = get_futex_value_locked(&curval, (int __user *)uaddr);
544
545 if (unlikely(ret)) {
546 queue_unlock(&q, bh);
547
548 /* If we would have faulted, release mmap_sem, fault it in and
549 * start all over again.
550 */
551 up_read(&current->mm->mmap_sem);
552
553 ret = get_user(curval, (int __user *)uaddr);
554
555 if (!ret)
556 goto retry;
557 return ret;
558 }
559 if (curval != val) {
560 ret = -EWOULDBLOCK;
561 queue_unlock(&q, bh);
562 goto out_release_sem;
563 }
564
565 /* Only actually queue if *uaddr contained val. */
566 __queue_me(&q, bh);
567
568 /*
569 * Now the futex is queued and we have checked the data, we
570 * don't want to hold mmap_sem while we sleep.
571 */
572 up_read(&current->mm->mmap_sem);
573
574 /*
575 * There might have been scheduling since the queue_me(), as we
576 * cannot hold a spinlock across the get_user() in case it
577 * faults, and we cannot just set TASK_INTERRUPTIBLE state when
578 * queueing ourselves into the futex hash. This code thus has to
579 * rely on the futex_wake() code removing us from hash when it
580 * wakes us up.
581 */
582
583 /* add_wait_queue is the barrier after __set_current_state. */
584 __set_current_state(TASK_INTERRUPTIBLE);
585 add_wait_queue(&q.waiters, &wait);
586 /*
587 * !list_empty() is safe here without any lock.
588 * q.lock_ptr != 0 is not safe, because of ordering against wakeup.
589 */
590 if (likely(!list_empty(&q.list)))
591 time = schedule_timeout(time);
592 __set_current_state(TASK_RUNNING);
593
594 /*
595 * NOTE: we don't remove ourselves from the waitqueue because
596 * we are the only user of it.
597 */
598
599 /* If we were woken (and unqueued), we succeeded, whatever. */
600 if (!unqueue_me(&q))
601 return 0;
602 if (time == 0)
603 return -ETIMEDOUT;
604 /* We expect signal_pending(current), but another thread may
605 * have handled it for us already. */
606 return -EINTR;
607
608 out_release_sem:
609 up_read(&current->mm->mmap_sem);
610 return ret;
611}
612
613static int futex_close(struct inode *inode, struct file *filp)
614{
615 struct futex_q *q = filp->private_data;
616
617 unqueue_me(q);
618 kfree(q);
619 return 0;
620}
621
622/* This is one-shot: once it's gone off you need a new fd */
623static unsigned int futex_poll(struct file *filp,
624 struct poll_table_struct *wait)
625{
626 struct futex_q *q = filp->private_data;
627 int ret = 0;
628
629 poll_wait(filp, &q->waiters, wait);
630
631 /*
632 * list_empty() is safe here without any lock.
633 * q->lock_ptr != 0 is not safe, because of ordering against wakeup.
634 */
635 if (list_empty(&q->list))
636 ret = POLLIN | POLLRDNORM;
637
638 return ret;
639}
640
641static struct file_operations futex_fops = {
642 .release = futex_close,
643 .poll = futex_poll,
644};
645
646/*
647 * Signal allows caller to avoid the race which would occur if they
648 * set the sigio stuff up afterwards.
649 */
650static int futex_fd(unsigned long uaddr, int signal)
651{
652 struct futex_q *q;
653 struct file *filp;
654 int ret, err;
655
656 ret = -EINVAL;
657 if (signal < 0 || signal > _NSIG)
658 goto out;
659
660 ret = get_unused_fd();
661 if (ret < 0)
662 goto out;
663 filp = get_empty_filp();
664 if (!filp) {
665 put_unused_fd(ret);
666 ret = -ENFILE;
667 goto out;
668 }
669 filp->f_op = &futex_fops;
670 filp->f_vfsmnt = mntget(futex_mnt);
671 filp->f_dentry = dget(futex_mnt->mnt_root);
672 filp->f_mapping = filp->f_dentry->d_inode->i_mapping;
673
674 if (signal) {
675 int err;
676 err = f_setown(filp, current->pid, 1);
677 if (err < 0) {
678 put_unused_fd(ret);
679 put_filp(filp);
680 ret = err;
681 goto out;
682 }
683 filp->f_owner.signum = signal;
684 }
685
686 q = kmalloc(sizeof(*q), GFP_KERNEL);
687 if (!q) {
688 put_unused_fd(ret);
689 put_filp(filp);
690 ret = -ENOMEM;
691 goto out;
692 }
693
694 down_read(&current->mm->mmap_sem);
695 err = get_futex_key(uaddr, &q->key);
696
697 if (unlikely(err != 0)) {
698 up_read(&current->mm->mmap_sem);
699 put_unused_fd(ret);
700 put_filp(filp);
701 kfree(q);
702 return err;
703 }
704
705 /*
706 * queue_me() must be called before releasing mmap_sem, because
707 * key->shared.inode needs to be referenced while holding it.
708 */
709 filp->private_data = q;
710
711 queue_me(q, ret, filp);
712 up_read(&current->mm->mmap_sem);
713
714 /* Now we map fd to filp, so userspace can access it */
715 fd_install(ret, filp);
716out:
717 return ret;
718}
719
720long do_futex(unsigned long uaddr, int op, int val, unsigned long timeout,
721 unsigned long uaddr2, int val2, int val3)
722{
723 int ret;
724
725 switch (op) {
726 case FUTEX_WAIT:
727 ret = futex_wait(uaddr, val, timeout);
728 break;
729 case FUTEX_WAKE:
730 ret = futex_wake(uaddr, val);
731 break;
732 case FUTEX_FD:
733 /* non-zero val means F_SETOWN(getpid()) & F_SETSIG(val) */
734 ret = futex_fd(uaddr, val);
735 break;
736 case FUTEX_REQUEUE:
737 ret = futex_requeue(uaddr, uaddr2, val, val2, NULL);
738 break;
739 case FUTEX_CMP_REQUEUE:
740 ret = futex_requeue(uaddr, uaddr2, val, val2, &val3);
741 break;
742 default:
743 ret = -ENOSYS;
744 }
745 return ret;
746}
747
748
749asmlinkage long sys_futex(u32 __user *uaddr, int op, int val,
750 struct timespec __user *utime, u32 __user *uaddr2,
751 int val3)
752{
753 struct timespec t;
754 unsigned long timeout = MAX_SCHEDULE_TIMEOUT;
755 int val2 = 0;
756
757 if ((op == FUTEX_WAIT) && utime) {
758 if (copy_from_user(&t, utime, sizeof(t)) != 0)
759 return -EFAULT;
760 timeout = timespec_to_jiffies(&t) + 1;
761 }
762 /*
763 * requeue parameter in 'utime' if op == FUTEX_REQUEUE.
764 */
765 if (op >= FUTEX_REQUEUE)
766 val2 = (int) (unsigned long) utime;
767
768 return do_futex((unsigned long)uaddr, op, val, timeout,
769 (unsigned long)uaddr2, val2, val3);
770}
771
772static struct super_block *
773futexfs_get_sb(struct file_system_type *fs_type,
774 int flags, const char *dev_name, void *data)
775{
776 return get_sb_pseudo(fs_type, "futex", NULL, 0xBAD1DEA);
777}
778
779static struct file_system_type futex_fs_type = {
780 .name = "futexfs",
781 .get_sb = futexfs_get_sb,
782 .kill_sb = kill_anon_super,
783};
784
785static int __init init(void)
786{
787 unsigned int i;
788
789 register_filesystem(&futex_fs_type);
790 futex_mnt = kern_mount(&futex_fs_type);
791
792 for (i = 0; i < ARRAY_SIZE(futex_queues); i++) {
793 INIT_LIST_HEAD(&futex_queues[i].chain);
794 spin_lock_init(&futex_queues[i].lock);
795 }
796 return 0;
797}
798__initcall(init);