blob: 7d1136e97c142d198b897dab1846acd99f1f655f [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001/*
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 *
Ingo Molnar0771dfe2006-03-27 01:16:22 -080011 * Robust futex support started by Ingo Molnar
12 * (C) Copyright 2006 Red Hat Inc, All Rights Reserved
13 * Thanks to Thomas Gleixner for suggestions, analysis and fixes.
14 *
Ingo Molnarc87e2832006-06-27 02:54:58 -070015 * PI-futex support started by Ingo Molnar and Thomas Gleixner
16 * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
17 * Copyright (C) 2006 Timesys Corp., Thomas Gleixner <tglx@timesys.com>
18 *
Eric Dumazet34f01cc2007-05-09 02:35:04 -070019 * PRIVATE futexes by Eric Dumazet
20 * Copyright (C) 2007 Eric Dumazet <dada1@cosmosbay.com>
21 *
Linus Torvalds1da177e2005-04-16 15:20:36 -070022 * Thanks to Ben LaHaise for yelling "hashed waitqueues" loudly
23 * enough at me, Linus for the original (flawed) idea, Matthew
24 * Kirkwood for proof-of-concept implementation.
25 *
26 * "The futexes are also cursed."
27 * "But they come in a choice of three flavours!"
28 *
29 * This program is free software; you can redistribute it and/or modify
30 * it under the terms of the GNU General Public License as published by
31 * the Free Software Foundation; either version 2 of the License, or
32 * (at your option) any later version.
33 *
34 * This program is distributed in the hope that it will be useful,
35 * but WITHOUT ANY WARRANTY; without even the implied warranty of
36 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
37 * GNU General Public License for more details.
38 *
39 * You should have received a copy of the GNU General Public License
40 * along with this program; if not, write to the Free Software
41 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
42 */
43#include <linux/slab.h>
44#include <linux/poll.h>
45#include <linux/fs.h>
46#include <linux/file.h>
47#include <linux/jhash.h>
48#include <linux/init.h>
49#include <linux/futex.h>
50#include <linux/mount.h>
51#include <linux/pagemap.h>
52#include <linux/syscalls.h>
Jesper Juhl7ed20e12005-05-01 08:59:14 -070053#include <linux/signal.h>
Rusty Russell9adef582007-05-08 00:26:42 -070054#include <linux/module.h>
Andrey Mirkinfd5eea42007-10-16 23:30:13 -070055#include <linux/magic.h>
Pavel Emelyanovb4888932007-10-18 23:40:14 -070056#include <linux/pid.h>
57#include <linux/nsproxy.h>
58
Jakub Jelinek4732efb2005-09-06 15:16:25 -070059#include <asm/futex.h>
Linus Torvalds1da177e2005-04-16 15:20:36 -070060
Ingo Molnarc87e2832006-06-27 02:54:58 -070061#include "rtmutex_common.h"
62
Thomas Gleixnera0c1e902008-02-23 15:23:57 -080063int __read_mostly futex_cmpxchg_enabled;
64
Linus Torvalds1da177e2005-04-16 15:20:36 -070065#define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8)
66
67/*
Ingo Molnarc87e2832006-06-27 02:54:58 -070068 * Priority Inheritance state:
69 */
70struct futex_pi_state {
71 /*
72 * list of 'owned' pi_state instances - these have to be
73 * cleaned up in do_exit() if the task exits prematurely:
74 */
75 struct list_head list;
76
77 /*
78 * The PI object:
79 */
80 struct rt_mutex pi_mutex;
81
82 struct task_struct *owner;
83 atomic_t refcount;
84
85 union futex_key key;
86};
87
88/*
Linus Torvalds1da177e2005-04-16 15:20:36 -070089 * We use this hashed waitqueue instead of a normal wait_queue_t, so
90 * we can wake only the relevant ones (hashed queues may be shared).
91 *
92 * A futex_q has a woken state, just like tasks have TASK_RUNNING.
Pierre Peifferec92d082007-05-09 02:35:00 -070093 * It is considered woken when plist_node_empty(&q->list) || q->lock_ptr == 0.
Linus Torvalds1da177e2005-04-16 15:20:36 -070094 * The order of wakup is always to make the first condition true, then
95 * wake up q->waiters, then make the second condition true.
96 */
97struct futex_q {
Pierre Peifferec92d082007-05-09 02:35:00 -070098 struct plist_node list;
Linus Torvalds1da177e2005-04-16 15:20:36 -070099 wait_queue_head_t waiters;
100
Ingo Molnare2970f22006-06-27 02:54:47 -0700101 /* Which hash list lock to use: */
Linus Torvalds1da177e2005-04-16 15:20:36 -0700102 spinlock_t *lock_ptr;
103
Ingo Molnare2970f22006-06-27 02:54:47 -0700104 /* Key which the futex is hashed on: */
Linus Torvalds1da177e2005-04-16 15:20:36 -0700105 union futex_key key;
106
Ingo Molnarc87e2832006-06-27 02:54:58 -0700107 /* Optional priority inheritance state: */
108 struct futex_pi_state *pi_state;
109 struct task_struct *task;
Thomas Gleixnercd689982008-02-01 17:45:14 +0100110
111 /* Bitset for the optional bitmasked wakeup */
112 u32 bitset;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700113};
114
115/*
116 * Split the global futex_lock into every hash list lock.
117 */
118struct futex_hash_bucket {
Pierre Peifferec92d082007-05-09 02:35:00 -0700119 spinlock_t lock;
120 struct plist_head chain;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700121};
122
123static struct futex_hash_bucket futex_queues[1<<FUTEX_HASHBITS];
124
Linus Torvalds1da177e2005-04-16 15:20:36 -0700125/*
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700126 * Take mm->mmap_sem, when futex is shared
127 */
128static inline void futex_lock_mm(struct rw_semaphore *fshared)
129{
130 if (fshared)
131 down_read(fshared);
132}
133
134/*
135 * Release mm->mmap_sem, when the futex is shared
136 */
137static inline void futex_unlock_mm(struct rw_semaphore *fshared)
138{
139 if (fshared)
140 up_read(fshared);
141}
142
143/*
Linus Torvalds1da177e2005-04-16 15:20:36 -0700144 * We hash on the keys returned from get_futex_key (see below).
145 */
146static struct futex_hash_bucket *hash_futex(union futex_key *key)
147{
148 u32 hash = jhash2((u32*)&key->both.word,
149 (sizeof(key->both.word)+sizeof(key->both.ptr))/4,
150 key->both.offset);
151 return &futex_queues[hash & ((1 << FUTEX_HASHBITS)-1)];
152}
153
154/*
155 * Return 1 if two futex_keys are equal, 0 otherwise.
156 */
157static inline int match_futex(union futex_key *key1, union futex_key *key2)
158{
159 return (key1->both.word == key2->both.word
160 && key1->both.ptr == key2->both.ptr
161 && key1->both.offset == key2->both.offset);
162}
163
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700164/**
165 * get_futex_key - Get parameters which are the keys for a futex.
166 * @uaddr: virtual address of the futex
167 * @shared: NULL for a PROCESS_PRIVATE futex,
168 * &current->mm->mmap_sem for a PROCESS_SHARED futex
169 * @key: address where result is stored.
170 *
171 * Returns a negative error code or 0
172 * The key words are stored in *key on success.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700173 *
Josef "Jeff" Sipekf3a43f32006-12-08 02:36:43 -0800174 * For shared mappings, it's (page->index, vma->vm_file->f_path.dentry->d_inode,
Linus Torvalds1da177e2005-04-16 15:20:36 -0700175 * offset_within_page). For private mappings, it's (uaddr, current->mm).
176 * We can usually work out the index without swapping in the page.
177 *
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700178 * fshared is NULL for PROCESS_PRIVATE futexes
179 * For other futexes, it points to &current->mm->mmap_sem and
180 * caller must have taken the reader lock. but NOT any spinlocks.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700181 */
Adrian Bunkfad23fc2007-11-02 16:43:22 +0100182static int get_futex_key(u32 __user *uaddr, struct rw_semaphore *fshared,
183 union futex_key *key)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700184{
Ingo Molnare2970f22006-06-27 02:54:47 -0700185 unsigned long address = (unsigned long)uaddr;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700186 struct mm_struct *mm = current->mm;
187 struct vm_area_struct *vma;
188 struct page *page;
189 int err;
190
191 /*
192 * The futex address must be "naturally" aligned.
193 */
Ingo Molnare2970f22006-06-27 02:54:47 -0700194 key->both.offset = address % PAGE_SIZE;
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700195 if (unlikely((address % sizeof(u32)) != 0))
Linus Torvalds1da177e2005-04-16 15:20:36 -0700196 return -EINVAL;
Ingo Molnare2970f22006-06-27 02:54:47 -0700197 address -= key->both.offset;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700198
199 /*
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700200 * PROCESS_PRIVATE futexes are fast.
201 * As the mm cannot disappear under us and the 'key' only needs
202 * virtual address, we dont even have to find the underlying vma.
203 * Note : We do have to check 'uaddr' is a valid user address,
204 * but access_ok() should be faster than find_vma()
205 */
206 if (!fshared) {
207 if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))))
208 return -EFAULT;
209 key->private.mm = mm;
210 key->private.address = address;
211 return 0;
212 }
213 /*
Linus Torvalds1da177e2005-04-16 15:20:36 -0700214 * The futex is hashed differently depending on whether
215 * it's in a shared or private mapping. So check vma first.
216 */
Ingo Molnare2970f22006-06-27 02:54:47 -0700217 vma = find_extend_vma(mm, address);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700218 if (unlikely(!vma))
219 return -EFAULT;
220
221 /*
222 * Permissions.
223 */
224 if (unlikely((vma->vm_flags & (VM_IO|VM_READ)) != VM_READ))
225 return (vma->vm_flags & VM_IO) ? -EPERM : -EACCES;
226
227 /*
228 * Private mappings are handled in a simple way.
229 *
230 * NOTE: When userspace waits on a MAP_SHARED mapping, even if
231 * it's a read-only handle, it's expected that futexes attach to
232 * the object not the particular process. Therefore we use
233 * VM_MAYSHARE here, not VM_SHARED which is restricted to shared
234 * mappings of _writable_ handles.
235 */
236 if (likely(!(vma->vm_flags & VM_MAYSHARE))) {
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700237 key->both.offset |= FUT_OFF_MMSHARED; /* reference taken on mm */
Linus Torvalds1da177e2005-04-16 15:20:36 -0700238 key->private.mm = mm;
Ingo Molnare2970f22006-06-27 02:54:47 -0700239 key->private.address = address;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700240 return 0;
241 }
242
243 /*
244 * Linear file mappings are also simple.
245 */
Josef "Jeff" Sipekf3a43f32006-12-08 02:36:43 -0800246 key->shared.inode = vma->vm_file->f_path.dentry->d_inode;
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700247 key->both.offset |= FUT_OFF_INODE; /* inode-based key. */
Linus Torvalds1da177e2005-04-16 15:20:36 -0700248 if (likely(!(vma->vm_flags & VM_NONLINEAR))) {
Ingo Molnare2970f22006-06-27 02:54:47 -0700249 key->shared.pgoff = (((address - vma->vm_start) >> PAGE_SHIFT)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700250 + vma->vm_pgoff);
251 return 0;
252 }
253
254 /*
255 * We could walk the page table to read the non-linear
256 * pte, and get the page index without fetching the page
257 * from swap. But that's a lot of code to duplicate here
258 * for a rare case, so we simply fetch the page.
259 */
Ingo Molnare2970f22006-06-27 02:54:47 -0700260 err = get_user_pages(current, mm, address, 1, 0, 0, &page, NULL);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700261 if (err >= 0) {
262 key->shared.pgoff =
263 page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
264 put_page(page);
265 return 0;
266 }
267 return err;
268}
269
270/*
271 * Take a reference to the resource addressed by a key.
272 * Can be called while holding spinlocks.
273 *
Linus Torvalds1da177e2005-04-16 15:20:36 -0700274 */
Adrian Bunkfad23fc2007-11-02 16:43:22 +0100275static void get_futex_key_refs(union futex_key *key)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700276{
Al Viro9dce07f2008-03-29 03:07:28 +0000277 if (key->both.ptr == NULL)
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700278 return;
279 switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) {
280 case FUT_OFF_INODE:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700281 atomic_inc(&key->shared.inode->i_count);
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700282 break;
283 case FUT_OFF_MMSHARED:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700284 atomic_inc(&key->private.mm->mm_count);
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700285 break;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700286 }
287}
288
289/*
290 * Drop a reference to the resource addressed by a key.
291 * The hash bucket spinlock must not be held.
292 */
Adrian Bunkfad23fc2007-11-02 16:43:22 +0100293static void drop_futex_key_refs(union futex_key *key)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700294{
Stephen Hemmingerc80544d2007-10-18 03:07:05 -0700295 if (!key->both.ptr)
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700296 return;
297 switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) {
298 case FUT_OFF_INODE:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700299 iput(key->shared.inode);
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700300 break;
301 case FUT_OFF_MMSHARED:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700302 mmdrop(key->private.mm);
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700303 break;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700304 }
305}
306
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700307static u32 cmpxchg_futex_value_locked(u32 __user *uaddr, u32 uval, u32 newval)
308{
309 u32 curval;
310
311 pagefault_disable();
312 curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval);
313 pagefault_enable();
314
315 return curval;
316}
317
318static int get_futex_value_locked(u32 *dest, u32 __user *from)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700319{
320 int ret;
321
Peter Zijlstraa8663742006-12-06 20:32:20 -0800322 pagefault_disable();
Ingo Molnare2970f22006-06-27 02:54:47 -0700323 ret = __copy_from_user_inatomic(dest, from, sizeof(u32));
Peter Zijlstraa8663742006-12-06 20:32:20 -0800324 pagefault_enable();
Linus Torvalds1da177e2005-04-16 15:20:36 -0700325
326 return ret ? -EFAULT : 0;
327}
328
329/*
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700330 * Fault handling.
331 * if fshared is non NULL, current->mm->mmap_sem is already held
Ingo Molnarc87e2832006-06-27 02:54:58 -0700332 */
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700333static int futex_handle_fault(unsigned long address,
334 struct rw_semaphore *fshared, int attempt)
Ingo Molnarc87e2832006-06-27 02:54:58 -0700335{
336 struct vm_area_struct * vma;
337 struct mm_struct *mm = current->mm;
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700338 int ret = -EFAULT;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700339
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700340 if (attempt > 2)
341 return ret;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700342
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700343 if (!fshared)
344 down_read(&mm->mmap_sem);
345 vma = find_vma(mm, address);
346 if (vma && address >= vma->vm_start &&
347 (vma->vm_flags & VM_WRITE)) {
Nick Piggin83c54072007-07-19 01:47:05 -0700348 int fault;
349 fault = handle_mm_fault(mm, vma, address, 1);
350 if (unlikely((fault & VM_FAULT_ERROR))) {
351#if 0
352 /* XXX: let's do this when we verify it is OK */
353 if (ret & VM_FAULT_OOM)
354 ret = -ENOMEM;
355#endif
356 } else {
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700357 ret = 0;
Nick Piggin83c54072007-07-19 01:47:05 -0700358 if (fault & VM_FAULT_MAJOR)
359 current->maj_flt++;
360 else
361 current->min_flt++;
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700362 }
Ingo Molnarc87e2832006-06-27 02:54:58 -0700363 }
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700364 if (!fshared)
365 up_read(&mm->mmap_sem);
366 return ret;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700367}
368
369/*
370 * PI code:
371 */
372static int refill_pi_state_cache(void)
373{
374 struct futex_pi_state *pi_state;
375
376 if (likely(current->pi_state_cache))
377 return 0;
378
Burman Yan4668edc2006-12-06 20:38:51 -0800379 pi_state = kzalloc(sizeof(*pi_state), GFP_KERNEL);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700380
381 if (!pi_state)
382 return -ENOMEM;
383
Ingo Molnarc87e2832006-06-27 02:54:58 -0700384 INIT_LIST_HEAD(&pi_state->list);
385 /* pi_mutex gets initialized later */
386 pi_state->owner = NULL;
387 atomic_set(&pi_state->refcount, 1);
388
389 current->pi_state_cache = pi_state;
390
391 return 0;
392}
393
394static struct futex_pi_state * alloc_pi_state(void)
395{
396 struct futex_pi_state *pi_state = current->pi_state_cache;
397
398 WARN_ON(!pi_state);
399 current->pi_state_cache = NULL;
400
401 return pi_state;
402}
403
404static void free_pi_state(struct futex_pi_state *pi_state)
405{
406 if (!atomic_dec_and_test(&pi_state->refcount))
407 return;
408
409 /*
410 * If pi_state->owner is NULL, the owner is most probably dying
411 * and has cleaned up the pi_state already
412 */
413 if (pi_state->owner) {
414 spin_lock_irq(&pi_state->owner->pi_lock);
415 list_del_init(&pi_state->list);
416 spin_unlock_irq(&pi_state->owner->pi_lock);
417
418 rt_mutex_proxy_unlock(&pi_state->pi_mutex, pi_state->owner);
419 }
420
421 if (current->pi_state_cache)
422 kfree(pi_state);
423 else {
424 /*
425 * pi_state->list is already empty.
426 * clear pi_state->owner.
427 * refcount is at 0 - put it back to 1.
428 */
429 pi_state->owner = NULL;
430 atomic_set(&pi_state->refcount, 1);
431 current->pi_state_cache = pi_state;
432 }
433}
434
435/*
436 * Look up the task based on what TID userspace gave us.
437 * We dont trust it.
438 */
439static struct task_struct * futex_find_get_task(pid_t pid)
440{
441 struct task_struct *p;
442
Oleg Nesterovd359b542006-09-29 02:00:55 -0700443 rcu_read_lock();
Pavel Emelyanov228ebcb2007-10-18 23:40:16 -0700444 p = find_task_by_vpid(pid);
Thomas Gleixnera06381f2007-06-23 11:48:40 +0200445 if (!p || ((current->euid != p->euid) && (current->euid != p->uid)))
446 p = ERR_PTR(-ESRCH);
447 else
448 get_task_struct(p);
449
Oleg Nesterovd359b542006-09-29 02:00:55 -0700450 rcu_read_unlock();
Ingo Molnarc87e2832006-06-27 02:54:58 -0700451
452 return p;
453}
454
455/*
456 * This task is holding PI mutexes at exit time => bad.
457 * Kernel cleans up PI-state, but userspace is likely hosed.
458 * (Robust-futex cleanup is separate and might save the day for userspace.)
459 */
460void exit_pi_state_list(struct task_struct *curr)
461{
Ingo Molnarc87e2832006-06-27 02:54:58 -0700462 struct list_head *next, *head = &curr->pi_state_list;
463 struct futex_pi_state *pi_state;
Ingo Molnar627371d2006-07-29 05:16:20 +0200464 struct futex_hash_bucket *hb;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700465 union futex_key key;
466
Thomas Gleixnera0c1e902008-02-23 15:23:57 -0800467 if (!futex_cmpxchg_enabled)
468 return;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700469 /*
470 * We are a ZOMBIE and nobody can enqueue itself on
471 * pi_state_list anymore, but we have to be careful
Ingo Molnar627371d2006-07-29 05:16:20 +0200472 * versus waiters unqueueing themselves:
Ingo Molnarc87e2832006-06-27 02:54:58 -0700473 */
474 spin_lock_irq(&curr->pi_lock);
475 while (!list_empty(head)) {
476
477 next = head->next;
478 pi_state = list_entry(next, struct futex_pi_state, list);
479 key = pi_state->key;
Ingo Molnar627371d2006-07-29 05:16:20 +0200480 hb = hash_futex(&key);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700481 spin_unlock_irq(&curr->pi_lock);
482
Ingo Molnarc87e2832006-06-27 02:54:58 -0700483 spin_lock(&hb->lock);
484
485 spin_lock_irq(&curr->pi_lock);
Ingo Molnar627371d2006-07-29 05:16:20 +0200486 /*
487 * We dropped the pi-lock, so re-check whether this
488 * task still owns the PI-state:
489 */
Ingo Molnarc87e2832006-06-27 02:54:58 -0700490 if (head->next != next) {
491 spin_unlock(&hb->lock);
492 continue;
493 }
494
Ingo Molnarc87e2832006-06-27 02:54:58 -0700495 WARN_ON(pi_state->owner != curr);
Ingo Molnar627371d2006-07-29 05:16:20 +0200496 WARN_ON(list_empty(&pi_state->list));
497 list_del_init(&pi_state->list);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700498 pi_state->owner = NULL;
499 spin_unlock_irq(&curr->pi_lock);
500
501 rt_mutex_unlock(&pi_state->pi_mutex);
502
503 spin_unlock(&hb->lock);
504
505 spin_lock_irq(&curr->pi_lock);
506 }
507 spin_unlock_irq(&curr->pi_lock);
508}
509
510static int
Pierre Peifferd0aa7a72007-05-09 02:35:02 -0700511lookup_pi_state(u32 uval, struct futex_hash_bucket *hb,
512 union futex_key *key, struct futex_pi_state **ps)
Ingo Molnarc87e2832006-06-27 02:54:58 -0700513{
514 struct futex_pi_state *pi_state = NULL;
515 struct futex_q *this, *next;
Pierre Peifferec92d082007-05-09 02:35:00 -0700516 struct plist_head *head;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700517 struct task_struct *p;
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700518 pid_t pid = uval & FUTEX_TID_MASK;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700519
520 head = &hb->chain;
521
Pierre Peifferec92d082007-05-09 02:35:00 -0700522 plist_for_each_entry_safe(this, next, head, list) {
Pierre Peifferd0aa7a72007-05-09 02:35:02 -0700523 if (match_futex(&this->key, key)) {
Ingo Molnarc87e2832006-06-27 02:54:58 -0700524 /*
525 * Another waiter already exists - bump up
526 * the refcount and return its pi_state:
527 */
528 pi_state = this->pi_state;
Thomas Gleixner06a9ec22006-07-10 04:44:30 -0700529 /*
530 * Userspace might have messed up non PI and PI futexes
531 */
532 if (unlikely(!pi_state))
533 return -EINVAL;
534
Ingo Molnar627371d2006-07-29 05:16:20 +0200535 WARN_ON(!atomic_read(&pi_state->refcount));
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700536 WARN_ON(pid && pi_state->owner &&
537 pi_state->owner->pid != pid);
Ingo Molnar627371d2006-07-29 05:16:20 +0200538
Ingo Molnarc87e2832006-06-27 02:54:58 -0700539 atomic_inc(&pi_state->refcount);
Pierre Peifferd0aa7a72007-05-09 02:35:02 -0700540 *ps = pi_state;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700541
542 return 0;
543 }
544 }
545
546 /*
Ingo Molnare3f2dde2006-07-29 05:17:57 +0200547 * We are the first waiter - try to look up the real owner and attach
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700548 * the new pi_state to it, but bail out when TID = 0
Ingo Molnarc87e2832006-06-27 02:54:58 -0700549 */
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700550 if (!pid)
Ingo Molnare3f2dde2006-07-29 05:17:57 +0200551 return -ESRCH;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700552 p = futex_find_get_task(pid);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700553 if (IS_ERR(p))
554 return PTR_ERR(p);
555
556 /*
557 * We need to look at the task state flags to figure out,
558 * whether the task is exiting. To protect against the do_exit
559 * change of the task flags, we do this protected by
560 * p->pi_lock:
561 */
562 spin_lock_irq(&p->pi_lock);
563 if (unlikely(p->flags & PF_EXITING)) {
564 /*
565 * The task is on the way out. When PF_EXITPIDONE is
566 * set, we know that the task has finished the
567 * cleanup:
568 */
569 int ret = (p->flags & PF_EXITPIDONE) ? -ESRCH : -EAGAIN;
570
571 spin_unlock_irq(&p->pi_lock);
572 put_task_struct(p);
573 return ret;
574 }
Ingo Molnarc87e2832006-06-27 02:54:58 -0700575
576 pi_state = alloc_pi_state();
577
578 /*
579 * Initialize the pi_mutex in locked state and make 'p'
580 * the owner of it:
581 */
582 rt_mutex_init_proxy_locked(&pi_state->pi_mutex, p);
583
584 /* Store the key for possible exit cleanups: */
Pierre Peifferd0aa7a72007-05-09 02:35:02 -0700585 pi_state->key = *key;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700586
Ingo Molnar627371d2006-07-29 05:16:20 +0200587 WARN_ON(!list_empty(&pi_state->list));
Ingo Molnarc87e2832006-06-27 02:54:58 -0700588 list_add(&pi_state->list, &p->pi_state_list);
589 pi_state->owner = p;
590 spin_unlock_irq(&p->pi_lock);
591
592 put_task_struct(p);
593
Pierre Peifferd0aa7a72007-05-09 02:35:02 -0700594 *ps = pi_state;
Ingo Molnarc87e2832006-06-27 02:54:58 -0700595
596 return 0;
597}
598
599/*
Linus Torvalds1da177e2005-04-16 15:20:36 -0700600 * The hash bucket lock must be held when this is called.
601 * Afterwards, the futex_q must not be accessed.
602 */
603static void wake_futex(struct futex_q *q)
604{
Pierre Peifferec92d082007-05-09 02:35:00 -0700605 plist_del(&q->list, &q->list.plist);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700606 /*
607 * The lock in wake_up_all() is a crucial memory barrier after the
Pierre Peifferec92d082007-05-09 02:35:00 -0700608 * plist_del() and also before assigning to q->lock_ptr.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700609 */
610 wake_up_all(&q->waiters);
611 /*
612 * The waiting task can free the futex_q as soon as this is written,
613 * without taking any locks. This must come last.
Andrew Morton8e311082005-12-23 19:54:46 -0800614 *
615 * A memory barrier is required here to prevent the following store
616 * to lock_ptr from getting ahead of the wakeup. Clearing the lock
617 * at the end of wake_up_all() does not prevent this store from
618 * moving.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700619 */
Ralf Baechleccdea2f2006-12-06 20:40:26 -0800620 smp_wmb();
Linus Torvalds1da177e2005-04-16 15:20:36 -0700621 q->lock_ptr = NULL;
622}
623
Ingo Molnarc87e2832006-06-27 02:54:58 -0700624static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this)
625{
626 struct task_struct *new_owner;
627 struct futex_pi_state *pi_state = this->pi_state;
628 u32 curval, newval;
629
630 if (!pi_state)
631 return -EINVAL;
632
Ingo Molnar217788672007-03-16 13:38:31 -0800633 spin_lock(&pi_state->pi_mutex.wait_lock);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700634 new_owner = rt_mutex_next_owner(&pi_state->pi_mutex);
635
636 /*
637 * This happens when we have stolen the lock and the original
638 * pending owner did not enqueue itself back on the rt_mutex.
639 * Thats not a tragedy. We know that way, that a lock waiter
640 * is on the fly. We make the futex_q waiter the pending owner.
641 */
642 if (!new_owner)
643 new_owner = this->task;
644
645 /*
646 * We pass it to the next owner. (The WAITERS bit is always
647 * kept enabled while there is PI state around. We must also
648 * preserve the owner died bit.)
649 */
Ingo Molnare3f2dde2006-07-29 05:17:57 +0200650 if (!(uval & FUTEX_OWNER_DIED)) {
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700651 int ret = 0;
652
Pavel Emelyanovb4888932007-10-18 23:40:14 -0700653 newval = FUTEX_WAITERS | task_pid_vnr(new_owner);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700654
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700655 curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700656
Ingo Molnare3f2dde2006-07-29 05:17:57 +0200657 if (curval == -EFAULT)
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700658 ret = -EFAULT;
Thomas Gleixnercde898f2007-12-05 15:46:09 +0100659 else if (curval != uval)
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700660 ret = -EINVAL;
661 if (ret) {
662 spin_unlock(&pi_state->pi_mutex.wait_lock);
663 return ret;
664 }
Ingo Molnare3f2dde2006-07-29 05:17:57 +0200665 }
Ingo Molnarc87e2832006-06-27 02:54:58 -0700666
Ingo Molnar627371d2006-07-29 05:16:20 +0200667 spin_lock_irq(&pi_state->owner->pi_lock);
668 WARN_ON(list_empty(&pi_state->list));
669 list_del_init(&pi_state->list);
670 spin_unlock_irq(&pi_state->owner->pi_lock);
671
672 spin_lock_irq(&new_owner->pi_lock);
673 WARN_ON(!list_empty(&pi_state->list));
Ingo Molnarc87e2832006-06-27 02:54:58 -0700674 list_add(&pi_state->list, &new_owner->pi_state_list);
675 pi_state->owner = new_owner;
Ingo Molnar627371d2006-07-29 05:16:20 +0200676 spin_unlock_irq(&new_owner->pi_lock);
677
Ingo Molnar217788672007-03-16 13:38:31 -0800678 spin_unlock(&pi_state->pi_mutex.wait_lock);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700679 rt_mutex_unlock(&pi_state->pi_mutex);
680
681 return 0;
682}
683
684static int unlock_futex_pi(u32 __user *uaddr, u32 uval)
685{
686 u32 oldval;
687
688 /*
689 * There is no waiter, so we unlock the futex. The owner died
690 * bit has not to be preserved here. We are the owner:
691 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700692 oldval = cmpxchg_futex_value_locked(uaddr, uval, 0);
Ingo Molnarc87e2832006-06-27 02:54:58 -0700693
694 if (oldval == -EFAULT)
695 return oldval;
696 if (oldval != uval)
697 return -EAGAIN;
698
699 return 0;
700}
701
Linus Torvalds1da177e2005-04-16 15:20:36 -0700702/*
Ingo Molnar8b8f3192006-07-03 00:25:05 -0700703 * Express the locking dependencies for lockdep:
704 */
705static inline void
706double_lock_hb(struct futex_hash_bucket *hb1, struct futex_hash_bucket *hb2)
707{
708 if (hb1 <= hb2) {
709 spin_lock(&hb1->lock);
710 if (hb1 < hb2)
711 spin_lock_nested(&hb2->lock, SINGLE_DEPTH_NESTING);
712 } else { /* hb1 > hb2 */
713 spin_lock(&hb2->lock);
714 spin_lock_nested(&hb1->lock, SINGLE_DEPTH_NESTING);
715 }
716}
717
718/*
Linus Torvalds1da177e2005-04-16 15:20:36 -0700719 * Wake up all waiters hashed on the physical page that is mapped
720 * to this virtual address:
721 */
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700722static int futex_wake(u32 __user *uaddr, struct rw_semaphore *fshared,
Thomas Gleixnercd689982008-02-01 17:45:14 +0100723 int nr_wake, u32 bitset)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700724{
Ingo Molnare2970f22006-06-27 02:54:47 -0700725 struct futex_hash_bucket *hb;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700726 struct futex_q *this, *next;
Pierre Peifferec92d082007-05-09 02:35:00 -0700727 struct plist_head *head;
Ingo Molnare2970f22006-06-27 02:54:47 -0700728 union futex_key key;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700729 int ret;
730
Thomas Gleixnercd689982008-02-01 17:45:14 +0100731 if (!bitset)
732 return -EINVAL;
733
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700734 futex_lock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700735
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700736 ret = get_futex_key(uaddr, fshared, &key);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700737 if (unlikely(ret != 0))
738 goto out;
739
Ingo Molnare2970f22006-06-27 02:54:47 -0700740 hb = hash_futex(&key);
741 spin_lock(&hb->lock);
742 head = &hb->chain;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700743
Pierre Peifferec92d082007-05-09 02:35:00 -0700744 plist_for_each_entry_safe(this, next, head, list) {
Linus Torvalds1da177e2005-04-16 15:20:36 -0700745 if (match_futex (&this->key, &key)) {
Ingo Molnared6f7b12006-07-01 04:35:46 -0700746 if (this->pi_state) {
747 ret = -EINVAL;
748 break;
749 }
Thomas Gleixnercd689982008-02-01 17:45:14 +0100750
751 /* Check if one of the bits is set in both bitsets */
752 if (!(this->bitset & bitset))
753 continue;
754
Linus Torvalds1da177e2005-04-16 15:20:36 -0700755 wake_futex(this);
756 if (++ret >= nr_wake)
757 break;
758 }
759 }
760
Ingo Molnare2970f22006-06-27 02:54:47 -0700761 spin_unlock(&hb->lock);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700762out:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700763 futex_unlock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700764 return ret;
765}
766
767/*
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700768 * Wake up all waiters hashed on the physical page that is mapped
769 * to this virtual address:
770 */
Ingo Molnare2970f22006-06-27 02:54:47 -0700771static int
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700772futex_wake_op(u32 __user *uaddr1, struct rw_semaphore *fshared,
773 u32 __user *uaddr2,
Ingo Molnare2970f22006-06-27 02:54:47 -0700774 int nr_wake, int nr_wake2, int op)
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700775{
776 union futex_key key1, key2;
Ingo Molnare2970f22006-06-27 02:54:47 -0700777 struct futex_hash_bucket *hb1, *hb2;
Pierre Peifferec92d082007-05-09 02:35:00 -0700778 struct plist_head *head;
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700779 struct futex_q *this, *next;
780 int ret, op_ret, attempt = 0;
781
782retryfull:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700783 futex_lock_mm(fshared);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700784
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700785 ret = get_futex_key(uaddr1, fshared, &key1);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700786 if (unlikely(ret != 0))
787 goto out;
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700788 ret = get_futex_key(uaddr2, fshared, &key2);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700789 if (unlikely(ret != 0))
790 goto out;
791
Ingo Molnare2970f22006-06-27 02:54:47 -0700792 hb1 = hash_futex(&key1);
793 hb2 = hash_futex(&key2);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700794
795retry:
Ingo Molnar8b8f3192006-07-03 00:25:05 -0700796 double_lock_hb(hb1, hb2);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700797
Ingo Molnare2970f22006-06-27 02:54:47 -0700798 op_ret = futex_atomic_op_inuser(op, uaddr2);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700799 if (unlikely(op_ret < 0)) {
Ingo Molnare2970f22006-06-27 02:54:47 -0700800 u32 dummy;
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700801
Ingo Molnare2970f22006-06-27 02:54:47 -0700802 spin_unlock(&hb1->lock);
803 if (hb1 != hb2)
804 spin_unlock(&hb2->lock);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700805
David Howells7ee1dd32006-01-06 00:11:44 -0800806#ifndef CONFIG_MMU
Ingo Molnare2970f22006-06-27 02:54:47 -0700807 /*
808 * we don't get EFAULT from MMU faults if we don't have an MMU,
809 * but we might get them from range checking
810 */
David Howells7ee1dd32006-01-06 00:11:44 -0800811 ret = op_ret;
812 goto out;
813#endif
814
David Gibson796f8d92005-11-07 00:59:33 -0800815 if (unlikely(op_ret != -EFAULT)) {
816 ret = op_ret;
817 goto out;
818 }
819
Ingo Molnare2970f22006-06-27 02:54:47 -0700820 /*
821 * futex_atomic_op_inuser needs to both read and write
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700822 * *(int __user *)uaddr2, but we can't modify it
823 * non-atomically. Therefore, if get_user below is not
824 * enough, we need to handle the fault ourselves, while
Ingo Molnare2970f22006-06-27 02:54:47 -0700825 * still holding the mmap_sem.
826 */
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700827 if (attempt++) {
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700828 ret = futex_handle_fault((unsigned long)uaddr2,
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700829 fshared, attempt);
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700830 if (ret)
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700831 goto out;
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700832 goto retry;
833 }
834
Ingo Molnare2970f22006-06-27 02:54:47 -0700835 /*
836 * If we would have faulted, release mmap_sem,
837 * fault it in and start all over again.
838 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700839 futex_unlock_mm(fshared);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700840
Ingo Molnare2970f22006-06-27 02:54:47 -0700841 ret = get_user(dummy, uaddr2);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700842 if (ret)
843 return ret;
844
845 goto retryfull;
846 }
847
Ingo Molnare2970f22006-06-27 02:54:47 -0700848 head = &hb1->chain;
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700849
Pierre Peifferec92d082007-05-09 02:35:00 -0700850 plist_for_each_entry_safe(this, next, head, list) {
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700851 if (match_futex (&this->key, &key1)) {
852 wake_futex(this);
853 if (++ret >= nr_wake)
854 break;
855 }
856 }
857
858 if (op_ret > 0) {
Ingo Molnare2970f22006-06-27 02:54:47 -0700859 head = &hb2->chain;
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700860
861 op_ret = 0;
Pierre Peifferec92d082007-05-09 02:35:00 -0700862 plist_for_each_entry_safe(this, next, head, list) {
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700863 if (match_futex (&this->key, &key2)) {
864 wake_futex(this);
865 if (++op_ret >= nr_wake2)
866 break;
867 }
868 }
869 ret += op_ret;
870 }
871
Ingo Molnare2970f22006-06-27 02:54:47 -0700872 spin_unlock(&hb1->lock);
873 if (hb1 != hb2)
874 spin_unlock(&hb2->lock);
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700875out:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700876 futex_unlock_mm(fshared);
877
Jakub Jelinek4732efb2005-09-06 15:16:25 -0700878 return ret;
879}
880
881/*
Linus Torvalds1da177e2005-04-16 15:20:36 -0700882 * Requeue all waiters hashed on one physical page to another
883 * physical page.
884 */
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700885static int futex_requeue(u32 __user *uaddr1, struct rw_semaphore *fshared,
886 u32 __user *uaddr2,
Ingo Molnare2970f22006-06-27 02:54:47 -0700887 int nr_wake, int nr_requeue, u32 *cmpval)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700888{
889 union futex_key key1, key2;
Ingo Molnare2970f22006-06-27 02:54:47 -0700890 struct futex_hash_bucket *hb1, *hb2;
Pierre Peifferec92d082007-05-09 02:35:00 -0700891 struct plist_head *head1;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700892 struct futex_q *this, *next;
893 int ret, drop_count = 0;
894
895 retry:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700896 futex_lock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700897
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700898 ret = get_futex_key(uaddr1, fshared, &key1);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700899 if (unlikely(ret != 0))
900 goto out;
Eric Dumazet34f01cc2007-05-09 02:35:04 -0700901 ret = get_futex_key(uaddr2, fshared, &key2);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700902 if (unlikely(ret != 0))
903 goto out;
904
Ingo Molnare2970f22006-06-27 02:54:47 -0700905 hb1 = hash_futex(&key1);
906 hb2 = hash_futex(&key2);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700907
Ingo Molnar8b8f3192006-07-03 00:25:05 -0700908 double_lock_hb(hb1, hb2);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700909
Ingo Molnare2970f22006-06-27 02:54:47 -0700910 if (likely(cmpval != NULL)) {
911 u32 curval;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700912
Ingo Molnare2970f22006-06-27 02:54:47 -0700913 ret = get_futex_value_locked(&curval, uaddr1);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700914
915 if (unlikely(ret)) {
Ingo Molnare2970f22006-06-27 02:54:47 -0700916 spin_unlock(&hb1->lock);
917 if (hb1 != hb2)
918 spin_unlock(&hb2->lock);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700919
Ingo Molnare2970f22006-06-27 02:54:47 -0700920 /*
921 * If we would have faulted, release mmap_sem, fault
Linus Torvalds1da177e2005-04-16 15:20:36 -0700922 * it in and start all over again.
923 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700924 futex_unlock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700925
Ingo Molnare2970f22006-06-27 02:54:47 -0700926 ret = get_user(curval, uaddr1);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700927
928 if (!ret)
929 goto retry;
930
931 return ret;
932 }
Ingo Molnare2970f22006-06-27 02:54:47 -0700933 if (curval != *cmpval) {
Linus Torvalds1da177e2005-04-16 15:20:36 -0700934 ret = -EAGAIN;
935 goto out_unlock;
936 }
937 }
938
Ingo Molnare2970f22006-06-27 02:54:47 -0700939 head1 = &hb1->chain;
Pierre Peifferec92d082007-05-09 02:35:00 -0700940 plist_for_each_entry_safe(this, next, head1, list) {
Linus Torvalds1da177e2005-04-16 15:20:36 -0700941 if (!match_futex (&this->key, &key1))
942 continue;
943 if (++ret <= nr_wake) {
944 wake_futex(this);
945 } else {
Sebastien Dugue59e0e0a2006-06-27 02:55:03 -0700946 /*
947 * If key1 and key2 hash to the same bucket, no need to
948 * requeue.
949 */
950 if (likely(head1 != &hb2->chain)) {
Pierre Peifferec92d082007-05-09 02:35:00 -0700951 plist_del(&this->list, &hb1->chain);
952 plist_add(&this->list, &hb2->chain);
Sebastien Dugue59e0e0a2006-06-27 02:55:03 -0700953 this->lock_ptr = &hb2->lock;
Pierre Peifferec92d082007-05-09 02:35:00 -0700954#ifdef CONFIG_DEBUG_PI_LIST
955 this->list.plist.lock = &hb2->lock;
956#endif
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -0700957 }
Linus Torvalds1da177e2005-04-16 15:20:36 -0700958 this->key = key2;
Rusty Russell9adef582007-05-08 00:26:42 -0700959 get_futex_key_refs(&key2);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700960 drop_count++;
961
962 if (ret - nr_wake >= nr_requeue)
963 break;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700964 }
965 }
966
967out_unlock:
Ingo Molnare2970f22006-06-27 02:54:47 -0700968 spin_unlock(&hb1->lock);
969 if (hb1 != hb2)
970 spin_unlock(&hb2->lock);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700971
Rusty Russell9adef582007-05-08 00:26:42 -0700972 /* drop_futex_key_refs() must be called outside the spinlocks. */
Linus Torvalds1da177e2005-04-16 15:20:36 -0700973 while (--drop_count >= 0)
Rusty Russell9adef582007-05-08 00:26:42 -0700974 drop_futex_key_refs(&key1);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700975
976out:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -0700977 futex_unlock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700978 return ret;
979}
980
981/* The key must be already stored in q->key. */
Eric Sesterhenn82af7ac2008-01-25 10:40:46 +0100982static inline struct futex_hash_bucket *queue_lock(struct futex_q *q)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700983{
Ingo Molnare2970f22006-06-27 02:54:47 -0700984 struct futex_hash_bucket *hb;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700985
Linus Torvalds1da177e2005-04-16 15:20:36 -0700986 init_waitqueue_head(&q->waiters);
987
Rusty Russell9adef582007-05-08 00:26:42 -0700988 get_futex_key_refs(&q->key);
Ingo Molnare2970f22006-06-27 02:54:47 -0700989 hb = hash_futex(&q->key);
990 q->lock_ptr = &hb->lock;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700991
Ingo Molnare2970f22006-06-27 02:54:47 -0700992 spin_lock(&hb->lock);
993 return hb;
Linus Torvalds1da177e2005-04-16 15:20:36 -0700994}
995
Eric Sesterhenn82af7ac2008-01-25 10:40:46 +0100996static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700997{
Pierre Peifferec92d082007-05-09 02:35:00 -0700998 int prio;
999
1000 /*
1001 * The priority used to register this element is
1002 * - either the real thread-priority for the real-time threads
1003 * (i.e. threads with a priority lower than MAX_RT_PRIO)
1004 * - or MAX_RT_PRIO for non-RT threads.
1005 * Thus, all RT-threads are woken first in priority order, and
1006 * the others are woken last, in FIFO order.
1007 */
1008 prio = min(current->normal_prio, MAX_RT_PRIO);
1009
1010 plist_node_init(&q->list, prio);
1011#ifdef CONFIG_DEBUG_PI_LIST
1012 q->list.plist.lock = &hb->lock;
1013#endif
1014 plist_add(&q->list, &hb->chain);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001015 q->task = current;
Ingo Molnare2970f22006-06-27 02:54:47 -07001016 spin_unlock(&hb->lock);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001017}
1018
1019static inline void
Ingo Molnare2970f22006-06-27 02:54:47 -07001020queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001021{
Ingo Molnare2970f22006-06-27 02:54:47 -07001022 spin_unlock(&hb->lock);
Rusty Russell9adef582007-05-08 00:26:42 -07001023 drop_futex_key_refs(&q->key);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001024}
1025
1026/*
1027 * queue_me and unqueue_me must be called as a pair, each
1028 * exactly once. They are called with the hashed spinlock held.
1029 */
1030
Linus Torvalds1da177e2005-04-16 15:20:36 -07001031/* Return 1 if we were still queued (ie. 0 means we were woken) */
1032static int unqueue_me(struct futex_q *q)
1033{
Linus Torvalds1da177e2005-04-16 15:20:36 -07001034 spinlock_t *lock_ptr;
Ingo Molnare2970f22006-06-27 02:54:47 -07001035 int ret = 0;
Linus Torvalds1da177e2005-04-16 15:20:36 -07001036
1037 /* In the common case we don't take the spinlock, which is nice. */
1038 retry:
1039 lock_ptr = q->lock_ptr;
Christian Borntraegere91467e2006-08-05 12:13:52 -07001040 barrier();
Stephen Hemmingerc80544d2007-10-18 03:07:05 -07001041 if (lock_ptr != NULL) {
Linus Torvalds1da177e2005-04-16 15:20:36 -07001042 spin_lock(lock_ptr);
1043 /*
1044 * q->lock_ptr can change between reading it and
1045 * spin_lock(), causing us to take the wrong lock. This
1046 * corrects the race condition.
1047 *
1048 * Reasoning goes like this: if we have the wrong lock,
1049 * q->lock_ptr must have changed (maybe several times)
1050 * between reading it and the spin_lock(). It can
1051 * change again after the spin_lock() but only if it was
1052 * already changed before the spin_lock(). It cannot,
1053 * however, change back to the original value. Therefore
1054 * we can detect whether we acquired the correct lock.
1055 */
1056 if (unlikely(lock_ptr != q->lock_ptr)) {
1057 spin_unlock(lock_ptr);
1058 goto retry;
1059 }
Pierre Peifferec92d082007-05-09 02:35:00 -07001060 WARN_ON(plist_node_empty(&q->list));
1061 plist_del(&q->list, &q->list.plist);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001062
1063 BUG_ON(q->pi_state);
1064
Linus Torvalds1da177e2005-04-16 15:20:36 -07001065 spin_unlock(lock_ptr);
1066 ret = 1;
1067 }
1068
Rusty Russell9adef582007-05-08 00:26:42 -07001069 drop_futex_key_refs(&q->key);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001070 return ret;
1071}
1072
Ingo Molnarc87e2832006-06-27 02:54:58 -07001073/*
1074 * PI futexes can not be requeued and must remove themself from the
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001075 * hash bucket. The hash bucket lock (i.e. lock_ptr) is held on entry
1076 * and dropped here.
Ingo Molnarc87e2832006-06-27 02:54:58 -07001077 */
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001078static void unqueue_me_pi(struct futex_q *q)
Ingo Molnarc87e2832006-06-27 02:54:58 -07001079{
Pierre Peifferec92d082007-05-09 02:35:00 -07001080 WARN_ON(plist_node_empty(&q->list));
1081 plist_del(&q->list, &q->list.plist);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001082
1083 BUG_ON(!q->pi_state);
1084 free_pi_state(q->pi_state);
1085 q->pi_state = NULL;
1086
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001087 spin_unlock(q->lock_ptr);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001088
Rusty Russell9adef582007-05-08 00:26:42 -07001089 drop_futex_key_refs(&q->key);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001090}
1091
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001092/*
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001093 * Fixup the pi_state owner with the new owner.
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001094 *
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001095 * Must be called with hash bucket lock held and mm->sem held for non
1096 * private futexes.
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001097 */
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001098static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001099 struct task_struct *newowner,
1100 struct rw_semaphore *fshared)
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001101{
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001102 u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001103 struct futex_pi_state *pi_state = q->pi_state;
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001104 struct task_struct *oldowner = pi_state->owner;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001105 u32 uval, curval, newval;
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001106 int ret, attempt = 0;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001107
1108 /* Owner died? */
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001109 if (!pi_state->owner)
1110 newtid |= FUTEX_OWNER_DIED;
1111
1112 /*
1113 * We are here either because we stole the rtmutex from the
1114 * pending owner or we are the pending owner which failed to
1115 * get the rtmutex. We have to replace the pending owner TID
1116 * in the user space variable. This must be atomic as we have
1117 * to preserve the owner died bit here.
1118 *
1119 * Note: We write the user space value _before_ changing the
1120 * pi_state because we can fault here. Imagine swapped out
1121 * pages or a fork, which was running right before we acquired
1122 * mmap_sem, that marked all the anonymous memory readonly for
1123 * cow.
1124 *
1125 * Modifying pi_state _before_ the user space value would
1126 * leave the pi_state in an inconsistent state when we fault
1127 * here, because we need to drop the hash bucket lock to
1128 * handle the fault. This might be observed in the PID check
1129 * in lookup_pi_state.
1130 */
1131retry:
1132 if (get_futex_value_locked(&uval, uaddr))
1133 goto handle_fault;
1134
1135 while (1) {
1136 newval = (uval & FUTEX_OWNER_DIED) | newtid;
1137
1138 curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
1139
1140 if (curval == -EFAULT)
1141 goto handle_fault;
1142 if (curval == uval)
1143 break;
1144 uval = curval;
1145 }
1146
1147 /*
1148 * We fixed up user space. Now we need to fix the pi_state
1149 * itself.
1150 */
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001151 if (pi_state->owner != NULL) {
1152 spin_lock_irq(&pi_state->owner->pi_lock);
1153 WARN_ON(list_empty(&pi_state->list));
1154 list_del_init(&pi_state->list);
1155 spin_unlock_irq(&pi_state->owner->pi_lock);
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001156 }
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001157
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001158 pi_state->owner = newowner;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001159
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001160 spin_lock_irq(&newowner->pi_lock);
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001161 WARN_ON(!list_empty(&pi_state->list));
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001162 list_add(&pi_state->list, &newowner->pi_state_list);
1163 spin_unlock_irq(&newowner->pi_lock);
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001164 return 0;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001165
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001166 /*
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001167 * To handle the page fault we need to drop the hash bucket
1168 * lock here. That gives the other task (either the pending
1169 * owner itself or the task which stole the rtmutex) the
1170 * chance to try the fixup of the pi_state. So once we are
1171 * back from handling the fault we need to check the pi_state
1172 * after reacquiring the hash bucket lock and before trying to
1173 * do another fixup. When the fixup has been done already we
1174 * simply return.
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001175 */
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001176handle_fault:
1177 spin_unlock(q->lock_ptr);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001178
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001179 ret = futex_handle_fault((unsigned long)uaddr, fshared, attempt++);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001180
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001181 spin_lock(q->lock_ptr);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001182
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001183 /*
1184 * Check if someone else fixed it for us:
1185 */
1186 if (pi_state->owner != oldowner)
1187 return 0;
1188
1189 if (ret)
1190 return ret;
1191
1192 goto retry;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001193}
1194
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001195/*
1196 * In case we must use restart_block to restart a futex_wait,
Steven Rostedtce6bd422007-12-05 15:46:09 +01001197 * we encode in the 'flags' shared capability
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001198 */
Steven Rostedtce6bd422007-12-05 15:46:09 +01001199#define FLAGS_SHARED 1
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001200
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001201static long futex_wait_restart(struct restart_block *restart);
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001202
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001203static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared,
Thomas Gleixnercd689982008-02-01 17:45:14 +01001204 u32 val, ktime_t *abs_time, u32 bitset)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001205{
Ingo Molnarc87e2832006-06-27 02:54:58 -07001206 struct task_struct *curr = current;
1207 DECLARE_WAITQUEUE(wait, curr);
Ingo Molnare2970f22006-06-27 02:54:47 -07001208 struct futex_hash_bucket *hb;
Linus Torvalds1da177e2005-04-16 15:20:36 -07001209 struct futex_q q;
Ingo Molnare2970f22006-06-27 02:54:47 -07001210 u32 uval;
1211 int ret;
Thomas Gleixnerbd197232007-06-17 21:11:10 +02001212 struct hrtimer_sleeper t;
Pierre Peifferc19384b2007-05-09 02:35:02 -07001213 int rem = 0;
Linus Torvalds1da177e2005-04-16 15:20:36 -07001214
Thomas Gleixnercd689982008-02-01 17:45:14 +01001215 if (!bitset)
1216 return -EINVAL;
1217
Ingo Molnarc87e2832006-06-27 02:54:58 -07001218 q.pi_state = NULL;
Thomas Gleixnercd689982008-02-01 17:45:14 +01001219 q.bitset = bitset;
Linus Torvalds1da177e2005-04-16 15:20:36 -07001220 retry:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001221 futex_lock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001222
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001223 ret = get_futex_key(uaddr, fshared, &q.key);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001224 if (unlikely(ret != 0))
1225 goto out_release_sem;
1226
Eric Sesterhenn82af7ac2008-01-25 10:40:46 +01001227 hb = queue_lock(&q);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001228
1229 /*
1230 * Access the page AFTER the futex is queued.
1231 * Order is important:
1232 *
1233 * Userspace waiter: val = var; if (cond(val)) futex_wait(&var, val);
1234 * Userspace waker: if (cond(var)) { var = new; futex_wake(&var); }
1235 *
1236 * The basic logical guarantee of a futex is that it blocks ONLY
1237 * if cond(var) is known to be true at the time of blocking, for
1238 * any cond. If we queued after testing *uaddr, that would open
1239 * a race condition where we could block indefinitely with
1240 * cond(var) false, which would violate the guarantee.
1241 *
1242 * A consequence is that futex_wait() can return zero and absorb
1243 * a wakeup when *uaddr != val on entry to the syscall. This is
1244 * rare, but normal.
1245 *
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001246 * for shared futexes, we hold the mmap semaphore, so the mapping
1247 * cannot have changed since we looked it up in get_futex_key.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001248 */
Ingo Molnare2970f22006-06-27 02:54:47 -07001249 ret = get_futex_value_locked(&uval, uaddr);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001250
1251 if (unlikely(ret)) {
Ingo Molnare2970f22006-06-27 02:54:47 -07001252 queue_unlock(&q, hb);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001253
Ingo Molnare2970f22006-06-27 02:54:47 -07001254 /*
1255 * If we would have faulted, release mmap_sem, fault it in and
Linus Torvalds1da177e2005-04-16 15:20:36 -07001256 * start all over again.
1257 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001258 futex_unlock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001259
Ingo Molnare2970f22006-06-27 02:54:47 -07001260 ret = get_user(uval, uaddr);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001261
1262 if (!ret)
1263 goto retry;
1264 return ret;
1265 }
Ingo Molnarc87e2832006-06-27 02:54:58 -07001266 ret = -EWOULDBLOCK;
1267 if (uval != val)
1268 goto out_unlock_release_sem;
Linus Torvalds1da177e2005-04-16 15:20:36 -07001269
1270 /* Only actually queue if *uaddr contained val. */
Eric Sesterhenn82af7ac2008-01-25 10:40:46 +01001271 queue_me(&q, hb);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001272
1273 /*
1274 * Now the futex is queued and we have checked the data, we
1275 * don't want to hold mmap_sem while we sleep.
Ingo Molnarc87e2832006-06-27 02:54:58 -07001276 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001277 futex_unlock_mm(fshared);
Linus Torvalds1da177e2005-04-16 15:20:36 -07001278
1279 /*
1280 * There might have been scheduling since the queue_me(), as we
1281 * cannot hold a spinlock across the get_user() in case it
1282 * faults, and we cannot just set TASK_INTERRUPTIBLE state when
1283 * queueing ourselves into the futex hash. This code thus has to
1284 * rely on the futex_wake() code removing us from hash when it
1285 * wakes us up.
1286 */
1287
1288 /* add_wait_queue is the barrier after __set_current_state. */
1289 __set_current_state(TASK_INTERRUPTIBLE);
1290 add_wait_queue(&q.waiters, &wait);
1291 /*
Pierre Peifferec92d082007-05-09 02:35:00 -07001292 * !plist_node_empty() is safe here without any lock.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001293 * q.lock_ptr != 0 is not safe, because of ordering against wakeup.
1294 */
Pierre Peifferec92d082007-05-09 02:35:00 -07001295 if (likely(!plist_node_empty(&q.list))) {
Pierre Peifferc19384b2007-05-09 02:35:02 -07001296 if (!abs_time)
1297 schedule();
1298 else {
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001299 hrtimer_init_on_stack(&t.timer, CLOCK_MONOTONIC,
1300 HRTIMER_MODE_ABS);
Pierre Peifferc19384b2007-05-09 02:35:02 -07001301 hrtimer_init_sleeper(&t, current);
1302 t.timer.expires = *abs_time;
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001303
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001304 hrtimer_start(&t.timer, t.timer.expires,
1305 HRTIMER_MODE_ABS);
Peter Zijlstra3588a082008-02-01 17:45:13 +01001306 if (!hrtimer_active(&t.timer))
1307 t.task = NULL;
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001308
Pierre Peifferc19384b2007-05-09 02:35:02 -07001309 /*
1310 * the timer could have already expired, in which
1311 * case current would be flagged for rescheduling.
1312 * Don't bother calling schedule.
1313 */
1314 if (likely(t.task))
1315 schedule();
1316
1317 hrtimer_cancel(&t.timer);
1318
1319 /* Flag if a timeout occured */
1320 rem = (t.task == NULL);
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001321
1322 destroy_hrtimer_on_stack(&t.timer);
Pierre Peifferc19384b2007-05-09 02:35:02 -07001323 }
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001324 }
Linus Torvalds1da177e2005-04-16 15:20:36 -07001325 __set_current_state(TASK_RUNNING);
1326
1327 /*
1328 * NOTE: we don't remove ourselves from the waitqueue because
1329 * we are the only user of it.
1330 */
1331
1332 /* If we were woken (and unqueued), we succeeded, whatever. */
1333 if (!unqueue_me(&q))
1334 return 0;
Pierre Peifferc19384b2007-05-09 02:35:02 -07001335 if (rem)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001336 return -ETIMEDOUT;
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001337
Ingo Molnare2970f22006-06-27 02:54:47 -07001338 /*
1339 * We expect signal_pending(current), but another thread may
1340 * have handled it for us already.
1341 */
Pierre Peifferc19384b2007-05-09 02:35:02 -07001342 if (!abs_time)
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001343 return -ERESTARTSYS;
1344 else {
1345 struct restart_block *restart;
1346 restart = &current_thread_info()->restart_block;
1347 restart->fn = futex_wait_restart;
Steven Rostedtce6bd422007-12-05 15:46:09 +01001348 restart->futex.uaddr = (u32 *)uaddr;
1349 restart->futex.val = val;
1350 restart->futex.time = abs_time->tv64;
Thomas Gleixnercd689982008-02-01 17:45:14 +01001351 restart->futex.bitset = bitset;
Steven Rostedtce6bd422007-12-05 15:46:09 +01001352 restart->futex.flags = 0;
1353
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001354 if (fshared)
Steven Rostedtce6bd422007-12-05 15:46:09 +01001355 restart->futex.flags |= FLAGS_SHARED;
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001356 return -ERESTART_RESTARTBLOCK;
1357 }
Linus Torvalds1da177e2005-04-16 15:20:36 -07001358
Ingo Molnarc87e2832006-06-27 02:54:58 -07001359 out_unlock_release_sem:
1360 queue_unlock(&q, hb);
1361
Linus Torvalds1da177e2005-04-16 15:20:36 -07001362 out_release_sem:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001363 futex_unlock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001364 return ret;
1365}
1366
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001367
1368static long futex_wait_restart(struct restart_block *restart)
1369{
Steven Rostedtce6bd422007-12-05 15:46:09 +01001370 u32 __user *uaddr = (u32 __user *)restart->futex.uaddr;
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001371 struct rw_semaphore *fshared = NULL;
Steven Rostedtce6bd422007-12-05 15:46:09 +01001372 ktime_t t;
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001373
Steven Rostedtce6bd422007-12-05 15:46:09 +01001374 t.tv64 = restart->futex.time;
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001375 restart->fn = do_no_restart_syscall;
Steven Rostedtce6bd422007-12-05 15:46:09 +01001376 if (restart->futex.flags & FLAGS_SHARED)
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001377 fshared = &current->mm->mmap_sem;
Thomas Gleixnercd689982008-02-01 17:45:14 +01001378 return (long)futex_wait(uaddr, fshared, restart->futex.val, &t,
1379 restart->futex.bitset);
Nick Piggin72c1bbf2007-05-08 00:26:43 -07001380}
1381
1382
Ingo Molnarc87e2832006-06-27 02:54:58 -07001383/*
1384 * Userspace tried a 0 -> TID atomic transition of the futex value
1385 * and failed. The kernel side here does the whole locking operation:
1386 * if there are waiters then it will block, it does PI, etc. (Due to
1387 * races the kernel might see a 0 value of the futex too.)
1388 */
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001389static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared,
1390 int detect, ktime_t *time, int trylock)
Ingo Molnarc87e2832006-06-27 02:54:58 -07001391{
Thomas Gleixnerc5780e92006-09-08 09:47:15 -07001392 struct hrtimer_sleeper timeout, *to = NULL;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001393 struct task_struct *curr = current;
1394 struct futex_hash_bucket *hb;
1395 u32 uval, newval, curval;
1396 struct futex_q q;
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001397 int ret, lock_taken, ownerdied = 0, attempt = 0;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001398
1399 if (refill_pi_state_cache())
1400 return -ENOMEM;
1401
Pierre Peifferc19384b2007-05-09 02:35:02 -07001402 if (time) {
Thomas Gleixnerc5780e92006-09-08 09:47:15 -07001403 to = &timeout;
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001404 hrtimer_init_on_stack(&to->timer, CLOCK_REALTIME,
1405 HRTIMER_MODE_ABS);
Thomas Gleixnerc5780e92006-09-08 09:47:15 -07001406 hrtimer_init_sleeper(to, current);
Pierre Peifferc19384b2007-05-09 02:35:02 -07001407 to->timer.expires = *time;
Thomas Gleixnerc5780e92006-09-08 09:47:15 -07001408 }
1409
Ingo Molnarc87e2832006-06-27 02:54:58 -07001410 q.pi_state = NULL;
1411 retry:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001412 futex_lock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001413
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001414 ret = get_futex_key(uaddr, fshared, &q.key);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001415 if (unlikely(ret != 0))
1416 goto out_release_sem;
1417
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001418 retry_unlocked:
Eric Sesterhenn82af7ac2008-01-25 10:40:46 +01001419 hb = queue_lock(&q);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001420
1421 retry_locked:
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001422 ret = lock_taken = 0;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001423
Ingo Molnarc87e2832006-06-27 02:54:58 -07001424 /*
1425 * To avoid races, we attempt to take the lock here again
1426 * (by doing a 0 -> TID atomic cmpxchg), while holding all
1427 * the locks. It will most likely not succeed.
1428 */
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001429 newval = task_pid_vnr(current);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001430
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001431 curval = cmpxchg_futex_value_locked(uaddr, 0, newval);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001432
1433 if (unlikely(curval == -EFAULT))
1434 goto uaddr_faulted;
1435
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001436 /*
1437 * Detect deadlocks. In case of REQUEUE_PI this is a valid
1438 * situation and we return success to user space.
1439 */
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001440 if (unlikely((curval & FUTEX_TID_MASK) == task_pid_vnr(current))) {
Thomas Gleixnerbd197232007-06-17 21:11:10 +02001441 ret = -EDEADLK;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001442 goto out_unlock_release_sem;
1443 }
1444
1445 /*
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001446 * Surprise - we got the lock. Just return to userspace:
Ingo Molnarc87e2832006-06-27 02:54:58 -07001447 */
1448 if (unlikely(!curval))
1449 goto out_unlock_release_sem;
1450
1451 uval = curval;
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001452
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001453 /*
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001454 * Set the WAITERS flag, so the owner will know it has someone
1455 * to wake at next unlock
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001456 */
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001457 newval = curval | FUTEX_WAITERS;
1458
1459 /*
1460 * There are two cases, where a futex might have no owner (the
Thomas Gleixnerbd197232007-06-17 21:11:10 +02001461 * owner TID is 0): OWNER_DIED. We take over the futex in this
1462 * case. We also do an unconditional take over, when the owner
1463 * of the futex died.
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001464 *
1465 * This is safe as we are protected by the hash bucket lock !
1466 */
1467 if (unlikely(ownerdied || !(curval & FUTEX_TID_MASK))) {
Thomas Gleixnerbd197232007-06-17 21:11:10 +02001468 /* Keep the OWNER_DIED bit */
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001469 newval = (curval & ~FUTEX_TID_MASK) | task_pid_vnr(current);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001470 ownerdied = 0;
1471 lock_taken = 1;
1472 }
Ingo Molnarc87e2832006-06-27 02:54:58 -07001473
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001474 curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001475
1476 if (unlikely(curval == -EFAULT))
1477 goto uaddr_faulted;
1478 if (unlikely(curval != uval))
1479 goto retry_locked;
1480
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001481 /*
Thomas Gleixnerbd197232007-06-17 21:11:10 +02001482 * We took the lock due to owner died take over.
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001483 */
Thomas Gleixnerbd197232007-06-17 21:11:10 +02001484 if (unlikely(lock_taken))
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001485 goto out_unlock_release_sem;
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001486
Ingo Molnarc87e2832006-06-27 02:54:58 -07001487 /*
1488 * We dont have the lock. Look up the PI state (or create it if
1489 * we are the first waiter):
1490 */
Pierre Peifferd0aa7a72007-05-09 02:35:02 -07001491 ret = lookup_pi_state(uval, hb, &q.key, &q.pi_state);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001492
1493 if (unlikely(ret)) {
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001494 switch (ret) {
Ingo Molnarc87e2832006-06-27 02:54:58 -07001495
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001496 case -EAGAIN:
1497 /*
1498 * Task is exiting and we just wait for the
1499 * exit to complete.
1500 */
1501 queue_unlock(&q, hb);
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001502 futex_unlock_mm(fshared);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001503 cond_resched();
1504 goto retry;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001505
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001506 case -ESRCH:
1507 /*
1508 * No owner found for this futex. Check if the
1509 * OWNER_DIED bit is set to figure out whether
1510 * this is a robust futex or not.
1511 */
1512 if (get_futex_value_locked(&curval, uaddr))
Ingo Molnarc87e2832006-06-27 02:54:58 -07001513 goto uaddr_faulted;
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001514
1515 /*
1516 * We simply start over in case of a robust
1517 * futex. The code above will take the futex
1518 * and return happy.
1519 */
1520 if (curval & FUTEX_OWNER_DIED) {
1521 ownerdied = 1;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001522 goto retry_locked;
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001523 }
1524 default:
1525 goto out_unlock_release_sem;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001526 }
Ingo Molnarc87e2832006-06-27 02:54:58 -07001527 }
1528
1529 /*
1530 * Only actually queue now that the atomic ops are done:
1531 */
Eric Sesterhenn82af7ac2008-01-25 10:40:46 +01001532 queue_me(&q, hb);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001533
1534 /*
1535 * Now the futex is queued and we have checked the data, we
1536 * don't want to hold mmap_sem while we sleep.
1537 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001538 futex_unlock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001539
1540 WARN_ON(!q.pi_state);
1541 /*
1542 * Block on the PI mutex:
1543 */
1544 if (!trylock)
1545 ret = rt_mutex_timed_lock(&q.pi_state->pi_mutex, to, 1);
1546 else {
1547 ret = rt_mutex_trylock(&q.pi_state->pi_mutex);
1548 /* Fixup the trylock return value: */
1549 ret = ret ? 0 : -EWOULDBLOCK;
1550 }
1551
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001552 futex_lock_mm(fshared);
Vernon Mauerya99e4e42006-07-01 04:35:42 -07001553 spin_lock(q.lock_ptr);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001554
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001555 if (!ret) {
1556 /*
1557 * Got the lock. We might not be the anticipated owner
1558 * if we did a lock-steal - fix up the PI-state in
1559 * that case:
1560 */
1561 if (q.pi_state->owner != curr)
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001562 ret = fixup_pi_state_owner(uaddr, &q, curr, fshared);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001563 } else {
Ingo Molnarc87e2832006-06-27 02:54:58 -07001564 /*
1565 * Catch the rare case, where the lock was released
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001566 * when we were on the way back before we locked the
1567 * hash bucket.
Ingo Molnarc87e2832006-06-27 02:54:58 -07001568 */
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001569 if (q.pi_state->owner == curr) {
1570 /*
1571 * Try to get the rt_mutex now. This might
1572 * fail as some other task acquired the
1573 * rt_mutex after we removed ourself from the
1574 * rt_mutex waiters list.
1575 */
1576 if (rt_mutex_trylock(&q.pi_state->pi_mutex))
1577 ret = 0;
1578 else {
1579 /*
1580 * pi_state is incorrect, some other
1581 * task did a lock steal and we
1582 * returned due to timeout or signal
1583 * without taking the rt_mutex. Too
1584 * late. We can access the
1585 * rt_mutex_owner without locking, as
1586 * the other task is now blocked on
1587 * the hash bucket lock. Fix the state
1588 * up.
1589 */
1590 struct task_struct *owner;
1591 int res;
1592
1593 owner = rt_mutex_owner(&q.pi_state->pi_mutex);
Thomas Gleixner1b7558e2008-06-23 11:21:58 +02001594 res = fixup_pi_state_owner(uaddr, &q, owner,
1595 fshared);
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001596
Thomas Gleixnercdf71a12008-01-08 19:47:38 +01001597 /* propagate -EFAULT, if the fixup failed */
1598 if (res)
1599 ret = res;
1600 }
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001601 } else {
1602 /*
1603 * Paranoia check. If we did not take the lock
1604 * in the trylock above, then we should not be
1605 * the owner of the rtmutex, neither the real
1606 * nor the pending one:
1607 */
1608 if (rt_mutex_owner(&q.pi_state->pi_mutex) == curr)
1609 printk(KERN_ERR "futex_lock_pi: ret = %d "
1610 "pi-mutex: %p pi-state %p\n", ret,
1611 q.pi_state->pi_mutex.owner,
1612 q.pi_state->owner);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001613 }
Ingo Molnarc87e2832006-06-27 02:54:58 -07001614 }
1615
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001616 /* Unqueue and drop the lock */
1617 unqueue_me_pi(&q);
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001618 futex_unlock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001619
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001620 if (to)
1621 destroy_hrtimer_on_stack(&to->timer);
Thomas Gleixnerc5780e92006-09-08 09:47:15 -07001622 return ret != -EINTR ? ret : -ERESTARTNOINTR;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001623
1624 out_unlock_release_sem:
1625 queue_unlock(&q, hb);
1626
1627 out_release_sem:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001628 futex_unlock_mm(fshared);
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001629 if (to)
1630 destroy_hrtimer_on_stack(&to->timer);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001631 return ret;
1632
1633 uaddr_faulted:
1634 /*
1635 * We have to r/w *(int __user *)uaddr, but we can't modify it
1636 * non-atomically. Therefore, if get_user below is not
1637 * enough, we need to handle the fault ourselves, while
1638 * still holding the mmap_sem.
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001639 *
1640 * ... and hb->lock. :-) --ANK
Ingo Molnarc87e2832006-06-27 02:54:58 -07001641 */
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001642 queue_unlock(&q, hb);
1643
Ingo Molnarc87e2832006-06-27 02:54:58 -07001644 if (attempt++) {
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001645 ret = futex_handle_fault((unsigned long)uaddr, fshared,
1646 attempt);
1647 if (ret)
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001648 goto out_release_sem;
1649 goto retry_unlocked;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001650 }
1651
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001652 futex_unlock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001653
1654 ret = get_user(uval, uaddr);
1655 if (!ret && (uval != -EFAULT))
1656 goto retry;
1657
Thomas Gleixner237fc6e2008-04-30 00:55:04 -07001658 if (to)
1659 destroy_hrtimer_on_stack(&to->timer);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001660 return ret;
1661}
1662
1663/*
Ingo Molnarc87e2832006-06-27 02:54:58 -07001664 * Userspace attempted a TID -> 0 atomic transition, and failed.
1665 * This is the in-kernel slowpath: we look up the PI state (if any),
1666 * and do the rt-mutex unlock.
1667 */
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001668static int futex_unlock_pi(u32 __user *uaddr, struct rw_semaphore *fshared)
Ingo Molnarc87e2832006-06-27 02:54:58 -07001669{
1670 struct futex_hash_bucket *hb;
1671 struct futex_q *this, *next;
1672 u32 uval;
Pierre Peifferec92d082007-05-09 02:35:00 -07001673 struct plist_head *head;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001674 union futex_key key;
1675 int ret, attempt = 0;
1676
1677retry:
1678 if (get_user(uval, uaddr))
1679 return -EFAULT;
1680 /*
1681 * We release only a lock we actually own:
1682 */
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001683 if ((uval & FUTEX_TID_MASK) != task_pid_vnr(current))
Ingo Molnarc87e2832006-06-27 02:54:58 -07001684 return -EPERM;
1685 /*
1686 * First take all the futex related locks:
1687 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001688 futex_lock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001689
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001690 ret = get_futex_key(uaddr, fshared, &key);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001691 if (unlikely(ret != 0))
1692 goto out;
1693
1694 hb = hash_futex(&key);
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001695retry_unlocked:
Ingo Molnarc87e2832006-06-27 02:54:58 -07001696 spin_lock(&hb->lock);
1697
Ingo Molnarc87e2832006-06-27 02:54:58 -07001698 /*
1699 * To avoid races, try to do the TID -> 0 atomic transition
1700 * again. If it succeeds then we can return without waking
1701 * anyone else up:
1702 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001703 if (!(uval & FUTEX_OWNER_DIED))
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001704 uval = cmpxchg_futex_value_locked(uaddr, task_pid_vnr(current), 0);
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001705
Ingo Molnarc87e2832006-06-27 02:54:58 -07001706
1707 if (unlikely(uval == -EFAULT))
1708 goto pi_faulted;
1709 /*
1710 * Rare case: we managed to release the lock atomically,
1711 * no need to wake anyone else up:
1712 */
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001713 if (unlikely(uval == task_pid_vnr(current)))
Ingo Molnarc87e2832006-06-27 02:54:58 -07001714 goto out_unlock;
1715
1716 /*
1717 * Ok, other tasks may need to be woken up - check waiters
1718 * and do the wakeup if necessary:
1719 */
1720 head = &hb->chain;
1721
Pierre Peifferec92d082007-05-09 02:35:00 -07001722 plist_for_each_entry_safe(this, next, head, list) {
Ingo Molnarc87e2832006-06-27 02:54:58 -07001723 if (!match_futex (&this->key, &key))
1724 continue;
1725 ret = wake_futex_pi(uaddr, uval, this);
1726 /*
1727 * The atomic access to the futex value
1728 * generated a pagefault, so retry the
1729 * user-access and the wakeup:
1730 */
1731 if (ret == -EFAULT)
1732 goto pi_faulted;
1733 goto out_unlock;
1734 }
1735 /*
1736 * No waiters - kernel unlocks the futex:
1737 */
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001738 if (!(uval & FUTEX_OWNER_DIED)) {
1739 ret = unlock_futex_pi(uaddr, uval);
1740 if (ret == -EFAULT)
1741 goto pi_faulted;
1742 }
Ingo Molnarc87e2832006-06-27 02:54:58 -07001743
1744out_unlock:
1745 spin_unlock(&hb->lock);
1746out:
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001747 futex_unlock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001748
1749 return ret;
1750
1751pi_faulted:
1752 /*
1753 * We have to r/w *(int __user *)uaddr, but we can't modify it
1754 * non-atomically. Therefore, if get_user below is not
1755 * enough, we need to handle the fault ourselves, while
1756 * still holding the mmap_sem.
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001757 *
1758 * ... and hb->lock. --ANK
Ingo Molnarc87e2832006-06-27 02:54:58 -07001759 */
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001760 spin_unlock(&hb->lock);
1761
Ingo Molnarc87e2832006-06-27 02:54:58 -07001762 if (attempt++) {
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001763 ret = futex_handle_fault((unsigned long)uaddr, fshared,
1764 attempt);
1765 if (ret)
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001766 goto out;
john stultz187226f2007-08-22 14:01:10 -07001767 uval = 0;
Alexey Kuznetsov778e9a92007-06-08 13:47:00 -07001768 goto retry_unlocked;
Ingo Molnarc87e2832006-06-27 02:54:58 -07001769 }
1770
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001771 futex_unlock_mm(fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07001772
1773 ret = get_user(uval, uaddr);
1774 if (!ret && (uval != -EFAULT))
1775 goto retry;
1776
Linus Torvalds1da177e2005-04-16 15:20:36 -07001777 return ret;
1778}
1779
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001780/*
1781 * Support for robust futexes: the kernel cleans up held futexes at
1782 * thread exit time.
1783 *
1784 * Implementation: user-space maintains a per-thread list of locks it
1785 * is holding. Upon do_exit(), the kernel carefully walks this list,
1786 * and marks all locks that are owned by this thread with the
Ingo Molnarc87e2832006-06-27 02:54:58 -07001787 * FUTEX_OWNER_DIED bit, and wakes up a waiter (if any). The list is
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001788 * always manipulated with the lock held, so the list is private and
1789 * per-thread. Userspace also maintains a per-thread 'list_op_pending'
1790 * field, to allow the kernel to clean up if the thread dies after
1791 * acquiring the lock, but just before it could have added itself to
1792 * the list. There can only be one such pending lock.
1793 */
1794
1795/**
1796 * sys_set_robust_list - set the robust-futex list head of a task
1797 * @head: pointer to the list-head
1798 * @len: length of the list-head, as userspace expects
1799 */
1800asmlinkage long
1801sys_set_robust_list(struct robust_list_head __user *head,
1802 size_t len)
1803{
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08001804 if (!futex_cmpxchg_enabled)
1805 return -ENOSYS;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001806 /*
1807 * The kernel knows only one size for now:
1808 */
1809 if (unlikely(len != sizeof(*head)))
1810 return -EINVAL;
1811
1812 current->robust_list = head;
1813
1814 return 0;
1815}
1816
1817/**
1818 * sys_get_robust_list - get the robust-futex list head of a task
1819 * @pid: pid of the process [zero for current task]
1820 * @head_ptr: pointer to a list-head pointer, the kernel fills it in
1821 * @len_ptr: pointer to a length field, the kernel fills in the header size
1822 */
1823asmlinkage long
Al Viroba46df92006-10-10 22:46:07 +01001824sys_get_robust_list(int pid, struct robust_list_head __user * __user *head_ptr,
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001825 size_t __user *len_ptr)
1826{
Al Viroba46df92006-10-10 22:46:07 +01001827 struct robust_list_head __user *head;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001828 unsigned long ret;
1829
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08001830 if (!futex_cmpxchg_enabled)
1831 return -ENOSYS;
1832
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001833 if (!pid)
1834 head = current->robust_list;
1835 else {
1836 struct task_struct *p;
1837
1838 ret = -ESRCH;
Oleg Nesterovaaa2a972006-09-29 02:00:55 -07001839 rcu_read_lock();
Pavel Emelyanov228ebcb2007-10-18 23:40:16 -07001840 p = find_task_by_vpid(pid);
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001841 if (!p)
1842 goto err_unlock;
1843 ret = -EPERM;
1844 if ((current->euid != p->euid) && (current->euid != p->uid) &&
1845 !capable(CAP_SYS_PTRACE))
1846 goto err_unlock;
1847 head = p->robust_list;
Oleg Nesterovaaa2a972006-09-29 02:00:55 -07001848 rcu_read_unlock();
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001849 }
1850
1851 if (put_user(sizeof(*head), len_ptr))
1852 return -EFAULT;
1853 return put_user(head, head_ptr);
1854
1855err_unlock:
Oleg Nesterovaaa2a972006-09-29 02:00:55 -07001856 rcu_read_unlock();
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001857
1858 return ret;
1859}
1860
1861/*
1862 * Process a futex-list entry, check whether it's owned by the
1863 * dying task, and do notification if so:
1864 */
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001865int handle_futex_death(u32 __user *uaddr, struct task_struct *curr, int pi)
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001866{
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001867 u32 uval, nval, mval;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001868
Ingo Molnar8f17d3a2006-03-27 01:16:27 -08001869retry:
1870 if (get_user(uval, uaddr))
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001871 return -1;
1872
Pavel Emelyanovb4888932007-10-18 23:40:14 -07001873 if ((uval & FUTEX_TID_MASK) == task_pid_vnr(curr)) {
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001874 /*
1875 * Ok, this dying thread is truly holding a futex
1876 * of interest. Set the OWNER_DIED bit atomically
1877 * via cmpxchg, and if the value had FUTEX_WAITERS
1878 * set, wake up a waiter (if any). (We have to do a
1879 * futex_wake() even if OWNER_DIED is already set -
1880 * to handle the rare but possible case of recursive
1881 * thread-death.) The rest of the cleanup is done in
1882 * userspace.
1883 */
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001884 mval = (uval & FUTEX_WAITERS) | FUTEX_OWNER_DIED;
1885 nval = futex_atomic_cmpxchg_inatomic(uaddr, uval, mval);
1886
Ingo Molnarc87e2832006-06-27 02:54:58 -07001887 if (nval == -EFAULT)
1888 return -1;
1889
1890 if (nval != uval)
Ingo Molnar8f17d3a2006-03-27 01:16:27 -08001891 goto retry;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001892
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001893 /*
1894 * Wake robust non-PI futexes here. The wakeup of
1895 * PI futexes happens in exit_pi_state():
1896 */
Thomas Gleixner36cf3b52007-07-15 23:41:20 -07001897 if (!pi && (uval & FUTEX_WAITERS))
Thomas Gleixnercd689982008-02-01 17:45:14 +01001898 futex_wake(uaddr, &curr->mm->mmap_sem, 1,
1899 FUTEX_BITSET_MATCH_ANY);
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001900 }
1901 return 0;
1902}
1903
1904/*
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001905 * Fetch a robust-list pointer. Bit 0 signals PI futexes:
1906 */
1907static inline int fetch_robust_entry(struct robust_list __user **entry,
Al Viroba46df92006-10-10 22:46:07 +01001908 struct robust_list __user * __user *head,
1909 int *pi)
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001910{
1911 unsigned long uentry;
1912
Al Viroba46df92006-10-10 22:46:07 +01001913 if (get_user(uentry, (unsigned long __user *)head))
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001914 return -EFAULT;
1915
Al Viroba46df92006-10-10 22:46:07 +01001916 *entry = (void __user *)(uentry & ~1UL);
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001917 *pi = uentry & 1;
1918
1919 return 0;
1920}
1921
1922/*
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001923 * Walk curr->robust_list (very carefully, it's a userspace list!)
1924 * and mark any locks found there dead, and notify any waiters.
1925 *
1926 * We silently return on any sign of list-walking problem.
1927 */
1928void exit_robust_list(struct task_struct *curr)
1929{
1930 struct robust_list_head __user *head = curr->robust_list;
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001931 struct robust_list __user *entry, *next_entry, *pending;
1932 unsigned int limit = ROBUST_LIST_LIMIT, pi, next_pi, pip;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001933 unsigned long futex_offset;
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001934 int rc;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001935
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08001936 if (!futex_cmpxchg_enabled)
1937 return;
1938
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001939 /*
1940 * Fetch the list head (which was registered earlier, via
1941 * sys_set_robust_list()):
1942 */
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001943 if (fetch_robust_entry(&entry, &head->list.next, &pi))
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001944 return;
1945 /*
1946 * Fetch the relative futex offset:
1947 */
1948 if (get_user(futex_offset, &head->futex_offset))
1949 return;
1950 /*
1951 * Fetch any possibly pending lock-add first, and handle it
1952 * if it exists:
1953 */
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001954 if (fetch_robust_entry(&pending, &head->list_op_pending, &pip))
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001955 return;
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001956
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001957 next_entry = NULL; /* avoid warning with gcc */
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001958 while (entry != &head->list) {
1959 /*
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001960 * Fetch the next entry in the list before calling
1961 * handle_futex_death:
1962 */
1963 rc = fetch_robust_entry(&next_entry, &entry->next, &next_pi);
1964 /*
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001965 * A pending lock might already be on the list, so
Ingo Molnarc87e2832006-06-27 02:54:58 -07001966 * don't process it twice:
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001967 */
1968 if (entry != pending)
Al Viroba46df92006-10-10 22:46:07 +01001969 if (handle_futex_death((void __user *)entry + futex_offset,
Ingo Molnare3f2dde2006-07-29 05:17:57 +02001970 curr, pi))
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001971 return;
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001972 if (rc)
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001973 return;
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001974 entry = next_entry;
1975 pi = next_pi;
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001976 /*
1977 * Avoid excessively long or circular lists:
1978 */
1979 if (!--limit)
1980 break;
1981
1982 cond_resched();
1983 }
Martin Schwidefsky9f96cb12007-10-01 01:20:13 -07001984
1985 if (pending)
1986 handle_futex_death((void __user *)pending + futex_offset,
1987 curr, pip);
Ingo Molnar0771dfe2006-03-27 01:16:22 -08001988}
1989
Pierre Peifferc19384b2007-05-09 02:35:02 -07001990long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout,
Ingo Molnare2970f22006-06-27 02:54:47 -07001991 u32 __user *uaddr2, u32 val2, u32 val3)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001992{
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08001993 int ret = -ENOSYS;
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001994 int cmd = op & FUTEX_CMD_MASK;
1995 struct rw_semaphore *fshared = NULL;
Linus Torvalds1da177e2005-04-16 15:20:36 -07001996
Eric Dumazet34f01cc2007-05-09 02:35:04 -07001997 if (!(op & FUTEX_PRIVATE_FLAG))
1998 fshared = &current->mm->mmap_sem;
1999
2000 switch (cmd) {
Linus Torvalds1da177e2005-04-16 15:20:36 -07002001 case FUTEX_WAIT:
Thomas Gleixnercd689982008-02-01 17:45:14 +01002002 val3 = FUTEX_BITSET_MATCH_ANY;
2003 case FUTEX_WAIT_BITSET:
2004 ret = futex_wait(uaddr, fshared, val, timeout, val3);
Linus Torvalds1da177e2005-04-16 15:20:36 -07002005 break;
2006 case FUTEX_WAKE:
Thomas Gleixnercd689982008-02-01 17:45:14 +01002007 val3 = FUTEX_BITSET_MATCH_ANY;
2008 case FUTEX_WAKE_BITSET:
2009 ret = futex_wake(uaddr, fshared, val, val3);
Linus Torvalds1da177e2005-04-16 15:20:36 -07002010 break;
Linus Torvalds1da177e2005-04-16 15:20:36 -07002011 case FUTEX_REQUEUE:
Eric Dumazet34f01cc2007-05-09 02:35:04 -07002012 ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL);
Linus Torvalds1da177e2005-04-16 15:20:36 -07002013 break;
2014 case FUTEX_CMP_REQUEUE:
Eric Dumazet34f01cc2007-05-09 02:35:04 -07002015 ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3);
Linus Torvalds1da177e2005-04-16 15:20:36 -07002016 break;
Jakub Jelinek4732efb2005-09-06 15:16:25 -07002017 case FUTEX_WAKE_OP:
Eric Dumazet34f01cc2007-05-09 02:35:04 -07002018 ret = futex_wake_op(uaddr, fshared, uaddr2, val, val2, val3);
Jakub Jelinek4732efb2005-09-06 15:16:25 -07002019 break;
Ingo Molnarc87e2832006-06-27 02:54:58 -07002020 case FUTEX_LOCK_PI:
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08002021 if (futex_cmpxchg_enabled)
2022 ret = futex_lock_pi(uaddr, fshared, val, timeout, 0);
Ingo Molnarc87e2832006-06-27 02:54:58 -07002023 break;
2024 case FUTEX_UNLOCK_PI:
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08002025 if (futex_cmpxchg_enabled)
2026 ret = futex_unlock_pi(uaddr, fshared);
Ingo Molnarc87e2832006-06-27 02:54:58 -07002027 break;
2028 case FUTEX_TRYLOCK_PI:
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08002029 if (futex_cmpxchg_enabled)
2030 ret = futex_lock_pi(uaddr, fshared, 0, timeout, 1);
Ingo Molnarc87e2832006-06-27 02:54:58 -07002031 break;
Linus Torvalds1da177e2005-04-16 15:20:36 -07002032 default:
2033 ret = -ENOSYS;
2034 }
2035 return ret;
2036}
2037
2038
Ingo Molnare2970f22006-06-27 02:54:47 -07002039asmlinkage long sys_futex(u32 __user *uaddr, int op, u32 val,
Linus Torvalds1da177e2005-04-16 15:20:36 -07002040 struct timespec __user *utime, u32 __user *uaddr2,
Ingo Molnare2970f22006-06-27 02:54:47 -07002041 u32 val3)
Linus Torvalds1da177e2005-04-16 15:20:36 -07002042{
Pierre Peifferc19384b2007-05-09 02:35:02 -07002043 struct timespec ts;
2044 ktime_t t, *tp = NULL;
Ingo Molnare2970f22006-06-27 02:54:47 -07002045 u32 val2 = 0;
Eric Dumazet34f01cc2007-05-09 02:35:04 -07002046 int cmd = op & FUTEX_CMD_MASK;
Linus Torvalds1da177e2005-04-16 15:20:36 -07002047
Thomas Gleixnercd689982008-02-01 17:45:14 +01002048 if (utime && (cmd == FUTEX_WAIT || cmd == FUTEX_LOCK_PI ||
2049 cmd == FUTEX_WAIT_BITSET)) {
Pierre Peifferc19384b2007-05-09 02:35:02 -07002050 if (copy_from_user(&ts, utime, sizeof(ts)) != 0)
Linus Torvalds1da177e2005-04-16 15:20:36 -07002051 return -EFAULT;
Pierre Peifferc19384b2007-05-09 02:35:02 -07002052 if (!timespec_valid(&ts))
Thomas Gleixner9741ef962006-03-31 02:31:32 -08002053 return -EINVAL;
Pierre Peifferc19384b2007-05-09 02:35:02 -07002054
2055 t = timespec_to_ktime(ts);
Eric Dumazet34f01cc2007-05-09 02:35:04 -07002056 if (cmd == FUTEX_WAIT)
Thomas Gleixner5a7780e2008-02-13 09:20:43 +01002057 t = ktime_add_safe(ktime_get(), t);
Pierre Peifferc19384b2007-05-09 02:35:02 -07002058 tp = &t;
Linus Torvalds1da177e2005-04-16 15:20:36 -07002059 }
2060 /*
Eric Dumazet34f01cc2007-05-09 02:35:04 -07002061 * requeue parameter in 'utime' if cmd == FUTEX_REQUEUE.
Andreas Schwabf54f0982007-07-31 00:38:51 -07002062 * number of waiters to wake in 'utime' if cmd == FUTEX_WAKE_OP.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002063 */
Andreas Schwabf54f0982007-07-31 00:38:51 -07002064 if (cmd == FUTEX_REQUEUE || cmd == FUTEX_CMP_REQUEUE ||
2065 cmd == FUTEX_WAKE_OP)
Ingo Molnare2970f22006-06-27 02:54:47 -07002066 val2 = (u32) (unsigned long) utime;
Linus Torvalds1da177e2005-04-16 15:20:36 -07002067
Pierre Peifferc19384b2007-05-09 02:35:02 -07002068 return do_futex(uaddr, op, val, tp, uaddr2, val2, val3);
Linus Torvalds1da177e2005-04-16 15:20:36 -07002069}
2070
Benjamin Herrenschmidtf6d107f2008-03-27 14:52:15 +11002071static int __init futex_init(void)
Linus Torvalds1da177e2005-04-16 15:20:36 -07002072{
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08002073 u32 curval;
Thomas Gleixner3e4ab742008-02-23 15:23:55 -08002074 int i;
Linus Torvalds1da177e2005-04-16 15:20:36 -07002075
Thomas Gleixnera0c1e902008-02-23 15:23:57 -08002076 /*
2077 * This will fail and we want it. Some arch implementations do
2078 * runtime detection of the futex_atomic_cmpxchg_inatomic()
2079 * functionality. We want to know that before we call in any
2080 * of the complex code paths. Also we want to prevent
2081 * registration of robust lists in that case. NULL is
2082 * guaranteed to fault and we get -EFAULT on functional
2083 * implementation, the non functional ones will return
2084 * -ENOSYS.
2085 */
2086 curval = cmpxchg_futex_value_locked(NULL, 0, 0);
2087 if (curval == -EFAULT)
2088 futex_cmpxchg_enabled = 1;
2089
Thomas Gleixner3e4ab742008-02-23 15:23:55 -08002090 for (i = 0; i < ARRAY_SIZE(futex_queues); i++) {
2091 plist_head_init(&futex_queues[i].chain, &futex_queues[i].lock);
2092 spin_lock_init(&futex_queues[i].lock);
2093 }
2094
Linus Torvalds1da177e2005-04-16 15:20:36 -07002095 return 0;
2096}
Benjamin Herrenschmidtf6d107f2008-03-27 14:52:15 +11002097__initcall(futex_init);