diff options
-rw-r--r-- | kernel/futex.c | 22 | ||||
-rw-r--r-- | kernel/rtmutex-debug.c | 1 | ||||
-rw-r--r-- | kernel/rtmutex.c | 318 | ||||
-rw-r--r-- | kernel/rtmutex_common.h | 16 |
4 files changed, 127 insertions, 230 deletions
diff --git a/kernel/futex.c b/kernel/futex.c index b766d28accd6..64c38115c7b6 100644 --- a/kernel/futex.c +++ b/kernel/futex.c | |||
@@ -1556,10 +1556,10 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, | |||
1556 | 1556 | ||
1557 | /* | 1557 | /* |
1558 | * We are here either because we stole the rtmutex from the | 1558 | * We are here either because we stole the rtmutex from the |
1559 | * pending owner or we are the pending owner which failed to | 1559 | * previous highest priority waiter or we are the highest priority |
1560 | * get the rtmutex. We have to replace the pending owner TID | 1560 | * waiter but failed to get the rtmutex the first time. |
1561 | * in the user space variable. This must be atomic as we have | 1561 | * We have to replace the newowner TID in the user space variable. |
1562 | * to preserve the owner died bit here. | 1562 | * This must be atomic as we have to preserve the owner died bit here. |
1563 | * | 1563 | * |
1564 | * Note: We write the user space value _before_ changing the pi_state | 1564 | * Note: We write the user space value _before_ changing the pi_state |
1565 | * because we can fault here. Imagine swapped out pages or a fork | 1565 | * because we can fault here. Imagine swapped out pages or a fork |
@@ -1608,8 +1608,8 @@ retry: | |||
1608 | 1608 | ||
1609 | /* | 1609 | /* |
1610 | * To handle the page fault we need to drop the hash bucket | 1610 | * To handle the page fault we need to drop the hash bucket |
1611 | * lock here. That gives the other task (either the pending | 1611 | * lock here. That gives the other task (either the highest priority |
1612 | * owner itself or the task which stole the rtmutex) the | 1612 | * waiter itself or the task which stole the rtmutex) the |
1613 | * chance to try the fixup of the pi_state. So once we are | 1613 | * chance to try the fixup of the pi_state. So once we are |
1614 | * back from handling the fault we need to check the pi_state | 1614 | * back from handling the fault we need to check the pi_state |
1615 | * after reacquiring the hash bucket lock and before trying to | 1615 | * after reacquiring the hash bucket lock and before trying to |
@@ -1685,18 +1685,20 @@ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) | |||
1685 | /* | 1685 | /* |
1686 | * pi_state is incorrect, some other task did a lock steal and | 1686 | * pi_state is incorrect, some other task did a lock steal and |
1687 | * we returned due to timeout or signal without taking the | 1687 | * we returned due to timeout or signal without taking the |
1688 | * rt_mutex. Too late. We can access the rt_mutex_owner without | 1688 | * rt_mutex. Too late. |
1689 | * locking, as the other task is now blocked on the hash bucket | ||
1690 | * lock. Fix the state up. | ||
1691 | */ | 1689 | */ |
1690 | raw_spin_lock(&q->pi_state->pi_mutex.wait_lock); | ||
1692 | owner = rt_mutex_owner(&q->pi_state->pi_mutex); | 1691 | owner = rt_mutex_owner(&q->pi_state->pi_mutex); |
1692 | if (!owner) | ||
1693 | owner = rt_mutex_next_owner(&q->pi_state->pi_mutex); | ||
1694 | raw_spin_unlock(&q->pi_state->pi_mutex.wait_lock); | ||
1693 | ret = fixup_pi_state_owner(uaddr, q, owner); | 1695 | ret = fixup_pi_state_owner(uaddr, q, owner); |
1694 | goto out; | 1696 | goto out; |
1695 | } | 1697 | } |
1696 | 1698 | ||
1697 | /* | 1699 | /* |
1698 | * Paranoia check. If we did not take the lock, then we should not be | 1700 | * Paranoia check. If we did not take the lock, then we should not be |
1699 | * the owner, nor the pending owner, of the rt_mutex. | 1701 | * the owner of the rt_mutex. |
1700 | */ | 1702 | */ |
1701 | if (rt_mutex_owner(&q->pi_state->pi_mutex) == current) | 1703 | if (rt_mutex_owner(&q->pi_state->pi_mutex) == current) |
1702 | printk(KERN_ERR "fixup_owner: ret = %d pi-mutex: %p " | 1704 | printk(KERN_ERR "fixup_owner: ret = %d pi-mutex: %p " |
diff --git a/kernel/rtmutex-debug.c b/kernel/rtmutex-debug.c index ddabb54bb5c8..3c7cbc2c33be 100644 --- a/kernel/rtmutex-debug.c +++ b/kernel/rtmutex-debug.c | |||
@@ -215,7 +215,6 @@ void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter) | |||
215 | put_pid(waiter->deadlock_task_pid); | 215 | put_pid(waiter->deadlock_task_pid); |
216 | TRACE_WARN_ON(!plist_node_empty(&waiter->list_entry)); | 216 | TRACE_WARN_ON(!plist_node_empty(&waiter->list_entry)); |
217 | TRACE_WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); | 217 | TRACE_WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); |
218 | TRACE_WARN_ON(waiter->task); | ||
219 | memset(waiter, 0x22, sizeof(*waiter)); | 218 | memset(waiter, 0x22, sizeof(*waiter)); |
220 | } | 219 | } |
221 | 220 | ||
diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index a9604815786a..ab449117aaf2 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c | |||
@@ -20,41 +20,34 @@ | |||
20 | /* | 20 | /* |
21 | * lock->owner state tracking: | 21 | * lock->owner state tracking: |
22 | * | 22 | * |
23 | * lock->owner holds the task_struct pointer of the owner. Bit 0 and 1 | 23 | * lock->owner holds the task_struct pointer of the owner. Bit 0 |
24 | * are used to keep track of the "owner is pending" and "lock has | 24 | * is used to keep track of the "lock has waiters" state. |
25 | * waiters" state. | ||
26 | * | 25 | * |
27 | * owner bit1 bit0 | 26 | * owner bit0 |
28 | * NULL 0 0 lock is free (fast acquire possible) | 27 | * NULL 0 lock is free (fast acquire possible) |
29 | * NULL 0 1 invalid state | 28 | * NULL 1 lock is free and has waiters and the top waiter |
30 | * NULL 1 0 Transitional State* | 29 | * is going to take the lock* |
31 | * NULL 1 1 invalid state | 30 | * taskpointer 0 lock is held (fast release possible) |
32 | * taskpointer 0 0 lock is held (fast release possible) | 31 | * taskpointer 1 lock is held and has waiters** |
33 | * taskpointer 0 1 task is pending owner | ||
34 | * taskpointer 1 0 lock is held and has waiters | ||
35 | * taskpointer 1 1 task is pending owner and lock has more waiters | ||
36 | * | ||
37 | * Pending ownership is assigned to the top (highest priority) | ||
38 | * waiter of the lock, when the lock is released. The thread is woken | ||
39 | * up and can now take the lock. Until the lock is taken (bit 0 | ||
40 | * cleared) a competing higher priority thread can steal the lock | ||
41 | * which puts the woken up thread back on the waiters list. | ||
42 | * | 32 | * |
43 | * The fast atomic compare exchange based acquire and release is only | 33 | * The fast atomic compare exchange based acquire and release is only |
44 | * possible when bit 0 and 1 of lock->owner are 0. | 34 | * possible when bit 0 of lock->owner is 0. |
35 | * | ||
36 | * (*) It also can be a transitional state when grabbing the lock | ||
37 | * with ->wait_lock is held. To prevent any fast path cmpxchg to the lock, | ||
38 | * we need to set the bit0 before looking at the lock, and the owner may be | ||
39 | * NULL in this small time, hence this can be a transitional state. | ||
45 | * | 40 | * |
46 | * (*) There's a small time where the owner can be NULL and the | 41 | * (**) There is a small time when bit 0 is set but there are no |
47 | * "lock has waiters" bit is set. This can happen when grabbing the lock. | 42 | * waiters. This can happen when grabbing the lock in the slow path. |
48 | * To prevent a cmpxchg of the owner releasing the lock, we need to set this | 43 | * To prevent a cmpxchg of the owner releasing the lock, we need to |
49 | * bit before looking at the lock, hence the reason this is a transitional | 44 | * set this bit before looking at the lock. |
50 | * state. | ||
51 | */ | 45 | */ |
52 | 46 | ||
53 | static void | 47 | static void |
54 | rt_mutex_set_owner(struct rt_mutex *lock, struct task_struct *owner, | 48 | rt_mutex_set_owner(struct rt_mutex *lock, struct task_struct *owner) |
55 | unsigned long mask) | ||
56 | { | 49 | { |
57 | unsigned long val = (unsigned long)owner | mask; | 50 | unsigned long val = (unsigned long)owner; |
58 | 51 | ||
59 | if (rt_mutex_has_waiters(lock)) | 52 | if (rt_mutex_has_waiters(lock)) |
60 | val |= RT_MUTEX_HAS_WAITERS; | 53 | val |= RT_MUTEX_HAS_WAITERS; |
@@ -203,15 +196,14 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, | |||
203 | * reached or the state of the chain has changed while we | 196 | * reached or the state of the chain has changed while we |
204 | * dropped the locks. | 197 | * dropped the locks. |
205 | */ | 198 | */ |
206 | if (!waiter || !waiter->task) | 199 | if (!waiter) |
207 | goto out_unlock_pi; | 200 | goto out_unlock_pi; |
208 | 201 | ||
209 | /* | 202 | /* |
210 | * Check the orig_waiter state. After we dropped the locks, | 203 | * Check the orig_waiter state. After we dropped the locks, |
211 | * the previous owner of the lock might have released the lock | 204 | * the previous owner of the lock might have released the lock. |
212 | * and made us the pending owner: | ||
213 | */ | 205 | */ |
214 | if (orig_waiter && !orig_waiter->task) | 206 | if (orig_waiter && !rt_mutex_owner(orig_lock)) |
215 | goto out_unlock_pi; | 207 | goto out_unlock_pi; |
216 | 208 | ||
217 | /* | 209 | /* |
@@ -254,6 +246,17 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, | |||
254 | 246 | ||
255 | /* Release the task */ | 247 | /* Release the task */ |
256 | raw_spin_unlock_irqrestore(&task->pi_lock, flags); | 248 | raw_spin_unlock_irqrestore(&task->pi_lock, flags); |
249 | if (!rt_mutex_owner(lock)) { | ||
250 | /* | ||
251 | * If the requeue above changed the top waiter, then we need | ||
252 | * to wake the new top waiter up to try to get the lock. | ||
253 | */ | ||
254 | |||
255 | if (top_waiter != rt_mutex_top_waiter(lock)) | ||
256 | wake_up_process(rt_mutex_top_waiter(lock)->task); | ||
257 | raw_spin_unlock(&lock->wait_lock); | ||
258 | goto out_put_task; | ||
259 | } | ||
257 | put_task_struct(task); | 260 | put_task_struct(task); |
258 | 261 | ||
259 | /* Grab the next task */ | 262 | /* Grab the next task */ |
@@ -296,78 +299,16 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, | |||
296 | } | 299 | } |
297 | 300 | ||
298 | /* | 301 | /* |
299 | * Optimization: check if we can steal the lock from the | ||
300 | * assigned pending owner [which might not have taken the | ||
301 | * lock yet]: | ||
302 | */ | ||
303 | static inline int try_to_steal_lock(struct rt_mutex *lock, | ||
304 | struct task_struct *task) | ||
305 | { | ||
306 | struct task_struct *pendowner = rt_mutex_owner(lock); | ||
307 | struct rt_mutex_waiter *next; | ||
308 | unsigned long flags; | ||
309 | |||
310 | if (!rt_mutex_owner_pending(lock)) | ||
311 | return 0; | ||
312 | |||
313 | if (pendowner == task) | ||
314 | return 1; | ||
315 | |||
316 | raw_spin_lock_irqsave(&pendowner->pi_lock, flags); | ||
317 | if (task->prio >= pendowner->prio) { | ||
318 | raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); | ||
319 | return 0; | ||
320 | } | ||
321 | |||
322 | /* | ||
323 | * Check if a waiter is enqueued on the pending owners | ||
324 | * pi_waiters list. Remove it and readjust pending owners | ||
325 | * priority. | ||
326 | */ | ||
327 | if (likely(!rt_mutex_has_waiters(lock))) { | ||
328 | raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); | ||
329 | return 1; | ||
330 | } | ||
331 | |||
332 | /* No chain handling, pending owner is not blocked on anything: */ | ||
333 | next = rt_mutex_top_waiter(lock); | ||
334 | plist_del(&next->pi_list_entry, &pendowner->pi_waiters); | ||
335 | __rt_mutex_adjust_prio(pendowner); | ||
336 | raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); | ||
337 | |||
338 | /* | ||
339 | * We are going to steal the lock and a waiter was | ||
340 | * enqueued on the pending owners pi_waiters queue. So | ||
341 | * we have to enqueue this waiter into | ||
342 | * task->pi_waiters list. This covers the case, | ||
343 | * where task is boosted because it holds another | ||
344 | * lock and gets unboosted because the booster is | ||
345 | * interrupted, so we would delay a waiter with higher | ||
346 | * priority as task->normal_prio. | ||
347 | * | ||
348 | * Note: in the rare case of a SCHED_OTHER task changing | ||
349 | * its priority and thus stealing the lock, next->task | ||
350 | * might be task: | ||
351 | */ | ||
352 | if (likely(next->task != task)) { | ||
353 | raw_spin_lock_irqsave(&task->pi_lock, flags); | ||
354 | plist_add(&next->pi_list_entry, &task->pi_waiters); | ||
355 | __rt_mutex_adjust_prio(task); | ||
356 | raw_spin_unlock_irqrestore(&task->pi_lock, flags); | ||
357 | } | ||
358 | return 1; | ||
359 | } | ||
360 | |||
361 | /* | ||
362 | * Try to take an rt-mutex | 302 | * Try to take an rt-mutex |
363 | * | 303 | * |
364 | * This fails | ||
365 | * - when the lock has a real owner | ||
366 | * - when a different pending owner exists and has higher priority than current | ||
367 | * | ||
368 | * Must be called with lock->wait_lock held. | 304 | * Must be called with lock->wait_lock held. |
305 | * | ||
306 | * @lock: the lock to be acquired. | ||
307 | * @task: the task which wants to acquire the lock | ||
308 | * @waiter: the waiter that is queued to the lock's wait list. (could be NULL) | ||
369 | */ | 309 | */ |
370 | static int try_to_take_rt_mutex(struct rt_mutex *lock) | 310 | static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, |
311 | struct rt_mutex_waiter *waiter) | ||
371 | { | 312 | { |
372 | /* | 313 | /* |
373 | * We have to be careful here if the atomic speedups are | 314 | * We have to be careful here if the atomic speedups are |
@@ -390,15 +331,52 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock) | |||
390 | */ | 331 | */ |
391 | mark_rt_mutex_waiters(lock); | 332 | mark_rt_mutex_waiters(lock); |
392 | 333 | ||
393 | if (rt_mutex_owner(lock) && !try_to_steal_lock(lock, current)) | 334 | if (rt_mutex_owner(lock)) |
394 | return 0; | 335 | return 0; |
395 | 336 | ||
337 | /* | ||
338 | * It will get the lock because of one of these conditions: | ||
339 | * 1) there is no waiter | ||
340 | * 2) higher priority than waiters | ||
341 | * 3) it is top waiter | ||
342 | */ | ||
343 | if (rt_mutex_has_waiters(lock)) { | ||
344 | if (task->prio >= rt_mutex_top_waiter(lock)->list_entry.prio) { | ||
345 | if (!waiter || waiter != rt_mutex_top_waiter(lock)) | ||
346 | return 0; | ||
347 | } | ||
348 | } | ||
349 | |||
350 | if (waiter || rt_mutex_has_waiters(lock)) { | ||
351 | unsigned long flags; | ||
352 | struct rt_mutex_waiter *top; | ||
353 | |||
354 | raw_spin_lock_irqsave(&task->pi_lock, flags); | ||
355 | |||
356 | /* remove the queued waiter. */ | ||
357 | if (waiter) { | ||
358 | plist_del(&waiter->list_entry, &lock->wait_list); | ||
359 | task->pi_blocked_on = NULL; | ||
360 | } | ||
361 | |||
362 | /* | ||
363 | * We have to enqueue the top waiter(if it exists) into | ||
364 | * task->pi_waiters list. | ||
365 | */ | ||
366 | if (rt_mutex_has_waiters(lock)) { | ||
367 | top = rt_mutex_top_waiter(lock); | ||
368 | top->pi_list_entry.prio = top->list_entry.prio; | ||
369 | plist_add(&top->pi_list_entry, &task->pi_waiters); | ||
370 | } | ||
371 | raw_spin_unlock_irqrestore(&task->pi_lock, flags); | ||
372 | } | ||
373 | |||
396 | /* We got the lock. */ | 374 | /* We got the lock. */ |
397 | debug_rt_mutex_lock(lock); | 375 | debug_rt_mutex_lock(lock); |
398 | 376 | ||
399 | rt_mutex_set_owner(lock, current, 0); | 377 | rt_mutex_set_owner(lock, task); |
400 | 378 | ||
401 | rt_mutex_deadlock_account_lock(lock, current); | 379 | rt_mutex_deadlock_account_lock(lock, task); |
402 | 380 | ||
403 | return 1; | 381 | return 1; |
404 | } | 382 | } |
@@ -436,6 +414,9 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, | |||
436 | 414 | ||
437 | raw_spin_unlock_irqrestore(&task->pi_lock, flags); | 415 | raw_spin_unlock_irqrestore(&task->pi_lock, flags); |
438 | 416 | ||
417 | if (!owner) | ||
418 | return 0; | ||
419 | |||
439 | if (waiter == rt_mutex_top_waiter(lock)) { | 420 | if (waiter == rt_mutex_top_waiter(lock)) { |
440 | raw_spin_lock_irqsave(&owner->pi_lock, flags); | 421 | raw_spin_lock_irqsave(&owner->pi_lock, flags); |
441 | plist_del(&top_waiter->pi_list_entry, &owner->pi_waiters); | 422 | plist_del(&top_waiter->pi_list_entry, &owner->pi_waiters); |
@@ -472,21 +453,18 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, | |||
472 | /* | 453 | /* |
473 | * Wake up the next waiter on the lock. | 454 | * Wake up the next waiter on the lock. |
474 | * | 455 | * |
475 | * Remove the top waiter from the current tasks waiter list and from | 456 | * Remove the top waiter from the current tasks waiter list and wake it up. |
476 | * the lock waiter list. Set it as pending owner. Then wake it up. | ||
477 | * | 457 | * |
478 | * Called with lock->wait_lock held. | 458 | * Called with lock->wait_lock held. |
479 | */ | 459 | */ |
480 | static void wakeup_next_waiter(struct rt_mutex *lock) | 460 | static void wakeup_next_waiter(struct rt_mutex *lock) |
481 | { | 461 | { |
482 | struct rt_mutex_waiter *waiter; | 462 | struct rt_mutex_waiter *waiter; |
483 | struct task_struct *pendowner; | ||
484 | unsigned long flags; | 463 | unsigned long flags; |
485 | 464 | ||
486 | raw_spin_lock_irqsave(¤t->pi_lock, flags); | 465 | raw_spin_lock_irqsave(¤t->pi_lock, flags); |
487 | 466 | ||
488 | waiter = rt_mutex_top_waiter(lock); | 467 | waiter = rt_mutex_top_waiter(lock); |
489 | plist_del(&waiter->list_entry, &lock->wait_list); | ||
490 | 468 | ||
491 | /* | 469 | /* |
492 | * Remove it from current->pi_waiters. We do not adjust a | 470 | * Remove it from current->pi_waiters. We do not adjust a |
@@ -495,43 +473,19 @@ static void wakeup_next_waiter(struct rt_mutex *lock) | |||
495 | * lock->wait_lock. | 473 | * lock->wait_lock. |
496 | */ | 474 | */ |
497 | plist_del(&waiter->pi_list_entry, ¤t->pi_waiters); | 475 | plist_del(&waiter->pi_list_entry, ¤t->pi_waiters); |
498 | pendowner = waiter->task; | ||
499 | waiter->task = NULL; | ||
500 | 476 | ||
501 | rt_mutex_set_owner(lock, pendowner, RT_MUTEX_OWNER_PENDING); | 477 | rt_mutex_set_owner(lock, NULL); |
502 | 478 | ||
503 | raw_spin_unlock_irqrestore(¤t->pi_lock, flags); | 479 | raw_spin_unlock_irqrestore(¤t->pi_lock, flags); |
504 | 480 | ||
505 | /* | 481 | wake_up_process(waiter->task); |
506 | * Clear the pi_blocked_on variable and enqueue a possible | ||
507 | * waiter into the pi_waiters list of the pending owner. This | ||
508 | * prevents that in case the pending owner gets unboosted a | ||
509 | * waiter with higher priority than pending-owner->normal_prio | ||
510 | * is blocked on the unboosted (pending) owner. | ||
511 | */ | ||
512 | raw_spin_lock_irqsave(&pendowner->pi_lock, flags); | ||
513 | |||
514 | WARN_ON(!pendowner->pi_blocked_on); | ||
515 | WARN_ON(pendowner->pi_blocked_on != waiter); | ||
516 | WARN_ON(pendowner->pi_blocked_on->lock != lock); | ||
517 | |||
518 | pendowner->pi_blocked_on = NULL; | ||
519 | |||
520 | if (rt_mutex_has_waiters(lock)) { | ||
521 | struct rt_mutex_waiter *next; | ||
522 | |||
523 | next = rt_mutex_top_waiter(lock); | ||
524 | plist_add(&next->pi_list_entry, &pendowner->pi_waiters); | ||
525 | } | ||
526 | raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); | ||
527 | |||
528 | wake_up_process(pendowner); | ||
529 | } | 482 | } |
530 | 483 | ||
531 | /* | 484 | /* |
532 | * Remove a waiter from a lock | 485 | * Remove a waiter from a lock and give up |
533 | * | 486 | * |
534 | * Must be called with lock->wait_lock held | 487 | * Must be called with lock->wait_lock held and |
488 | * have just failed to try_to_take_rt_mutex(). | ||
535 | */ | 489 | */ |
536 | static void remove_waiter(struct rt_mutex *lock, | 490 | static void remove_waiter(struct rt_mutex *lock, |
537 | struct rt_mutex_waiter *waiter) | 491 | struct rt_mutex_waiter *waiter) |
@@ -543,11 +497,13 @@ static void remove_waiter(struct rt_mutex *lock, | |||
543 | 497 | ||
544 | raw_spin_lock_irqsave(¤t->pi_lock, flags); | 498 | raw_spin_lock_irqsave(¤t->pi_lock, flags); |
545 | plist_del(&waiter->list_entry, &lock->wait_list); | 499 | plist_del(&waiter->list_entry, &lock->wait_list); |
546 | waiter->task = NULL; | ||
547 | current->pi_blocked_on = NULL; | 500 | current->pi_blocked_on = NULL; |
548 | raw_spin_unlock_irqrestore(¤t->pi_lock, flags); | 501 | raw_spin_unlock_irqrestore(¤t->pi_lock, flags); |
549 | 502 | ||
550 | if (first && owner != current) { | 503 | if (!owner) |
504 | return; | ||
505 | |||
506 | if (first) { | ||
551 | 507 | ||
552 | raw_spin_lock_irqsave(&owner->pi_lock, flags); | 508 | raw_spin_lock_irqsave(&owner->pi_lock, flags); |
553 | 509 | ||
@@ -614,21 +570,19 @@ void rt_mutex_adjust_pi(struct task_struct *task) | |||
614 | * or TASK_UNINTERRUPTIBLE) | 570 | * or TASK_UNINTERRUPTIBLE) |
615 | * @timeout: the pre-initialized and started timer, or NULL for none | 571 | * @timeout: the pre-initialized and started timer, or NULL for none |
616 | * @waiter: the pre-initialized rt_mutex_waiter | 572 | * @waiter: the pre-initialized rt_mutex_waiter |
617 | * @detect_deadlock: passed to task_blocks_on_rt_mutex | ||
618 | * | 573 | * |
619 | * lock->wait_lock must be held by the caller. | 574 | * lock->wait_lock must be held by the caller. |
620 | */ | 575 | */ |
621 | static int __sched | 576 | static int __sched |
622 | __rt_mutex_slowlock(struct rt_mutex *lock, int state, | 577 | __rt_mutex_slowlock(struct rt_mutex *lock, int state, |
623 | struct hrtimer_sleeper *timeout, | 578 | struct hrtimer_sleeper *timeout, |
624 | struct rt_mutex_waiter *waiter, | 579 | struct rt_mutex_waiter *waiter) |
625 | int detect_deadlock) | ||
626 | { | 580 | { |
627 | int ret = 0; | 581 | int ret = 0; |
628 | 582 | ||
629 | for (;;) { | 583 | for (;;) { |
630 | /* Try to acquire the lock: */ | 584 | /* Try to acquire the lock: */ |
631 | if (try_to_take_rt_mutex(lock)) | 585 | if (try_to_take_rt_mutex(lock, current, waiter)) |
632 | break; | 586 | break; |
633 | 587 | ||
634 | /* | 588 | /* |
@@ -645,39 +599,11 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state, | |||
645 | break; | 599 | break; |
646 | } | 600 | } |
647 | 601 | ||
648 | /* | ||
649 | * waiter->task is NULL the first time we come here and | ||
650 | * when we have been woken up by the previous owner | ||
651 | * but the lock got stolen by a higher prio task. | ||
652 | */ | ||
653 | if (!waiter->task) { | ||
654 | ret = task_blocks_on_rt_mutex(lock, waiter, current, | ||
655 | detect_deadlock); | ||
656 | /* | ||
657 | * If we got woken up by the owner then start loop | ||
658 | * all over without going into schedule to try | ||
659 | * to get the lock now: | ||
660 | */ | ||
661 | if (unlikely(!waiter->task)) { | ||
662 | /* | ||
663 | * Reset the return value. We might | ||
664 | * have returned with -EDEADLK and the | ||
665 | * owner released the lock while we | ||
666 | * were walking the pi chain. | ||
667 | */ | ||
668 | ret = 0; | ||
669 | continue; | ||
670 | } | ||
671 | if (unlikely(ret)) | ||
672 | break; | ||
673 | } | ||
674 | |||
675 | raw_spin_unlock(&lock->wait_lock); | 602 | raw_spin_unlock(&lock->wait_lock); |
676 | 603 | ||
677 | debug_rt_mutex_print_deadlock(waiter); | 604 | debug_rt_mutex_print_deadlock(waiter); |
678 | 605 | ||
679 | if (waiter->task) | 606 | schedule_rt_mutex(lock); |
680 | schedule_rt_mutex(lock); | ||
681 | 607 | ||
682 | raw_spin_lock(&lock->wait_lock); | 608 | raw_spin_lock(&lock->wait_lock); |
683 | set_current_state(state); | 609 | set_current_state(state); |
@@ -698,12 +624,11 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, | |||
698 | int ret = 0; | 624 | int ret = 0; |
699 | 625 | ||
700 | debug_rt_mutex_init_waiter(&waiter); | 626 | debug_rt_mutex_init_waiter(&waiter); |
701 | waiter.task = NULL; | ||
702 | 627 | ||
703 | raw_spin_lock(&lock->wait_lock); | 628 | raw_spin_lock(&lock->wait_lock); |
704 | 629 | ||
705 | /* Try to acquire the lock again: */ | 630 | /* Try to acquire the lock again: */ |
706 | if (try_to_take_rt_mutex(lock)) { | 631 | if (try_to_take_rt_mutex(lock, current, NULL)) { |
707 | raw_spin_unlock(&lock->wait_lock); | 632 | raw_spin_unlock(&lock->wait_lock); |
708 | return 0; | 633 | return 0; |
709 | } | 634 | } |
@@ -717,12 +642,14 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, | |||
717 | timeout->task = NULL; | 642 | timeout->task = NULL; |
718 | } | 643 | } |
719 | 644 | ||
720 | ret = __rt_mutex_slowlock(lock, state, timeout, &waiter, | 645 | ret = task_blocks_on_rt_mutex(lock, &waiter, current, detect_deadlock); |
721 | detect_deadlock); | 646 | |
647 | if (likely(!ret)) | ||
648 | ret = __rt_mutex_slowlock(lock, state, timeout, &waiter); | ||
722 | 649 | ||
723 | set_current_state(TASK_RUNNING); | 650 | set_current_state(TASK_RUNNING); |
724 | 651 | ||
725 | if (unlikely(waiter.task)) | 652 | if (unlikely(ret)) |
726 | remove_waiter(lock, &waiter); | 653 | remove_waiter(lock, &waiter); |
727 | 654 | ||
728 | /* | 655 | /* |
@@ -737,14 +664,6 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, | |||
737 | if (unlikely(timeout)) | 664 | if (unlikely(timeout)) |
738 | hrtimer_cancel(&timeout->timer); | 665 | hrtimer_cancel(&timeout->timer); |
739 | 666 | ||
740 | /* | ||
741 | * Readjust priority, when we did not get the lock. We might | ||
742 | * have been the pending owner and boosted. Since we did not | ||
743 | * take the lock, the PI boost has to go. | ||
744 | */ | ||
745 | if (unlikely(ret)) | ||
746 | rt_mutex_adjust_prio(current); | ||
747 | |||
748 | debug_rt_mutex_free_waiter(&waiter); | 667 | debug_rt_mutex_free_waiter(&waiter); |
749 | 668 | ||
750 | return ret; | 669 | return ret; |
@@ -762,7 +681,7 @@ rt_mutex_slowtrylock(struct rt_mutex *lock) | |||
762 | 681 | ||
763 | if (likely(rt_mutex_owner(lock) != current)) { | 682 | if (likely(rt_mutex_owner(lock) != current)) { |
764 | 683 | ||
765 | ret = try_to_take_rt_mutex(lock); | 684 | ret = try_to_take_rt_mutex(lock, current, NULL); |
766 | /* | 685 | /* |
767 | * try_to_take_rt_mutex() sets the lock waiters | 686 | * try_to_take_rt_mutex() sets the lock waiters |
768 | * bit unconditionally. Clean this up. | 687 | * bit unconditionally. Clean this up. |
@@ -992,7 +911,7 @@ void rt_mutex_init_proxy_locked(struct rt_mutex *lock, | |||
992 | { | 911 | { |
993 | __rt_mutex_init(lock, NULL); | 912 | __rt_mutex_init(lock, NULL); |
994 | debug_rt_mutex_proxy_lock(lock, proxy_owner); | 913 | debug_rt_mutex_proxy_lock(lock, proxy_owner); |
995 | rt_mutex_set_owner(lock, proxy_owner, 0); | 914 | rt_mutex_set_owner(lock, proxy_owner); |
996 | rt_mutex_deadlock_account_lock(lock, proxy_owner); | 915 | rt_mutex_deadlock_account_lock(lock, proxy_owner); |
997 | } | 916 | } |
998 | 917 | ||
@@ -1008,7 +927,7 @@ void rt_mutex_proxy_unlock(struct rt_mutex *lock, | |||
1008 | struct task_struct *proxy_owner) | 927 | struct task_struct *proxy_owner) |
1009 | { | 928 | { |
1010 | debug_rt_mutex_proxy_unlock(lock); | 929 | debug_rt_mutex_proxy_unlock(lock); |
1011 | rt_mutex_set_owner(lock, NULL, 0); | 930 | rt_mutex_set_owner(lock, NULL); |
1012 | rt_mutex_deadlock_account_unlock(proxy_owner); | 931 | rt_mutex_deadlock_account_unlock(proxy_owner); |
1013 | } | 932 | } |
1014 | 933 | ||
@@ -1034,20 +953,14 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock, | |||
1034 | 953 | ||
1035 | raw_spin_lock(&lock->wait_lock); | 954 | raw_spin_lock(&lock->wait_lock); |
1036 | 955 | ||
1037 | mark_rt_mutex_waiters(lock); | 956 | if (try_to_take_rt_mutex(lock, task, NULL)) { |
1038 | |||
1039 | if (!rt_mutex_owner(lock) || try_to_steal_lock(lock, task)) { | ||
1040 | /* We got the lock for task. */ | ||
1041 | debug_rt_mutex_lock(lock); | ||
1042 | rt_mutex_set_owner(lock, task, 0); | ||
1043 | raw_spin_unlock(&lock->wait_lock); | 957 | raw_spin_unlock(&lock->wait_lock); |
1044 | rt_mutex_deadlock_account_lock(lock, task); | ||
1045 | return 1; | 958 | return 1; |
1046 | } | 959 | } |
1047 | 960 | ||
1048 | ret = task_blocks_on_rt_mutex(lock, waiter, task, detect_deadlock); | 961 | ret = task_blocks_on_rt_mutex(lock, waiter, task, detect_deadlock); |
1049 | 962 | ||
1050 | if (ret && !waiter->task) { | 963 | if (ret && !rt_mutex_owner(lock)) { |
1051 | /* | 964 | /* |
1052 | * Reset the return value. We might have | 965 | * Reset the return value. We might have |
1053 | * returned with -EDEADLK and the owner | 966 | * returned with -EDEADLK and the owner |
@@ -1056,6 +969,10 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock, | |||
1056 | */ | 969 | */ |
1057 | ret = 0; | 970 | ret = 0; |
1058 | } | 971 | } |
972 | |||
973 | if (unlikely(ret)) | ||
974 | remove_waiter(lock, waiter); | ||
975 | |||
1059 | raw_spin_unlock(&lock->wait_lock); | 976 | raw_spin_unlock(&lock->wait_lock); |
1060 | 977 | ||
1061 | debug_rt_mutex_print_deadlock(waiter); | 978 | debug_rt_mutex_print_deadlock(waiter); |
@@ -1110,12 +1027,11 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, | |||
1110 | 1027 | ||
1111 | set_current_state(TASK_INTERRUPTIBLE); | 1028 | set_current_state(TASK_INTERRUPTIBLE); |
1112 | 1029 | ||
1113 | ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter, | 1030 | ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter); |
1114 | detect_deadlock); | ||
1115 | 1031 | ||
1116 | set_current_state(TASK_RUNNING); | 1032 | set_current_state(TASK_RUNNING); |
1117 | 1033 | ||
1118 | if (unlikely(waiter->task)) | 1034 | if (unlikely(ret)) |
1119 | remove_waiter(lock, waiter); | 1035 | remove_waiter(lock, waiter); |
1120 | 1036 | ||
1121 | /* | 1037 | /* |
@@ -1126,13 +1042,5 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, | |||
1126 | 1042 | ||
1127 | raw_spin_unlock(&lock->wait_lock); | 1043 | raw_spin_unlock(&lock->wait_lock); |
1128 | 1044 | ||
1129 | /* | ||
1130 | * Readjust priority, when we did not get the lock. We might have been | ||
1131 | * the pending owner and boosted. Since we did not take the lock, the | ||
1132 | * PI boost has to go. | ||
1133 | */ | ||
1134 | if (unlikely(ret)) | ||
1135 | rt_mutex_adjust_prio(current); | ||
1136 | |||
1137 | return ret; | 1045 | return ret; |
1138 | } | 1046 | } |
diff --git a/kernel/rtmutex_common.h b/kernel/rtmutex_common.h index 97a2f81866af..53a66c85261b 100644 --- a/kernel/rtmutex_common.h +++ b/kernel/rtmutex_common.h | |||
@@ -91,9 +91,8 @@ task_top_pi_waiter(struct task_struct *p) | |||
91 | /* | 91 | /* |
92 | * lock->owner state tracking: | 92 | * lock->owner state tracking: |
93 | */ | 93 | */ |
94 | #define RT_MUTEX_OWNER_PENDING 1UL | 94 | #define RT_MUTEX_HAS_WAITERS 1UL |
95 | #define RT_MUTEX_HAS_WAITERS 2UL | 95 | #define RT_MUTEX_OWNER_MASKALL 1UL |
96 | #define RT_MUTEX_OWNER_MASKALL 3UL | ||
97 | 96 | ||
98 | static inline struct task_struct *rt_mutex_owner(struct rt_mutex *lock) | 97 | static inline struct task_struct *rt_mutex_owner(struct rt_mutex *lock) |
99 | { | 98 | { |
@@ -101,17 +100,6 @@ static inline struct task_struct *rt_mutex_owner(struct rt_mutex *lock) | |||
101 | ((unsigned long)lock->owner & ~RT_MUTEX_OWNER_MASKALL); | 100 | ((unsigned long)lock->owner & ~RT_MUTEX_OWNER_MASKALL); |
102 | } | 101 | } |
103 | 102 | ||
104 | static inline struct task_struct *rt_mutex_real_owner(struct rt_mutex *lock) | ||
105 | { | ||
106 | return (struct task_struct *) | ||
107 | ((unsigned long)lock->owner & ~RT_MUTEX_HAS_WAITERS); | ||
108 | } | ||
109 | |||
110 | static inline unsigned long rt_mutex_owner_pending(struct rt_mutex *lock) | ||
111 | { | ||
112 | return (unsigned long)lock->owner & RT_MUTEX_OWNER_PENDING; | ||
113 | } | ||
114 | |||
115 | /* | 103 | /* |
116 | * PI-futex support (proxy locking functions, etc.): | 104 | * PI-futex support (proxy locking functions, etc.): |
117 | */ | 105 | */ |