rtmutex.c 29.4 KB
Newer Older
1 2 3 4 5 6 7 8 9
/*
 * RT-Mutexes: simple blocking mutual exclusion locks with PI support
 *
 * started by Ingo Molnar and Thomas Gleixner.
 *
 *  Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
 *  Copyright (C) 2005-2006 Timesys Corp., Thomas Gleixner <tglx@timesys.com>
 *  Copyright (C) 2005 Kihon Technologies Inc., Steven Rostedt
 *  Copyright (C) 2006 Esben Nielsen
10 11
 *
 *  See Documentation/rt-mutex-design.txt for details.
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
 */
#include <linux/spinlock.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/timer.h>

#include "rtmutex_common.h"

/*
 * lock->owner state tracking:
 *
 * lock->owner holds the task_struct pointer of the owner. Bit 0 and 1
 * are used to keep track of the "owner is pending" and "lock has
 * waiters" state.
 *
 * owner	bit1	bit0
 * NULL		0	0	lock is free (fast acquire possible)
 * NULL		0	1	invalid state
 * NULL		1	0	Transitional State*
 * NULL		1	1	invalid state
 * taskpointer	0	0	lock is held (fast release possible)
 * taskpointer	0	1	task is pending owner
 * taskpointer	1	0	lock is held and has waiters
 * taskpointer	1	1	task is pending owner and lock has more waiters
 *
 * Pending ownership is assigned to the top (highest priority)
 * waiter of the lock, when the lock is released. The thread is woken
 * up and can now take the lock. Until the lock is taken (bit 0
 * cleared) a competing higher priority thread can steal the lock
 * which puts the woken up thread back on the waiters list.
 *
 * The fast atomic compare exchange based acquire and release is only
 * possible when bit 0 and 1 of lock->owner are 0.
 *
 * (*) There's a small time where the owner can be NULL and the
 * "lock has waiters" bit is set.  This can happen when grabbing the lock.
 * To prevent a cmpxchg of the owner releasing the lock, we need to set this
 * bit before looking at the lock, hence the reason this is a transitional
 * state.
 */

53
static void
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
rt_mutex_set_owner(struct rt_mutex *lock, struct task_struct *owner,
		   unsigned long mask)
{
	unsigned long val = (unsigned long)owner | mask;

	if (rt_mutex_has_waiters(lock))
		val |= RT_MUTEX_HAS_WAITERS;

	lock->owner = (struct task_struct *)val;
}

static inline void clear_rt_mutex_waiters(struct rt_mutex *lock)
{
	lock->owner = (struct task_struct *)
			((unsigned long)lock->owner & ~RT_MUTEX_HAS_WAITERS);
}

static void fixup_rt_mutex_waiters(struct rt_mutex *lock)
{
	if (!rt_mutex_has_waiters(lock))
		clear_rt_mutex_waiters(lock);
}

77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
/*
 * We can speed up the acquire/release, if the architecture
 * supports cmpxchg and if there's no debugging state to be set up
 */
#if defined(__HAVE_ARCH_CMPXCHG) && !defined(CONFIG_DEBUG_RT_MUTEXES)
# define rt_mutex_cmpxchg(l,c,n)	(cmpxchg(&l->owner, c, n) == c)
static inline void mark_rt_mutex_waiters(struct rt_mutex *lock)
{
	unsigned long owner, *p = (unsigned long *) &lock->owner;

	do {
		owner = *p;
	} while (cmpxchg(p, owner, owner | RT_MUTEX_HAS_WAITERS) != owner);
}
#else
# define rt_mutex_cmpxchg(l,c,n)	(0)
static inline void mark_rt_mutex_waiters(struct rt_mutex *lock)
{
	lock->owner = (struct task_struct *)
			((unsigned long)lock->owner | RT_MUTEX_HAS_WAITERS);
}
#endif

100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
/*
 * Calculate task priority from the waiter list priority
 *
 * Return task->normal_prio when the waiter list is empty or when
 * the waiter is not allowed to do priority boosting
 */
int rt_mutex_getprio(struct task_struct *task)
{
	if (likely(!task_has_pi_waiters(task)))
		return task->normal_prio;

	return min(task_top_pi_waiter(task)->pi_list_entry.prio,
		   task->normal_prio);
}

/*
 * Adjust the priority of a task, after its pi_waiters got modified.
 *
 * This can be both boosting and unboosting. task->pi_lock must be held.
 */
120
static void __rt_mutex_adjust_prio(struct task_struct *task)
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
{
	int prio = rt_mutex_getprio(task);

	if (task->prio != prio)
		rt_mutex_setprio(task, prio);
}

/*
 * Adjust task priority (undo boosting). Called from the exit path of
 * rt_mutex_slowunlock() and rt_mutex_slowlock().
 *
 * (Note: We do this outside of the protection of lock->wait_lock to
 * allow the lock to be taken while or before we readjust the priority
 * of task. We do not use the spin_xx_mutex() variants here as we are
 * outside of the debug path.)
 */
static void rt_mutex_adjust_prio(struct task_struct *task)
{
	unsigned long flags;

141
	raw_spin_lock_irqsave(&task->pi_lock, flags);
142
	__rt_mutex_adjust_prio(task);
143
	raw_spin_unlock_irqrestore(&task->pi_lock, flags);
144 145 146 147 148 149 150 151 152 153 154 155
}

/*
 * Max number of times we'll walk the boosting chain:
 */
int max_lock_depth = 1024;

/*
 * Adjust the priority chain. Also used for deadlock detection.
 * Decreases task's usage by one - may thus free the task.
 * Returns 0 or -EDEADLK.
 */
156 157 158 159 160
static int rt_mutex_adjust_prio_chain(struct task_struct *task,
				      int deadlock_detect,
				      struct rt_mutex *orig_lock,
				      struct rt_mutex_waiter *orig_waiter,
				      struct task_struct *top_task)
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
{
	struct rt_mutex *lock;
	struct rt_mutex_waiter *waiter, *top_waiter = orig_waiter;
	int detect_deadlock, ret = 0, depth = 0;
	unsigned long flags;

	detect_deadlock = debug_rt_mutex_detect_deadlock(orig_waiter,
							 deadlock_detect);

	/*
	 * The (de)boosting is a step by step approach with a lot of
	 * pitfalls. We want this to be preemptible and we want hold a
	 * maximum of two locks per step. So we have to check
	 * carefully whether things change under us.
	 */
 again:
	if (++depth > max_lock_depth) {
		static int prev_max;

		/*
		 * Print this only once. If the admin changes the limit,
		 * print a new message when reaching the limit again.
		 */
		if (prev_max != max_lock_depth) {
			prev_max = max_lock_depth;
			printk(KERN_WARNING "Maximum lock depth %d reached "
			       "task: %s (%d)\n", max_lock_depth,
188
			       top_task->comm, task_pid_nr(top_task));
189 190 191 192 193 194 195 196 197
		}
		put_task_struct(task);

		return deadlock_detect ? -EDEADLK : 0;
	}
 retry:
	/*
	 * Task can not go away as we did a get_task() before !
	 */
198
	raw_spin_lock_irqsave(&task->pi_lock, flags);
199 200 201 202 203 204 205 206 207 208

	waiter = task->pi_blocked_on;
	/*
	 * Check whether the end of the boosting chain has been
	 * reached or the state of the chain has changed while we
	 * dropped the locks.
	 */
	if (!waiter || !waiter->task)
		goto out_unlock_pi;

209 210 211 212 213 214 215 216 217 218 219 220 221
	/*
	 * Check the orig_waiter state. After we dropped the locks,
	 * the previous owner of the lock might have released the lock
	 * and made us the pending owner:
	 */
	if (orig_waiter && !orig_waiter->task)
		goto out_unlock_pi;

	/*
	 * Drop out, when the task has no waiters. Note,
	 * top_waiter can be NULL, when we are in the deboosting
	 * mode!
	 */
222 223 224 225 226 227 228 229 230 231 232 233
	if (top_waiter && (!task_has_pi_waiters(task) ||
			   top_waiter != task_top_pi_waiter(task)))
		goto out_unlock_pi;

	/*
	 * When deadlock detection is off then we check, if further
	 * priority adjustment is necessary.
	 */
	if (!detect_deadlock && waiter->list_entry.prio == task->prio)
		goto out_unlock_pi;

	lock = waiter->lock;
234
	if (!raw_spin_trylock(&lock->wait_lock)) {
235
		raw_spin_unlock_irqrestore(&task->pi_lock, flags);
236 237 238 239 240
		cpu_relax();
		goto retry;
	}

	/* Deadlock detection */
241
	if (lock == orig_lock || rt_mutex_owner(lock) == top_task) {
242
		debug_rt_mutex_deadlock(deadlock_detect, orig_waiter, lock);
243
		raw_spin_unlock(&lock->wait_lock);
244 245 246 247 248 249 250 251 252 253 254 255
		ret = deadlock_detect ? -EDEADLK : 0;
		goto out_unlock_pi;
	}

	top_waiter = rt_mutex_top_waiter(lock);

	/* Requeue the waiter */
	plist_del(&waiter->list_entry, &lock->wait_list);
	waiter->list_entry.prio = task->prio;
	plist_add(&waiter->list_entry, &lock->wait_list);

	/* Release the task */
256
	raw_spin_unlock_irqrestore(&task->pi_lock, flags);
257 258 259 260
	put_task_struct(task);

	/* Grab the next task */
	task = rt_mutex_owner(lock);
261
	get_task_struct(task);
262
	raw_spin_lock_irqsave(&task->pi_lock, flags);
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279

	if (waiter == rt_mutex_top_waiter(lock)) {
		/* Boost the owner */
		plist_del(&top_waiter->pi_list_entry, &task->pi_waiters);
		waiter->pi_list_entry.prio = waiter->list_entry.prio;
		plist_add(&waiter->pi_list_entry, &task->pi_waiters);
		__rt_mutex_adjust_prio(task);

	} else if (top_waiter == waiter) {
		/* Deboost the owner */
		plist_del(&waiter->pi_list_entry, &task->pi_waiters);
		waiter = rt_mutex_top_waiter(lock);
		waiter->pi_list_entry.prio = waiter->list_entry.prio;
		plist_add(&waiter->pi_list_entry, &task->pi_waiters);
		__rt_mutex_adjust_prio(task);
	}

280
	raw_spin_unlock_irqrestore(&task->pi_lock, flags);
281 282

	top_waiter = rt_mutex_top_waiter(lock);
283
	raw_spin_unlock(&lock->wait_lock);
284 285 286 287 288 289 290

	if (!detect_deadlock && waiter != top_waiter)
		goto out_put_task;

	goto again;

 out_unlock_pi:
291
	raw_spin_unlock_irqrestore(&task->pi_lock, flags);
292 293
 out_put_task:
	put_task_struct(task);
294

295 296 297 298 299 300 301 302
	return ret;
}

/*
 * Optimization: check if we can steal the lock from the
 * assigned pending owner [which might not have taken the
 * lock yet]:
 */
303 304
static inline int try_to_steal_lock(struct rt_mutex *lock,
				    struct task_struct *task)
305 306 307 308 309 310 311 312
{
	struct task_struct *pendowner = rt_mutex_owner(lock);
	struct rt_mutex_waiter *next;
	unsigned long flags;

	if (!rt_mutex_owner_pending(lock))
		return 0;

313
	if (pendowner == task)
314 315
		return 1;

316
	raw_spin_lock_irqsave(&pendowner->pi_lock, flags);
317
	if (task->prio >= pendowner->prio) {
318
		raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags);
319 320 321 322 323 324 325 326 327
		return 0;
	}

	/*
	 * Check if a waiter is enqueued on the pending owners
	 * pi_waiters list. Remove it and readjust pending owners
	 * priority.
	 */
	if (likely(!rt_mutex_has_waiters(lock))) {
328
		raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags);
329 330 331 332 333 334 335
		return 1;
	}

	/* No chain handling, pending owner is not blocked on anything: */
	next = rt_mutex_top_waiter(lock);
	plist_del(&next->pi_list_entry, &pendowner->pi_waiters);
	__rt_mutex_adjust_prio(pendowner);
336
	raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags);
337 338 339 340 341

	/*
	 * We are going to steal the lock and a waiter was
	 * enqueued on the pending owners pi_waiters queue. So
	 * we have to enqueue this waiter into
342 343
	 * task->pi_waiters list. This covers the case,
	 * where task is boosted because it holds another
344 345
	 * lock and gets unboosted because the booster is
	 * interrupted, so we would delay a waiter with higher
346
	 * priority as task->normal_prio.
347 348 349
	 *
	 * Note: in the rare case of a SCHED_OTHER task changing
	 * 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);
354 355
		plist_add(&next->pi_list_entry, &task->pi_waiters);
		__rt_mutex_adjust_prio(task);
356
		raw_spin_unlock_irqrestore(&task->pi_lock, flags);
357 358 359 360 361 362 363 364 365 366 367 368 369
	}
	return 1;
}

/*
 * Try to take an rt-mutex
 *
 * This fails
 * - when the lock has a real owner
 * - when a different pending owner exists and has higher priority than current
 *
 * Must be called with lock->wait_lock held.
 */
370
static int try_to_take_rt_mutex(struct rt_mutex *lock)
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
{
	/*
	 * We have to be careful here if the atomic speedups are
	 * enabled, such that, when
	 *  - no other waiter is on the lock
	 *  - the lock has been released since we did the cmpxchg
	 * the lock can be released or taken while we are doing the
	 * checks and marking the lock with RT_MUTEX_HAS_WAITERS.
	 *
	 * The atomic acquire/release aware variant of
	 * mark_rt_mutex_waiters uses a cmpxchg loop. After setting
	 * the WAITERS bit, the atomic release / acquire can not
	 * happen anymore and lock->wait_lock protects us from the
	 * non-atomic case.
	 *
	 * Note, that this might set lock->owner =
	 * RT_MUTEX_HAS_WAITERS in the case the lock is not contended
	 * any more. This is fixed up when we take the ownership.
	 * This is the transitional state explained at the top of this file.
	 */
	mark_rt_mutex_waiters(lock);

393
	if (rt_mutex_owner(lock) && !try_to_steal_lock(lock, current))
394 395 396
		return 0;

	/* We got the lock. */
397
	debug_rt_mutex_lock(lock);
398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414

	rt_mutex_set_owner(lock, current, 0);

	rt_mutex_deadlock_account_lock(lock, current);

	return 1;
}

/*
 * Task blocks on lock.
 *
 * Prepare waiter and propagate pi chain
 *
 * This must be called with lock->wait_lock held.
 */
static int task_blocks_on_rt_mutex(struct rt_mutex *lock,
				   struct rt_mutex_waiter *waiter,
415
				   struct task_struct *task,
416
				   int detect_deadlock)
417
{
418
	struct task_struct *owner = rt_mutex_owner(lock);
419 420
	struct rt_mutex_waiter *top_waiter = waiter;
	unsigned long flags;
421
	int chain_walk = 0, res;
422

423
	raw_spin_lock_irqsave(&task->pi_lock, flags);
424 425
	__rt_mutex_adjust_prio(task);
	waiter->task = task;
426
	waiter->lock = lock;
427 428
	plist_node_init(&waiter->list_entry, task->prio);
	plist_node_init(&waiter->pi_list_entry, task->prio);
429 430 431 432 433 434

	/* Get the top priority waiter on the lock */
	if (rt_mutex_has_waiters(lock))
		top_waiter = rt_mutex_top_waiter(lock);
	plist_add(&waiter->list_entry, &lock->wait_list);

435
	task->pi_blocked_on = waiter;
436

437
	raw_spin_unlock_irqrestore(&task->pi_lock, flags);
438 439

	if (waiter == rt_mutex_top_waiter(lock)) {
440
		raw_spin_lock_irqsave(&owner->pi_lock, flags);
441 442 443 444
		plist_del(&top_waiter->pi_list_entry, &owner->pi_waiters);
		plist_add(&waiter->pi_list_entry, &owner->pi_waiters);

		__rt_mutex_adjust_prio(owner);
445 446
		if (owner->pi_blocked_on)
			chain_walk = 1;
447
		raw_spin_unlock_irqrestore(&owner->pi_lock, flags);
448
	}
449 450 451 452
	else if (debug_rt_mutex_detect_deadlock(waiter, detect_deadlock))
		chain_walk = 1;

	if (!chain_walk)
453 454
		return 0;

455 456 457 458 459 460 461
	/*
	 * The owner can't disappear while holding a lock,
	 * so the owner struct is protected by wait_lock.
	 * Gets dropped in rt_mutex_adjust_prio_chain()!
	 */
	get_task_struct(owner);

462
	raw_spin_unlock(&lock->wait_lock);
463

464
	res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, waiter,
465
					 task);
466

467
	raw_spin_lock(&lock->wait_lock);
468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485

	return res;
}

/*
 * Wake up the next waiter on the lock.
 *
 * Remove the top waiter from the current tasks waiter list and from
 * the lock waiter list. Set it as pending owner. Then wake it up.
 *
 * Called with lock->wait_lock held.
 */
static void wakeup_next_waiter(struct rt_mutex *lock)
{
	struct rt_mutex_waiter *waiter;
	struct task_struct *pendowner;
	unsigned long flags;

486
	raw_spin_lock_irqsave(&current->pi_lock, flags);
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502

	waiter = rt_mutex_top_waiter(lock);
	plist_del(&waiter->list_entry, &lock->wait_list);

	/*
	 * Remove it from current->pi_waiters. We do not adjust a
	 * possible priority boost right now. We execute wakeup in the
	 * boosted mode and go back to normal after releasing
	 * lock->wait_lock.
	 */
	plist_del(&waiter->pi_list_entry, &current->pi_waiters);
	pendowner = waiter->task;
	waiter->task = NULL;

	rt_mutex_set_owner(lock, pendowner, RT_MUTEX_OWNER_PENDING);

503
	raw_spin_unlock_irqrestore(&current->pi_lock, flags);
504 505 506 507 508 509 510 511

	/*
	 * Clear the pi_blocked_on variable and enqueue a possible
	 * waiter into the pi_waiters list of the pending owner. This
	 * prevents that in case the pending owner gets unboosted a
	 * waiter with higher priority than pending-owner->normal_prio
	 * is blocked on the unboosted (pending) owner.
	 */
512
	raw_spin_lock_irqsave(&pendowner->pi_lock, flags);
513 514 515 516 517 518 519 520 521 522 523 524 525

	WARN_ON(!pendowner->pi_blocked_on);
	WARN_ON(pendowner->pi_blocked_on != waiter);
	WARN_ON(pendowner->pi_blocked_on->lock != lock);

	pendowner->pi_blocked_on = NULL;

	if (rt_mutex_has_waiters(lock)) {
		struct rt_mutex_waiter *next;

		next = rt_mutex_top_waiter(lock);
		plist_add(&next->pi_list_entry, &pendowner->pi_waiters);
	}
526
	raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags);
527 528 529 530 531 532 533 534 535

	wake_up_process(pendowner);
}

/*
 * Remove a waiter from a lock
 *
 * Must be called with lock->wait_lock held
 */
536 537
static void remove_waiter(struct rt_mutex *lock,
			  struct rt_mutex_waiter *waiter)
538 539
{
	int first = (waiter == rt_mutex_top_waiter(lock));
540
	struct task_struct *owner = rt_mutex_owner(lock);
541
	unsigned long flags;
542
	int chain_walk = 0;
543

544
	raw_spin_lock_irqsave(&current->pi_lock, flags);
545 546 547
	plist_del(&waiter->list_entry, &lock->wait_list);
	waiter->task = NULL;
	current->pi_blocked_on = NULL;
548
	raw_spin_unlock_irqrestore(&current->pi_lock, flags);
549 550 551

	if (first && owner != current) {

552
		raw_spin_lock_irqsave(&owner->pi_lock, flags);
553 554 555 556 557 558 559 560 561 562 563

		plist_del(&waiter->pi_list_entry, &owner->pi_waiters);

		if (rt_mutex_has_waiters(lock)) {
			struct rt_mutex_waiter *next;

			next = rt_mutex_top_waiter(lock);
			plist_add(&next->pi_list_entry, &owner->pi_waiters);
		}
		__rt_mutex_adjust_prio(owner);

564 565 566
		if (owner->pi_blocked_on)
			chain_walk = 1;

567
		raw_spin_unlock_irqrestore(&owner->pi_lock, flags);
568 569 570 571
	}

	WARN_ON(!plist_node_empty(&waiter->pi_list_entry));

572
	if (!chain_walk)
573 574
		return;

575 576 577
	/* gets dropped in rt_mutex_adjust_prio_chain()! */
	get_task_struct(owner);

578
	raw_spin_unlock(&lock->wait_lock);
579

580
	rt_mutex_adjust_prio_chain(owner, 0, lock, NULL, current);
581

582
	raw_spin_lock(&lock->wait_lock);
583 584
}

585 586 587 588 589 590 591 592 593 594
/*
 * Recheck the pi chain, in case we got a priority setting
 *
 * Called from sched_setscheduler
 */
void rt_mutex_adjust_pi(struct task_struct *task)
{
	struct rt_mutex_waiter *waiter;
	unsigned long flags;

595
	raw_spin_lock_irqsave(&task->pi_lock, flags);
596 597 598

	waiter = task->pi_blocked_on;
	if (!waiter || waiter->list_entry.prio == task->prio) {
599
		raw_spin_unlock_irqrestore(&task->pi_lock, flags);
600 601 602
		return;
	}

603
	raw_spin_unlock_irqrestore(&task->pi_lock, flags);
604

605 606
	/* gets dropped in rt_mutex_adjust_prio_chain()! */
	get_task_struct(task);
607
	rt_mutex_adjust_prio_chain(task, 0, NULL, NULL, task);
608 609
}

610 611 612 613 614 615 616 617 618 619
/**
 * __rt_mutex_slowlock() - Perform the wait-wake-try-to-take loop
 * @lock:		 the rt_mutex to take
 * @state:		 the state the task should block in (TASK_INTERRUPTIBLE
 * 			 or TASK_UNINTERRUPTIBLE)
 * @timeout:		 the pre-initialized and started timer, or NULL for none
 * @waiter:		 the pre-initialized rt_mutex_waiter
 * @detect_deadlock:	 passed to task_blocks_on_rt_mutex
 *
 * lock->wait_lock must be held by the caller.
620 621
 */
static int __sched
622 623 624 625
__rt_mutex_slowlock(struct rt_mutex *lock, int state,
		    struct hrtimer_sleeper *timeout,
		    struct rt_mutex_waiter *waiter,
		    int detect_deadlock)
626 627 628 629 630
{
	int ret = 0;

	for (;;) {
		/* Try to acquire the lock: */
631
		if (try_to_take_rt_mutex(lock))
632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648
			break;

		/*
		 * TASK_INTERRUPTIBLE checks for signals and
		 * timeout. Ignored otherwise.
		 */
		if (unlikely(state == TASK_INTERRUPTIBLE)) {
			/* Signal pending? */
			if (signal_pending(current))
				ret = -EINTR;
			if (timeout && !timeout->task)
				ret = -ETIMEDOUT;
			if (ret)
				break;
		}

		/*
649
		 * waiter->task is NULL the first time we come here and
650 651 652
		 * when we have been woken up by the previous owner
		 * but the lock got stolen by a higher prio task.
		 */
653 654
		if (!waiter->task) {
			ret = task_blocks_on_rt_mutex(lock, waiter, current,
655
						      detect_deadlock);
656 657 658 659 660
			/*
			 * If we got woken up by the owner then start loop
			 * all over without going into schedule to try
			 * to get the lock now:
			 */
661
			if (unlikely(!waiter->task)) {
662 663 664 665 666 667 668
				/*
				 * Reset the return value. We might
				 * have returned with -EDEADLK and the
				 * owner released the lock while we
				 * were walking the pi chain.
				 */
				ret = 0;
669
				continue;
670
			}
671 672 673
			if (unlikely(ret))
				break;
		}
674

675
		raw_spin_unlock(&lock->wait_lock);
676

677
		debug_rt_mutex_print_deadlock(waiter);
678

679
		if (waiter->task)
680
			schedule_rt_mutex(lock);
681

682
		raw_spin_lock(&lock->wait_lock);
683 684 685
		set_current_state(state);
	}

686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702
	return ret;
}

/*
 * Slow path lock function:
 */
static int __sched
rt_mutex_slowlock(struct rt_mutex *lock, int state,
		  struct hrtimer_sleeper *timeout,
		  int detect_deadlock)
{
	struct rt_mutex_waiter waiter;
	int ret = 0;

	debug_rt_mutex_init_waiter(&waiter);
	waiter.task = NULL;

703
	raw_spin_lock(&lock->wait_lock);
704 705 706

	/* Try to acquire the lock again: */
	if (try_to_take_rt_mutex(lock)) {
707
		raw_spin_unlock(&lock->wait_lock);
708 709 710 711 712 713 714 715 716 717 718 719 720 721 722
		return 0;
	}

	set_current_state(state);

	/* Setup the timer, when timeout != NULL */
	if (unlikely(timeout)) {
		hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS);
		if (!hrtimer_active(&timeout->timer))
			timeout->task = NULL;
	}

	ret = __rt_mutex_slowlock(lock, state, timeout, &waiter,
				  detect_deadlock);

723 724 725
	set_current_state(TASK_RUNNING);

	if (unlikely(waiter.task))
726
		remove_waiter(lock, &waiter);
727 728 729 730 731 732 733

	/*
	 * try_to_take_rt_mutex() sets the waiter bit
	 * unconditionally. We might have to fix that up.
	 */
	fixup_rt_mutex_waiters(lock);

734
	raw_spin_unlock(&lock->wait_lock);
735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756

	/* Remove pending timer: */
	if (unlikely(timeout))
		hrtimer_cancel(&timeout->timer);

	/*
	 * Readjust priority, when we did not get the lock. We might
	 * have been the pending owner and boosted. Since we did not
	 * take the lock, the PI boost has to go.
	 */
	if (unlikely(ret))
		rt_mutex_adjust_prio(current);

	debug_rt_mutex_free_waiter(&waiter);

	return ret;
}

/*
 * Slow path try-lock function:
 */
static inline int
757
rt_mutex_slowtrylock(struct rt_mutex *lock)
758 759 760
{
	int ret = 0;

761
	raw_spin_lock(&lock->wait_lock);
762 763 764

	if (likely(rt_mutex_owner(lock) != current)) {

765
		ret = try_to_take_rt_mutex(lock);
766 767 768 769 770 771 772
		/*
		 * try_to_take_rt_mutex() sets the lock waiters
		 * bit unconditionally. Clean this up.
		 */
		fixup_rt_mutex_waiters(lock);
	}

773
	raw_spin_unlock(&lock->wait_lock);
774 775 776 777 778 779 780 781 782 783

	return ret;
}

/*
 * Slow path to release a rt-mutex:
 */
static void __sched
rt_mutex_slowunlock(struct rt_mutex *lock)
{
784
	raw_spin_lock(&lock->wait_lock);
785 786 787 788 789 790 791

	debug_rt_mutex_unlock(lock);

	rt_mutex_deadlock_account_unlock(current);

	if (!rt_mutex_has_waiters(lock)) {
		lock->owner = NULL;
792
		raw_spin_unlock(&lock->wait_lock);
793 794 795 796 797
		return;
	}

	wakeup_next_waiter(lock);

798
	raw_spin_unlock(&lock->wait_lock);
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814

	/* Undo pi boosting if necessary: */
	rt_mutex_adjust_prio(current);
}

/*
 * debug aware fast / slowpath lock,trylock,unlock
 *
 * The atomic acquire/release ops are compiled away, when either the
 * architecture does not support cmpxchg or when debugging is enabled.
 */
static inline int
rt_mutex_fastlock(struct rt_mutex *lock, int state,
		  int detect_deadlock,
		  int (*slowfn)(struct rt_mutex *lock, int state,
				struct hrtimer_sleeper *timeout,
815
				int detect_deadlock))
816 817 818 819 820
{
	if (!detect_deadlock && likely(rt_mutex_cmpxchg(lock, NULL, current))) {
		rt_mutex_deadlock_account_lock(lock, current);
		return 0;
	} else
821
		return slowfn(lock, state, NULL, detect_deadlock);
822 823 824 825 826 827 828
}

static inline int
rt_mutex_timed_fastlock(struct rt_mutex *lock, int state,
			struct hrtimer_sleeper *timeout, int detect_deadlock,
			int (*slowfn)(struct rt_mutex *lock, int state,
				      struct hrtimer_sleeper *timeout,
829
				      int detect_deadlock))
830 831 832 833 834
{
	if (!detect_deadlock && likely(rt_mutex_cmpxchg(lock, NULL, current))) {
		rt_mutex_deadlock_account_lock(lock, current);
		return 0;
	} else
835
		return slowfn(lock, state, timeout, detect_deadlock);
836 837 838 839
}

static inline int
rt_mutex_fasttrylock(struct rt_mutex *lock,
840
		     int (*slowfn)(struct rt_mutex *lock))
841 842 843 844 845
{
	if (likely(rt_mutex_cmpxchg(lock, NULL, current))) {
		rt_mutex_deadlock_account_lock(lock, current);
		return 1;
	}
846
	return slowfn(lock);
847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893
}

static inline void
rt_mutex_fastunlock(struct rt_mutex *lock,
		    void (*slowfn)(struct rt_mutex *lock))
{
	if (likely(rt_mutex_cmpxchg(lock, current, NULL)))
		rt_mutex_deadlock_account_unlock(current);
	else
		slowfn(lock);
}

/**
 * rt_mutex_lock - lock a rt_mutex
 *
 * @lock: the rt_mutex to be locked
 */
void __sched rt_mutex_lock(struct rt_mutex *lock)
{
	might_sleep();

	rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, 0, rt_mutex_slowlock);
}
EXPORT_SYMBOL_GPL(rt_mutex_lock);

/**
 * rt_mutex_lock_interruptible - lock a rt_mutex interruptible
 *
 * @lock: 		the rt_mutex to be locked
 * @detect_deadlock:	deadlock detection on/off
 *
 * Returns:
 *  0 		on success
 * -EINTR 	when interrupted by a signal
 * -EDEADLK	when the lock would deadlock (when deadlock detection is on)
 */
int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock,
						 int detect_deadlock)
{
	might_sleep();

	return rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE,
				 detect_deadlock, rt_mutex_slowlock);
}
EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible);

/**
894 895 896
 * rt_mutex_timed_lock - lock a rt_mutex interruptible
 *			the timeout structure is provided
 *			by the caller
897 898 899 900 901 902 903 904
 *
 * @lock: 		the rt_mutex to be locked
 * @timeout:		timeout structure or NULL (no timeout)
 * @detect_deadlock:	deadlock detection on/off
 *
 * Returns:
 *  0 		on success
 * -EINTR 	when interrupted by a signal
905
 * -ETIMEDOUT	when the timeout expired
906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942
 * -EDEADLK	when the lock would deadlock (when deadlock detection is on)
 */
int
rt_mutex_timed_lock(struct rt_mutex *lock, struct hrtimer_sleeper *timeout,
		    int detect_deadlock)
{
	might_sleep();

	return rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout,
				       detect_deadlock, rt_mutex_slowlock);
}
EXPORT_SYMBOL_GPL(rt_mutex_timed_lock);

/**
 * rt_mutex_trylock - try to lock a rt_mutex
 *
 * @lock:	the rt_mutex to be locked
 *
 * Returns 1 on success and 0 on contention
 */
int __sched rt_mutex_trylock(struct rt_mutex *lock)
{
	return rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock);
}
EXPORT_SYMBOL_GPL(rt_mutex_trylock);

/**
 * rt_mutex_unlock - unlock a rt_mutex
 *
 * @lock: the rt_mutex to be unlocked
 */
void __sched rt_mutex_unlock(struct rt_mutex *lock)
{
	rt_mutex_fastunlock(lock, rt_mutex_slowunlock);
}
EXPORT_SYMBOL_GPL(rt_mutex_unlock);

943
/**
944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
 * rt_mutex_destroy - mark a mutex unusable
 * @lock: the mutex to be destroyed
 *
 * This function marks the mutex uninitialized, and any subsequent
 * use of the mutex is forbidden. The mutex must not be locked when
 * this function is called.
 */
void rt_mutex_destroy(struct rt_mutex *lock)
{
	WARN_ON(rt_mutex_is_locked(lock));
#ifdef CONFIG_DEBUG_RT_MUTEXES
	lock->magic = NULL;
#endif
}

EXPORT_SYMBOL_GPL(rt_mutex_destroy);

/**
 * __rt_mutex_init - initialize the rt lock
 *
 * @lock: the rt lock to be initialized
 *
 * Initialize the rt lock to unlocked state.
 *
 * Initializing of a locked rt lock is not allowed
 */
void __rt_mutex_init(struct rt_mutex *lock, const char *name)
{
	lock->owner = NULL;
973 974
	raw_spin_lock_init(&lock->wait_lock);
	plist_head_init_raw(&lock->wait_list, &lock->wait_lock);
975 976 977 978

	debug_rt_mutex_init(lock, name);
}
EXPORT_SYMBOL_GPL(__rt_mutex_init);
979 980 981 982 983 984 985 986 987 988 989 990 991 992 993

/**
 * rt_mutex_init_proxy_locked - initialize and lock a rt_mutex on behalf of a
 *				proxy owner
 *
 * @lock: 	the rt_mutex to be locked
 * @proxy_owner:the task to set as owner
 *
 * No locking. Caller has to do serializing itself
 * Special API call for PI-futex support
 */
void rt_mutex_init_proxy_locked(struct rt_mutex *lock,
				struct task_struct *proxy_owner)
{
	__rt_mutex_init(lock, NULL);
994
	debug_rt_mutex_proxy_lock(lock, proxy_owner);
995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014
	rt_mutex_set_owner(lock, proxy_owner, 0);
	rt_mutex_deadlock_account_lock(lock, proxy_owner);
}

/**
 * rt_mutex_proxy_unlock - release a lock on behalf of owner
 *
 * @lock: 	the rt_mutex to be locked
 *
 * No locking. Caller has to do serializing itself
 * Special API call for PI-futex support
 */
void rt_mutex_proxy_unlock(struct rt_mutex *lock,
			   struct task_struct *proxy_owner)
{
	debug_rt_mutex_proxy_unlock(lock);
	rt_mutex_set_owner(lock, NULL, 0);
	rt_mutex_deadlock_account_unlock(proxy_owner);
}

1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034
/**
 * rt_mutex_start_proxy_lock() - Start lock acquisition for another task
 * @lock:		the rt_mutex to take
 * @waiter:		the pre-initialized rt_mutex_waiter
 * @task:		the task to prepare
 * @detect_deadlock:	perform deadlock detection (1) or not (0)
 *
 * Returns:
 *  0 - task blocked on lock
 *  1 - acquired the lock for task, caller should wake it up
 * <0 - error
 *
 * Special API call for FUTEX_REQUEUE_PI support.
 */
int rt_mutex_start_proxy_lock(struct rt_mutex *lock,
			      struct rt_mutex_waiter *waiter,
			      struct task_struct *task, int detect_deadlock)
{
	int ret;

1035
	raw_spin_lock(&lock->wait_lock);
1036 1037 1038 1039 1040 1041 1042

	mark_rt_mutex_waiters(lock);

	if (!rt_mutex_owner(lock) || try_to_steal_lock(lock, task)) {
		/* We got the lock for task. */
		debug_rt_mutex_lock(lock);
		rt_mutex_set_owner(lock, task, 0);
1043
		raw_spin_unlock(&lock->wait_lock);
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
		rt_mutex_deadlock_account_lock(lock, task);
		return 1;
	}

	ret = task_blocks_on_rt_mutex(lock, waiter, task, detect_deadlock);

	if (ret && !waiter->task) {
		/*
		 * Reset the return value. We might have
		 * returned with -EDEADLK and the owner
		 * released the lock while we were walking the
		 * pi chain.  Let the waiter sort it out.
		 */
		ret = 0;
	}
1059
	raw_spin_unlock(&lock->wait_lock);
1060 1061 1062 1063 1064 1065

	debug_rt_mutex_print_deadlock(waiter);

	return ret;
}

1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084
/**
 * rt_mutex_next_owner - return the next owner of the lock
 *
 * @lock: the rt lock query
 *
 * Returns the next owner of the lock or NULL
 *
 * Caller has to serialize against other accessors to the lock
 * itself.
 *
 * Special API call for PI-futex support
 */
struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock)
{
	if (!rt_mutex_has_waiters(lock))
		return NULL;

	return rt_mutex_top_waiter(lock)->task;
}
1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108

/**
 * rt_mutex_finish_proxy_lock() - Complete lock acquisition
 * @lock:		the rt_mutex we were woken on
 * @to:			the timeout, null if none. hrtimer should already have
 * 			been started.
 * @waiter:		the pre-initialized rt_mutex_waiter
 * @detect_deadlock:	perform deadlock detection (1) or not (0)
 *
 * Complete the lock acquisition started our behalf by another thread.
 *
 * Returns:
 *  0 - success
 * <0 - error, one of -EINTR, -ETIMEDOUT, or -EDEADLK
 *
 * Special API call for PI-futex requeue support
 */
int rt_mutex_finish_proxy_lock(struct rt_mutex *lock,
			       struct hrtimer_sleeper *to,
			       struct rt_mutex_waiter *waiter,
			       int detect_deadlock)
{
	int ret;

1109
	raw_spin_lock(&lock->wait_lock);
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126

	set_current_state(TASK_INTERRUPTIBLE);

	ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter,
				  detect_deadlock);

	set_current_state(TASK_RUNNING);

	if (unlikely(waiter->task))
		remove_waiter(lock, waiter);

	/*
	 * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might
	 * have to fix that up.
	 */
	fixup_rt_mutex_waiters(lock);

1127
	raw_spin_unlock(&lock->wait_lock);
1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138

	/*
	 * Readjust priority, when we did not get the lock. We might have been
	 * the pending owner and boosted. Since we did not take the lock, the
	 * PI boost has to go.
	 */
	if (unlikely(ret))
		rt_mutex_adjust_prio(current);

	return ret;
}