mutex.c 25.1 KB
Newer Older
I
Ingo Molnar 已提交
1
/*
2
 * kernel/locking/mutex.c
I
Ingo Molnar 已提交
3 4 5 6 7 8 9 10 11 12
 *
 * Mutexes: blocking mutual exclusion locks
 *
 * Started by Ingo Molnar:
 *
 *  Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
 *
 * Many thanks to Arjan van de Ven, Thomas Gleixner, Steven Rostedt and
 * David Howells for suggestions and improvements.
 *
13 14 15 16 17
 *  - Adaptive spinning for mutexes by Peter Zijlstra. (Ported to mainline
 *    from the -rt tree, where it was originally implemented for rtmutexes
 *    by Steven Rostedt, based on work by Gregory Haskins, Peter Morreale
 *    and Sven Dietrich.
 *
18
 * Also see Documentation/locking/mutex-design.txt.
I
Ingo Molnar 已提交
19 20
 */
#include <linux/mutex.h>
21
#include <linux/ww_mutex.h>
I
Ingo Molnar 已提交
22
#include <linux/sched.h>
23
#include <linux/sched/rt.h>
24
#include <linux/export.h>
I
Ingo Molnar 已提交
25 26
#include <linux/spinlock.h>
#include <linux/interrupt.h>
27
#include <linux/debug_locks.h>
28
#include <linux/osq_lock.h>
I
Ingo Molnar 已提交
29 30 31 32 33 34 35 36

/*
 * In the DEBUG case we are using the "NULL fastpath" for mutexes,
 * which forces all calls into the slowpath:
 */
#ifdef CONFIG_DEBUG_MUTEXES
# include "mutex-debug.h"
# include <asm-generic/mutex-null.h>
P
Peter Zijlstra 已提交
37 38 39 40 41 42 43
/*
 * Must be 0 for the debug case so we do not do the unlock outside of the
 * wait_lock region. debug_mutex_unlock() will do the actual unlock in this
 * case.
 */
# undef __mutex_slowpath_needs_to_unlock
# define  __mutex_slowpath_needs_to_unlock()	0
I
Ingo Molnar 已提交
44 45 46 47 48
#else
# include "mutex.h"
# include <asm/mutex.h>
#endif

49 50
void
__mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key)
I
Ingo Molnar 已提交
51 52 53 54
{
	atomic_set(&lock->count, 1);
	spin_lock_init(&lock->wait_lock);
	INIT_LIST_HEAD(&lock->wait_list);
55
	mutex_clear_owner(lock);
56
#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
57
	osq_lock_init(&lock->osq);
58
#endif
I
Ingo Molnar 已提交
59

60
	debug_mutex_init(lock, name, key);
I
Ingo Molnar 已提交
61 62 63 64
}

EXPORT_SYMBOL(__mutex_init);

P
Peter Zijlstra 已提交
65
#ifndef CONFIG_DEBUG_LOCK_ALLOC
I
Ingo Molnar 已提交
66 67 68 69 70 71
/*
 * We split the mutex lock/unlock logic into separate fastpath and
 * slowpath functions, to reduce the register pressure on the fastpath.
 * We also put the fastpath first in the kernel image, to make sure the
 * branch is predicted by the CPU as default-untaken.
 */
A
Andi Kleen 已提交
72
__visible void __sched __mutex_lock_slowpath(atomic_t *lock_count);
I
Ingo Molnar 已提交
73

74
/**
I
Ingo Molnar 已提交
75 76 77 78 79 80 81 82 83
 * mutex_lock - acquire the mutex
 * @lock: the mutex to be acquired
 *
 * Lock the mutex exclusively for this task. If the mutex is not
 * available right now, it will sleep until it can get it.
 *
 * The mutex must later on be released by the same task that
 * acquired it. Recursive locking is not allowed. The task
 * may not exit without first unlocking the mutex. Also, kernel
84
 * memory where the mutex resides must not be freed with
I
Ingo Molnar 已提交
85 86 87 88 89 90 91 92 93 94
 * the mutex still locked. The mutex must first be initialized
 * (or statically defined) before it can be locked. memset()-ing
 * the mutex to 0 is not allowed.
 *
 * ( The CONFIG_DEBUG_MUTEXES .config option turns on debugging
 *   checks that will enforce the restrictions and will also do
 *   deadlock debugging. )
 *
 * This function is similar to (but not equivalent to) down().
 */
95
void __sched mutex_lock(struct mutex *lock)
I
Ingo Molnar 已提交
96
{
97
	might_sleep();
I
Ingo Molnar 已提交
98 99 100 101 102
	/*
	 * The locking fastpath is the 1->0 transition from
	 * 'unlocked' into 'locked' state.
	 */
	__mutex_fastpath_lock(&lock->count, __mutex_lock_slowpath);
103
	mutex_set_owner(lock);
I
Ingo Molnar 已提交
104 105 106
}

EXPORT_SYMBOL(mutex_lock);
P
Peter Zijlstra 已提交
107
#endif
I
Ingo Molnar 已提交
108

109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149
static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww,
						   struct ww_acquire_ctx *ww_ctx)
{
#ifdef CONFIG_DEBUG_MUTEXES
	/*
	 * If this WARN_ON triggers, you used ww_mutex_lock to acquire,
	 * but released with a normal mutex_unlock in this call.
	 *
	 * This should never happen, always use ww_mutex_unlock.
	 */
	DEBUG_LOCKS_WARN_ON(ww->ctx);

	/*
	 * Not quite done after calling ww_acquire_done() ?
	 */
	DEBUG_LOCKS_WARN_ON(ww_ctx->done_acquire);

	if (ww_ctx->contending_lock) {
		/*
		 * After -EDEADLK you tried to
		 * acquire a different ww_mutex? Bad!
		 */
		DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock != ww);

		/*
		 * You called ww_mutex_lock after receiving -EDEADLK,
		 * but 'forgot' to unlock everything else first?
		 */
		DEBUG_LOCKS_WARN_ON(ww_ctx->acquired > 0);
		ww_ctx->contending_lock = NULL;
	}

	/*
	 * Naughty, using a different class will lead to undefined behavior!
	 */
	DEBUG_LOCKS_WARN_ON(ww_ctx->ww_class != ww->ww_class);
#endif
	ww_ctx->acquired++;
}

/*
150
 * After acquiring lock with fastpath or when we lost out in contested
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193
 * slowpath, set ctx and wake up any waiters so they can recheck.
 *
 * This function is never called when CONFIG_DEBUG_LOCK_ALLOC is set,
 * as the fastpath and opportunistic spinning are disabled in that case.
 */
static __always_inline void
ww_mutex_set_context_fastpath(struct ww_mutex *lock,
			       struct ww_acquire_ctx *ctx)
{
	unsigned long flags;
	struct mutex_waiter *cur;

	ww_mutex_lock_acquired(lock, ctx);

	lock->ctx = ctx;

	/*
	 * The lock->ctx update should be visible on all cores before
	 * the atomic read is done, otherwise contended waiters might be
	 * missed. The contended waiters will either see ww_ctx == NULL
	 * and keep spinning, or it will acquire wait_lock, add itself
	 * to waiter list and sleep.
	 */
	smp_mb(); /* ^^^ */

	/*
	 * Check if lock is contended, if not there is nobody to wake up
	 */
	if (likely(atomic_read(&lock->base.count) == 0))
		return;

	/*
	 * Uh oh, we raced in fastpath, wake up everyone in this case,
	 * so they can see the new lock->ctx.
	 */
	spin_lock_mutex(&lock->base.wait_lock, flags);
	list_for_each_entry(cur, &lock->base.wait_list, list) {
		debug_mutex_wake_waiter(&lock->base, cur);
		wake_up_process(cur->task);
	}
	spin_unlock_mutex(&lock->base.wait_lock, flags);
}

194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217
/*
 * After acquiring lock in the slowpath set ctx and wake up any
 * waiters so they can recheck.
 *
 * Callers must hold the mutex wait_lock.
 */
static __always_inline void
ww_mutex_set_context_slowpath(struct ww_mutex *lock,
			      struct ww_acquire_ctx *ctx)
{
	struct mutex_waiter *cur;

	ww_mutex_lock_acquired(lock, ctx);
	lock->ctx = ctx;

	/*
	 * Give any possible sleeping processes the chance to wake up,
	 * so they can recheck if they have to back off.
	 */
	list_for_each_entry(cur, &lock->base.wait_list, list) {
		debug_mutex_wake_waiter(&lock->base, cur);
		wake_up_process(cur->task);
	}
}
218

219 220 221 222 223 224
#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
/*
 * Look out! "owner" is an entirely speculative pointer
 * access and not reliable.
 */
static noinline
225
bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner)
226
{
227
	bool ret = true;
228

229
	rcu_read_lock();
230
	while (lock->owner == owner) {
231 232
		/*
		 * Ensure we emit the owner->on_cpu, dereference _after_
233 234
		 * checking lock->owner still matches owner. If that fails,
		 * owner might point to freed memory. If it still matches,
235 236 237 238 239 240 241 242
		 * the rcu_read_lock() ensures the memory stays valid.
		 */
		barrier();

		if (!owner->on_cpu || need_resched()) {
			ret = false;
			break;
		}
243

244
		cpu_relax_lowlatency();
245 246 247
	}
	rcu_read_unlock();

248
	return ret;
249
}
250 251 252 253 254 255

/*
 * Initial check for entering the mutex spinning loop
 */
static inline int mutex_can_spin_on_owner(struct mutex *lock)
{
256
	struct task_struct *owner;
257 258
	int retval = 1;

259 260 261
	if (need_resched())
		return 0;

262
	rcu_read_lock();
263
	owner = READ_ONCE(lock->owner);
264 265
	if (owner)
		retval = owner->on_cpu;
266 267 268 269 270 271 272
	rcu_read_unlock();
	/*
	 * if lock->owner is not set, the mutex owner may have just acquired
	 * it and not set the owner yet or the mutex has been released.
	 */
	return retval;
}
273 274 275 276 277 278 279

/*
 * Atomically try to take the lock when it is available
 */
static inline bool mutex_try_to_acquire(struct mutex *lock)
{
	return !mutex_is_locked(lock) &&
280
		(atomic_cmpxchg_acquire(&lock->count, 1, 0) == 1);
281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
}

/*
 * Optimistic spinning.
 *
 * We try to spin for acquisition when we find that the lock owner
 * is currently running on a (different) CPU and while we don't
 * need to reschedule. The rationale is that if the lock owner is
 * running, it is likely to release the lock soon.
 *
 * Since this needs the lock owner, and this mutex implementation
 * doesn't track the owner atomically in the lock field, we need to
 * track it non-atomically.
 *
 * We can't do this for DEBUG_MUTEXES because that relies on wait_lock
 * to serialize everything.
 *
 * The mutex spinners are queued up using MCS lock so that only one
 * spinner can compete for the mutex. However, if mutex spinning isn't
 * going to happen, there is no point in going through the lock/unlock
 * overhead.
 *
 * Returns true when the lock was taken, otherwise false, indicating
 * that we need to jump to the slowpath and sleep.
 */
static bool mutex_optimistic_spin(struct mutex *lock,
				  struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx)
{
	struct task_struct *task = current;

	if (!mutex_can_spin_on_owner(lock))
		goto done;

314 315 316 317 318
	/*
	 * In order to avoid a stampede of mutex spinners trying to
	 * acquire the mutex all at once, the spinners need to take a
	 * MCS (queued) lock first before spinning on the owner field.
	 */
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
	if (!osq_lock(&lock->osq))
		goto done;

	while (true) {
		struct task_struct *owner;

		if (use_ww_ctx && ww_ctx->acquired > 0) {
			struct ww_mutex *ww;

			ww = container_of(lock, struct ww_mutex, base);
			/*
			 * If ww->ctx is set the contents are undefined, only
			 * by acquiring wait_lock there is a guarantee that
			 * they are not invalid when reading.
			 *
			 * As such, when deadlock detection needs to be
			 * performed the optimistic spinning cannot be done.
			 */
337
			if (READ_ONCE(ww->ctx))
338 339 340 341 342 343 344
				break;
		}

		/*
		 * If there's an owner, wait for it to either
		 * release the lock or go to sleep.
		 */
345
		owner = READ_ONCE(lock->owner);
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389
		if (owner && !mutex_spin_on_owner(lock, owner))
			break;

		/* Try to acquire the mutex if it is unlocked. */
		if (mutex_try_to_acquire(lock)) {
			lock_acquired(&lock->dep_map, ip);

			if (use_ww_ctx) {
				struct ww_mutex *ww;
				ww = container_of(lock, struct ww_mutex, base);

				ww_mutex_set_context_fastpath(ww, ww_ctx);
			}

			mutex_set_owner(lock);
			osq_unlock(&lock->osq);
			return true;
		}

		/*
		 * When there's no owner, we might have preempted between the
		 * owner acquiring the lock and setting the owner field. If
		 * we're an RT task that will live-lock because we won't let
		 * the owner complete.
		 */
		if (!owner && (need_resched() || rt_task(task)))
			break;

		/*
		 * The cpu_relax() call is a compiler barrier which forces
		 * everything in this loop to be re-loaded. We don't need
		 * memory barriers as we'll eventually observe the right
		 * values at the cost of a few extra spins.
		 */
		cpu_relax_lowlatency();
	}

	osq_unlock(&lock->osq);
done:
	/*
	 * If we fell out of the spin path because of need_resched(),
	 * reschedule now, before we try-lock the mutex. This avoids getting
	 * scheduled out right after we obtained the mutex.
	 */
390 391 392 393 394 395
	if (need_resched()) {
		/*
		 * We _should_ have TASK_RUNNING here, but just in case
		 * we do not, make it so, otherwise we might get stuck.
		 */
		__set_current_state(TASK_RUNNING);
396
		schedule_preempt_disabled();
397
	}
398 399 400 401 402 403 404 405 406

	return false;
}
#else
static bool mutex_optimistic_spin(struct mutex *lock,
				  struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx)
{
	return false;
}
407 408
#endif

A
Andi Kleen 已提交
409 410
__visible __used noinline
void __sched __mutex_unlock_slowpath(atomic_t *lock_count);
I
Ingo Molnar 已提交
411

412
/**
I
Ingo Molnar 已提交
413 414 415 416 417 418 419 420 421 422
 * mutex_unlock - release the mutex
 * @lock: the mutex to be released
 *
 * Unlock a mutex that has been locked by this task previously.
 *
 * This function must not be used in interrupt context. Unlocking
 * of a not locked mutex is not allowed.
 *
 * This function is similar to (but not equivalent to) up().
 */
423
void __sched mutex_unlock(struct mutex *lock)
I
Ingo Molnar 已提交
424 425 426 427 428
{
	/*
	 * The unlocking fastpath is the 0->1 transition from 'locked'
	 * into 'unlocked' state:
	 */
429 430 431 432 433 434 435 436
#ifndef CONFIG_DEBUG_MUTEXES
	/*
	 * When debugging is enabled we must not clear the owner before time,
	 * the slow path will always be taken, and that clears the owner field
	 * after verifying that it was indeed current.
	 */
	mutex_clear_owner(lock);
#endif
I
Ingo Molnar 已提交
437 438 439 440 441
	__mutex_fastpath_unlock(&lock->count, __mutex_unlock_slowpath);
}

EXPORT_SYMBOL(mutex_unlock);

442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480
/**
 * ww_mutex_unlock - release the w/w mutex
 * @lock: the mutex to be released
 *
 * Unlock a mutex that has been locked by this task previously with any of the
 * ww_mutex_lock* functions (with or without an acquire context). It is
 * forbidden to release the locks after releasing the acquire context.
 *
 * This function must not be used in interrupt context. Unlocking
 * of a unlocked mutex is not allowed.
 */
void __sched ww_mutex_unlock(struct ww_mutex *lock)
{
	/*
	 * The unlocking fastpath is the 0->1 transition from 'locked'
	 * into 'unlocked' state:
	 */
	if (lock->ctx) {
#ifdef CONFIG_DEBUG_MUTEXES
		DEBUG_LOCKS_WARN_ON(!lock->ctx->acquired);
#endif
		if (lock->ctx->acquired > 0)
			lock->ctx->acquired--;
		lock->ctx = NULL;
	}

#ifndef CONFIG_DEBUG_MUTEXES
	/*
	 * When debugging is enabled we must not clear the owner before time,
	 * the slow path will always be taken, and that clears the owner field
	 * after verifying that it was indeed current.
	 */
	mutex_clear_owner(&lock->base);
#endif
	__mutex_fastpath_unlock(&lock->base.count, __mutex_unlock_slowpath);
}
EXPORT_SYMBOL(ww_mutex_unlock);

static inline int __sched
481
__ww_mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx)
482 483
{
	struct ww_mutex *ww = container_of(lock, struct ww_mutex, base);
484
	struct ww_acquire_ctx *hold_ctx = READ_ONCE(ww->ctx);
485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503

	if (!hold_ctx)
		return 0;

	if (unlikely(ctx == hold_ctx))
		return -EALREADY;

	if (ctx->stamp - hold_ctx->stamp <= LONG_MAX &&
	    (ctx->stamp != hold_ctx->stamp || ctx > hold_ctx)) {
#ifdef CONFIG_DEBUG_MUTEXES
		DEBUG_LOCKS_WARN_ON(ctx->contending_lock);
		ctx->contending_lock = ww;
#endif
		return -EDEADLK;
	}

	return 0;
}

I
Ingo Molnar 已提交
504 505 506
/*
 * Lock a mutex (possibly interruptible), slowpath:
 */
507
static __always_inline int __sched
P
Peter Zijlstra 已提交
508
__mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
509
		    struct lockdep_map *nest_lock, unsigned long ip,
510
		    struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx)
I
Ingo Molnar 已提交
511 512 513
{
	struct task_struct *task = current;
	struct mutex_waiter waiter;
514
	unsigned long flags;
515
	int ret;
I
Ingo Molnar 已提交
516

P
Peter Zijlstra 已提交
517
	preempt_disable();
518
	mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip);
519

520 521 522 523
	if (mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx)) {
		/* got the lock, yay! */
		preempt_enable();
		return 0;
524
	}
525

526
	spin_lock_mutex(&lock->wait_lock, flags);
I
Ingo Molnar 已提交
527

528 529
	/*
	 * Once more, try to acquire the lock. Only try-lock the mutex if
530
	 * it is unlocked to reduce unnecessary xchg() operations.
531
	 */
532 533
	if (!mutex_is_locked(lock) &&
	    (atomic_xchg_acquire(&lock->count, 0) == 1))
534 535
		goto skip_wait;

536
	debug_mutex_lock_common(lock, &waiter);
R
Roman Zippel 已提交
537
	debug_mutex_add_waiter(lock, &waiter, task_thread_info(task));
I
Ingo Molnar 已提交
538 539 540 541 542

	/* add waiting tasks to the end of the waitqueue (FIFO): */
	list_add_tail(&waiter.list, &lock->wait_list);
	waiter.task = task;

P
Peter Zijlstra 已提交
543
	lock_contended(&lock->dep_map, ip);
544

I
Ingo Molnar 已提交
545 546 547 548 549 550 551 552
	for (;;) {
		/*
		 * Lets try to take the lock again - this is needed even if
		 * we get here for the first time (shortly after failing to
		 * acquire the lock), to make sure that we get a wakeup once
		 * it's unlocked. Later on, if we sleep, this is the
		 * operation that gives us the lock. We xchg it to -1, so
		 * that when we release the lock, we properly wake up the
553 554
		 * other waiters. We only attempt the xchg if the count is
		 * non-negative in order to avoid unnecessary xchg operations:
I
Ingo Molnar 已提交
555
		 */
556
		if (atomic_read(&lock->count) >= 0 &&
557
		    (atomic_xchg_acquire(&lock->count, -1) == 1))
I
Ingo Molnar 已提交
558 559 560 561 562 563
			break;

		/*
		 * got a signal? (This code gets eliminated in the
		 * TASK_UNINTERRUPTIBLE case.)
		 */
564
		if (unlikely(signal_pending_state(state, task))) {
565 566 567
			ret = -EINTR;
			goto err;
		}
I
Ingo Molnar 已提交
568

569
		if (use_ww_ctx && ww_ctx->acquired > 0) {
570
			ret = __ww_mutex_lock_check_stamp(lock, ww_ctx);
571 572
			if (ret)
				goto err;
I
Ingo Molnar 已提交
573
		}
574

I
Ingo Molnar 已提交
575 576
		__set_task_state(task, state);

L
Lucas De Marchi 已提交
577
		/* didn't get the lock, go to sleep: */
578
		spin_unlock_mutex(&lock->wait_lock, flags);
579
		schedule_preempt_disabled();
580
		spin_lock_mutex(&lock->wait_lock, flags);
I
Ingo Molnar 已提交
581
	}
582 583
	__set_task_state(task, TASK_RUNNING);

584 585 586 587 588
	mutex_remove_waiter(lock, &waiter, current_thread_info());
	/* set it to 0 if there are no waiters left: */
	if (likely(list_empty(&lock->wait_list)))
		atomic_set(&lock->count, 0);
	debug_mutex_free_waiter(&waiter);
I
Ingo Molnar 已提交
589

590 591
skip_wait:
	/* got the lock - cleanup and rejoice! */
P
Peter Zijlstra 已提交
592
	lock_acquired(&lock->dep_map, ip);
593
	mutex_set_owner(lock);
I
Ingo Molnar 已提交
594

595
	if (use_ww_ctx) {
596
		struct ww_mutex *ww = container_of(lock, struct ww_mutex, base);
597
		ww_mutex_set_context_slowpath(ww, ww_ctx);
598 599
	}

600
	spin_unlock_mutex(&lock->wait_lock, flags);
P
Peter Zijlstra 已提交
601
	preempt_enable();
I
Ingo Molnar 已提交
602
	return 0;
603 604 605 606 607 608 609 610

err:
	mutex_remove_waiter(lock, &waiter, task_thread_info(task));
	spin_unlock_mutex(&lock->wait_lock, flags);
	debug_mutex_free_waiter(&waiter);
	mutex_release(&lock->dep_map, 1, ip);
	preempt_enable();
	return ret;
I
Ingo Molnar 已提交
611 612
}

613 614 615 616 617
#ifdef CONFIG_DEBUG_LOCK_ALLOC
void __sched
mutex_lock_nested(struct mutex *lock, unsigned int subclass)
{
	might_sleep();
618
	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE,
619
			    subclass, NULL, _RET_IP_, NULL, 0);
620 621 622
}

EXPORT_SYMBOL_GPL(mutex_lock_nested);
623

624 625 626 627
void __sched
_mutex_lock_nest_lock(struct mutex *lock, struct lockdep_map *nest)
{
	might_sleep();
628
	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE,
629
			    0, nest, _RET_IP_, NULL, 0);
630 631 632 633
}

EXPORT_SYMBOL_GPL(_mutex_lock_nest_lock);

L
Liam R. Howlett 已提交
634 635 636 637
int __sched
mutex_lock_killable_nested(struct mutex *lock, unsigned int subclass)
{
	might_sleep();
638
	return __mutex_lock_common(lock, TASK_KILLABLE,
639
				   subclass, NULL, _RET_IP_, NULL, 0);
L
Liam R. Howlett 已提交
640 641 642
}
EXPORT_SYMBOL_GPL(mutex_lock_killable_nested);

643 644 645 646
int __sched
mutex_lock_interruptible_nested(struct mutex *lock, unsigned int subclass)
{
	might_sleep();
647
	return __mutex_lock_common(lock, TASK_INTERRUPTIBLE,
648
				   subclass, NULL, _RET_IP_, NULL, 0);
649 650 651
}

EXPORT_SYMBOL_GPL(mutex_lock_interruptible_nested);
652

653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677
static inline int
ww_mutex_deadlock_injection(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
{
#ifdef CONFIG_DEBUG_WW_MUTEX_SLOWPATH
	unsigned tmp;

	if (ctx->deadlock_inject_countdown-- == 0) {
		tmp = ctx->deadlock_inject_interval;
		if (tmp > UINT_MAX/4)
			tmp = UINT_MAX;
		else
			tmp = tmp*2 + tmp + tmp/2;

		ctx->deadlock_inject_interval = tmp;
		ctx->deadlock_inject_countdown = tmp;
		ctx->contending_lock = lock;

		ww_mutex_unlock(lock);

		return -EDEADLK;
	}
#endif

	return 0;
}
678 679 680 681

int __sched
__ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
{
682 683
	int ret;

684
	might_sleep();
685
	ret =  __mutex_lock_common(&lock->base, TASK_UNINTERRUPTIBLE,
686
				   0, &ctx->dep_map, _RET_IP_, ctx, 1);
687
	if (!ret && ctx->acquired > 1)
688 689 690
		return ww_mutex_deadlock_injection(lock, ctx);

	return ret;
691 692 693 694 695 696
}
EXPORT_SYMBOL_GPL(__ww_mutex_lock);

int __sched
__ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
{
697 698
	int ret;

699
	might_sleep();
700
	ret = __mutex_lock_common(&lock->base, TASK_INTERRUPTIBLE,
701
				  0, &ctx->dep_map, _RET_IP_, ctx, 1);
702

703
	if (!ret && ctx->acquired > 1)
704 705 706
		return ww_mutex_deadlock_injection(lock, ctx);

	return ret;
707 708 709
}
EXPORT_SYMBOL_GPL(__ww_mutex_lock_interruptible);

710 711
#endif

I
Ingo Molnar 已提交
712 713 714
/*
 * Release the lock, slowpath:
 */
715
static inline void
716
__mutex_unlock_common_slowpath(struct mutex *lock, int nested)
I
Ingo Molnar 已提交
717
{
718
	unsigned long flags;
I
Ingo Molnar 已提交
719 720

	/*
721 722 723 724 725 726 727 728
	 * As a performance measurement, release the lock before doing other
	 * wakeup related duties to follow. This allows other tasks to acquire
	 * the lock sooner, while still handling cleanups in past unlock calls.
	 * This can be done as we do not enforce strict equivalence between the
	 * mutex counter and wait_list.
	 *
	 *
	 * Some architectures leave the lock unlocked in the fastpath failure
I
Ingo Molnar 已提交
729
	 * case, others need to leave it locked. In the later case we have to
730
	 * unlock it here - as the lock counter is currently 0 or negative.
I
Ingo Molnar 已提交
731 732 733 734
	 */
	if (__mutex_slowpath_needs_to_unlock())
		atomic_set(&lock->count, 1);

735 736 737 738
	spin_lock_mutex(&lock->wait_lock, flags);
	mutex_release(&lock->dep_map, nested, _RET_IP_);
	debug_mutex_unlock(lock);

I
Ingo Molnar 已提交
739 740 741 742 743 744 745 746 747 748 749
	if (!list_empty(&lock->wait_list)) {
		/* get the first entry from the wait-list: */
		struct mutex_waiter *waiter =
				list_entry(lock->wait_list.next,
					   struct mutex_waiter, list);

		debug_mutex_wake_waiter(lock, waiter);

		wake_up_process(waiter->task);
	}

750
	spin_unlock_mutex(&lock->wait_lock, flags);
I
Ingo Molnar 已提交
751 752
}

753 754 755
/*
 * Release the lock, slowpath:
 */
A
Andi Kleen 已提交
756
__visible void
757 758
__mutex_unlock_slowpath(atomic_t *lock_count)
{
759 760 761
	struct mutex *lock = container_of(lock_count, struct mutex, count);

	__mutex_unlock_common_slowpath(lock, 1);
762 763
}

P
Peter Zijlstra 已提交
764
#ifndef CONFIG_DEBUG_LOCK_ALLOC
I
Ingo Molnar 已提交
765 766 767 768
/*
 * Here come the less common (and hence less performance-critical) APIs:
 * mutex_lock_interruptible() and mutex_trylock().
 */
769
static noinline int __sched
770
__mutex_lock_killable_slowpath(struct mutex *lock);
L
Liam R. Howlett 已提交
771

772
static noinline int __sched
773
__mutex_lock_interruptible_slowpath(struct mutex *lock);
I
Ingo Molnar 已提交
774

775 776
/**
 * mutex_lock_interruptible - acquire the mutex, interruptible
I
Ingo Molnar 已提交
777 778 779 780 781 782 783 784 785
 * @lock: the mutex to be acquired
 *
 * Lock the mutex like mutex_lock(), and return 0 if the mutex has
 * been acquired or sleep until the mutex becomes available. If a
 * signal arrives while waiting for the lock then this function
 * returns -EINTR.
 *
 * This function is similar to (but not equivalent to) down_interruptible().
 */
786
int __sched mutex_lock_interruptible(struct mutex *lock)
I
Ingo Molnar 已提交
787
{
788 789
	int ret;

790
	might_sleep();
791 792
	ret =  __mutex_fastpath_lock_retval(&lock->count);
	if (likely(!ret)) {
793
		mutex_set_owner(lock);
794 795 796
		return 0;
	} else
		return __mutex_lock_interruptible_slowpath(lock);
I
Ingo Molnar 已提交
797 798 799 800
}

EXPORT_SYMBOL(mutex_lock_interruptible);

801
int __sched mutex_lock_killable(struct mutex *lock)
L
Liam R. Howlett 已提交
802
{
803 804
	int ret;

L
Liam R. Howlett 已提交
805
	might_sleep();
806 807
	ret = __mutex_fastpath_lock_retval(&lock->count);
	if (likely(!ret)) {
808
		mutex_set_owner(lock);
809 810 811
		return 0;
	} else
		return __mutex_lock_killable_slowpath(lock);
L
Liam R. Howlett 已提交
812 813 814
}
EXPORT_SYMBOL(mutex_lock_killable);

A
Andi Kleen 已提交
815
__visible void __sched
P
Peter Zijlstra 已提交
816 817 818 819
__mutex_lock_slowpath(atomic_t *lock_count)
{
	struct mutex *lock = container_of(lock_count, struct mutex, count);

820
	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, 0,
821
			    NULL, _RET_IP_, NULL, 0);
P
Peter Zijlstra 已提交
822 823
}

824
static noinline int __sched
825
__mutex_lock_killable_slowpath(struct mutex *lock)
L
Liam R. Howlett 已提交
826
{
827
	return __mutex_lock_common(lock, TASK_KILLABLE, 0,
828
				   NULL, _RET_IP_, NULL, 0);
L
Liam R. Howlett 已提交
829 830
}

831
static noinline int __sched
832
__mutex_lock_interruptible_slowpath(struct mutex *lock)
I
Ingo Molnar 已提交
833
{
834
	return __mutex_lock_common(lock, TASK_INTERRUPTIBLE, 0,
835
				   NULL, _RET_IP_, NULL, 0);
836 837 838 839 840 841
}

static noinline int __sched
__ww_mutex_lock_slowpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
{
	return __mutex_lock_common(&lock->base, TASK_UNINTERRUPTIBLE, 0,
842
				   NULL, _RET_IP_, ctx, 1);
I
Ingo Molnar 已提交
843
}
844 845 846 847 848 849

static noinline int __sched
__ww_mutex_lock_interruptible_slowpath(struct ww_mutex *lock,
					    struct ww_acquire_ctx *ctx)
{
	return __mutex_lock_common(&lock->base, TASK_INTERRUPTIBLE, 0,
850
				   NULL, _RET_IP_, ctx, 1);
851 852
}

P
Peter Zijlstra 已提交
853
#endif
I
Ingo Molnar 已提交
854 855 856 857 858 859 860 861

/*
 * Spinlock based trylock, we take the spinlock and check whether we
 * can get the lock:
 */
static inline int __mutex_trylock_slowpath(atomic_t *lock_count)
{
	struct mutex *lock = container_of(lock_count, struct mutex, count);
862
	unsigned long flags;
I
Ingo Molnar 已提交
863 864
	int prev;

865 866 867 868
	/* No need to trylock if the mutex is locked. */
	if (mutex_is_locked(lock))
		return 0;

869
	spin_lock_mutex(&lock->wait_lock, flags);
I
Ingo Molnar 已提交
870

871
	prev = atomic_xchg_acquire(&lock->count, -1);
872
	if (likely(prev == 1)) {
873
		mutex_set_owner(lock);
874 875
		mutex_acquire(&lock->dep_map, 0, 1, _RET_IP_);
	}
876

I
Ingo Molnar 已提交
877 878 879 880
	/* Set it back to 0 if there are no waiters: */
	if (likely(list_empty(&lock->wait_list)))
		atomic_set(&lock->count, 0);

881
	spin_unlock_mutex(&lock->wait_lock, flags);
I
Ingo Molnar 已提交
882 883 884 885

	return prev == 1;
}

886 887
/**
 * mutex_trylock - try to acquire the mutex, without waiting
I
Ingo Molnar 已提交
888 889 890 891 892 893
 * @lock: the mutex to be acquired
 *
 * Try to acquire the mutex atomically. Returns 1 if the mutex
 * has been acquired successfully, and 0 on contention.
 *
 * NOTE: this function follows the spin_trylock() convention, so
894
 * it is negated from the down_trylock() return values! Be careful
I
Ingo Molnar 已提交
895 896 897 898 899
 * about this when converting semaphore users to mutexes.
 *
 * This function must not be used in interrupt context. The
 * mutex must be released by the same task that acquired it.
 */
900
int __sched mutex_trylock(struct mutex *lock)
I
Ingo Molnar 已提交
901
{
902 903 904 905 906 907 908
	int ret;

	ret = __mutex_fastpath_trylock(&lock->count, __mutex_trylock_slowpath);
	if (ret)
		mutex_set_owner(lock);

	return ret;
I
Ingo Molnar 已提交
909 910
}
EXPORT_SYMBOL(mutex_trylock);
911

912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950
#ifndef CONFIG_DEBUG_LOCK_ALLOC
int __sched
__ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
{
	int ret;

	might_sleep();

	ret = __mutex_fastpath_lock_retval(&lock->base.count);

	if (likely(!ret)) {
		ww_mutex_set_context_fastpath(lock, ctx);
		mutex_set_owner(&lock->base);
	} else
		ret = __ww_mutex_lock_slowpath(lock, ctx);
	return ret;
}
EXPORT_SYMBOL(__ww_mutex_lock);

int __sched
__ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
{
	int ret;

	might_sleep();

	ret = __mutex_fastpath_lock_retval(&lock->base.count);

	if (likely(!ret)) {
		ww_mutex_set_context_fastpath(lock, ctx);
		mutex_set_owner(&lock->base);
	} else
		ret = __ww_mutex_lock_interruptible_slowpath(lock, ctx);
	return ret;
}
EXPORT_SYMBOL(__ww_mutex_lock_interruptible);

#endif

951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973
/**
 * atomic_dec_and_mutex_lock - return holding mutex if we dec to 0
 * @cnt: the atomic which we are to dec
 * @lock: the mutex to return holding if we dec to 0
 *
 * return true and hold lock if we dec to 0, return false otherwise
 */
int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock)
{
	/* dec if we can't possibly hit 0 */
	if (atomic_add_unless(cnt, -1, 1))
		return 0;
	/* we might hit 0, so take the lock */
	mutex_lock(lock);
	if (!atomic_dec_and_test(cnt)) {
		/* when we actually did the dec, we didn't hit 0 */
		mutex_unlock(lock);
		return 0;
	}
	/* we hit 0, and we hold the lock */
	return 1;
}
EXPORT_SYMBOL(atomic_dec_and_mutex_lock);