Merge branch 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull core locking updates from Ingo Molnar:
 "The main updates in this cycle were:

   - mutex MCS refactoring finishing touches: improve comments, refactor
     and clean up code, reduce debug data structure footprint, etc.

   - qrwlock finishing touches: remove old code, self-test updates.

   - small rwsem optimization

   - various smaller fixes/cleanups"

* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/lockdep: Revert qrwlock recusive stuff
  locking/rwsem: Avoid double checking before try acquiring write lock
  locking/rwsem: Move EXPORT_SYMBOL() lines to follow function definition
  locking/rwlock, x86: Delete unused asm/rwlock.h and rwlock.S
  locking/rwlock, x86: Clean up asm/spinlock*.h to remove old rwlock code
  locking/semaphore: Resolve some shadow warnings
  locking/selftest: Support queued rwlock
  locking/lockdep: Restrict the use of recursive read_lock() with qrwlock
  locking/spinlocks: Always evaluate the second argument of spin_lock_nested()
  locking/Documentation: Update locking/mutex-design.txt disadvantages
  locking/Documentation: Move locking related docs into Documentation/locking/
  locking/mutexes: Use MUTEX_SPIN_ON_OWNER when appropriate
  locking/mutexes: Refactor optimistic spinning code
  locking/mcs: Remove obsolete comment
  locking/mutexes: Document quick lock release when unlocking
  locking/mutexes: Standardize arguments in lock/unlock slowpaths
  locking: Remove deprecated smp_mb__() barriers

Documentation/00-INDEX

@@ -287,6 +287,8 @@ local_ops.txt
 	- semantics and behavior of local atomic operations.
 lockdep-design.txt
 	- documentation on the runtime locking correctness validator.
+locking/
+	- directory with info about kernel locking primitives
 lockstat.txt
 	- info on collecting statistics on locks (and contention).
 lockup-watchdogs.txt

Documentation/DocBook/kernel-locking.tmpl

@@ -1972,7 +1972,7 @@ machines due to caching.
    <itemizedlist>
     <listitem>
      <para>
-       <filename>Documentation/spinlocks.txt</filename>: 
+       <filename>Documentation/locking/spinlocks.txt</filename>:
        Linus Torvalds' spinlocking tutorial in the kernel sources.
      </para>
     </listitem>

Documentation/lockstat.txt → Documentation/locking/lockstat.txt

@@ -12,7 +12,7 @@ Because things like lock contention can severely impact performance.
 - HOW
 
 Lockdep already has hooks in the lock functions and maps lock instances to
-lock classes. We build on that (see Documentation/lockdep-design.txt).
+lock classes. We build on that (see Documentation/lokcing/lockdep-design.txt).
 The graph below shows the relation between the lock functions and the various
 hooks therein.
 

Documentation/mutex-design.txt → Documentation/locking/mutex-design.txt

@@ -145,9 +145,9 @@ Disadvantages
 
 Unlike its original design and purpose, 'struct mutex' is larger than
 most locks in the kernel. E.g: on x86-64 it is 40 bytes, almost twice
-as large as 'struct semaphore' (24 bytes) and 8 bytes shy of the
-'struct rw_semaphore' variant. Larger structure sizes mean more CPU
-cache and memory footprint.
+as large as 'struct semaphore' (24 bytes) and tied, along with rwsems,
+for the largest lock in the kernel. Larger structure sizes mean more
+CPU cache and memory footprint.
 
 When to use mutexes
 -------------------

MAINTAINERS

@@ -5680,8 +5680,8 @@ M:	Ingo Molnar <mingo@redhat.com>
 L:	linux-kernel@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git core/locking
 S:	Maintained
-F:	Documentation/lockdep*.txt
-F:	Documentation/lockstat.txt
+F:	Documentation/locking/lockdep*.txt
+F:	Documentation/locking/lockstat.txt
 F:	include/linux/lockdep.h
 F:	kernel/locking/
 

arch/x86/include/asm/rwlock.h

-#ifndef _ASM_X86_RWLOCK_H
-#define _ASM_X86_RWLOCK_H
-
-#include <asm/asm.h>
-
-#if CONFIG_NR_CPUS <= 2048
-
-#ifndef __ASSEMBLY__
-typedef union {
-	s32 lock;
-	s32 write;
-} arch_rwlock_t;
-#endif
-
-#define RW_LOCK_BIAS		0x00100000
-#define READ_LOCK_SIZE(insn)	__ASM_FORM(insn##l)
-#define READ_LOCK_ATOMIC(n)	atomic_##n
-#define WRITE_LOCK_ADD(n)	__ASM_FORM_COMMA(addl n)
-#define WRITE_LOCK_SUB(n)	__ASM_FORM_COMMA(subl n)
-#define WRITE_LOCK_CMP		RW_LOCK_BIAS
-
-#else /* CONFIG_NR_CPUS > 2048 */
-
-#include <linux/const.h>
-
-#ifndef __ASSEMBLY__
-typedef union {
-	s64 lock;
-	struct {
-		u32 read;
-		s32 write;
-	};
-} arch_rwlock_t;
-#endif
-
-#define RW_LOCK_BIAS		(_AC(1,L) << 32)
-#define READ_LOCK_SIZE(insn)	__ASM_FORM(insn##q)
-#define READ_LOCK_ATOMIC(n)	atomic64_##n
-#define WRITE_LOCK_ADD(n)	__ASM_FORM(incl)
-#define WRITE_LOCK_SUB(n)	__ASM_FORM(decl)
-#define WRITE_LOCK_CMP		1
-
-#endif /* CONFIG_NR_CPUS */
-
-#define __ARCH_RW_LOCK_UNLOCKED		{ RW_LOCK_BIAS }
-
-/* Actual code is in asm/spinlock.h or in arch/x86/lib/rwlock.S */
-
-#endif /* _ASM_X86_RWLOCK_H */

arch/x86/include/asm/spinlock.h

@@ -187,7 +187,6 @@ static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
 		cpu_relax();
 }
 
-#ifndef CONFIG_QUEUE_RWLOCK
 /*
  * Read-write spinlocks, allowing multiple readers
  * but only one writer.
@@ -198,91 +197,15 @@ static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
  * irq-safe write-lock, but readers can get non-irqsafe
  * read-locks.
  *
- * On x86, we implement read-write locks as a 32-bit counter
- * with the high bit (sign) being the "contended" bit.
+ * On x86, we implement read-write locks using the generic qrwlock with
+ * x86 specific optimization.
  */
 
-/**
- * read_can_lock - would read_trylock() succeed?
- * @lock: the rwlock in question.
- */
-static inline int arch_read_can_lock(arch_rwlock_t *lock)
-{
-	return lock->lock > 0;
-}
-
-/**
- * write_can_lock - would write_trylock() succeed?
- * @lock: the rwlock in question.
- */
-static inline int arch_write_can_lock(arch_rwlock_t *lock)
-{
-	return lock->write == WRITE_LOCK_CMP;
-}
-
-static inline void arch_read_lock(arch_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX READ_LOCK_SIZE(dec) " (%0)\n\t"
-		     "jns 1f\n"
-		     "call __read_lock_failed\n\t"
-		     "1:\n"
-		     ::LOCK_PTR_REG (rw) : "memory");
-}
-
-static inline void arch_write_lock(arch_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX WRITE_LOCK_SUB(%1) "(%0)\n\t"
-		     "jz 1f\n"
-		     "call __write_lock_failed\n\t"
-		     "1:\n"
-		     ::LOCK_PTR_REG (&rw->write), "i" (RW_LOCK_BIAS)
-		     : "memory");
-}
-
-static inline int arch_read_trylock(arch_rwlock_t *lock)
-{
-	READ_LOCK_ATOMIC(t) *count = (READ_LOCK_ATOMIC(t) *)lock;
-
-	if (READ_LOCK_ATOMIC(dec_return)(count) >= 0)
-		return 1;
-	READ_LOCK_ATOMIC(inc)(count);
-	return 0;
-}
-
-static inline int arch_write_trylock(arch_rwlock_t *lock)
-{
-	atomic_t *count = (atomic_t *)&lock->write;
-
-	if (atomic_sub_and_test(WRITE_LOCK_CMP, count))
-		return 1;
-	atomic_add(WRITE_LOCK_CMP, count);
-	return 0;
-}
-
-static inline void arch_read_unlock(arch_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX READ_LOCK_SIZE(inc) " %0"
-		     :"+m" (rw->lock) : : "memory");
-}
-
-static inline void arch_write_unlock(arch_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX WRITE_LOCK_ADD(%1) "%0"
-		     : "+m" (rw->write) : "i" (RW_LOCK_BIAS) : "memory");
-}
-#else
 #include <asm/qrwlock.h>
-#endif /* CONFIG_QUEUE_RWLOCK */
 
 #define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
 #define arch_write_lock_flags(lock, flags) arch_write_lock(lock)
 
-#undef READ_LOCK_SIZE
-#undef READ_LOCK_ATOMIC
-#undef WRITE_LOCK_ADD
-#undef WRITE_LOCK_SUB
-#undef WRITE_LOCK_CMP
-
 #define arch_spin_relax(lock)	cpu_relax()
 #define arch_read_relax(lock)	cpu_relax()
 #define arch_write_relax(lock)	cpu_relax()

arch/x86/include/asm/spinlock_types.h

@@ -34,10 +34,6 @@ typedef struct arch_spinlock {
 
 #define __ARCH_SPIN_LOCK_UNLOCKED	{ { 0 } }
 
-#ifdef CONFIG_QUEUE_RWLOCK
 #include <asm-generic/qrwlock_types.h>
-#else
-#include <asm/rwlock.h>
-#endif
 
 #endif /* _ASM_X86_SPINLOCK_TYPES_H */

arch/x86/lib/Makefile

@@ -20,7 +20,6 @@ lib-y := delay.o misc.o cmdline.o
 lib-y += thunk_$(BITS).o
 lib-y += usercopy_$(BITS).o usercopy.o getuser.o putuser.o
 lib-y += memcpy_$(BITS).o
-lib-$(CONFIG_SMP) += rwlock.o
 lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o
 lib-$(CONFIG_INSTRUCTION_DECODER) += insn.o inat.o
 

arch/x86/lib/rwlock.S

-/* Slow paths of read/write spinlocks. */
-
-#include <linux/linkage.h>
-#include <asm/alternative-asm.h>
-#include <asm/frame.h>
-#include <asm/rwlock.h>
-
-#ifdef CONFIG_X86_32
-# define __lock_ptr eax
-#else
-# define __lock_ptr rdi
-#endif
-
-ENTRY(__write_lock_failed)
-	CFI_STARTPROC
-	FRAME
-0:	LOCK_PREFIX
-	WRITE_LOCK_ADD($RW_LOCK_BIAS) (%__lock_ptr)
-1:	rep; nop
-	cmpl	$WRITE_LOCK_CMP, (%__lock_ptr)
-	jne	1b
-	LOCK_PREFIX
-	WRITE_LOCK_SUB($RW_LOCK_BIAS) (%__lock_ptr)
-	jnz	0b
-	ENDFRAME
-	ret
-	CFI_ENDPROC
-END(__write_lock_failed)
-
-ENTRY(__read_lock_failed)
-	CFI_STARTPROC
-	FRAME
-0:	LOCK_PREFIX
-	READ_LOCK_SIZE(inc) (%__lock_ptr)
-1:	rep; nop
-	READ_LOCK_SIZE(cmp) $1, (%__lock_ptr)
-	js	1b
-	LOCK_PREFIX
-	READ_LOCK_SIZE(dec) (%__lock_ptr)
-	js	0b
-	ENDFRAME
-	ret
-	CFI_ENDPROC
-END(__read_lock_failed)

drivers/gpu/drm/drm_modeset_lock.c

@@ -35,7 +35,7 @@
  * of extra utility/tracking out of our acquire-ctx.  This is provided
  * by drm_modeset_lock / drm_modeset_acquire_ctx.
  *
- * For basic principles of ww_mutex, see: Documentation/ww-mutex-design.txt
+ * For basic principles of ww_mutex, see: Documentation/locking/ww-mutex-design.txt
  *
  * The basic usage pattern is to:
  *

include/linux/atomic.h

@@ -3,42 +3,6 @@
 #define _LINUX_ATOMIC_H
 #include <asm/atomic.h>
 
-/*
- * Provide __deprecated wrappers for the new interface, avoid flag day changes.
- * We need the ugly external functions to break header recursion hell.
- */
-#ifndef smp_mb__before_atomic_inc
-static inline void __deprecated smp_mb__before_atomic_inc(void)
-{
-	extern void __smp_mb__before_atomic(void);
-	__smp_mb__before_atomic();
-}
-#endif
-
-#ifndef smp_mb__after_atomic_inc
-static inline void __deprecated smp_mb__after_atomic_inc(void)
-{
-	extern void __smp_mb__after_atomic(void);
-	__smp_mb__after_atomic();
-}
-#endif
-
-#ifndef smp_mb__before_atomic_dec
-static inline void __deprecated smp_mb__before_atomic_dec(void)
-{
-	extern void __smp_mb__before_atomic(void);
-	__smp_mb__before_atomic();
-}
-#endif
-
-#ifndef smp_mb__after_atomic_dec
-static inline void __deprecated smp_mb__after_atomic_dec(void)
-{
-	extern void __smp_mb__after_atomic(void);
-	__smp_mb__after_atomic();
-}
-#endif
-
 /**
  * atomic_add_unless - add unless the number is already a given value
  * @v: pointer of type atomic_t

include/linux/bitops.h

@@ -32,26 +32,6 @@ extern unsigned long __sw_hweight64(__u64 w);
  */
 #include <asm/bitops.h>
 
-/*
- * Provide __deprecated wrappers for the new interface, avoid flag day changes.
- * We need the ugly external functions to break header recursion hell.
- */
-#ifndef smp_mb__before_clear_bit
-static inline void __deprecated smp_mb__before_clear_bit(void)
-{
-	extern void __smp_mb__before_atomic(void);
-	__smp_mb__before_atomic();
-}
-#endif
-
-#ifndef smp_mb__after_clear_bit
-static inline void __deprecated smp_mb__after_clear_bit(void)
-{
-	extern void __smp_mb__after_atomic(void);
-	__smp_mb__after_atomic();
-}
-#endif
-
 #define for_each_set_bit(bit, addr, size) \
 	for ((bit) = find_first_bit((addr), (size));		\
 	     (bit) < (size);					\

include/linux/lockdep.h

@@ -4,7 +4,7 @@
  *  Copyright (C) 2006,2007 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
  *  Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
  *
- * see Documentation/lockdep-design.txt for more details.
+ * see Documentation/locking/lockdep-design.txt for more details.
  */
 #ifndef __LINUX_LOCKDEP_H
 #define __LINUX_LOCKDEP_H

include/linux/mutex.h

@@ -52,7 +52,7 @@ struct mutex {
 	atomic_t		count;
 	spinlock_t		wait_lock;
 	struct list_head	wait_list;
-#if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_SMP)
+#if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_MUTEX_SPIN_ON_OWNER)
 	struct task_struct	*owner;
 #endif
 #ifdef CONFIG_MUTEX_SPIN_ON_OWNER
@@ -133,7 +133,7 @@ static inline int mutex_is_locked(struct mutex *lock)
 
 /*
  * See kernel/locking/mutex.c for detailed documentation of these APIs.
- * Also see Documentation/mutex-design.txt.
+ * Also see Documentation/locking/mutex-design.txt.
  */
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 extern void mutex_lock_nested(struct mutex *lock, unsigned int subclass);

include/linux/rwsem.h

@@ -149,7 +149,7 @@ extern void downgrade_write(struct rw_semaphore *sem);
  * static then another method for expressing nested locking is
  * the explicit definition of lock class keys and the use of
  * lockdep_set_class() at lock initialization time.
- * See Documentation/lockdep-design.txt for more details.)
+ * See Documentation/locking/lockdep-design.txt for more details.)
  */
 extern void down_read_nested(struct rw_semaphore *sem, int subclass);
 extern void down_write_nested(struct rw_semaphore *sem, int subclass);

include/linux/spinlock.h

@@ -197,7 +197,13 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) __releases(lock)
 		 _raw_spin_lock_nest_lock(lock, &(nest_lock)->dep_map);	\
 	 } while (0)
 #else
-# define raw_spin_lock_nested(lock, subclass)		_raw_spin_lock(lock)
+/*
+ * Always evaluate the 'subclass' argument to avoid that the compiler
+ * warns about set-but-not-used variables when building with
+ * CONFIG_DEBUG_LOCK_ALLOC=n and with W=1.
+ */
+# define raw_spin_lock_nested(lock, subclass)		\
+	_raw_spin_lock(((void)(subclass), (lock)))
 # define raw_spin_lock_nest_lock(lock, nest_lock)	_raw_spin_lock(lock)
 #endif
 

kernel/locking/mcs_spinlock.h

@@ -56,9 +56,6 @@ do {									\
  * If the lock has already been acquired, then this will proceed to spin
  * on this node->locked until the previous lock holder sets the node->locked
  * in mcs_spin_unlock().
- *
- * We don't inline mcs_spin_lock() so that perf can correctly account for the
- * time spent in this lock function.
  */
 static inline
 void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node)

kernel/locking/mutex.c

@@ -15,7 +15,7 @@
  *    by Steven Rostedt, based on work by Gregory Haskins, Peter Morreale
  *    and Sven Dietrich.
  *
- * Also see Documentation/mutex-design.txt.
+ * Also see Documentation/locking/mutex-design.txt.
  */
 #include <linux/mutex.h>
 #include <linux/ww_mutex.h>
@@ -106,6 +106,92 @@ void __sched mutex_lock(struct mutex *lock)
 EXPORT_SYMBOL(mutex_lock);
 #endif
 
+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++;
+}
+
+/*
+ * after acquiring lock with fastpath or when we lost out in contested
+ * 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);
+}
+
+
 #ifdef CONFIG_MUTEX_SPIN_ON_OWNER
 /*
  * In order to avoid a stampede of mutex spinners from acquiring the mutex
@@ -180,6 +266,129 @@ static inline int mutex_can_spin_on_owner(struct mutex *lock)
 	 */
 	return retval;
 }
+
+/*
+ * 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) &&
+		(atomic_cmpxchg(&lock->count, 1, 0) == 1);
+}
+
+/*
+ * 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;
+
+	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.
+			 */
+			if (ACCESS_ONCE(ww->ctx))
+				break;
+		}
+
+		/*
+		 * If there's an owner, wait for it to either
+		 * release the lock or go to sleep.
+		 */
+		owner = ACCESS_ONCE(lock->owner);
+		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.
+	 */
+	if (need_resched())
+		schedule_preempt_disabled();
+
+	return false;
+}
+#else
+static bool mutex_optimistic_spin(struct mutex *lock,
+				  struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx)
+{
+	return false;
+}
 #endif
 
 __visible __used noinline
@@ -277,91 +486,6 @@ __mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx)
 	return 0;
 }
 
-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++;
-}
-
-/*
- * after acquiring lock with fastpath or when we lost out in contested
- * 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);
-}
-
 /*
  * Lock a mutex (possibly interruptible), slowpath:
  */
@@ -378,104 +502,12 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 	preempt_disable();
 	mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip);
 
-#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
-	/*
-	 * 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.
-	 */
-	if (!mutex_can_spin_on_owner(lock))
-		goto slowpath;
-
-	if (!osq_lock(&lock->osq))
-		goto slowpath;
-
-	for (;;) {
-		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.
-			 */
-			if (ACCESS_ONCE(ww->ctx))
-				break;
-		}
-
-		/*
-		 * If there's an owner, wait for it to either
-		 * release the lock or go to sleep.
-		 */
-		owner = ACCESS_ONCE(lock->owner);
-		if (owner && !mutex_spin_on_owner(lock, owner))
-			break;
-
-		/* Try to acquire the mutex if it is unlocked. */
-		if (!mutex_is_locked(lock) &&
-		    (atomic_cmpxchg(&lock->count, 1, 0) == 1)) {
-			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);
+	if (mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx)) {
+		/* got the lock, yay! */
 		preempt_enable();
 		return 0;
 	}
 
-		/*
-		 * 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);
-slowpath:
-	/*
-	 * 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.
-	 */
-	if (need_resched())
-		schedule_preempt_disabled();
-#endif
 	spin_lock_mutex(&lock->wait_lock, flags);
 
 	/*
@@ -679,15 +711,21 @@ EXPORT_SYMBOL_GPL(__ww_mutex_lock_interruptible);
  * Release the lock, slowpath:
  */
 static inline void
-__mutex_unlock_common_slowpath(atomic_t *lock_count, int nested)
+__mutex_unlock_common_slowpath(struct mutex *lock, int nested)
 {
-	struct mutex *lock = container_of(lock_count, struct mutex, count);
 	unsigned long flags;
 
 	/*
-	 * some architectures leave the lock unlocked in the fastpath failure
+	 * 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
 	 * case, others need to leave it locked. In the later case we have to
-	 * unlock it here
+	 * unlock it here - as the lock counter is currently 0 or negative.
 	 */
 	if (__mutex_slowpath_needs_to_unlock())
 		atomic_set(&lock->count, 1);
@@ -716,7 +754,9 @@ __mutex_unlock_common_slowpath(atomic_t *lock_count, int nested)
 __visible void
 __mutex_unlock_slowpath(atomic_t *lock_count)
 {
-	__mutex_unlock_common_slowpath(lock_count, 1);
+	struct mutex *lock = container_of(lock_count, struct mutex, count);
+
+	__mutex_unlock_common_slowpath(lock, 1);
 }
 
 #ifndef CONFIG_DEBUG_LOCK_ALLOC

kernel/locking/mutex.h

@@ -16,7 +16,7 @@
 #define mutex_remove_waiter(lock, waiter, ti) \
 		__list_del((waiter)->list.prev, (waiter)->list.next)
 
-#ifdef CONFIG_SMP
+#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
 static inline void mutex_set_owner(struct mutex *lock)
 {
 	lock->owner = current;

kernel/locking/rtmutex.c

@@ -8,7 +8,7 @@
  *  Copyright (C) 2005 Kihon Technologies Inc., Steven Rostedt
  *  Copyright (C) 2006 Esben Nielsen
  *
- *  See Documentation/rt-mutex-design.txt for details.
+ *  See Documentation/locking/rt-mutex-design.txt for details.
  */
 #include <linux/spinlock.h>
 #include <linux/export.h>

kernel/locking/rwsem-xadd.c

@@ -246,19 +246,22 @@ struct rw_semaphore __sched *rwsem_down_read_failed(struct rw_semaphore *sem)
 
 	return sem;
 }
+EXPORT_SYMBOL(rwsem_down_read_failed);
 
 static inline bool rwsem_try_write_lock(long count, struct rw_semaphore *sem)
 {
-	if (!(count & RWSEM_ACTIVE_MASK)) {
-		/* try acquiring the write lock */
-		if (sem->count == RWSEM_WAITING_BIAS &&
+	/*
+	 * Try acquiring the write lock. Check count first in order
+	 * to reduce unnecessary expensive cmpxchg() operations.
+	 */
+	if (count == RWSEM_WAITING_BIAS &&
 	    cmpxchg(&sem->count, RWSEM_WAITING_BIAS,
 		    RWSEM_ACTIVE_WRITE_BIAS) == RWSEM_WAITING_BIAS) {
 		if (!list_is_singular(&sem->wait_list))
 			rwsem_atomic_update(RWSEM_WAITING_BIAS, sem);
 		return true;
 	}
-	}
+
 	return false;
 }
 
@@ -465,6 +468,7 @@ struct rw_semaphore __sched *rwsem_down_write_failed(struct rw_semaphore *sem)
 
 	return sem;
 }
+EXPORT_SYMBOL(rwsem_down_write_failed);
 
 /*
  * handle waking up a waiter on the semaphore
@@ -485,6 +489,7 @@ struct rw_semaphore *rwsem_wake(struct rw_semaphore *sem)
 
 	return sem;
 }
+EXPORT_SYMBOL(rwsem_wake);
 
 /*
  * downgrade a write lock into a read lock
@@ -506,8 +511,4 @@ struct rw_semaphore *rwsem_downgrade_wake(struct rw_semaphore *sem)
 
 	return sem;
 }
-
-EXPORT_SYMBOL(rwsem_down_read_failed);
-EXPORT_SYMBOL(rwsem_down_write_failed);
-EXPORT_SYMBOL(rwsem_wake);
 EXPORT_SYMBOL(rwsem_downgrade_wake);

kernel/locking/semaphore.c

@@ -36,7 +36,7 @@
 static noinline void __down(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
 static noinline int __down_killable(struct semaphore *sem);
-static noinline int __down_timeout(struct semaphore *sem, long jiffies);
+static noinline int __down_timeout(struct semaphore *sem, long timeout);
 static noinline void __up(struct semaphore *sem);
 
 /**
@@ -145,14 +145,14 @@ EXPORT_SYMBOL(down_trylock);
 /**
  * down_timeout - acquire the semaphore within a specified time
  * @sem: the semaphore to be acquired
- * @jiffies: how long to wait before failing
+ * @timeout: how long to wait before failing
  *
  * Attempts to acquire the semaphore.  If no more tasks are allowed to
  * acquire the semaphore, calling this function will put the task to sleep.
  * If the semaphore is not released within the specified number of jiffies,
  * this function returns -ETIME.  It returns 0 if the semaphore was acquired.
  */
-int down_timeout(struct semaphore *sem, long jiffies)
+int down_timeout(struct semaphore *sem, long timeout)
 {
 	unsigned long flags;
 	int result = 0;
@@ -161,7 +161,7 @@ int down_timeout(struct semaphore *sem, long jiffies)
 	if (likely(sem->count > 0))
 		sem->count--;
 	else
-		result = __down_timeout(sem, jiffies);
+		result = __down_timeout(sem, timeout);
 	raw_spin_unlock_irqrestore(&sem->lock, flags);
 
 	return result;
@@ -248,9 +248,9 @@ static noinline int __sched __down_killable(struct semaphore *sem)
 	return __down_common(sem, TASK_KILLABLE, MAX_SCHEDULE_TIMEOUT);
 }
 
-static noinline int __sched __down_timeout(struct semaphore *sem, long jiffies)
+static noinline int __sched __down_timeout(struct semaphore *sem, long timeout)
 {
-	return __down_common(sem, TASK_UNINTERRUPTIBLE, jiffies);
+	return __down_common(sem, TASK_UNINTERRUPTIBLE, timeout);
 }
 
 static noinline void __sched __up(struct semaphore *sem)

kernel/sched/core.c

@@ -90,22 +90,6 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/sched.h>
 
-#ifdef smp_mb__before_atomic
-void __smp_mb__before_atomic(void)
-{
-	smp_mb__before_atomic();
-}
-EXPORT_SYMBOL(__smp_mb__before_atomic);
-#endif
-
-#ifdef smp_mb__after_atomic
-void __smp_mb__after_atomic(void)
-{
-	smp_mb__after_atomic();
-}
-EXPORT_SYMBOL(__smp_mb__after_atomic);
-#endif
-
 void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period)
 {
 	unsigned long delta;

lib/Kconfig.debug

@@ -952,7 +952,7 @@ config PROVE_LOCKING
 	 the proof of observed correctness is also maintained for an
 	 arbitrary combination of these separate locking variants.
 
-	 For more details, see Documentation/lockdep-design.txt.
+	 For more details, see Documentation/locking/lockdep-design.txt.
 
 config LOCKDEP
 	bool
@@ -973,7 +973,7 @@ config LOCK_STAT
 	help
 	 This feature enables tracking lock contention points
 
-	 For more details, see Documentation/lockstat.txt
+	 For more details, see Documentation/locking/lockstat.txt
 
 	 This also enables lock events required by "perf lock",
 	 subcommand of perf.