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

Pull locking changes from Ingo Molnar:
 "The most noticeable change are mutex speedups from Waiman Long, for
  higher loads.  These scalability changes should be most noticeable on
  larger server systems.

  There are also cleanups, fixes and debuggability improvements."

* 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  lockdep: Consolidate bug messages into a single print_lockdep_off() function
  lockdep: Print out additional debugging advice when we hit lockdep BUGs
  mutex: Back out architecture specific check for negative mutex count
  mutex: Queue mutex spinners with MCS lock to reduce cacheline contention
  mutex: Make more scalable by doing less atomic operations
  mutex: Move mutex spinning code from sched/core.c back to mutex.c
  locking/rtmutex/tester: Set correct permissions on sysfs files
  lockdep: Remove unnecessary 'hlock_next' variable

5 files changed, 165 insertions, 72 deletions

diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 8a0efac4f99d..6a3bccba7e7d 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -380,6 +380,13 @@ static int verbose(struct lock_class *class)
 unsigned long nr_stack_trace_entries;
 static unsigned long stack_trace[MAX_STACK_TRACE_ENTRIES];
 
+static void print_lockdep_off(const char *bug_msg)
+{
+	printk(KERN_DEBUG "%s\n", bug_msg);
+	printk(KERN_DEBUG "turning off the locking correctness validator.\n");
+	printk(KERN_DEBUG "Please attach the output of /proc/lock_stat to the bug report\n");
+}
+
 static int save_trace(struct stack_trace *trace)
 {
 	trace->nr_entries = 0;
@@ -409,8 +416,7 @@ static int save_trace(struct stack_trace *trace)
 		if (!debug_locks_off_graph_unlock())
 			return 0;
 
-		printk("BUG: MAX_STACK_TRACE_ENTRIES too low!\n");
-		printk("turning off the locking correctness validator.\n");
+		print_lockdep_off("BUG: MAX_STACK_TRACE_ENTRIES too low!");
 		dump_stack();
 
 		return 0;
@@ -763,8 +769,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
 		}
 		raw_local_irq_restore(flags);
 
-		printk("BUG: MAX_LOCKDEP_KEYS too low!\n");
-		printk("turning off the locking correctness validator.\n");
+		print_lockdep_off("BUG: MAX_LOCKDEP_KEYS too low!");
 		dump_stack();
 		return NULL;
 	}
@@ -834,8 +839,7 @@ static struct lock_list *alloc_list_entry(void)
 		if (!debug_locks_off_graph_unlock())
 			return NULL;
 
-		printk("BUG: MAX_LOCKDEP_ENTRIES too low!\n");
-		printk("turning off the locking correctness validator.\n");
+		print_lockdep_off("BUG: MAX_LOCKDEP_ENTRIES too low!");
 		dump_stack();
 		return NULL;
 	}
@@ -2000,7 +2004,7 @@ static inline int lookup_chain_cache(struct task_struct *curr,
 	struct lock_class *class = hlock_class(hlock);
 	struct list_head *hash_head = chainhashentry(chain_key);
 	struct lock_chain *chain;
-	struct held_lock *hlock_curr, *hlock_next;
+	struct held_lock *hlock_curr;
 	int i, j;
 
 	/*
@@ -2048,8 +2052,7 @@ cache_hit:
 		if (!debug_locks_off_graph_unlock())
 			return 0;
 
-		printk("BUG: MAX_LOCKDEP_CHAINS too low!\n");
-		printk("turning off the locking correctness validator.\n");
+		print_lockdep_off("BUG: MAX_LOCKDEP_CHAINS too low!");
 		dump_stack();
 		return 0;
 	}
@@ -2057,12 +2060,10 @@ cache_hit:
 	chain->chain_key = chain_key;
 	chain->irq_context = hlock->irq_context;
 	/* Find the first held_lock of current chain */
-	hlock_next = hlock;
 	for (i = curr->lockdep_depth - 1; i >= 0; i--) {
 		hlock_curr = curr->held_locks + i;
-		if (hlock_curr->irq_context != hlock_next->irq_context)
+		if (hlock_curr->irq_context != hlock->irq_context)
 			break;
-		hlock_next = hlock;
 	}
 	i++;
 	chain->depth = curr->lockdep_depth + 1 - i;
@@ -3190,9 +3191,9 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass,
 #endif
 	if (unlikely(curr->lockdep_depth >= MAX_LOCK_DEPTH)) {
 		debug_locks_off();
-		printk("BUG: MAX_LOCK_DEPTH too low, depth: %i  max: %lu!\n",
+		print_lockdep_off("BUG: MAX_LOCK_DEPTH too low!");
+		printk(KERN_DEBUG "depth: %i  max: %lu!\n",
 		       curr->lockdep_depth, MAX_LOCK_DEPTH);
-		printk("turning off the locking correctness validator.\n");
 
 		lockdep_print_held_locks(current);
 		debug_show_all_locks();
diff --git a/kernel/mutex.c b/kernel/mutex.c
index 52f23011b6e0..ad53a664f113 100644
--- a/kernel/mutex.c
+++ b/kernel/mutex.c
@@ -37,6 +37,12 @@
 # include <asm/mutex.h>
 #endif
 
+/*
+ * A negative mutex count indicates that waiters are sleeping waiting for the
+ * mutex.
+ */
+#define	MUTEX_SHOW_NO_WAITER(mutex)	(atomic_read(&(mutex)->count) >= 0)
+
 void
 __mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key)
 {
@@ -44,6 +50,9 @@ __mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key)
 	spin_lock_init(&lock->wait_lock);
 	INIT_LIST_HEAD(&lock->wait_list);
 	mutex_clear_owner(lock);
+#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
+	lock->spin_mlock = NULL;
+#endif
 
 	debug_mutex_init(lock, name, key);
 }
@@ -95,6 +104,124 @@ void __sched mutex_lock(struct mutex *lock)
 EXPORT_SYMBOL(mutex_lock);
 #endif
 
+#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
+/*
+ * In order to avoid a stampede of mutex spinners from acquiring the mutex
+ * more or less simultaneously, the spinners need to acquire a MCS lock
+ * first before spinning on the owner field.
+ *
+ * We don't inline mspin_lock() so that perf can correctly account for the
+ * time spent in this lock function.
+ */
+struct mspin_node {
+	struct mspin_node *next ;
+	int		  locked;	/* 1 if lock acquired */
+};
+#define	MLOCK(mutex)	((struct mspin_node **)&((mutex)->spin_mlock))
+
+static noinline
+void mspin_lock(struct mspin_node **lock, struct mspin_node *node)
+{
+	struct mspin_node *prev;
+
+	/* Init node */
+	node->locked = 0;
+	node->next   = NULL;
+
+	prev = xchg(lock, node);
+	if (likely(prev == NULL)) {
+		/* Lock acquired */
+		node->locked = 1;
+		return;
+	}
+	ACCESS_ONCE(prev->next) = node;
+	smp_wmb();
+	/* Wait until the lock holder passes the lock down */
+	while (!ACCESS_ONCE(node->locked))
+		arch_mutex_cpu_relax();
+}
+
+static void mspin_unlock(struct mspin_node **lock, struct mspin_node *node)
+{
+	struct mspin_node *next = ACCESS_ONCE(node->next);
+
+	if (likely(!next)) {
+		/*
+		 * Release the lock by setting it to NULL
+		 */
+		if (cmpxchg(lock, node, NULL) == node)
+			return;
+		/* Wait until the next pointer is set */
+		while (!(next = ACCESS_ONCE(node->next)))
+			arch_mutex_cpu_relax();
+	}
+	ACCESS_ONCE(next->locked) = 1;
+	smp_wmb();
+}
+
+/*
+ * Mutex spinning code migrated from kernel/sched/core.c
+ */
+
+static inline bool owner_running(struct mutex *lock, struct task_struct *owner)
+{
+	if (lock->owner != owner)
+		return false;
+
+	/*
+	 * Ensure we emit the owner->on_cpu, dereference _after_ checking
+	 * lock->owner still matches owner, if that fails, owner might
+	 * point to free()d memory, if it still matches, the rcu_read_lock()
+	 * ensures the memory stays valid.
+	 */
+	barrier();
+
+	return owner->on_cpu;
+}
+
+/*
+ * Look out! "owner" is an entirely speculative pointer
+ * access and not reliable.
+ */
+static noinline
+int mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner)
+{
+	rcu_read_lock();
+	while (owner_running(lock, owner)) {
+		if (need_resched())
+			break;
+
+		arch_mutex_cpu_relax();
+	}
+	rcu_read_unlock();
+
+	/*
+	 * We break out the loop above on need_resched() and when the
+	 * owner changed, which is a sign for heavy contention. Return
+	 * success only when lock->owner is NULL.
+	 */
+	return lock->owner == NULL;
+}
+
+/*
+ * Initial check for entering the mutex spinning loop
+ */
+static inline int mutex_can_spin_on_owner(struct mutex *lock)
+{
+	int retval = 1;
+
+	rcu_read_lock();
+	if (lock->owner)
+		retval = lock->owner->on_cpu;
+	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;
+}
+#endif
+
 static __used noinline void __sched __mutex_unlock_slowpath(atomic_t *lock_count);
 
 /**
@@ -158,25 +285,39 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 	 *
 	 * 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;
 
 	for (;;) {
 		struct task_struct *owner;
+		struct mspin_node  node;
 
 		/*
 		 * If there's an owner, wait for it to either
 		 * release the lock or go to sleep.
 		 */
+		mspin_lock(MLOCK(lock), &node);
 		owner = ACCESS_ONCE(lock->owner);
-		if (owner && !mutex_spin_on_owner(lock, owner))
+		if (owner && !mutex_spin_on_owner(lock, owner)) {
+			mspin_unlock(MLOCK(lock), &node);
 			break;
+		}
 
-		if (atomic_cmpxchg(&lock->count, 1, 0) == 1) {
+		if ((atomic_read(&lock->count) == 1) &&
+		    (atomic_cmpxchg(&lock->count, 1, 0) == 1)) {
 			lock_acquired(&lock->dep_map, ip);
 			mutex_set_owner(lock);
+			mspin_unlock(MLOCK(lock), &node);
 			preempt_enable();
 			return 0;
 		}
+		mspin_unlock(MLOCK(lock), &node);
 
 		/*
 		 * When there's no owner, we might have preempted between the
@@ -195,6 +336,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 		 */
 		arch_mutex_cpu_relax();
 	}
+slowpath:
 #endif
 	spin_lock_mutex(&lock->wait_lock, flags);
 
@@ -205,7 +347,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 	list_add_tail(&waiter.list, &lock->wait_list);
 	waiter.task = task;
 
-	if (atomic_xchg(&lock->count, -1) == 1)
+	if (MUTEX_SHOW_NO_WAITER(lock) && (atomic_xchg(&lock->count, -1) == 1))
 		goto done;
 
 	lock_contended(&lock->dep_map, ip);
@@ -220,7 +362,8 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 		 * that when we release the lock, we properly wake up the
 		 * other waiters:
 		 */
-		if (atomic_xchg(&lock->count, -1) == 1)
+		if (MUTEX_SHOW_NO_WAITER(lock) &&
+		   (atomic_xchg(&lock->count, -1) == 1))
 			break;
 
 		/*
diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c
index 7890b10084a7..1d96dd0d93c1 100644
--- a/kernel/rtmutex-tester.c
+++ b/kernel/rtmutex-tester.c
@@ -14,6 +14,7 @@
 #include <linux/spinlock.h>
 #include <linux/timer.h>
 #include <linux/freezer.h>
+#include <linux/stat.h>
 
 #include "rtmutex.h"
 
@@ -366,8 +367,8 @@ static ssize_t sysfs_test_status(struct device *dev, struct device_attribute *at
 	return curr - buf;
 }
 
-static DEVICE_ATTR(status, 0600, sysfs_test_status, NULL);
-static DEVICE_ATTR(command, 0600, NULL, sysfs_test_command);
+static DEVICE_ATTR(status, S_IRUSR, sysfs_test_status, NULL);
+static DEVICE_ATTR(command, S_IWUSR, NULL, sysfs_test_command);
 
 static struct bus_type rttest_subsys = {
 	.name = "rttest",
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 67d04651f44b..42053547e0f5 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2999,51 +2999,6 @@ void __sched schedule_preempt_disabled(void)
 	preempt_disable();
 }
 
-#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
-
-static inline bool owner_running(struct mutex *lock, struct task_struct *owner)
-{
-	if (lock->owner != owner)
-		return false;
-
-	/*
-	 * Ensure we emit the owner->on_cpu, dereference _after_ checking
-	 * lock->owner still matches owner, if that fails, owner might
-	 * point to free()d memory, if it still matches, the rcu_read_lock()
-	 * ensures the memory stays valid.
-	 */
-	barrier();
-
-	return owner->on_cpu;
-}
-
-/*
- * Look out! "owner" is an entirely speculative pointer
- * access and not reliable.
- */
-int mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner)
-{
-	if (!sched_feat(OWNER_SPIN))
-		return 0;
-
-	rcu_read_lock();
-	while (owner_running(lock, owner)) {
-		if (need_resched())
-			break;
-
-		arch_mutex_cpu_relax();
-	}
-	rcu_read_unlock();
-
-	/*
-	 * We break out the loop above on need_resched() and when the
-	 * owner changed, which is a sign for heavy contention. Return
-	 * success only when lock->owner is NULL.
-	 */
-	return lock->owner == NULL;
-}
-#endif
-
 #ifdef CONFIG_PREEMPT
 /*
  * this is the entry point to schedule() from in-kernel preemption
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 1ad1d2b5395f..99399f8e4799 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -46,13 +46,6 @@ SCHED_FEAT(DOUBLE_TICK, false)
 SCHED_FEAT(LB_BIAS, true)
 
 /*
- * Spin-wait on mutex acquisition when the mutex owner is running on
- * another cpu -- assumes that when the owner is running, it will soon
- * release the lock. Decreases scheduling overhead.
- */
-SCHED_FEAT(OWNER_SPIN, true)
-
-/*
  * Decrement CPU power based on time not spent running tasks
  */
 SCHED_FEAT(NONTASK_POWER, true)