1 files changed, 67 insertions, 23 deletions

diff --git a/kernel/futex.c b/kernel/futex.c
index 44a1261cb9ff..67dacaf93e56 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -157,7 +157,9 @@
  * enqueue.
  */
 
+#ifndef CONFIG_HAVE_FUTEX_CMPXCHG
 int __read_mostly futex_cmpxchg_enabled;
+#endif
 
 /*
  * Futex flags used to encode options to functions and preserve them across
@@ -234,6 +236,7 @@ static const struct futex_q futex_q_init = {
  * waiting on a futex.
  */
 struct futex_hash_bucket {
+	atomic_t waiters;
 	spinlock_t lock;
 	struct plist_head chain;
 } ____cacheline_aligned_in_smp;
@@ -253,22 +256,37 @@ static inline void futex_get_mm(union futex_key *key)
 	smp_mb__after_atomic_inc();
 }
 
-static inline bool hb_waiters_pending(struct futex_hash_bucket *hb)
+/*
+ * Reflects a new waiter being added to the waitqueue.
+ */
+static inline void hb_waiters_inc(struct futex_hash_bucket *hb)
 {
 #ifdef CONFIG_SMP
+	atomic_inc(&hb->waiters);
 	/*
-	 * Tasks trying to enter the critical region are most likely
-	 * potential waiters that will be added to the plist. Ensure
-	 * that wakers won't miss to-be-slept tasks in the window between
-	 * the wait call and the actual plist_add.
+	 * Full barrier (A), see the ordering comment above.
 	 */
-	if (spin_is_locked(&hb->lock))
-		return true;
-	smp_rmb(); /* Make sure we check the lock state first */
+	smp_mb__after_atomic_inc();
+#endif
+}
+
+/*
+ * Reflects a waiter being removed from the waitqueue by wakeup
+ * paths.
+ */
+static inline void hb_waiters_dec(struct futex_hash_bucket *hb)
+{
+#ifdef CONFIG_SMP
+	atomic_dec(&hb->waiters);
+#endif
+}
 
-	return !plist_head_empty(&hb->chain);
+static inline int hb_waiters_pending(struct futex_hash_bucket *hb)
+{
+#ifdef CONFIG_SMP
+	return atomic_read(&hb->waiters);
 #else
-	return true;
+	return 1;
 #endif
 }
 
@@ -954,6 +972,7 @@ static void __unqueue_futex(struct futex_q *q)
 
 	hb = container_of(q->lock_ptr, struct futex_hash_bucket, lock);
 	plist_del(&q->list, &hb->chain);
+	hb_waiters_dec(hb);
 }
 
 /*
@@ -1257,7 +1276,9 @@ void requeue_futex(struct futex_q *q, struct futex_hash_bucket *hb1,
 	 */
 	if (likely(&hb1->chain != &hb2->chain)) {
 		plist_del(&q->list, &hb1->chain);
+		hb_waiters_dec(hb1);
 		plist_add(&q->list, &hb2->chain);
+		hb_waiters_inc(hb2);
 		q->lock_ptr = &hb2->lock;
 	}
 	get_futex_key_refs(key2);
@@ -1600,6 +1621,17 @@ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q)
 	struct futex_hash_bucket *hb;
 
 	hb = hash_futex(&q->key);
+
+	/*
+	 * Increment the counter before taking the lock so that
+	 * a potential waker won't miss a to-be-slept task that is
+	 * waiting for the spinlock. This is safe as all queue_lock()
+	 * users end up calling queue_me(). Similarly, for housekeeping,
+	 * decrement the counter at queue_unlock() when some error has
+	 * occurred and we don't end up adding the task to the list.
+	 */
+	hb_waiters_inc(hb);
+
 	q->lock_ptr = &hb->lock;
 
 	spin_lock(&hb->lock); /* implies MB (A) */
@@ -1611,6 +1643,7 @@ queue_unlock(struct futex_hash_bucket *hb)
 	__releases(&hb->lock)
 {
 	spin_unlock(&hb->lock);
+	hb_waiters_dec(hb);
 }
 
 /**
@@ -2342,6 +2375,7 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb,
 		 * Unqueue the futex_q and determine which it was.
 		 */
 		plist_del(&q->list, &hb->chain);
+		hb_waiters_dec(hb);
 
 		/* Handle spurious wakeups gracefully */
 		ret = -EWOULDBLOCK;
@@ -2843,9 +2877,28 @@ SYSCALL_DEFINE6(futex, u32 __user *, uaddr, int, op, u32, val,
 	return do_futex(uaddr, op, val, tp, uaddr2, val2, val3);
 }
 
-static int __init futex_init(void)
+static void __init futex_detect_cmpxchg(void)
 {
+#ifndef CONFIG_HAVE_FUTEX_CMPXCHG
 	u32 curval;
+
+	/*
+	 * This will fail and we want it. Some arch implementations do
+	 * runtime detection of the futex_atomic_cmpxchg_inatomic()
+	 * functionality. We want to know that before we call in any
+	 * of the complex code paths. Also we want to prevent
+	 * registration of robust lists in that case. NULL is
+	 * guaranteed to fault and we get -EFAULT on functional
+	 * implementation, the non-functional ones will return
+	 * -ENOSYS.
+	 */
+	if (cmpxchg_futex_value_locked(&curval, NULL, 0, 0) == -EFAULT)
+		futex_cmpxchg_enabled = 1;
+#endif
+}
+
+static int __init futex_init(void)
+{
 	unsigned int futex_shift;
 	unsigned long i;
 
@@ -2861,20 +2914,11 @@ static int __init futex_init(void)
 					       &futex_shift, NULL,
 					       futex_hashsize, futex_hashsize);
 	futex_hashsize = 1UL << futex_shift;
-	/*
-	 * This will fail and we want it. Some arch implementations do
-	 * runtime detection of the futex_atomic_cmpxchg_inatomic()
-	 * functionality. We want to know that before we call in any
-	 * of the complex code paths. Also we want to prevent
-	 * registration of robust lists in that case. NULL is
-	 * guaranteed to fault and we get -EFAULT on functional
-	 * implementation, the non-functional ones will return
-	 * -ENOSYS.
-	 */
-	if (cmpxchg_futex_value_locked(&curval, NULL, 0, 0) == -EFAULT)
-		futex_cmpxchg_enabled = 1;
+
+	futex_detect_cmpxchg();
 
 	for (i = 0; i < futex_hashsize; i++) {
+		atomic_set(&futex_queues[i].waiters, 0);
 		plist_head_init(&futex_queues[i].chain);
 		spin_lock_init(&futex_queues[i].lock);
 	}