futex: Fix errors in nested key ref-counting

futex_wait() is leaking key references due to futex_wait_setup()
acquiring an additional reference via the queue_lock() routine. The
nested key ref-counting has been masking bugs and complicating code
analysis. queue_lock() is only called with a previously ref-counted
key, so remove the additional ref-counting from the queue_(un)lock()
functions.

Also futex_wait_requeue_pi() drops one key reference too many in
unqueue_me_pi(). Remove the key reference handling from
unqueue_me_pi(). This was paired with a queue_lock() in
futex_lock_pi(), so the count remains unchanged.

Document remaining nested key ref-counting sites.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Reported-and-tested-by: Matthieu Fertré<matthieu.fertre@kerlabs.com>
Reported-by: Louis Rilling<louis.rilling@kerlabs.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <4CBB17A8.70401@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@kernel.org

kernel/futex.c

@@ -1363,7 +1363,6 @@ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q)
 {
 	struct futex_hash_bucket *hb;
 
-	get_futex_key_refs(&q->key);
 	hb = hash_futex(&q->key);
 	q->lock_ptr = &hb->lock;
 
@@ -1375,7 +1374,6 @@ static inline void
 queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb)
 {
 	spin_unlock(&hb->lock);
-	drop_futex_key_refs(&q->key);
 }
 
 /**
@@ -1480,8 +1478,6 @@ static void unqueue_me_pi(struct futex_q *q)
 	q->pi_state = NULL;
 
 	spin_unlock(q->lock_ptr);
-
-	drop_futex_key_refs(&q->key);
 }
 
 /*
@@ -1812,7 +1808,10 @@ static int futex_wait(u32 __user *uaddr, int fshared,
 	}
 
 retry:
-	/* Prepare to wait on uaddr. */
+	/*
+	 * Prepare to wait on uaddr. On success, holds hb lock and increments
+	 * q.key refs.
+	 */
 	ret = futex_wait_setup(uaddr, val, fshared, &q, &hb);
 	if (ret)
 		goto out;
@@ -1822,24 +1821,23 @@ static int futex_wait(u32 __user *uaddr, int fshared,
 
 	/* If we were woken (and unqueued), we succeeded, whatever. */
 	ret = 0;
+	/* unqueue_me() drops q.key ref */
 	if (!unqueue_me(&q))
-		goto out_put_key;
+		goto out;
 	ret = -ETIMEDOUT;
 	if (to && !to->task)
-		goto out_put_key;
+		goto out;
 
 	/*
 	 * We expect signal_pending(current), but we might be the
 	 * victim of a spurious wakeup as well.
 	 */
-	if (!signal_pending(current)) {
-		put_futex_key(fshared, &q.key);
+	if (!signal_pending(current))
 		goto retry;
-	}
 
 	ret = -ERESTARTSYS;
 	if (!abs_time)
-		goto out_put_key;
+		goto out;
 
 	restart = &current_thread_info()->restart_block;
 	restart->fn = futex_wait_restart;
@@ -1856,8 +1854,6 @@ static int futex_wait(u32 __user *uaddr, int fshared,
 
 	ret = -ERESTART_RESTARTBLOCK;
 
-out_put_key:
-	put_futex_key(fshared, &q.key);
 out:
 	if (to) {
 		hrtimer_cancel(&to->timer);
@@ -2236,7 +2232,10 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared,
 	q.rt_waiter = &rt_waiter;
 	q.requeue_pi_key = &key2;
 
-	/* Prepare to wait on uaddr. */
+	/*
+	 * Prepare to wait on uaddr. On success, increments q.key (key1) ref
+	 * count.
+	 */
 	ret = futex_wait_setup(uaddr, val, fshared, &q, &hb);
 	if (ret)
 		goto out_key2;
@@ -2254,7 +2253,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared,
 	 * In order for us to be here, we know our q.key == key2, and since
 	 * we took the hb->lock above, we also know that futex_requeue() has
 	 * completed and we no longer have to concern ourselves with a wakeup
-	 * race with the atomic proxy lock acquition by the requeue code.
+	 * race with the atomic proxy lock acquisition by the requeue code. The
+	 * futex_requeue dropped our key1 reference and incremented our key2
+	 * reference count.
 	 */
 
 	/* Check if the requeue code acquired the second futex for us. */