futex: Implement lockless wakeups

Davidlohr Bueso · Ingo Molnar · commit 1d0dcb3ad9d3 · 2015-05-08T12:21:40.000+02:00
Given the overall futex architecture, any chance of reducing hb->lock contention is welcome. In this particular case, using wake-queues to enable lockless wakeups addresses very much real world performance concerns, even cases of soft-lockups in cases of large amounts of blocked tasks (which is not hard to find in large boxes, using but just a handful of futex). At the lowest level, this patch can reduce latency of a single thread attempting to acquire hb->lock in highly contended scenarios by a up to 2x. At lower counts of nr_wake there are no regressions, confirming, of course, that the wake_q handling overhead is practically non existent. For instance, while a fair amount of variation, the extended pef-bench wakeup benchmark shows for a 20 core machine the following avg per-thread time to wakeup its share of tasks: nr_thr ms-before ms-after 16 0.0590 0.0215 32 0.0396 0.0220 48 0.0417 0.0182 64 0.0536 0.0236 80 0.0414 0.0097 96 0.0672 0.0152 Naturally, this can cause spurious wakeups. However there is no core code that cannot handle them afaict, and furthermore tglx does have the point that other events can already trigger them anyway. Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Chris Mason <clm@fb.com> Cc: Davidlohr Bueso <dave@stgolabs.net> Cc: George Spelvin <linux@horizon.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Manfred Spraul <manfred@colorfullife.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/1430494072-30283-3-git-send-email-dave@stgolabs.net Signed-off-by: Ingo Molnar <mingo@kernel.org>
diff --git a/kernel/futex.c b/kernel/futex.c
@@ -1090,24 +1090,22 @@ static void __unqueue_futex(struct futex_q *q)
 
 /*
  * The hash bucket lock must be held when this is called.
- * Afterwards, the futex_q must not be accessed.
+ * Afterwards, the futex_q must not be accessed. Callers
+ * must ensure to later call wake_up_q() for the actual
+ * wakeups to occur.
  */
-static void wake_futex(struct futex_q *q)
+static void mark_wake_futex(struct wake_q_head *wake_q, struct futex_q *q)
 {
 	struct task_struct *p = q->task;
 
 	if (WARN(q->pi_state || q->rt_waiter, "refusing to wake PI futex\n"))
 		return;
 
 	/*
-	 * We set q->lock_ptr = NULL _before_ we wake up the task. If
-	 * a non-futex wake up happens on another CPU then the task
-	 * might exit and p would dereference a non-existing task
-	 * struct. Prevent this by holding a reference on p across the
-	 * wake up.
+	 * Queue the task for later wakeup for after we've released
+	 * the hb->lock. wake_q_add() grabs reference to p.
 	 */
-	get_task_struct(p);
-
+	wake_q_add(wake_q, p);
 	__unqueue_futex(q);
 	/*
 	 * The waiting task can free the futex_q as soon as
@@ -1117,9 +1115,6 @@ static void wake_futex(struct futex_q *q)
 	 */
 	smp_wmb();
 	q->lock_ptr = NULL;
-
-	wake_up_state(p, TASK_NORMAL);
-	put_task_struct(p);
 }
 
 static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this)
@@ -1217,6 +1212,7 @@ futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset)
 	struct futex_q *this, *next;
 	union futex_key key = FUTEX_KEY_INIT;
 	int ret;
+	WAKE_Q(wake_q);
 
 	if (!bitset)
 		return -EINVAL;
@@ -1244,13 +1240,14 @@ futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset)
 			if (!(this->bitset & bitset))
 				continue;
 
-			wake_futex(this);
+			mark_wake_futex(&wake_q, this);
 			if (++ret >= nr_wake)
 				break;
 		}
 	}
 
 	spin_unlock(&hb->lock);
+	wake_up_q(&wake_q);
 out_put_key:
 	put_futex_key(&key);
 out:
@@ -1269,6 +1266,7 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2,
 	struct futex_hash_bucket *hb1, *hb2;
 	struct futex_q *this, *next;
 	int ret, op_ret;
+	WAKE_Q(wake_q);
 
 retry:
 	ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1, VERIFY_READ);
@@ -1320,7 +1318,7 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2,
 				ret = -EINVAL;
 				goto out_unlock;
 			}
-			wake_futex(this);
+			mark_wake_futex(&wake_q, this);
 			if (++ret >= nr_wake)
 				break;
 		}
@@ -1334,7 +1332,7 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2,
 					ret = -EINVAL;
 					goto out_unlock;
 				}
-				wake_futex(this);
+				mark_wake_futex(&wake_q, this);
 				if (++op_ret >= nr_wake2)
 					break;
 			}
@@ -1344,6 +1342,7 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2,
 
 out_unlock:
 	double_unlock_hb(hb1, hb2);
+	wake_up_q(&wake_q);
 out_put_keys:
 	put_futex_key(&key2);
 out_put_key1:
@@ -1503,6 +1502,7 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags,
 	struct futex_pi_state *pi_state = NULL;
 	struct futex_hash_bucket *hb1, *hb2;
 	struct futex_q *this, *next;
+	WAKE_Q(wake_q);
 
 	if (requeue_pi) {
 		/*
@@ -1679,7 +1679,7 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags,
 		 * woken by futex_unlock_pi().
 		 */
 		if (++task_count <= nr_wake && !requeue_pi) {
-			wake_futex(this);
+			mark_wake_futex(&wake_q, this);
 			continue;
 		}
 
@@ -1719,6 +1719,7 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags,
 out_unlock:
 	free_pi_state(pi_state);
 	double_unlock_hb(hb1, hb2);
+	wake_up_q(&wake_q);
 	hb_waiters_dec(hb2);
 
 	/*
