Re: [RFC] locking/rwsem: Avoid issuing wakeup before setting the reader waiter to nil

[Peter Zijlstra]

On Thu, Nov 29, 2018 at 06:27:00PM +0100, Peter Zijlstra wrote:
> 
> wake_up_q() should, per the barriers in wake_up_process, ensure that if
> wake_a_add() fails, there will be a wakeup of that task after that
> point.
> 
> So if we put wake_up_q() at the location where wake_up_process() should
> be, it should all work.
> 
> The bug in question is that it can happen at any time after
> wake_q_add(), not necessarily at wake_up_q().

Hmm, I think we're missing a barrier in wake_q_add(); when cmpxchg()
fails we still need an smp_mb().

Something like so.

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 3d87a28da378..69def558edf6 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -400,6 +400,13 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
 {
 	struct wake_q_node *node = &task->wake_q;
 
+	/*
+	 * Ensure, that when the below cmpxchg() fails, the corresponding
+	 * wake_up_q() will observe our prior state.
+	 *
+	 * Pairs with the smp_mb() from wake_up_q()'s wake_up_process().
+	 */
+	smp_mb();
 	/*
 	 * Atomically grab the task, if ->wake_q is !nil already it means
 	 * its already queued (either by us or someone else) and will get the
@@ -408,7 +415,7 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
 	 * This cmpxchg() executes a full barrier, which pairs with the full
 	 * barrier executed by the wakeup in wake_up_q().
 	 */
-	if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL))
+	if (cmpxchg_relaxed(&node->next, NULL, WAKE_Q_TAIL))
 		return;
 
 	get_task_struct(task);