sched: Fix rq->nr_iowait ordering

schedule() ttwu()
deactivate_task(); if (p->on_rq && ...) // false
atomic_dec(&task_rq(p)->nr_iowait);
if (prev->in_iowait)
atomic_inc(&rq->nr_iowait);

Allows nr_iowait to be decremented before it gets incremented,
resulting in more dodgy IO-wait numbers than usual.

Note that because we can now do ttwu_queue_wakelist() before
p->on_cpu==0, we lose the natural ordering and have to further delay
the decrement.

Fixes: c6e7bd7afaeb ("sched/core: Optimize ttwu() spinning on p->on_cpu")
Reported-by: Tejun Heo <[e-mail]>
Signed-off-by: Peter Zijlstra (Intel) <[e-mail]>
Acked-by: Mel Gorman <[e-mail]>
Link: https://lkml.kernel.org/r/[e-mail]

kernel/sched/core.c

···
 #ifdef CONFIG_SMP
 	if (wake_flags & WF_MIGRATED)
 		en_flags |= ENQUEUE_MIGRATED;
+	else
 #endif
+	if (p->in_iowait) {
+		delayacct_blkio_end(p);
+		atomic_dec(&task_rq(p)->nr_iowait);
+	}
 
 	activate_task(rq, p, en_flags);
 	ttwu_do_wakeup(rq, p, wake_flags, rf);
···
 	if (READ_ONCE(p->on_rq) && ttwu_runnable(p, wake_flags))
 		goto unlock;
 
-	if (p->in_iowait) {
-		delayacct_blkio_end(p);
-		atomic_dec(&task_rq(p)->nr_iowait);
-	}
-
 #ifdef CONFIG_SMP
 	/*
 	 * Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be
···
 
 	cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags);
 	if (task_cpu(p) != cpu) {
+		if (p->in_iowait) {
+			delayacct_blkio_end(p);
+			atomic_dec(&task_rq(p)->nr_iowait);
+		}
+
 		wake_flags |= WF_MIGRATED;
 		psi_ttwu_dequeue(p);
 		set_task_cpu(p, cpu);