[PATCH 5.10 083/146] sched/pelt: Ensure that *_sum is always synced with *_avg

From: Greg Kroah-Hartman
Date: Mon Jun 21 2021 - 12:33:53 EST


From: Vincent Guittot <vincent.guittot@xxxxxxxxxx>

[ Upstream commit fcf6631f3736985ec89bdd76392d3c7bfb60119f ]

Rounding in PELT calculation happening when entities are attached/detached
of a cfs_rq can result into situations where util/runnable_avg is not null
but util/runnable_sum is. This is normally not possible so we need to
ensure that util/runnable_sum stays synced with util/runnable_avg.

detach_entity_load_avg() is the last place where we don't sync
util/runnable_sum with util/runnbale_avg when moving some sched_entities

Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Link: https://lkml.kernel.org/r/20210601085832.12626-1-vincent.guittot@xxxxxxxxxx
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
kernel/sched/fair.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index ff8a172a69ca..d6e1c90de570 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3767,11 +3767,17 @@ static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s
*/
static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
+ /*
+ * cfs_rq->avg.period_contrib can be used for both cfs_rq and se.
+ * See ___update_load_avg() for details.
+ */
+ u32 divider = get_pelt_divider(&cfs_rq->avg);
+
dequeue_load_avg(cfs_rq, se);
sub_positive(&cfs_rq->avg.util_avg, se->avg.util_avg);
- sub_positive(&cfs_rq->avg.util_sum, se->avg.util_sum);
+ cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * divider;
sub_positive(&cfs_rq->avg.runnable_avg, se->avg.runnable_avg);
- sub_positive(&cfs_rq->avg.runnable_sum, se->avg.runnable_sum);
+ cfs_rq->avg.runnable_sum = cfs_rq->avg.runnable_avg * divider;

add_tg_cfs_propagate(cfs_rq, -se->avg.load_sum);

--
2.30.2