The following commit has been merged into the sched/urgent branch of tip:
Commit-ID: 91dcf1e8068e9a8823e419a7a34ff4341275fb70
Gitweb: https://git.kernel.org/tip/91dcf1e8068e9a8823e419a7a34ff4341275fb70
Author: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
AuthorDate: Tue, 11 Apr 2023 11:06:11 +02:00
Committer: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
CommitterDate: Wed, 12 Apr 2023 16:46:30 +02:00
sched/fair: Fix imbalance overflow
When local group is fully busy but its average load is above system load,
computing the imbalance will overflow and local group is not the best
target for pulling this load.
Fixes: 0b0695f2b34a ("sched/fair: Rework load_balance()")
Reported-by: Tingjia Cao <tjcao980311@xxxxxxxxx>
Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Tested-by: Tingjia Cao <tjcao980311@xxxxxxxxx>
Link: https://lore.kernel.org/lkml/CABcWv9_DAhVBOq2=W=2ypKE9dKM5s2DvoV8-U0+GDwwuKZ89jQ@xxxxxxxxxxxxxx/T/
---
kernel/sched/fair.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6986ea3..5f6587d 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -10238,6 +10238,16 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
sds->avg_load = (sds->total_load * SCHED_CAPACITY_SCALE) /
sds->total_capacity;
+
+ /*
+ * If the local group is more loaded than the average system
+ * load, don't try to pull any tasks.
+ */
+ if (local->avg_load >= sds->avg_load) {
+ env->imbalance = 0;
+ return;
+ }
+
}
/*
|