linux-kernel - [PATCH 1/2] sched/fair: Skip updating PELT for !fair task

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20220924160429.71312-1-zhouchengming@bytedance.com>
Date:   Sun, 25 Sep 2022 00:04:28 +0800
From:   Chengming Zhou <zhouchengming@...edance.com>
To:     mingo@...hat.com, peterz@...radead.org, juri.lelli@...hat.com,
        vincent.guittot@...aro.org, dietmar.eggemann@....com,
        rostedt@...dmis.org, bsegall@...gle.com
Cc:     linux-kernel@...r.kernel.org,
        Chengming Zhou <zhouchengming@...edance.com>
Subject: [PATCH 1/2] sched/fair: Skip updating PELT for !fair task

The set_task_rq_fair() is introduced by the commit ad936d8658fd
("sched/fair: Make it possible to account fair load avg consistently")
to update PELT for !fair task when it got migrated or moved to another
cgroup.

Since we don't really care about !fair task's PELT values until when
it switched_to_fair(), so we can just adjust its last_update_time,
its PELT values will be updated correctly when switched_to_fair().

Signed-off-by: Chengming Zhou <zhouchengming@...edance.com>
---
 kernel/sched/fair.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e4a0b8bd941c..dea86d8a6c02 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3636,8 +3636,7 @@ void set_task_rq_fair(struct sched_entity *se,
 	p_last_update_time = cfs_rq_last_update_time(prev);
 	n_last_update_time = cfs_rq_last_update_time(next);
 
-	__update_load_avg_blocked_se(p_last_update_time, se);
-	se->avg.last_update_time = n_last_update_time;
+	se->avg.last_update_time += n_last_update_time - p_last_update_time;
 }
 
 /*
-- 
2.37.2