[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20241014031057.8199-4-laoar.shao@gmail.com>
Date: Mon, 14 Oct 2024 11:10:56 +0800
From: Yafang Shao <laoar.shao@...il.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,
mgorman@...e.de,
vschneid@...hat.com,
hannes@...xchg.org,
surenb@...gle.com
Cc: cgroups@...r.kernel.org,
linux-kernel@...r.kernel.org,
Yafang Shao <laoar.shao@...il.com>
Subject: [PATCH v3 bpf-next 3/4] sched, psi: Don't account irq time if sched_clock_irqtime is disabled
sched_clock_irqtime may be disabled due to the clock source. When disabled,
irq_time_read() won't change over time, so there is nothing to account. We
can save iterating the whole hierarchy on every tick and context switch.
Signed-off-by: Yafang Shao <laoar.shao@...il.com>
Acked-by: Johannes Weiner <hannes@...xchg.org>
---
kernel/sched/psi.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c
index 020d58967d4e..49d9c75be0c8 100644
--- a/kernel/sched/psi.c
+++ b/kernel/sched/psi.c
@@ -1000,7 +1000,8 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st
u64 now, irq;
s64 delta;
- if (static_branch_likely(&psi_disabled))
+ if (static_branch_likely(&psi_disabled) ||
+ !static_branch_likely(&sched_clock_irqtime))
return;
if (!curr->pid)
--
2.43.5
Powered by blists - more mailing lists