[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191127101456.2c814108@gandalf.local.home>
Date: Wed, 27 Nov 2019 10:14:56 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: Zhenzhong Duan <zhenzhong.duan@...cle.com>
Cc: linux-kernel@...r.kernel.org, mingo@...hat.com,
peterz@...radead.org, juri.lelli@...hat.com,
vincent.guittot@...aro.org, dietmar.eggemann@....com,
bsegall@...gle.com, mgorman@...e.de
Subject: Re: [PATCH] sched/clock: use static_branch_likely() check at
sched_clock_running
On Wed, 27 Nov 2019 16:37:28 +0800
Zhenzhong Duan <zhenzhong.duan@...cle.com> wrote:
> sched_clock_running is enabled early at bootup stage and never
> disabled. So hints that to compiler by using static_branch_likely()
> rather than static_branch_unlikely().
Looks like the confusion was the moving of the "!":
- if (unlikely(!sched_clock_running))
+ if (!static_branch_unlikely(&sched_clock_running))
Where, it was unlikely that !sched_clock_running would be true, but
because the "!" was moved outside the "unlikely()" it makes the test
"likely()". That is, if we added an intermediate step, it would have
been:
if (!likely(sched_clock_running))
which would have prevented the mistake that this patch fixes.
Reviewed-by: Steven Rostedt (VMware) <rostedt@...dmis.org>
-- Steve
>
> Fixes: 46457ea464f5 ("sched/clock: Use static key for sched_clock_running")
> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@...cle.com>
> ---
> kernel/sched/clock.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c
> index 1152259..12bca64 100644
> --- a/kernel/sched/clock.c
> +++ b/kernel/sched/clock.c
> @@ -370,7 +370,7 @@ u64 sched_clock_cpu(int cpu)
> if (sched_clock_stable())
> return sched_clock() + __sched_clock_offset;
>
> - if (!static_branch_unlikely(&sched_clock_running))
> + if (!static_branch_likely(&sched_clock_running))
> return sched_clock();
>
> preempt_disable_notrace();
> @@ -393,7 +393,7 @@ void sched_clock_tick(void)
> if (sched_clock_stable())
> return;
>
> - if (!static_branch_unlikely(&sched_clock_running))
> + if (!static_branch_likely(&sched_clock_running))
> return;
>
> lockdep_assert_irqs_disabled();
> @@ -460,7 +460,7 @@ void __init sched_clock_init(void)
>
> u64 sched_clock_cpu(int cpu)
> {
> - if (!static_branch_unlikely(&sched_clock_running))
> + if (!static_branch_likely(&sched_clock_running))
> return 0;
>
> return sched_clock();
Powered by blists - more mailing lists