[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJZ5v0h+hoA0QBQKdnnxWErNGZ0Fd-sYjjH_oTdH6No9Uz2B2A@mail.gmail.com>
Date: Fri, 17 Jun 2016 17:32:25 +0200
From: "Rafael J. Wysocki" <rafael@...nel.org>
To: Peter Zijlstra <peterz@...radead.org>,
Jisheng Zhang <jszhang@...vell.com>
Cc: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
"Rafael J. Wysocki" <rafael@...nel.org>,
"Rafael J. Wysocki" <rjw@...ysocki.net>,
Viresh Kumar <viresh.kumar@...aro.org>,
"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: regression caused by 08f511fd41c3 ("cpufreq: Reduce
cpufreq_update_util() overhead a bit")
On Fri, Jun 17, 2016 at 4:03 PM, Peter Zijlstra <peterz@...radead.org> wrote:
> On Fri, Jun 17, 2016 at 06:16:51AM -0700, Paul E. McKenney wrote:
>> > Paul, Peter, any ideas about what may be going on here?
>>
>> Looks to me like this commit moved some code from synchronize_rcu() to
>> synchronize_sched(). Assuming that this is a CONFIG_PREEMPT=y system,
>> might there have been a decrease in the wakeups from the rcu_preempt
>> kthread?
>
> The 'funny' thing is though; those synchronize thingies are only reached
> when we change cpufreq policy, so things like:
>
> for i in /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor ; do echo performance > $i ; done
Or if you change min or max for a policy and the driver is intel_pstate.
> Something which is hardly possible when idle. Weird.
Well, exactly.
Jisheng, do you use intel_pstate or acpi-cpufreq as the scaling driver?
Powered by blists - more mailing lists