[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJWu+orGGUOiAXUxEaOcUAbBbOE-5Am_DkwfrZBaSN+Qy5kq5g@mail.gmail.com>
Date: Thu, 18 May 2017 21:59:19 -0700
From: Joel Fernandes <joelaf@...gle.com>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: Peter Zijlstra <peterz@...radead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"Rafael J. Wysocki" <rafael@...nel.org>
Subject: Re: [PATCH 2/2] sched: Use iowait boost policy option in schedutil
On Thu, May 18, 2017 at 9:39 PM, Viresh Kumar <viresh.kumar@...aro.org> wrote:
[..]
>> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
>> index 76877a62b5fa..6915925bc947 100644
>> --- a/kernel/sched/cpufreq_schedutil.c
>> +++ b/kernel/sched/cpufreq_schedutil.c
>> @@ -24,6 +24,7 @@
>> struct sugov_tunables {
>> struct gov_attr_set attr_set;
>> unsigned int rate_limit_us;
>> + unsigned int iowait_boost_enable;
>
> Maybe just:
>
> bool iowait_boost;
Yes, Rafael mentioned this for the cpufreq framework part, I will do
it this way in v2.
>
>> };
>>
>> struct sugov_policy {
>> @@ -47,6 +48,7 @@ struct sugov_policy {
>> bool work_in_progress;
>>
>> bool need_freq_update;
>> + unsigned int iowait_boost_enable;
>
> I don't see this being used at all. Am I missing something ?
>
Ah! Yes you're right, this is spurious, I will drop it. Thanks for spotting it.
-Joel
Powered by blists - more mailing lists