[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFgnuDd+Cw-09S65i52QhpUQ-ZwHtNv5yDuu4xTwsiCFbm5fQA@mail.gmail.com>
Date: Fri, 8 Nov 2013 13:14:39 +0800
From: Xiaoguang Chen <chenxg.marvell@...il.com>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: Stratos Karafotis <skarafotis@...il.com>,
Xiaoguang Chen <chenxg@...vell.com>,
"Rafael J. Wysocki" <rjw@...ysocki.net>,
"cpufreq@...r.kernel.org" <cpufreq@...r.kernel.org>,
"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] cpufreq: conservative: fix requested_freq reduction issue
2013/11/8 Viresh Kumar <viresh.kumar@...aro.org>:
> On 8 November 2013 10:31, Xiaoguang Chen <chenxg.marvell@...il.com> wrote:
>> Hi, Viresh, Sorry for the late reply.
>
> That's fine :)
>
>> I'll prepare the patch.
>
> Thanks.
>
>> BTW, do you think we should set requeste_freq to policy->max when such
>> condition happens?
>
> I thought about that earlier, but then thought this would be a cleaner solution.
> And guess what, I was wrong.. There is one scenario for which we need to
> set requested_freq correctly..
>
> Suppose we have requested_freq currently greater than policy->max and load
> decreases. We will start decreasing requested_freq but it will take
> one iteration
> to make it equal to max, which is not we want.. So we actually need to set
> it to max when it gets over it.
>
> So, don't do the change I asked for, i.e. replacing == with >=. But
> update existing
> code:
>
> dbs_info->requested_freq += get_freq_target(cs_tuners, policy);
>
> this way:
>
> dbs_info->requested_freq += get_freq_target(cs_tuners, policy);
> if (dbs_info->requested_freq > policy->max)
> dbs_info->requested_freq = policy->max;
OK, patch will be sent out later
Xiaoguang
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists