linux-kernel - Re: [PATCH RFC v5] cpufreq: schedutil: Make iowait boost more energy efficient

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 18 Jul 2017 23:12:29 -0700
From:   Joel Fernandes <joelaf@...gle.com>
To:     Viresh Kumar <viresh.kumar@...aro.org>
Cc:     Juri Lelli <juri.lelli@....com>,
        LKML <linux-kernel@...r.kernel.org>,
        Patrick Bellasi <patrick.bellasi@....com>,
        Andres Oportus <andresoportus@...gle.com>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>,
        Len Brown <lenb@...nel.org>,
        "Rafael J . Wysocki" <rjw@...ysocki.net>,
        Ingo Molnar <mingo@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>
Subject: Re: [PATCH RFC v5] cpufreq: schedutil: Make iowait boost more energy efficient

On Tue, Jul 18, 2017 at 10:37 PM, Viresh Kumar <viresh.kumar@...aro.org> wrote:
> On 18-07-17, 15:02, Juri Lelli wrote:
>> Mmm, seems to make sense to me. :/
>>
>> Would the following work (on top of Joel's v5)? Rationale being that
>> only in sugov_set_iowait_boost we might bump freq up (if no iowait_boost
>> was set) or start from policy->min. In sugov_iowait_boost (consumer)
>> instead we do the decay (if no boosting was pending).
>>
>> ---
>>  kernel/sched/cpufreq_schedutil.c | 29 +++++++++++++++++++----------
>>  1 file changed, 19 insertions(+), 10 deletions(-)
>>
>> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
>> index 46b2479641cc..b270563c15a5 100644
>> --- a/kernel/sched/cpufreq_schedutil.c
>> +++ b/kernel/sched/cpufreq_schedutil.c
>> @@ -171,8 +171,14 @@ static void sugov_set_iowait_boost(struct sugov_cpu *sg_cpu, u64 time,
>>  {
>>       if (flags & SCHED_CPUFREQ_IOWAIT) {
>>               sg_cpu->iowait_boost_pending = true;
>> -             sg_cpu->iowait_boost = max(sg_cpu->iowait_boost,
>> -                                        sg_cpu->sg_policy->policy->min);
>> +             if (sg_cpu->iowait_boost) {
>> +                     /* Bump up 2*current_boost until hitting max */
>> +                     sg_cpu->iowait_boost = max(sg_cpu->iowait_boost << 1,
>> +                                                sg_cpu->iowait_boost_max);
>
> And we are back at where we started :)
>
> This wouldn't work because sugov_set_iowait_boost() gets called a lot.
> Maybe 10 times within a rate_limit_us period.
>
> The thumb rule is, never double/half the boost from
> sugov_set_iowait_boost() :)

Ok so tomorrow I'll post something slightly different. In
sugov_iowait_boost, if the flag is set and current iowait_boost < min,
then set it to min, otherwise double it. If the flag is not set, halve
it. And both these would be done, *before* consuming the boost (as in
Viresh's last patch). I think that's reasonable and handles all cases.
Hopefully that will put it to rest, let me know if any objections.

thanks,

-Joel