[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <DM4PR12MB52786E4AEE69C9AF9E3A975B9C859@DM4PR12MB5278.namprd12.prod.outlook.com>
Date: Sat, 9 Jul 2022 08:49:50 +0000
From: "Yuan, Perry" <Perry.Yuan@....com>
To: "Fontenot, Nathan" <Nathan.Fontenot@....com>,
"rafael.j.wysocki@...el.com" <rafael.j.wysocki@...el.com>,
"viresh.kumar@...aro.org" <viresh.kumar@...aro.org>,
"Huang, Ray" <Ray.Huang@....com>,
"Rafael J. Wysocki" <rafael@...nel.org>,
"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
CC: "Sharma, Deepak" <Deepak.Sharma@....com>,
"Limonciello, Mario" <Mario.Limonciello@....com>,
"Deucher, Alexander" <Alexander.Deucher@....com>,
"Su, Jinzhou (Joe)" <Jinzhou.Su@....com>,
"Huang, Shimmer" <Shimmer.Huang@....com>,
"Du, Xiaojian" <Xiaojian.Du@....com>,
"Meng, Li (Jassmine)" <Li.Meng@....com>
Subject: RE: [PATCH 07/12] cpufreq: amd_pstate: map desired perf into pstate
scope for powersave governor
[AMD Official Use Only - General]
Hi Nathan:
> -----Original Message-----
> From: Fontenot, Nathan <Nathan.Fontenot@....com>
> Sent: Friday, July 8, 2022 3:57 AM
> To: Yuan, Perry <Perry.Yuan@....com>; rafael.j.wysocki@...el.com;
> viresh.kumar@...aro.org; Huang, Ray <Ray.Huang@....com>; Rafael J.
> Wysocki <rafael@...nel.org>; linux-pm@...r.kernel.org; linux-
> kernel@...r.kernel.org
> Cc: Sharma, Deepak <Deepak.Sharma@....com>; Limonciello, Mario
> <Mario.Limonciello@....com>; Fontenot, Nathan
> <Nathan.Fontenot@....com>; Deucher, Alexander
> <Alexander.Deucher@....com>; Su, Jinzhou (Joe) <Jinzhou.Su@....com>;
> Huang, Shimmer <Shimmer.Huang@....com>; Du, Xiaojian
> <Xiaojian.Du@....com>; Meng, Li (Jassmine) <Li.Meng@....com>
> Subject: Re: [PATCH 07/12] cpufreq: amd_pstate: map desired perf into
> pstate scope for powersave governor
>
> On 7/7/22 12:00, Perry Yuan wrote:
> > The patch will fix the invalid desired perf value for powersave
> > governor. This issue is found when testing on one AMD EPYC system, the
> > actual des_perf is smaller than the min_perf value, that is invalid
> > value. because the min_perf is the lowest_perf system can support in
> > idle state.
> >
> > Signed-off-by: Perry Yuan <Perry.Yuan@....com>
> > ---
> > drivers/cpufreq/amd-pstate.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/cpufreq/amd-pstate.c
> > b/drivers/cpufreq/amd-pstate.c index 7c51f4125263..154eed849f38
> 100644
> > --- a/drivers/cpufreq/amd-pstate.c
> > +++ b/drivers/cpufreq/amd-pstate.c
> > @@ -317,6 +317,7 @@ static int amd_pstate_target(struct cpufreq_policy
> *policy,
> > des_perf = DIV_ROUND_CLOSEST(target_freq * cap_perf,
> > cpudata->max_freq);
> >
> > + des_perf = clamp_t(unsigned long, des_perf, min_perf, max_perf);
> > cpufreq_freq_transition_begin(policy, &freqs);
> > amd_pstate_update(cpudata, min_perf, des_perf,
> > max_perf, false);
>
> The clamping of the desired perf value should be moved to
> amd_pstate_update(). The only other caller of amd_pstate_update() is
> amd_pstate_adjust_perf() which already clamps the desired perf value
> before making the call.
>
> -Nathan
Thanks for your suggestion.
Add this change into V2.
Perry.
Powered by blists - more mailing lists