[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 9 Sep 2021 17:59:33 +0800
From: Huang Rui <ray.huang@....com>
To: "Fontenot, Nathan" <Nathan.Fontenot@....com>
CC: "Rafael J . Wysocki" <rafael.j.wysocki@...el.com>,
Viresh Kumar <viresh.kumar@...aro.org>,
Shuah Khan <skhan@...uxfoundation.org>,
"Borislav Petkov" <bp@...e.de>, Ingo Molnar <mingo@...nel.org>,
"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
"Sharma, Deepak" <Deepak.Sharma@....com>,
"Deucher, Alexander" <Alexander.Deucher@....com>,
"Limonciello, Mario" <Mario.Limonciello@....com>,
"Su, Jinzhou (Joe)" <Jinzhou.Su@....com>,
"Du, Xiaojian" <Xiaojian.Du@....com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"x86@...nel.org" <x86@...nel.org>
Subject: Re: [PATCH 16/19] cpupower: enable boost state support for
amd-pstate module
On Thu, Sep 09, 2021 at 01:32:37AM +0800, Fontenot, Nathan wrote:
> On 9/8/2021 9:59 AM, Huang Rui wrote:
> > The AMD P-state boost API is different from ACPI hardware P-states, so
> > implement the support for amd-pstate kernel module.
> >
> > Signed-off-by: Huang Rui <ray.huang@....com>
> > ---
> > tools/power/cpupower/lib/cpufreq.c | 20 ++++++++++++++++++++
> > tools/power/cpupower/lib/cpufreq.h | 3 +++
> > tools/power/cpupower/utils/helpers/misc.c | 7 +++++++
> > 3 files changed, 30 insertions(+)
> >
> > diff --git a/tools/power/cpupower/lib/cpufreq.c b/tools/power/cpupower/lib/cpufreq.c
> > index 3f92ddadaad2..37da87bdcfb1 100644
> > --- a/tools/power/cpupower/lib/cpufreq.c
> > +++ b/tools/power/cpupower/lib/cpufreq.c
> > @@ -790,3 +790,23 @@ unsigned long cpufreq_get_transitions(unsigned int cpu)
> > {
> > return sysfs_cpufreq_get_one_value(cpu, STATS_NUM_TRANSITIONS);
> > }
> > +
> > +int amd_pstate_boost_support(unsigned int cpu)
> > +{
> > + unsigned int highest_perf, nominal_perf;
> > +
> > + highest_perf = sysfs_cpufreq_get_one_value(cpu, AMD_PSTATE_HIGHEST_PERF);
> > + nominal_perf = sysfs_cpufreq_get_one_value(cpu, AMD_PSTATE_NOMINAL_PERF);
> > +
> > + return highest_perf > nominal_perf ? 1 : 0;
> > +}
> > +
> > +int amd_pstate_boost_enabled(unsigned int cpu)
> > +{
> > + unsigned int cpuinfo_max, amd_pstate_max;
> > +
> > + cpuinfo_max = sysfs_cpufreq_get_one_value(cpu, CPUINFO_MAX_FREQ);
> > + amd_pstate_max = sysfs_cpufreq_get_one_value(cpu, AMD_PSTATE_MAX_FREQ);
> > +
> > + return cpuinfo_max == amd_pstate_max ? 1 : 0;
> > +}
> > diff --git a/tools/power/cpupower/lib/cpufreq.h b/tools/power/cpupower/lib/cpufreq.h
> > index 95f4fd9e2656..d54d02a7a4f4 100644
> > --- a/tools/power/cpupower/lib/cpufreq.h
> > +++ b/tools/power/cpupower/lib/cpufreq.h
> > @@ -203,6 +203,9 @@ int cpufreq_modify_policy_governor(unsigned int cpu, char *governor);
> > int cpufreq_set_frequency(unsigned int cpu,
> > unsigned long target_frequency);
> >
> > +int amd_pstate_boost_support(unsigned int cpu);
> > +int amd_pstate_boost_enabled(unsigned int cpu);
> > +
> > #ifdef __cplusplus
> > }
> > #endif
> > diff --git a/tools/power/cpupower/utils/helpers/misc.c b/tools/power/cpupower/utils/helpers/misc.c
> > index 07d80775fb68..aba979320760 100644
> > --- a/tools/power/cpupower/utils/helpers/misc.c
> > +++ b/tools/power/cpupower/utils/helpers/misc.c
> > @@ -10,6 +10,7 @@
> > #if defined(__i386__) || defined(__x86_64__)
> >
> > #include "cpupower_intern.h"
> > +#include "cpufreq.h"
> >
> > #define MSR_AMD_HWCR 0xc0010015
> >
> > @@ -39,6 +40,12 @@ int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active,
> > if (ret)
> > return ret;
> > }
> > + } if ((cpupower_cpu_info.caps & CPUPOWER_CAP_AMD_PSTATE) &&
>
> Shouldn't this be
>
> } else if ((cpupower_cpu_info.caps & CPUPOWER_CAP_AMD_PSTATE) &&
>
Nice catch, it's my typo.
Thanks,
Ray
Powered by blists - more mailing lists