Re: [PATCH v2 1/1] x86,sched: On AMD EPYC set freq_max = max_boost in schedutil invariant formula

[Rafael J. Wysocki]

On Tue, Feb 2, 2021 at 7:59 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
>
> On Fri, Jan 22, 2021 at 9:47 PM Giovanni Gherdovich <ggherdovich@xxxxxxx> wrote:
> >
>
> [cut]
>
> >
> > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems")
> > Fixes: 976df7e5730e ("x86, sched: Use midpoint of max_boost and max_P for frequency invariance on AMD EPYC")
> > Reported-by: Michael Larabel <Michael@xxxxxxxxxxxx>
> > Signed-off-by: Giovanni Gherdovich <ggherdovich@xxxxxxx>
> > ---
> >  drivers/cpufreq/acpi-cpufreq.c   | 64 +++++++++++++++++++++++++++++++-
> >  drivers/cpufreq/cpufreq.c        |  3 ++
> >  include/linux/cpufreq.h          |  5 +++
> >  kernel/sched/cpufreq_schedutil.c |  8 +++-
> >  4 files changed, 76 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
> > index 1e4fbb002a31..2378bc1bf2c4 100644
> > --- a/drivers/cpufreq/acpi-cpufreq.c
> > +++ b/drivers/cpufreq/acpi-cpufreq.c
> > @@ -27,6 +27,10 @@
> >
> >  #include <acpi/processor.h>
> >
> > +#ifdef CONFIG_ACPI_CPPC_LIB
>
> Why is the #ifdef needed here?
>
> > +#include <acpi/cppc_acpi.h>
> > +#endif
> > +
> >  #include <asm/msr.h>
> >  #include <asm/processor.h>
> >  #include <asm/cpufeature.h>
> > @@ -628,11 +632,57 @@ static int acpi_cpufreq_blacklist(struct cpuinfo_x86 *c)
> >  }
> >  #endif
> >
> > +#ifdef CONFIG_ACPI_CPPC_LIB
> > +static bool amd_max_boost(unsigned int max_freq,
> > +                         unsigned int *max_boost)
> > +{
> > +       struct cppc_perf_caps perf_caps;
> > +       u64 highest_perf, nominal_perf, perf_ratio;
> > +       int ret;
> > +
> > +       ret = cppc_get_perf_caps(0, &perf_caps);
> > +       if (ret) {
> > +               pr_debug("Could not retrieve perf counters (%d)\n", ret);
> > +               return false;
> > +       }
> > +
> > +       highest_perf = perf_caps.highest_perf;
> > +       nominal_perf = perf_caps.nominal_perf;
> > +
> > +       if (!highest_perf || !nominal_perf) {
> > +               pr_debug("Could not retrieve highest or nominal performance\n");
> > +               return false;
> > +       }
> > +
> > +       perf_ratio = div_u64(highest_perf * SCHED_CAPACITY_SCALE, nominal_perf);
> > +       if (perf_ratio <= SCHED_CAPACITY_SCALE) {
> > +               pr_debug("Either perf_ratio is 0, or nominal >= highest performance\n");
> > +               return false;
> > +       }
> > +
> > +       *max_boost = max_freq * perf_ratio >> SCHED_CAPACITY_SHIFT;
> > +       if (!*max_boost) {
> > +               pr_debug("max_boost seems to be zero\n");
> > +               return false;
> > +       }
> > +
> > +       return true;
> > +}
> > +#else
> > +static bool amd_max_boost(unsigned int max_freq,
> > +                         unsigned int *max_boost)
> > +{
> > +       return false;
> > +}
> > +#endif
> > +
> >  static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
> >  {
> >         unsigned int i;
> >         unsigned int valid_states = 0;
> >         unsigned int cpu = policy->cpu;
> > +       unsigned int freq, max_freq = 0;
> > +       unsigned int max_boost;
> >         struct acpi_cpufreq_data *data;
> >         unsigned int result = 0;
> >         struct cpuinfo_x86 *c = &cpu_data(policy->cpu);
> > @@ -779,15 +829,25 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
> >                     freq_table[valid_states-1].frequency / 1000)
> >                         continue;
> >
> > +               freq = perf->states[i].core_frequency * 1000;
> >                 freq_table[valid_states].driver_data = i;
> > -               freq_table[valid_states].frequency =
> > -                   perf->states[i].core_frequency * 1000;
> > +               freq_table[valid_states].frequency = freq;
> > +
> > +               if (freq > max_freq)
> > +                       max_freq = freq;
> > +
> >                 valid_states++;
> >         }
> >         freq_table[valid_states].frequency = CPUFREQ_TABLE_END;
> >         policy->freq_table = freq_table;
> >         perf->state = 0;
> >
> > +       if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
> > +           amd_max_boost(max_freq, &max_boost)) {
> > +               policy->cpuinfo.max_boost = max_boost;
>
> Why not to set max_freq to max_boost instead?

I mean, would setting the frequency in the last table entry to max_boost work?

Alternatively, one more (artificial) entry with the frequency equal to
max_boost could be added.

> This value is set once at the init time anyway and schedutil would use
> max_boost instead of max_freq anyway.
>
> Also notice that the static branch is global and the max_boost value
> for different CPUs may be different, at least in theory.
>
> > +               static_branch_enable(&cpufreq_amd_max_boost);
> > +       }
> > +
> >         switch (perf->control_register.space_id) {
> >         case ACPI_ADR_SPACE_SYSTEM_IO:
> >                 /*
> > diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> > index d0a3525ce27f..b96677f6b57e 100644
> > --- a/drivers/cpufreq/cpufreq.c
> > +++ b/drivers/cpufreq/cpufreq.c
> > @@ -2721,6 +2721,9 @@ int cpufreq_boost_enabled(void)
> >  }
> >  EXPORT_SYMBOL_GPL(cpufreq_boost_enabled);
> >
> > +DEFINE_STATIC_KEY_FALSE(cpufreq_amd_max_boost);
> > +EXPORT_SYMBOL_GPL(cpufreq_amd_max_boost);
> > +
> >  /*********************************************************************
> >   *               REGISTER / UNREGISTER CPUFREQ DRIVER                *
> >   *********************************************************************/
> > diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> > index 9c8b7437b6cd..341cac76d254 100644
> > --- a/include/linux/cpufreq.h
> > +++ b/include/linux/cpufreq.h
> > @@ -40,9 +40,14 @@ enum cpufreq_table_sorting {
> >         CPUFREQ_TABLE_SORTED_DESCENDING
> >  };
> >
> > +DECLARE_STATIC_KEY_FALSE(cpufreq_amd_max_boost);
> > +
> > +#define cpufreq_driver_has_max_boost() static_branch_unlikely(&cpufreq_amd_max_boost)
> > +
> >  struct cpufreq_cpuinfo {
> >         unsigned int            max_freq;
> >         unsigned int            min_freq;
> > +       unsigned int            max_boost;
> >
> >         /* in 10^(-9) s = nanoseconds */
> >         unsigned int            transition_latency;
> > diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> > index 6931f0cdeb80..541f3db3f576 100644
> > --- a/kernel/sched/cpufreq_schedutil.c
> > +++ b/kernel/sched/cpufreq_schedutil.c
> > @@ -159,8 +159,12 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy,
> >                                   unsigned long util, unsigned long max)
> >  {
> >         struct cpufreq_policy *policy = sg_policy->policy;
> > -       unsigned int freq = arch_scale_freq_invariant() ?
> > -                               policy->cpuinfo.max_freq : policy->cur;
> > +       unsigned int freq, max_freq;
> > +
> > +       max_freq = cpufreq_driver_has_max_boost() ?
> > +                       policy->cpuinfo.max_boost : policy->cpuinfo.max_freq;
> > +
> > +       freq = arch_scale_freq_invariant() ? max_freq : policy->cur;
> >
> >         freq = map_util_freq(util, freq, max);
> >
> > --
> > 2.26.2
> >