Re: [PATCH v4 6/7] cpufreq/cppc: set the frequency used for computing the capacity

[Vincent Guittot]

Hi Pierre,

On Thu, 2 Nov 2023 at 10:07, Pierre Gondois <pierre.gondois@xxxxxxx> wrote:
>
> Hello Vincent,
>
> On 10/27/23 10:03, Vincent Guittot wrote:
> > Save the frequency associated to the performance that has been used when
> > initializing the capacity of CPUs.
> > Also, cppc cpufreq driver can register an artificial energy model. In such
> > case, it needs the frequency for this compute capacity.
> >
> > Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> > ---
> >   drivers/base/arch_topology.c | 15 ++++++++++++++-
> >   1 file changed, 14 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
> > index 9a073c2d2086..d4bef370feb3 100644
> > --- a/drivers/base/arch_topology.c
> > +++ b/drivers/base/arch_topology.c
> > @@ -349,6 +349,7 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
> >
> >   void topology_init_cpu_capacity_cppc(void)
> >   {
> > +     u64 capacity, capacity_scale = 0;
> >       struct cppc_perf_caps perf_caps;
> >       int cpu;
> >
> > @@ -365,6 +366,10 @@ void topology_init_cpu_capacity_cppc(void)
> >                   (perf_caps.highest_perf >= perf_caps.nominal_perf) &&
> >                   (perf_caps.highest_perf >= perf_caps.lowest_perf)) {
> >                       raw_capacity[cpu] = perf_caps.highest_perf;
> > +                     capacity_scale = max_t(u64, capacity_scale, raw_capacity[cpu]);
> > +
> > +                     per_cpu(capacity_ref_freq, cpu) = cppc_perf_to_khz(&perf_caps, raw_capacity[cpu]);
> > +
>
> To continue Beata's point, I think there is an issue with the following:
> cppc_perf_to_khz() and cppc_khz_to_perf() were previously used with a struct containing
> frequencies in KHz, cf. [1].
> In the original _CPC object, frequencies are in MHz. It means that the perf_caps struct
> here contains frequencies in MHz, and per_cpu(capacity_ref_freq, cpu) is in MHz aswell.

Yeah, I haven't noticed this intermediate step in cppc_cpufreq. I'm
going to fix this

>
> [1] https://github.com/torvalds/linux/blob/master/drivers/cpufreq/cppc_cpufreq.c#L682
>
>
> >                       pr_debug("cpu_capacity: CPU%d cpu_capacity=%u (raw).\n",
> >                                cpu, raw_capacity[cpu]);
> >                       continue;
> > @@ -375,7 +380,15 @@ void topology_init_cpu_capacity_cppc(void)
> >               goto exit;
> >       }
> >
> > -     topology_normalize_cpu_scale();
> > +     for_each_possible_cpu(cpu) {
> > +             capacity = raw_capacity[cpu];
> > +             capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
> > +                                  capacity_scale);
> > +             topology_set_cpu_scale(cpu, capacity);
> > +             pr_debug("cpu_capacity: CPU%d cpu_capacity=%lu\n",
> > +                     cpu, topology_get_cpu_scale(cpu));
> > +     }
> > +
> >       schedule_work(&update_topology_flags_work);
> >       pr_debug("cpu_capacity: cpu_capacity initialization done\n");
> >