Re: [PATCH v4 6/7] arm64: use activity monitors for frequency invariance

[Valentin Schneider]

Ionela Voinescu writes:

> Signed-off-by: Ionela Voinescu <ionela.voinescu@xxxxxxx>
> Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
> Cc: Will Deacon <will@xxxxxxxxxx>
> Cc: Sudeep Holla <sudeep.holla@xxxxxxx>

With the small nits below:

Reviewed-by: Valentin Schneider <valentin.schneider@xxxxxxx>

> diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
> index fa9528dfd0ce..7606cbd63517 100644
> --- a/arch/arm64/kernel/topology.c
> +++ b/arch/arm64/kernel/topology.c
> +
> +static inline int

That should be bool, seeing what it returns.

> +enable_policy_freq_counters(int cpu, cpumask_var_t valid_cpus)
> +{
> +	struct cpufreq_policy *policy = cpufreq_cpu_get(cpu);
> +
> +	if (!policy) {
> +		pr_debug("CPU%d: No cpufreq policy found.\n", cpu);
> +		return false;
> +	}
> +
> +	if (cpumask_subset(policy->related_cpus, valid_cpus))
> +		cpumask_or(amu_fie_cpus, policy->related_cpus,
> +			   amu_fie_cpus);
> +
> +	cpufreq_cpu_put(policy);
> +
> +	return true;
> +}
> diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
> index 1eb81f113786..1ab2b7503d63 100644
> --- a/drivers/base/arch_topology.c
> +++ b/drivers/base/arch_topology.c
> @@ -29,6 +29,14 @@ void arch_set_freq_scale(struct cpumask *cpus, unsigned long cur_freq,
>       unsigned long scale;
>       int i;
>
> +	/*
> +	 * If the use of counters for FIE is enabled, just return as we don't
> +	 * want to update the scale factor with information from CPUFREQ.
> +	 * Instead the scale factor will be updated from arch_scale_freq_tick.
> +	 */
> +	if (arch_cpu_freq_counters(cpus))
> +		return;
> +
>       scale = (cur_freq << SCHED_CAPACITY_SHIFT) / max_freq;
>
>       for_each_cpu(i, cpus)
> diff --git a/include/linux/topology.h b/include/linux/topology.h
> index eb2fe6edd73c..397aad6ae163 100644
> --- a/include/linux/topology.h
> +++ b/include/linux/topology.h
> @@ -227,5 +227,12 @@ static inline const struct cpumask *cpu_cpu_mask(int cpu)
>       return cpumask_of_node(cpu_to_node(cpu));
>  }
>
> +#ifndef arch_cpu_freq_counters
> +static __always_inline
> +bool arch_cpu_freq_counters(struct cpumask *cpus)
> +{
> +	return false;
> +}
> +#endif
>

Apologies for commenting on this only now, I had missed it in my earlier
round of review.

I would've liked to keep this contained within arm64 stuff until we agreed
on a more generic counter-driven FIE interface, but seems like we can't evade
it due to the arch_topology situation.

Would it make sense to relocate this stub to arch_topology.h instead, at
least for the time being? That way the only non-arm64 changes are condensed
in arch_topology (even if it doesn't change much in terms of header files,
since topology.h imports arch_topology.h)

>  #endif /* _LINUX_TOPOLOGY_H */