Re: [PATCH v2 1/5] arch_topology: Introduce thermal pressure update function
From: Dietmar Eggemann
Date: Tue Oct 26 2021 - 12:51:10 EST
On 15/10/2021 16:45, Lukasz Luba wrote:
> The thermal pressure is a mechanism which is used for providing
> information about reduced CPU performance to the scheduler. Usually code
> has to convert the value from frequency units into capacity units,
> which are understandable by the scheduler. Create a common conversion code
> which can be just used via a handy API.
>
> Signed-off-by: Lukasz Luba <lukasz.luba@xxxxxxx>
> ---
> arch/arm/include/asm/topology.h | 1 +
> arch/arm64/include/asm/topology.h | 1 +
> drivers/base/arch_topology.c | 36 ++++++++++++++++++++++++++++++-
> include/linux/arch_topology.h | 3 +++
> include/linux/sched/topology.h | 7 ++++++
> 5 files changed, 47 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm/include/asm/topology.h b/arch/arm/include/asm/topology.h
> index 470299ee2fba..aee6c456c085 100644
> --- a/arch/arm/include/asm/topology.h
> +++ b/arch/arm/include/asm/topology.h
> @@ -24,6 +24,7 @@
> /* Replace task scheduler's default thermal pressure API */
> #define arch_scale_thermal_pressure topology_get_thermal_pressure
> #define arch_set_thermal_pressure topology_set_thermal_pressure
> +#define arch_thermal_pressure_update topology_thermal_pressure_update
>
> #else
>
> diff --git a/arch/arm64/include/asm/topology.h b/arch/arm64/include/asm/topology.h
> index ec2db3419c41..c997015402bc 100644
> --- a/arch/arm64/include/asm/topology.h
> +++ b/arch/arm64/include/asm/topology.h
> @@ -33,6 +33,7 @@ void update_freq_counters_refs(void);
> /* Replace task scheduler's default thermal pressure API */
> #define arch_scale_thermal_pressure topology_get_thermal_pressure
> #define arch_set_thermal_pressure topology_set_thermal_pressure
> +#define arch_thermal_pressure_update topology_thermal_pressure_update
s/thermal_pressure_update/update_thermal_pressure ?
The scheme seems to be {arch|topology}_*foo*_thermal_pressure
But ...
>
> #include <asm-generic/topology.h>
>
> diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
> index 43407665918f..1fa28b5afdb2 100644
> --- a/drivers/base/arch_topology.c
> +++ b/drivers/base/arch_topology.c
> @@ -25,6 +25,7 @@
> static DEFINE_PER_CPU(struct scale_freq_data __rcu *, sft_data);
> static struct cpumask scale_freq_counters_mask;
> static bool scale_freq_invariant;
> +static DEFINE_PER_CPU(u32, freq_factor) = 1;
>
> static bool supports_scale_freq_counters(const struct cpumask *cpus)
> {
> @@ -168,6 +169,40 @@ void topology_set_thermal_pressure(const struct cpumask *cpus,
> }
> EXPORT_SYMBOL_GPL(topology_set_thermal_pressure);
>
> +/**
> + * topology_thermal_pressure_update() - Update thermal pressure for CPUs
> + * @cpus : The related CPUs for which capacity has been reduced
> + * @capped_freq : The maximum allowed frequency that CPUs can run at
> + *
> + * Update the value of thermal pressure for all @cpus in the mask. The
> + * cpumask should include all (online+offline) affected CPUs, to avoid
> + * operating on stale data when hot-plug is used for some CPUs. The
> + * @capped_freq must be less or equal to the max possible frequency and
> + * reflects the currently allowed max CPUs frequency due to thermal capping.
> + * The @capped_freq must be provided in kHz.
> + */
> +void topology_thermal_pressure_update(const struct cpumask *cpus,
> + unsigned long capped_freq)
> +{
... why not just s/unsigned long th_pressure/unsigned long capped_freq
in existing topology_set_thermal_pressure() and move code the
frequency/capacity conversion in there? The patch set will become
considerably smaller.
void topology_set_thermal_pressure(const struct cpumask *cpus,
- unsigned long th_pressure)
+ unsigned long capped_freq)
{
+ unsigned long max_capacity, capacity;
int cpu;
- for_each_cpu(cpu, cpus)
- WRITE_ONCE(per_cpu(thermal_pressure, cpu), th_pressure);
+ if (!cpus)
+ return;
+
+ cpu = cpumask_first(cpus);
+ max_capacity = arch_scale_cpu_capacity(cpu);
+
+ /* Convert to MHz scale which is used in 'freq_factor' */
+ capped_freq /= 1000;
+
+ capacity = mult_frac(capped_freq, max_capacity,
+ per_cpu(freq_factor, cpu));
+
+ for_each_cpu(cpu, cpus) {
+ WRITE_ONCE(per_cpu(thermal_pressure, cpu),
+ max_capacity - capacity);
+ }
}
EXPORT_SYMBOL_GPL(topology_set_thermal_pressure);
And a user like [drivers/thermal/cpufreq_cooling.c] can call
arch_set_thermal_pressure(cpus, frequency).
[...]