2021-01-05 18:23:41

by Rafael J. Wysocki

[permalink] [raw]
Subject: [PATCH] cpufreq: intel_pstate: Use HWP capabilities in intel_cpufreq_adjust_perf()

From: Rafael J. Wysocki <[email protected]>

If turbo P-states cannot be used, either due to the configuration of
the processor, or because intel_pstate is not allowed to used them,
the maximum available P-state with HWP enabled corresponds to the
HWP_CAP.GUARANTEED value which is not static. It can be adjusted by
an out-of-band agent or during an Intel Speed Select performance
level change, so long as it remains less than or equal to
HWP_CAP.MAX.

However, if turbo P-states cannot be used, intel_cpufreq_adjust_perf()
always uses pstate.max_pstate (set during the initialization of the
driver only) as the maximum available P-state, so it may miss a change
of the HWP_CAP.GUARANTEED value.

Prevent that from happening by modifyig intel_cpufreq_adjust_perf()
to always read the "guaranteed" and "maximum turbo" performance
levels from the cached HWP_CAP value.

Signed-off-by: Rafael J. Wysocki <[email protected]>
---
drivers/cpufreq/intel_pstate.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

Index: linux-pm/drivers/cpufreq/intel_pstate.c
===================================================================
--- linux-pm.orig/drivers/cpufreq/intel_pstate.c
+++ linux-pm/drivers/cpufreq/intel_pstate.c
@@ -2653,12 +2653,13 @@ static void intel_cpufreq_adjust_perf(un
unsigned long capacity)
{
struct cpudata *cpu = all_cpu_data[cpunum];
+ u64 hwp_cap = READ_ONCE(cpu->hwp_cap_cached);
int old_pstate = cpu->pstate.current_pstate;
int cap_pstate, min_pstate, max_pstate, target_pstate;

update_turbo_state();
- cap_pstate = global.turbo_disabled ? cpu->pstate.max_pstate :
- cpu->pstate.turbo_pstate;
+ cap_pstate = global.turbo_disabled ? HWP_GUARANTEED_PERF(hwp_cap) :
+ HWP_HIGHEST_PERF(hwp_cap);

/* Optimization: Avoid unnecessary divisions. */





2021-01-06 07:03:04

by srinivas pandruvada

[permalink] [raw]
Subject: Re: [PATCH] cpufreq: intel_pstate: Use HWP capabilities in intel_cpufreq_adjust_perf()

On Tue, 2021-01-05 at 19:20 +0100, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <[email protected]>
>
> If turbo P-states cannot be used, either due to the configuration of
> the processor, or because intel_pstate is not allowed to used them,
> the maximum available P-state with HWP enabled corresponds to the
> HWP_CAP.GUARANTEED value which is not static.  It can be adjusted by
> an out-of-band agent or during an Intel Speed Select performance
> level change, so long as it remains less than or equal to
> HWP_CAP.MAX.
>
> However, if turbo P-states cannot be used,
> intel_cpufreq_adjust_perf()
> always uses pstate.max_pstate (set during the initialization of the
> driver only) as the maximum available P-state, so it may miss a
> change
> of the HWP_CAP.GUARANTEED value.
>
> Prevent that from happening by modifyig intel_cpufreq_adjust_perf()
> to always read the "guaranteed" and "maximum turbo" performance
> levels from the cached HWP_CAP value.
>
> Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Srinivas Pandruvada <[email protected]>

> ---
>  drivers/cpufreq/intel_pstate.c |    5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
>
> Index: linux-pm/drivers/cpufreq/intel_pstate.c
> ===================================================================
> --- linux-pm.orig/drivers/cpufreq/intel_pstate.c
> +++ linux-pm/drivers/cpufreq/intel_pstate.c
> @@ -2653,12 +2653,13 @@ static void intel_cpufreq_adjust_perf(un
>                                       unsigned long capacity)
>  {
>         struct cpudata *cpu = all_cpu_data[cpunum];
> +       u64 hwp_cap = READ_ONCE(cpu->hwp_cap_cached);
>         int old_pstate = cpu->pstate.current_pstate;
>         int cap_pstate, min_pstate, max_pstate, target_pstate;
>  
>         update_turbo_state();
> -       cap_pstate = global.turbo_disabled ? cpu->pstate.max_pstate :
> -                                            cpu-
> >pstate.turbo_pstate;
> +       cap_pstate = global.turbo_disabled ?
> HWP_GUARANTEED_PERF(hwp_cap) :
> +                                           
> HWP_HIGHEST_PERF(hwp_cap);
>  
>         /* Optimization: Avoid unnecessary divisions. */
>  
>
>
>