cpufreq: intel_pstate: Show different max frequency with turbo 3 and HWP
diff mbox

Message ID 20180718215159.28698-1-srinivas.pandruvada@linux.intel.com
State Mainlined
Delegated to: Rafael Wysocki
Headers show

Commit Message

Srinivas Pandruvada July 18, 2018, 9:51 p.m. UTC
On HWP platforms with Turbo 3.0, the HWP capability max ratio shows the
maximum ratio of that core, which can be different than other cores. If
we show the correct maximum frequency in cpufreq sysfs via
cpuinfo_max_freq and scaling_max_freq then, user can know which cores
can run faster for pinning some high priority tasks.

Currently the max turbo frequency is shown as max frequency, which is
the max of all cores, even if some cores can't reach that frequency
even for single threaded workload.

But it is possible that max ratio in HWP capabilities is set as 0xFF or
some high invalid value (E.g. One KBL NUC). Since the actual performance
can never exceed 1 core turbo frequency from MSR TURBO_RATIO_LIMIT, we
use this as a bound check.

Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
---
 drivers/cpufreq/intel_pstate.c | 9 +++++++++
 1 file changed, 9 insertions(+)

Comments

Rafael J. Wysocki July 23, 2018, 11:35 a.m. UTC | #1
On Wednesday, July 18, 2018 11:51:59 PM CEST Srinivas Pandruvada wrote:
> On HWP platforms with Turbo 3.0, the HWP capability max ratio shows the
> maximum ratio of that core, which can be different than other cores. If
> we show the correct maximum frequency in cpufreq sysfs via
> cpuinfo_max_freq and scaling_max_freq then, user can know which cores
> can run faster for pinning some high priority tasks.
> 
> Currently the max turbo frequency is shown as max frequency, which is
> the max of all cores, even if some cores can't reach that frequency
> even for single threaded workload.
> 
> But it is possible that max ratio in HWP capabilities is set as 0xFF or
> some high invalid value (E.g. One KBL NUC). Since the actual performance
> can never exceed 1 core turbo frequency from MSR TURBO_RATIO_LIMIT, we
> use this as a bound check.
> 
> Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
> ---
>  drivers/cpufreq/intel_pstate.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
> index ece120da3353..97096127d015 100644
> --- a/drivers/cpufreq/intel_pstate.c
> +++ b/drivers/cpufreq/intel_pstate.c
> @@ -2072,6 +2072,15 @@ static int __intel_pstate_cpu_init(struct cpufreq_policy *policy)
>  			cpu->pstate.max_pstate : cpu->pstate.turbo_pstate;
>  	policy->cpuinfo.max_freq *= cpu->pstate.scaling;
>  
> +	if (hwp_active) {
> +		unsigned int max_freq;
> +
> +		max_freq = global.turbo_disabled ?
> +			cpu->pstate.max_freq : cpu->pstate.turbo_freq;
> +		if (max_freq < policy->cpuinfo.max_freq)
> +			policy->cpuinfo.max_freq = max_freq;
> +	}
> +
>  	intel_pstate_init_acpi_perf_limits(policy);
>  
>  	policy->fast_switch_possible = true;
> 

Applied, thanks!

Patch
diff mbox

diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index ece120da3353..97096127d015 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -2072,6 +2072,15 @@  static int __intel_pstate_cpu_init(struct cpufreq_policy *policy)
 			cpu->pstate.max_pstate : cpu->pstate.turbo_pstate;
 	policy->cpuinfo.max_freq *= cpu->pstate.scaling;
 
+	if (hwp_active) {
+		unsigned int max_freq;
+
+		max_freq = global.turbo_disabled ?
+			cpu->pstate.max_freq : cpu->pstate.turbo_freq;
+		if (max_freq < policy->cpuinfo.max_freq)
+			policy->cpuinfo.max_freq = max_freq;
+	}
+
 	intel_pstate_init_acpi_perf_limits(policy);
 
 	policy->fast_switch_possible = true;