[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAJZ5v0i8PKdaF5mt4tHsqWy5wO2+c6sRxmW+UiR7_HOn=w4Y6A@mail.gmail.com>
Date: Wed, 16 Mar 2022 19:16:04 +0100
From: "Rafael J. Wysocki" <rafael@...nel.org>
To: Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
Cc: Len Brown <lenb@...nel.org>,
"Rafael J. Wysocki" <rafael@...nel.org>,
Viresh Kumar <viresh.kumar@...aro.org>,
Linux PM <linux-pm@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] cpufreq: intel_pstate: Use firmware default EPP
On Thu, Mar 10, 2022 at 11:42 PM Srinivas Pandruvada
<srinivas.pandruvada@...ux.intel.com> wrote:
>
> For some specific platforms (E.g. AlderLake) the balance performance
> EPP is updated from the hard coded value in the driver. This acts as
> the default and balance_performance EPP. The purpose of this EPP
> update is to reach maximum 1 core turbo frequency (when possible) out
> of the box.
>
> Although we can achieve the objective by using hard coded value in the
> driver, there can be other EPP which can be better in terms of power.
> But that will be very subjective based on platform and use cases.
> This is not practical to have a per platform specific default hard coded
> in the driver.
>
> If a platform wants to specify default EPP, it can be set in the firmware.
> If this EPP is not the chipset default of 0x80 (balance_perf_epp unless
> driver changed it) and more performance oriented but not 0, the driver
> can use this as the default and balanced_perf EPP. In this case no driver
> update is required every time there is some new platform and default EPP.
>
> If the firmware didn't update the EPP from the chipset default then
> the hard coded value is used as per existing implementation.
>
> Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
> ---
> drivers/cpufreq/intel_pstate.c | 38 ++++++++++++++++++++++++++++------
> 1 file changed, 32 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
> index bc7f7e6759bd..846bb3a78788 100644
> --- a/drivers/cpufreq/intel_pstate.c
> +++ b/drivers/cpufreq/intel_pstate.c
> @@ -1692,6 +1692,37 @@ static void intel_pstate_enable_hwp_interrupt(struct cpudata *cpudata)
> }
> }
>
> +static void intel_pstate_update_epp_defaults(struct cpudata *cpudata)
> +{
> + cpudata->epp_default = intel_pstate_get_epp(cpudata, 0);
> +
> + /*
> + * If this CPU gen doesn't call for change in balance_perf
> + * EPP return.
> + */
> + if (epp_values[EPP_INDEX_BALANCE_PERFORMANCE] == HWP_EPP_BALANCE_PERFORMANCE)
> + return;
> +
> + /*
> + * If powerup EPP is something other than chipset default 0x80 and
> + * - is more performance oriented than 0x80 (default balance_perf EPP)
> + * - But less performance oriented than performance EPP
> + * then use this as new balance_perf EPP.
> + */
> + if (cpudata->epp_default < HWP_EPP_BALANCE_PERFORMANCE &&
> + cpudata->epp_default > HWP_EPP_PERFORMANCE) {
> + epp_values[EPP_INDEX_BALANCE_PERFORMANCE] = cpudata->epp_default;
> + return;
> + }
> +
> + /*
> + * Use hard coded value per gen to update the balance_perf
> + * and default EPP.
> + */
> + cpudata->epp_default = epp_values[EPP_INDEX_BALANCE_PERFORMANCE];
> + intel_pstate_set_epp(cpudata, cpudata->epp_default);
> +}
> +
> static void intel_pstate_hwp_enable(struct cpudata *cpudata)
> {
> /* First disable HWP notification interrupt till we activate again */
> @@ -1705,12 +1736,7 @@ static void intel_pstate_hwp_enable(struct cpudata *cpudata)
> if (cpudata->epp_default >= 0)
> return;
>
> - if (epp_values[EPP_INDEX_BALANCE_PERFORMANCE] == HWP_EPP_BALANCE_PERFORMANCE) {
> - cpudata->epp_default = intel_pstate_get_epp(cpudata, 0);
> - } else {
> - cpudata->epp_default = epp_values[EPP_INDEX_BALANCE_PERFORMANCE];
> - intel_pstate_set_epp(cpudata, cpudata->epp_default);
> - }
> + intel_pstate_update_epp_defaults(cpudata);
> }
>
> static int atom_get_min_pstate(void)
> --
Applied as 5.18 material, thanks!
Powered by blists - more mailing lists