mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: srinivas pandruvada <srinivas.pandruvada@linux.intel.com>
To: Xueqin Luo <luoxueqin@kylinos.cn>,
	rafael@kernel.org, lenb@kernel.org,  viresh.kumar@linaro.org,
	saket.dumbre@intel.com, linux-acpi@vger.kernel.org,
		linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
		acpica-devel@lists.linux.dev
Subject: Re: [PATCH] ACPI: CPPC: Add helper to get effective guaranteed performance
Date: Fri, 29 May 2026 08:33:19 -0700	[thread overview]
Message-ID: <60119eb883057391dc6def200be119e56a6e4fb5.camel@linux.intel.com> (raw)
In-Reply-To: <20260529081801.29346-1-luoxueqin@kylinos.cn>

On Fri, 2026-05-29 at 16:18 +0800, Xueqin Luo wrote:
> According to the ACPI CPPC specification, if the Guaranteed
> Performance register is not implemented, OSPM assumes the
> guaranteed performance is equal to nominal performance.
> 
> Add cppc_get_effective_guaranteed_perf() helper to provide a
> spec-compliant effective guaranteed performance value while
> preserving the raw values returned by cppc_get_perf_caps().
> 
Hi Luo,

Are you trying to fix any issue where you have issue with
"base_frequency" display? intel_pstate will revert to native MSR, which
is always present when we are using CPPC.

I worry about some bad nominal perf in some platforms as we will be
dependent on OEM BIOS.

Thanks,
Srinivas


> Signed-off-by: Xueqin Luo <luoxueqin@kylinos.cn>
> ---
>  drivers/acpi/cppc_acpi.c       | 26 ++++++++++++++++++++++++++
>  drivers/cpufreq/intel_pstate.c | 10 ++++------
>  include/acpi/cppc_acpi.h       |  5 +++++
>  3 files changed, 35 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c
> index f370be8715ae..6dc11030707a 100644
> --- a/drivers/acpi/cppc_acpi.c
> +++ b/drivers/acpi/cppc_acpi.c
> @@ -1328,6 +1328,32 @@ int cppc_get_highest_perf(int cpunum, u64
> *highest_perf)
>  }
>  EXPORT_SYMBOL_GPL(cppc_get_highest_perf);
>  
> +/**
> + * cppc_get_effective_guaranteed_perf - get effective guaranteed
> performance
> + * @cpunum: CPU number
> + * @guaranteed_perf: Effective guaranteed performance value
> + *
> + * The ACPI CPPC specification states that if the Guaranteed
> + * Performance register is not implemented, OSPM assumes the
> + * guaranteed performance is equal to nominal performance.
> + *
> + * Return: 0 on success, negative errno on failure.
> + */
> +int cppc_get_effective_guaranteed_perf(int cpunum, u64
> *guaranteed_perf)
> +{
> +	struct cppc_perf_caps perf_caps;
> +	int ret;
> +
> +	ret = cppc_get_perf_caps(cpunum, &perf_caps);
> +	if (ret)
> +		return ret;
> +	*guaranteed_perf = perf_caps.guaranteed_perf ?:
> +			   perf_caps.nominal_perf;
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(cppc_get_effective_guaranteed_perf);
> +
>  /**
>   * cppc_get_epp_perf - Get the epp register value.
>   * @cpunum: CPU from which to get epp preference value.
> diff --git a/drivers/cpufreq/intel_pstate.c
> b/drivers/cpufreq/intel_pstate.c
> index 8dc22a65e025..8934e48f71a2 100644
> --- a/drivers/cpufreq/intel_pstate.c
> +++ b/drivers/cpufreq/intel_pstate.c
> @@ -404,17 +404,15 @@ static void intel_pstate_set_itmt_prio(int cpu)
>  
>  static int intel_pstate_get_cppc_guaranteed(int cpu)
>  {
> -	struct cppc_perf_caps cppc_perf;
> +	u64 guaranteed_perf;
>  	int ret;
>  
> -	ret = cppc_get_perf_caps(cpu, &cppc_perf);
> +	ret = cppc_get_effective_guaranteed_perf(cpu,
> &guaranteed_perf);
> +
>  	if (ret)
>  		return ret;
>  
> -	if (cppc_perf.guaranteed_perf)
> -		return cppc_perf.guaranteed_perf;
> -
> -	return cppc_perf.nominal_perf;
> +	return guaranteed_perf;
>  }
>  
>  static int intel_pstate_cppc_get_scaling(int cpu)
> diff --git a/include/acpi/cppc_acpi.h b/include/acpi/cppc_acpi.h
> index d1f02ceec4f9..c02f3bc1fc1a 100644
> --- a/include/acpi/cppc_acpi.h
> +++ b/include/acpi/cppc_acpi.h
> @@ -155,6 +155,7 @@ struct cppc_cpudata {
>  extern int cppc_get_desired_perf(int cpunum, u64 *desired_perf);
>  extern int cppc_get_nominal_perf(int cpunum, u64 *nominal_perf);
>  extern int cppc_get_highest_perf(int cpunum, u64 *highest_perf);
> +extern int cppc_get_effective_guaranteed_perf(int cpunum, u64
> *guaranteed_perf);
>  extern int cppc_get_perf_ctrs(int cpu, struct cppc_perf_fb_ctrs
> *perf_fb_ctrs);
>  extern int cppc_get_perf(int cpu, struct cppc_perf_ctrls
> *perf_ctrls);
>  extern int cppc_set_perf(int cpu, struct cppc_perf_ctrls
> *perf_ctrls);
> @@ -198,6 +199,10 @@ static inline int cppc_get_highest_perf(int
> cpunum, u64 *highest_perf)
>  {
>  	return -EOPNOTSUPP;
>  }
> +static inline int cppc_get_effective_guaranteed_perf(int cpunum, u64
> *guaranteed_perf)
> +{
> +	return -EOPNOTSUPP;
> +}
>  static inline int cppc_get_perf_ctrs(int cpu, struct
> cppc_perf_fb_ctrs *perf_fb_ctrs)
>  {
>  	return -EOPNOTSUPP;

  reply	other threads:[~2026-05-29 15:33 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-29  8:18 Xueqin Luo
2026-05-29 15:33 ` srinivas pandruvada [this message]
2026-06-01  2:31   ` luoxueqin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=60119eb883057391dc6def200be119e56a6e4fb5.camel@linux.intel.com \
    --to=srinivas.pandruvada@linux.intel.com \
    --cc=acpica-devel@lists.linux.dev \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=luoxueqin@kylinos.cn \
    --cc=rafael@kernel.org \
    --cc=saket.dumbre@intel.com \
    --cc=viresh.kumar@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®