Linux Power Management development
 help / color / mirror / Atom feed
From: Mario Limonciello <mario.limonciello@amd.com>
To: Naresh Solanki <naresh.solanki@9elements.com>
Cc: Huang Rui <ray.huang@amd.com>,
	"Gautham R. Shenoy" <gautham.shenoy@amd.com>,
	Perry Yuan <perry.yuan@amd.com>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Viresh Kumar <viresh.kumar@linaro.org>,
	linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] cpufreq/amd-pstate: Refactor max frequency calculation
Date: Wed, 18 Dec 2024 13:20:09 -0600	[thread overview]
Message-ID: <b1210cbf-22c8-48f1-ad50-d81395abc8f3@amd.com> (raw)
In-Reply-To: <20241218190030.1228868-1-naresh.solanki@9elements.com>

On 12/18/2024 13:00, Naresh Solanki wrote:
> Refactor to calculate max-freq more accurately.

Can you add some more detail about what you're finding?
What was it before, what is it now, why is it more accurate?

> 
> Signed-off-by: Naresh Solanki <naresh.solanki@9elements.com>
> ---
>   drivers/cpufreq/amd-pstate.c | 5 ++---
>   1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
> index d7630bab2516..78a2cbd14952 100644
> --- a/drivers/cpufreq/amd-pstate.c
> +++ b/drivers/cpufreq/amd-pstate.c
> @@ -892,7 +892,7 @@ static int amd_pstate_init_freq(struct amd_cpudata *cpudata)
>   	u64 numerator;
>   	u32 nominal_perf, nominal_freq;
>   	u32 lowest_nonlinear_perf, lowest_nonlinear_freq;
> -	u32 boost_ratio, lowest_nonlinear_ratio;
> +	u32 lowest_nonlinear_ratio;
>   	struct cppc_perf_caps cppc_perf;
>   
>   	ret = cppc_get_perf_caps(cpudata->cpu, &cppc_perf);
> @@ -914,8 +914,7 @@ static int amd_pstate_init_freq(struct amd_cpudata *cpudata)
>   	ret = amd_get_boost_ratio_numerator(cpudata->cpu, &numerator);
>   	if (ret)
>   		return ret;
> -	boost_ratio = div_u64(numerator << SCHED_CAPACITY_SHIFT, nominal_perf);
> -	max_freq = (nominal_freq * boost_ratio >> SCHED_CAPACITY_SHIFT) * 1000;
> +	max_freq = div_u64(numerator * nominal_freq * 1000, nominal_perf);

This doesn't apply currently, because of some changes in the 
superm1.git/linux-next branch; specifically:

https://git.kernel.org/superm1/c/68cb0e77b6439

I haven't sent this out to linux-pm yet so it could be in linux-next, 
but will be doing that soon.  So can you please rebase on that branch if 
this change still makes sense?

>   
>   	lowest_nonlinear_perf = READ_ONCE(cpudata->lowest_nonlinear_perf);
>   	lowest_nonlinear_ratio = div_u64(lowest_nonlinear_perf << SCHED_CAPACITY_SHIFT,


  reply	other threads:[~2024-12-18 19:20 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-18 19:00 [PATCH] cpufreq/amd-pstate: Refactor max frequency calculation Naresh Solanki
2024-12-18 19:20 ` Mario Limonciello [this message]
2024-12-19  9:21   ` Naresh Solanki
2024-12-19 15:01     ` Mario Limonciello
2024-12-19 16:33       ` Naresh Solanki

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=b1210cbf-22c8-48f1-ad50-d81395abc8f3@amd.com \
    --to=mario.limonciello@amd.com \
    --cc=gautham.shenoy@amd.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=naresh.solanki@9elements.com \
    --cc=perry.yuan@amd.com \
    --cc=rafael@kernel.org \
    --cc=ray.huang@amd.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