From: "Gautham R. Shenoy" <gautham.shenoy@amd.com>
To: Mario Limonciello <superm1@kernel.org>
Cc: Perry Yuan <perry.yuan@amd.com>,
Dhananjay Ugwekar <Dhananjay.Ugwekar@amd.com>,
"open list:X86 ARCHITECTURE (32-BIT AND 64-BIT)"
<linux-kernel@vger.kernel.org>,
"open list:CPU FREQUENCY SCALING FRAMEWORK"
<linux-pm@vger.kernel.org>,
Mario Limonciello <mario.limonciello@amd.com>
Subject: Re: [PATCH v2 02/17] cpufreq/amd-pstate: Drop min and max cached frequencies
Date: Mon, 17 Feb 2025 12:08:59 +0530 [thread overview]
Message-ID: <Z7LZg8oq3LqfKZID@BLRRASHENOY1.amd.com> (raw)
In-Reply-To: <20250215005244.1212285-3-superm1@kernel.org>
Hello Mario,
On Fri, Feb 14, 2025 at 06:52:29PM -0600, Mario Limonciello wrote:
> From: Mario Limonciello <mario.limonciello@amd.com>
>
> Use the perf_to_freq helpers to calculate this on the fly.
>
> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> --
> v2:
> * Keep cached limits
> ---
[..snip..]
> --- a/drivers/cpufreq/amd-pstate.c
> +++ b/drivers/cpufreq/amd-pstate.c
> @@ -717,7 +717,7 @@ static int amd_pstate_cpu_boost_update(struct cpufreq_policy *policy, bool on)
> int ret = 0;
>
> nominal_freq = READ_ONCE(cpudata->nominal_freq);
> - max_freq = READ_ONCE(cpudata->max_freq);
> + max_freq = perf_to_freq(cpudata, READ_ONCE(cpudata->highest_perf));
>
> if (on)
> policy->cpuinfo.max_freq = max_freq;
> @@ -901,35 +901,25 @@ static u32 amd_pstate_get_transition_latency(unsigned int cpu)
> static int amd_pstate_init_freq(struct amd_cpudata *cpudata)
> {
> int ret;
> - u32 min_freq, max_freq;
> - u32 nominal_freq, lowest_nonlinear_freq;
> + u32 min_freq, nominal_freq, lowest_nonlinear_freq;
> struct cppc_perf_caps cppc_perf;
>
> ret = cppc_get_perf_caps(cpudata->cpu, &cppc_perf);
> if (ret)
> return ret;
>
> - if (quirks && quirks->lowest_freq)
> - min_freq = quirks->lowest_freq;
> - else
> - min_freq = cppc_perf.lowest_freq;
> -
> if (quirks && quirks->nominal_freq)
> nominal_freq = quirks->nominal_freq;
> else
> nominal_freq = cppc_perf.nominal_freq;
>
> - min_freq *= 1000;
> nominal_freq *= 1000;
> -
> WRITE_ONCE(cpudata->nominal_freq, nominal_freq);
So cpudata->nominal_freq will be in KHz here.
> - WRITE_ONCE(cpudata->min_freq, min_freq);
> -
> - max_freq = perf_to_freq(cpudata, cpudata->highest_perf);
> - lowest_nonlinear_freq = perf_to_freq(cpudata, cpudata->lowest_nonlinear_perf);
>
> - WRITE_ONCE(cpudata->lowest_nonlinear_freq, lowest_nonlinear_freq);
> - WRITE_ONCE(cpudata->max_freq, max_freq);
> + if (quirks && quirks->lowest_freq) {
> + min_freq = quirks->lowest_freq;
> + } else
> + min_freq = cppc_perf.lowest_freq;
>
Since cppc_perf exposes the frequency values in MHz, min_freq is in MHz.
> /**
> * Below values need to be initialized correctly, otherwise driver will fail to load
> @@ -937,12 +927,15 @@ static int amd_pstate_init_freq(struct amd_cpudata *cpudata)
> * lowest_nonlinear_freq is a value between [min_freq, nominal_freq]
> * Check _CPC in ACPI table objects if any values are incorrect
> */
> - if (min_freq <= 0 || max_freq <= 0 || nominal_freq <= 0 || min_freq > max_freq) {
> - pr_err("min_freq(%d) or max_freq(%d) or nominal_freq(%d) value is incorrect\n",
> - min_freq, max_freq, nominal_freq);
> + if (nominal_freq <= 0) {
> + pr_err("nominal_freq(%d) value is incorrect\n",
> + nominal_freq);
> return -EINVAL;
> }
>
> + lowest_nonlinear_freq = perf_to_freq(cpudata, cpudata->lowest_nonlinear_perf);
Since lowest_nonlinear_freq will be computed using cpudata->nominal_freq, the former will be in KHz.
> + WRITE_ONCE(cpudata->lowest_nonlinear_freq, lowest_nonlinear_freq);
> +
> if (lowest_nonlinear_freq <= min_freq || lowest_nonlinear_freq > nominal_freq) {
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
And thus since lowest_nonlinear_freq is in KHz and min_freq is in MHz, this check will always be true.
Shouldn't the min_freq be multiplied by 1000 ?
> pr_err("lowest_nonlinear_freq(%d) value is out of range [min_freq(%d), nominal_freq(%d)]\n",
> lowest_nonlinear_freq, min_freq, nominal_freq);
--
Thanks and Regards
gautham.
next prev parent reply other threads:[~2025-02-17 6:39 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-15 0:52 [PATCH v2 00/17] amd-pstate cleanups Mario Limonciello
2025-02-15 0:52 ` [PATCH v2 01/17] cpufreq/amd-pstate: Show a warning when a CPU fails to setup Mario Limonciello
2025-02-17 5:53 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 02/17] cpufreq/amd-pstate: Drop min and max cached frequencies Mario Limonciello
2025-02-17 6:38 ` Gautham R. Shenoy [this message]
2025-02-17 21:06 ` Mario Limonciello
2025-02-15 0:52 ` [PATCH v2 03/17] cpufreq/amd-pstate: Move perf values into a union Mario Limonciello
2025-02-17 9:01 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 04/17] cpufreq/amd-pstate: Overhaul locking Mario Limonciello
2025-02-17 9:12 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 05/17] cpufreq/amd-pstate: Drop `cppc_cap1_cached` Mario Limonciello
2025-02-17 9:12 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 06/17] cpufreq/amd-pstate-ut: Use _free macro to free put policy Mario Limonciello
2025-02-17 10:11 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 07/17] cpufreq/amd-pstate-ut: Allow lowest nonlinear and lowest to be the same Mario Limonciello
2025-02-17 10:14 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 08/17] cpufreq/amd-pstate-ut: Drop SUCCESS and FAIL enums Mario Limonciello
2025-02-17 11:34 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 09/17] cpufreq/amd-pstate-ut: Continue on missing policies Mario Limonciello
2025-02-17 11:38 ` Gautham R. Shenoy
2025-02-17 17:04 ` Mario Limonciello
2025-02-15 0:52 ` [PATCH v2 10/17] cpufreq/amd-pstate-ut: Adjust variable scope for amd_pstate_ut_check_freq() Mario Limonciello
2025-02-17 11:39 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 11/17] cpufreq/amd-pstate: Replace all AMD_CPPC_* macros with masks Mario Limonciello
2025-02-17 13:48 ` Gautham R. Shenoy
2025-02-17 17:12 ` kernel test robot
2025-02-15 0:52 ` [PATCH v2 12/17] cpufreq/amd-pstate: Cache CPPC request in shared mem case too Mario Limonciello
2025-02-17 13:54 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 13/17] cpufreq/amd-pstate: Move all EPP tracing into *_update_perf and *_set_epp functions Mario Limonciello
2025-02-17 14:14 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 14/17] cpufreq/amd-pstate: Update cppc_req_cached for shared mem EPP writes Mario Limonciello
2025-02-17 14:30 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 15/17] cpufreq/amd-pstate: Drop debug statements for policy setting Mario Limonciello
2025-02-17 14:31 ` Gautham R. Shenoy
2025-02-15 0:52 ` [PATCH v2 16/17] cpufreq/amd-pstate: Rework CPPC enabling Mario Limonciello
2025-02-15 0:52 ` [PATCH v2 17/17] cpufreq/amd-pstate: Stop caching EPP Mario Limonciello
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=Z7LZg8oq3LqfKZID@BLRRASHENOY1.amd.com \
--to=gautham.shenoy@amd.com \
--cc=Dhananjay.Ugwekar@amd.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mario.limonciello@amd.com \
--cc=perry.yuan@amd.com \
--cc=superm1@kernel.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®