Re: [PATCH v2 2/4] cpufreq: amd-pstate: Set a fallback policy based on preferred_profile

From: "Limonciello, Mario" <mario.limonciello@amd.com>
To: Huang Rui <ray.huang@amd.com>
Cc: "Yuan, Perry" <Perry.Yuan@amd.com>,
	"rafael.j.wysocki@intel.com" <rafael.j.wysocki@intel.com>,
	"viresh.kumar@linaro.org" <viresh.kumar@linaro.org>,
	"Sharma, Deepak" <Deepak.Sharma@amd.com>,
	"Karny, Wyes" <Wyes.Karny@amd.com>,
	"Shenoy, Gautham Ranjal" <gautham.shenoy@amd.com>,
	"Li, Sun peng (Leo)" <Sunpeng.Li@amd.com>,
	"Huang, Shimmer" <Shimmer.Huang@amd.com>,
	"Du, Xiaojian" <Xiaojian.Du@amd.com>,
	"Meng, Li (Jassmine)" <Li.Meng@amd.com>,
	"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 2/4] cpufreq: amd-pstate: Set a fallback policy based on preferred_profile
Date: Tue, 20 Jun 2023 10:18:30 -0500	[thread overview]
Message-ID: <9e3577fa-2bd2-da2a-bc35-5e70f49e011d@amd.com> (raw)
In-Reply-To: <ZJHAatXe+iVzM/JS@amd.com>

On 6/20/2023 10:06 AM, Huang Rui wrote:
> On Tue, Jun 20, 2023 at 11:02:00PM +0800, Limonciello, Mario wrote:
>> On 6/20/2023 9:58 AM, Huang Rui wrote:
>>> On Thu, Jun 15, 2023 at 02:32:25PM +0800, Yuan, Perry wrote:
>>>> From: Mario Limonciello <mario.limonciello@amd.com>
>>>>
>>>> If a user's configuration doesn't explicitly specify the cpufreq
>>>> scaling governor then the code currently explicitly falls back to
>>>> 'powersave'. This default is fine for notebooks and desktops, but
>>> May I know if the processor is powerful desktop such as threadripper,
>>> whether it will be default to 'performance' or 'powersave'?
>> It's currently defaulting to 'powersave' for desktops and
>> workstations.
>>
>> Do you think we should adopt performance for these?
> Yes, I didn't see any different use cases here between server and
> threadripper here. Or I missed anything?
Workstations and Desktops usually have to go through energy
consumption certifications.  Couldn't setting it to performance be
inappropriate for those?
> Do we have a way to separate them?

If Threadripper identified as

3 Workstation

I'd agree; but I'd think we're going to lump AM4/AM5 desktops
along with Threadripper.  So should we still set all those to performance?

>
> Thanks,
> Ray
>
>>> Thanks,
>>> Ray
>>>
>>>> servers and undefined machines should default to 'performance'.
>>>>
>>>> Look at the 'preferred_profile' field from the FADT to set this
>>>> policy accordingly.
>>>>
>>>> Link: https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/05_ACPI_Software_Programming_Model/ACPI_Software_Programming_Model.html#fixed-acpi-description-table-fadt
>>>> Suggested-by: Wyes Karny <Wyes.Karny@amd.com>
>>>> Reviewed-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
>>>> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
>>>> ---
>>>>    drivers/cpufreq/amd-pstate.c | 7 +++++--
>>>>    1 file changed, 5 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
>>>> index ddd346a239e0..c9d296ebf81e 100644
>>>> --- a/drivers/cpufreq/amd-pstate.c
>>>> +++ b/drivers/cpufreq/amd-pstate.c
>>>> @@ -1102,10 +1102,13 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy)
>>>>    	policy->max = policy->cpuinfo.max_freq;
>>>>    
>>>>    	/*
>>>> -	 * Set the policy to powersave to provide a valid fallback value in case
>>>> +	 * Set the policy to provide a valid fallback value in case
>>>>    	 * the default cpufreq governor is neither powersave nor performance.
>>>>    	 */
>>>> -	policy->policy = CPUFREQ_POLICY_POWERSAVE;
>>>> +	if (acpi_pm_profile_server() || acpi_pm_profile_undefined())
>>>> +		policy->policy = CPUFREQ_POLICY_PERFORMANCE;
>>>> +	else
>>>> +		policy->policy = CPUFREQ_POLICY_POWERSAVE;
>>>>    
>>>>    	if (boot_cpu_has(X86_FEATURE_CPPC)) {
>>>>    		ret = rdmsrl_on_cpu(cpudata->cpu, MSR_AMD_CPPC_REQ, &value);
>>>> -- 
>>>> 2.34.1
>>>>