Re: [PATCH 1/3] cpufreq: Add a callback to update the min_freq_req from drivers

[Dhananjay Ugwekar <Dhananjay.Ugwekar@amd.com>]

Hello Rafael,

On 10/7/2024 9:18 PM, Rafael J. Wysocki wrote:
> On Mon, Oct 7, 2024 at 5:46 PM Rafael J. Wysocki <rafael@kernel.org> wrote:
>>
>> Hi,
>>
>> On Mon, Oct 7, 2024 at 6:40 AM Dhananjay Ugwekar
>> <Dhananjay.Ugwekar@amd.com> wrote:
>>>
>>> Hello Rafael,
>>>
>>> On 10/4/2024 11:47 PM, Rafael J. Wysocki wrote:
>>>> On Thu, Oct 3, 2024 at 10:44 AM Dhananjay Ugwekar
>>>> <Dhananjay.Ugwekar@amd.com> wrote:
>>>>>
>>>>> Currently, there is no proper way to update the initial lower frequency
>>>>> limit from cpufreq drivers.
>>>>
>>>> Why do you want to do it?
>>>
>>> We want to set the initial lower frequency limit at a more efficient level
>>> (lowest_nonlinear_freq) than the lowest frequency, which helps save power in
>>> some idle scenarios, and also improves benchmark results in some scenarios.
>>> At the same time, we want to allow the user to set the lower limit back to
>>> the inefficient lowest frequency.
>>
>> So you want the default value of scaling_min_freq to be greater than
>> the total floor.

Yes, we want to set the default min value to what we think is best for the platform.

>>
>> I have to say that I'm not particularly fond of this approach because
>> it is adding a new meaning to scaling_min_freq: Setting it below the
>> default would not cause the driver to use inefficient frequencies
> 
> s/not/now/ (sorry)

I believe we are not changing the meaning of the scaling_min_freq just setting it 
to the best value at boot and then allowing the user to have access to the entire 
frequency range for the platform. Also, we have cpuinfo_min_freq/max_freq to 
indicate to the user as to what the entire frequency range is for the platform 
(depending on boost enabled/disabled).

> 
> I should have double checked this before sending.
> 
>> which user space may not be aware of.

I guess, this part we can fix by documenting correctly ?

>> Moreover, it would tell the
>> driver how far it could go with that.

Sorry, I didnt understand this part.

>>
>> IMV it would be bettwr to have a separate interface for this kind of tuning.

I feel like we can incorporate this change cleanly enough into scaling_min_freq, 
without adding a new interface which might further confuse the user. But please 
let me know your concerns and thoughts.

Thanks,
Dhananjay