Re: RFC: offering a standardized (/sys/class) userspace API for selecting system/laptop performance-profiles

[Benjamin Berg <bberg@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1154 bytes --]

On Thu, 2020-09-17 at 15:24 +0200, Hans de Goede wrote:
> > Excuse my ignorance, but why does "lap_mode" need to be here?
> > I understand the implications of it regarding performance, but
> > I think it would be more sense to export it via the hwmon (or
> > something similar) subsystem? What am I missing?
> 
> Well hwmon has very clearly defined sensor types, like voltage,
> fan-speed and temperature. lap_mode does not match any of them.

Not sure. I think you can consider it a simple sensor. In this case it
just affects thermal limits that are imposed on the system (and some
other stuff like RF for modems).

It does have a major impact on something like a "performance" mode
because of those imposed limits and causes these modes to be degraded.
But I think this interface can just show the configured state, even if
the component knows that the performance state is severely limited due
to other factors.

Benjamin

> Also registering another-type class device just for the lap_mode
> boolean seems overkill, esp. since lap_mode is inherently coupled
> to the performance-profile stuff.
> 
> Regards,
> 
> Hans
> 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]