[Intel-gfx] [PATCH] drm/i915: Extend i915_powersave parameter.
Rodrigo Vivi
rodrigo.vivi at gmail.com
Thu Jul 18 18:38:41 CEST 2013
On Wed, Jul 17, 2013 at 6:30 PM, Daniel Vetter <daniel at ffwll.ch> wrote:
> On Wed, Jul 17, 2013 at 11:23 PM, Rodrigo Vivi <rodrigo.vivi at gmail.com> wrote:
>>>> My opinion is that we respect the specific module parameters, and if
>>>> they are left to default values, then apply the global powersave
>>>> parameter. If that too is default, then we apply the module default.
>>>
>>> Jumping in a bit late, but: I've honestly never understood why we have
>>> two levels of module options. Imo having individual knobs for each
>>> delicate feature makes more sense, strange dependencies in module
>>> option will only confuse dim-witted developers like me when looking at
>>> a bug ;-)
>>>
>>> So could we just reduce powersave to the few things that we haven't
>>> touched yet (iirc only DRRS)?
>>
>> That is fine for me too... either add all features under this umbrella
>> or make it be only one feature like drrs that doesn't have its own
>> parameter...
>> the only cons I see in this case is the name of parameter that is too
>> generic...
>
> We're allowed to kill module options now, so if we fix up drrs and
> make powersave completely useless, we can just remove.
I'm in favor of remove powersave parameter completely. Just let me
know what are the issues with drrs and how I can help to fix them and
remove this parameter completely.
>
>> But honestly I don't have a stronger position I just wanted to start
>> the discussion because I don't like the way it is today... So it is up
>> to you... I can either send v2 or a new simple patch that removes fbc
>> from this i915_powersave. Just let me know what is better...
>
> I vote for moving fbc out of powersave as a separate option. Worst
> case we need to educate a users and tell him that frobbing around with
> random module options isn't a good idea ;-)
Agree
> -Daniel
> --
> Daniel Vetter
> Software Engineer, Intel Corporation
> +41 (0) 79 365 57 48 - http://blog.ffwll.ch
--
Rodrigo Vivi
Blog: http://blog.vivi.eng.br
More information about the Intel-gfx
mailing list