[Intel-gfx] [PATCH 3/4] drm/i915: Do not acquire crtc state to check clock during modeset, v3.

[Mika Kahola]

On Tue, 2015-12-15 at 10:26 +0000, Chris Wilson wrote:
> On Tue, Dec 15, 2015 at 12:22:40PM +0200, Mika Kahola wrote:
> > On Tue, 2015-11-24 at 11:29 +0100, Maarten Lankhorst wrote:
> > > diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> > > index 3c46037b6e55..178a042f917e 100644
> > > --- a/drivers/gpu/drm/i915/intel_display.c
> > > +++ b/drivers/gpu/drm/i915/intel_display.c
> > > @@ -5991,22 +5991,31 @@ static int broxton_calc_cdclk(struct drm_i915_private *dev_priv,
> > >  static int intel_mode_max_pixclk(struct drm_device *dev,
> > >  				 struct drm_atomic_state *state)
> > >  {
> > > -	struct intel_crtc *intel_crtc;
> > > -	struct intel_crtc_state *crtc_state;
> > > -	int max_pixclk = 0;
> > > +	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
> > > +	struct drm_i915_private *dev_priv = dev->dev_private;
> > This is a nitpick but we should use nowadays to_i915() 
> 
> If you're going to bring that up, we should be passing in the right
> pointer to begin with!
> -Chris
This was more meant be as a notification. I was advised that we should
use 'to_i915()' in a new code. Personally, I don't have any preference one
way or another. Both ways will work. Maybe this would be a topic for a follow
up patch?

Cheers,
Mika