[Intel-gfx] [PATCH 01/10] drm/i915: Remove "VLV magic" from irq setup

[Ville Syrjälä]

On Mon, Apr 11, 2016 at 06:20:04PM +0300, Imre Deak wrote:
> On ma, 2016-04-11 at 16:56 +0300, ville.syrjala at linux.intel.com wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > 
> > No clue what this is supposed to achieve. I think it's been there
> > since
> > the very beginning, so presumably some kind of kludge for very early
> > silicon. Let's just throw it out.
> > 
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > ---
> >  drivers/gpu/drm/i915/i915_irq.c | 6 ------
> >  1 file changed, 6 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_irq.c
> > b/drivers/gpu/drm/i915/i915_irq.c
> > index 679f08c944ef..1d21ebfffd4d 100644
> > --- a/drivers/gpu/drm/i915/i915_irq.c
> > +++ b/drivers/gpu/drm/i915/i915_irq.c
> > @@ -3319,12 +3319,6 @@ static void valleyview_irq_preinstall(struct
> > drm_device *dev)
> >  {
> >  	struct drm_i915_private *dev_priv = dev->dev_private;
> >  
> > -	/* VLV magic */
> > -	I915_WRITE(VLV_IMR, 0);
> > -	I915_WRITE(RING_IMR(RENDER_RING_BASE), 0);
> > -	I915_WRITE(RING_IMR(GEN6_BSD_RING_BASE), 0);
> > -	I915_WRITE(RING_IMR(BLT_RING_BASE), 0);
> > -
> 
> AFAICS this would also leave the ring interrupts unmasked before we
> called ring->get_irq(), so at least the change is a fix for that.

On a related note, in ringbuffer mode we don't actually make sure they're
masked. The execlist code has explicit ring IMR initialization.

> I
> haven't found any explanation for the above either, so I guess the best
> we can do at this point is to see if things continue to work without
> it:
> 
> Reviewed-by: Imre Deak <imre.deak at intel.com>
> 
> >  	gen5_gt_irq_reset(dev);
> >  
> >  	I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK);

-- 
Ville Syrjälä
Intel OTC