On Thu, Feb 9, 2012 at 14:39, Chris Wilson <ch...@chris-wilson.co.uk> wrote:
> On Thu, 9 Feb 2012 17:15:50 +0100, Daniel Vetter <daniel.vet...@ffwll.ch>
> wrote:
> > v3: Pimp commit message and make Chris Wilson less grumpy by adding a
> > module option.
>
> > +void i915_gem_init_ppgtt(struct drm_device *dev)
> > +{
> > + drm_i915_private_t *dev_priv = dev->dev_private;
> > + uint32_t pd_offset;
> > + struct intel_ring_buffer *ring;
> > + int i;
> > +
> > + if (i915_enable_ppgtt && !HAS_ALIASING_PPGTT(dev))
> > + return;
> > +
>
> This is quite a late check since we have already partitioned the
> aperture and run i915_gem_init_aliasing_ppgtt. I think the enable check
> should be done earlier and this check reduced to
> if (!dev_priv->mm.aliasing_ppgtt)
> return;
>
I think we need to keep the enable check here, because this can be called
from within i915_reset as well.
To fix this and prevent clash between i915_gem_init_ppgtt and
i915_gem_init_ppgtt, I guess you could add the 'i915_enable_ppgtt && '
check to the i915_gem_init_ppgtt call in Patch3 as well.
--
Eugeni Dodonov
<http://eugeni.dodonov.net/>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
