[Intel-gfx] [PATCH] drm/i915: Make 48bit full ppgtt configuration generic (v4)

[Ville Syrjälä]

On Thu, Sep 13, 2018 at 10:12:06AM -0700, Bob Paauwe wrote:
> On Thu, 13 Sep 2018 20:05:54 +0300
> Ville Syrjälä <ville.syrjala at linux.intel.com> wrote:
> 
> > On Thu, Sep 13, 2018 at 10:02:57AM -0700, Bob Paauwe wrote:
> > > On Wed, 12 Sep 2018 17:10:58 +0100
> > > Chris Wilson <chris at chris-wilson.co.uk> wrote:
> > >   
> > > > Quoting Bob Paauwe (2018-09-12 17:04:30)  
> > > > > diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
> > > > > index 43ed8b28aeaa..33d7225edbbb 100644
> > > > > --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c
> > > > > +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
> > > > > @@ -181,6 +181,8 @@ struct drm_i915_private *mock_gem_device(void)
> > > > >                 I915_GTT_PAGE_SIZE_64K |
> > > > >                 I915_GTT_PAGE_SIZE_2M;
> > > > >  
> > > > > +       mkwrite_device_info(i915)->full_ppgtt_bits = 48;    
> > > > 
> > > > Actually the mock ppgtt is 64b.
> > > > -Chris  
> > > 
> > > Setting it 64 bit causes mock_hugepages to fail.  
> > 
> > 1<<64 somewhere?
> > 
> 
> Doh, yeah, there is that. So what does 64b mean for ppgtt->vm.total?
> Should it really be 63b?

GENMASK() & ~(GTT_PAGE_SIZE-1) maybe?

-- 
Ville Syrjälä
Intel