[igt-dev] [PATCH i-g-t] tests/i915/gem_tiled_wc: skip on platforms without GGTT

[Chris Wilson]

Quoting Zbigniew Kempczyński (2020-01-28 09:22:04)
> On Tue, Jan 28, 2020 at 09:11:50AM +0000, Chris Wilson wrote:
> > Quoting Zbigniew Kempczyński (2020-01-28 09:00:22)
> > > On platforms without aperture ensure test is skipped. Localize
> > > GEM_MMAP ioctl() to make sure this interface is tested.
> > 
> > Actually this would be ok with device coherent. Not sure if it's
> > relevant... You could certainly write the equivalent for mmap_offset,
> > that is the GTT path takes the fence, and the others do not.
> > 
> > > Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski at intel.com>
> > > Cc: Chris Wilson <chris at chris-wilson.co.uk>
> > Reviewed-by: Chris Wilson <chris at chris-wilson.co.uk>
> > 
> > I would move this into gem_mmap_wc just to make it more self-contained,
> > igt_subtest("tiled").
> 
> You think just to remove this test after migrating to gem_mmap_wc? 

I think this would make a valid subtest for gem_mmap_wc.c (replacing this
.c) and also applies to gem_mmap_offset.c

> Only cons I see now is that we don't support offset in gem_mmap_offset__wc()
> so I would defer it when this argument will be handled.

That's not a huge deal wrt to the test, all it is doing is verifying
that even though the object has a tiling, the WC/WB/UC views are untiled
whereas GTT is.
-Chris