[igt-dev] [PATCH v2 4/8] tests/kms_draw_crc: Skip GTT subtests on platforms w/o aperture

[Dixit, Ashutosh]

On Tue, 11 Feb 2020 16:17:28 -0800, Dixit, Ashutosh wrote:
>
> On Mon, 10 Feb 2020 18:31:04 -0800, Imre Deak wrote:
> >
> > Subtests drawing through a GTT mapping are not relevant on platforms w/o
> > a GTT aperture, so skip them.
> >
> > Cc: Matt Roper <matthew.d.roper at intel.com>
> > Signed-off-by: Imre Deak <imre.deak at intel.com>
> > Reviewed-by: Matt Roper <matthew.d.roper at intel.com>
> > ---
> >  tests/kms_draw_crc.c | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/tests/kms_draw_crc.c b/tests/kms_draw_crc.c
> > index ea14db9a..6de9feae 100644
> > --- a/tests/kms_draw_crc.c
> > +++ b/tests/kms_draw_crc.c
> > @@ -178,6 +178,8 @@ static void draw_method_subtest(enum igt_draw_method method,
> >	igt_crc_t crc;
> >
> >	igt_skip_on(method == IGT_DRAW_MMAP_WC && !gem_mmap__has_wc(drm_fd));
> > +	igt_skip_on(method == IGT_DRAW_MMAP_GTT &&
> > +		    !gem_has_mappable_ggtt(drm_fd));
>
> Can't we add the skip directly in draw_rect_mmap_gtt() so we don't have to
> go skip in each individual test which ends up calling draw_rect_mmap_gtt()?

Sorry, so yes we cannot put a igt_skip() in draw_rect_mmap_gtt() in case
say we are looping over all the methods in a subtest and igt_skip() will
end up skipping the entire subtest, i.e. all the other methods too. At most
we can return without doing anything from draw_rect_mmap_gtt() but we can't
igt_skip().

I am assuming that the tests are passing in spite of not being able to set
the tiling in igt_create_fb()?