Re: [PATCH v4 3/3] drm/format-helper: Add KUnit tests for drm_fb_xrgb8888_to_gray8()

From: Maxime Ripard
Date: Mon Sep 19 2022 - 04:25:33 EST


On Mon, Sep 19, 2022 at 10:18:01AM +0200, José Expósito wrote:
> Hi Maxime,
>
> Thanks for looking into the patches.
>
> On Mon, Sep 19, 2022 at 09:36:45AM +0200, Maxime Ripard wrote:
> > Hi,
> >
> > On Mon, Sep 19, 2022 at 09:15:31AM +0200, José Expósito wrote:
> > > Extend the existing test cases to test the conversion from XRGB8888 to
> > > grayscale.
> > >
> > > Tested-by: Maíra Canal <mairacanal@xxxxxxxxxx>
> > > Reviewed-by: David Gow <davidgow@xxxxxxxxxx>
> > > Signed-off-by: José Expósito <jose.exposito89@xxxxxxxxx>
> > > ---
> > > .../gpu/drm/tests/drm_format_helper_test.c | 62 +++++++++++++++++++
> > > 1 file changed, 62 insertions(+)
> > >
> > > diff --git a/drivers/gpu/drm/tests/drm_format_helper_test.c b/drivers/gpu/drm/tests/drm_format_helper_test.c
> > > index 09d358b54da0..71722e828abe 100644
> > > --- a/drivers/gpu/drm/tests/drm_format_helper_test.c
> > > +++ b/drivers/gpu/drm/tests/drm_format_helper_test.c
> > > @@ -37,6 +37,11 @@ struct convert_to_xrgb2101010_result {
> > > const u32 expected[TEST_BUF_SIZE];
> > > };
> > >
> > > +struct convert_to_gray8_result {
> > > + unsigned int dst_pitch;
> > > + const u8 expected[TEST_BUF_SIZE];
> > > +};
> > > +
> > >
> > > [...]
> > >
> > > static struct kunit_case drm_format_helper_test_cases[] = {
> > > KUNIT_CASE_PARAM(drm_test_fb_xrgb8888_to_rgb332, convert_xrgb8888_gen_params),
> > > KUNIT_CASE_PARAM(drm_test_fb_xrgb8888_to_rgb565, convert_xrgb8888_gen_params),
> > > KUNIT_CASE_PARAM(drm_test_fb_xrgb8888_to_rgb888_test, convert_xrgb8888_gen_params),
> > > KUNIT_CASE_PARAM(drm_test_fb_xrgb8888_to_xrgb2101010_test, convert_xrgb8888_gen_params),
> > > + KUNIT_CASE_PARAM(drm_test_fb_xrgb8888_to_gray8_test, convert_xrgb8888_gen_params),
> >
> > The trailing test feels redundant,
>
> Why do you feel like it is redundant? Under the hood, most conversion
> functions reuse the same code, but the *_line() function is unique and
> I think that it is worth testing it.

Sure, I wasn't commenting on the code itself, but the trailing "test" in
the test name: drm_test_fb_xrgb8888_to_gray8_test
^

Which is redundant since we already have "test" in the prefix.

Maxime