[igt-dev] [PATCH i-g-t 2/2] lib/igt_kms: reset plane alpha and blend in igt_plane_reset()

[Pandiyan, Dhinakaran]

On Fri, 2019-03-15 at 17:42 +0200, Juha-Pekka Heikkila wrote:
> plane alpha and blend would affect other tests if left
> in unexpected states
> 
> Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila at gmail.com>

This explains the kms_ccs failures that follow kms_plane_alpha_blend
(pipe-*-coverage-7efc)

I wrote a similar patch and also included a reset of the alpha props.
in the test itself.

diff --git a/tests/kms_plane_alpha_blend.c
b/tests/kms_plane_alpha_blend.c
index d7dc44ab..48717c51 100644
--- a/tests/kms_plane_alpha_blend.c
+++ b/tests/kms_plane_alpha_blend.c
@@ -506,6 +506,7 @@ static void run_test_on_pipe_planes(data_t *data,
enum pipe pipe, bool blend,
                test(data, pipe, plane);
                igt_plane_set_fb(plane, NULL);
        }
+       reset_alpha(display, pipe);


Your fix looks good as all tests are expected (?) to call
igt_display_require()


Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan at intel.com>
> ---
>  lib/igt_kms.c | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/lib/igt_kms.c b/lib/igt_kms.c
> index e1eacc1..cdb5aba 100644
> --- a/lib/igt_kms.c
> +++ b/lib/igt_kms.c
> @@ -1781,6 +1781,13 @@ static void igt_plane_reset(igt_plane_t
> *plane)
>  	if (igt_plane_has_prop(plane, IGT_PLANE_ROTATION))
>  		igt_plane_set_prop_value(plane, IGT_PLANE_ROTATION,
> IGT_ROTATION_0);
>  
> +	if (igt_plane_has_prop(plane, IGT_PLANE_PIXEL_BLEND_MODE))
> +		igt_plane_set_prop_enum(plane,
> IGT_PLANE_PIXEL_BLEND_MODE, "None");
> +
> +	if (igt_plane_has_prop(plane, IGT_PLANE_ALPHA))
> +		igt_plane_set_prop_value(plane, IGT_PLANE_ALPHA,
> 0xffff);
> +
> +
>  	igt_plane_clear_prop_changed(plane, IGT_PLANE_IN_FENCE_FD);
>  	plane->values[IGT_PLANE_IN_FENCE_FD] = ~0ULL;
>  	plane->gem_handle = 0;