[igt-dev] [PATCH i-g-t] tests/kms_rotation_crc: MPO subtest- mixing pixel formats and tiling on different planes.

[Juha-Pekka Heikkila]

On 8.11.2021 7.28, Ananya Sharma wrote:
>   Added P010 pixel format with already existing pixel formats like RGB565, XRGB8888, NV12.
> 
> Signed-off-by: Ananya Sharma <ananya.sharma at intel.com>
> ---
>   tests/kms_rotation_crc.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/kms_rotation_crc.c b/tests/kms_rotation_crc.c
> index 1497120c..9c208315 100644
> --- a/tests/kms_rotation_crc.c
> +++ b/tests/kms_rotation_crc.c
> @@ -666,7 +666,7 @@ static void test_multi_plane_rotation(data_t *data, enum pipe pipe)
>   	 * case with tiling are 2 bpp, 4 bpp and NV12.
>   	 */
>   	static const uint32_t formatlist[] = {DRM_FORMAT_RGB565,
> -		DRM_FORMAT_XRGB8888, DRM_FORMAT_NV12};
> +		DRM_FORMAT_XRGB8888, DRM_FORMAT_NV12, DRM_FORMAT_P010};

I suspect for this to work you will also need to change in 
test_multi_plane_rotation(..) in inner loops anywhere comparison for 
DRM_FORMAT_NV12 take into account P0xx format in some way with 
igt_format_is_yuv_semiplanar(..). Some of those inner loop rules are 
excluding, some including.

I think otherwise those runtime cached crcs will cause mismatches

/Juha-Pekka

>   
>   	static struct {
>   		igt_rotation_t rotation;
>