Re: [PATCH i-g-t v3 1/1] tests/kms_color: Add 10-bit color depth support to gamma test for MediaTek

["Jason-JH Lin (林睿祥)" <Jason-JH.Lin@mediatek.com>]

On Mon, 2026-04-20 at 21:43 +0800, Jay Liu wrote:
> 1. Add 10-bit color depth flow to gamma test for MediaTek devices,
> which only support bit-true
> results with 10-bit data. The test now selects 8-bit or 10-bit color
> depth and framebuffer
> format based on device type.
> 
> 2. To satisfy the legacy gamma test requirements, the MTK Gamma LUT
> limit should be set to
> 4092, in the current 16-bit LUT design, setting the entry to 0xffd0
> ensures the driver maps
> it correctly to 4092.
> 
> Signed-off-by: Jay Liu <jay.liu@mediatek.com>
> ---
>  tests/kms_color.c | 33 ++++++++++++++++++++++++++-------
>  1 file changed, 26 insertions(+), 7 deletions(-)
> 
> diff --git a/tests/kms_color.c b/tests/kms_color.c
> index 565edb2dc..35b0106b7 100644
> --- a/tests/kms_color.c
> +++ b/tests/kms_color.c
> @@ -288,7 +288,7 @@ static bool test_pipe_legacy_gamma(data_t *data,
>  	fb_id = igt_create_fb(data->drm_fd,
>  			      mode->hdisplay,
>  			      mode->vdisplay,
> -			      DRM_FORMAT_XRGB8888,
> +			      data->drm_format,
>  			      DRM_FORMAT_MOD_LINEAR,
>  			      &fb);
>  	igt_assert(fb_id);
> @@ -296,7 +296,7 @@ static bool test_pipe_legacy_gamma(data_t *data,
>  	fb_modeset_id = igt_create_fb(data->drm_fd,
>  				      mode->hdisplay,
>  				      mode->vdisplay,
> -				      DRM_FORMAT_XRGB8888,
> +				      data->drm_format,
>  				      DRM_FORMAT_MOD_LINEAR,
>  				      &fb_modeset);
>  	igt_assert(fb_modeset_id);
> @@ -322,8 +322,19 @@ static bool test_pipe_legacy_gamma(data_t *data,
>  	igt_plane_set_fb(primary, &fb);
>  
>  	red_lut[0] = green_lut[0] = blue_lut[0] = 0;
> -	for (i = 1; i < legacy_lut_size; i++)
> -		red_lut[i] = green_lut[i] = blue_lut[i] = 0xffff;
> +	for (i = 1; i < legacy_lut_size; i++) {
> +		/*
> +		 * The Medaitek Gamma LUT maps 10bit input data to
> 12bit output.
> +		 * In this test case, the maximum input value is
> 1023.
> +		 * When the Gamma function is disabled, the data
> defaults to a 4092 output.
> +		 * To achieve bit true results, the Gamma LUT
> maximum must be set to 4092.
> +		 * Note that 0xFFFD0 represents 16-bit,which maps to
> 4092 in 12-bit depth.
> +		 */
> +		if (is_mtk_device(data->drm_fd))
> +			red_lut[i] = green_lut[i] = blue_lut[i] =
> 0xffd0;
> +		else
> +			red_lut[i] = green_lut[i] = blue_lut[i] =
> 0xffff;
> +	}
>  	igt_assert_eq(drmModeCrtcSetGamma(data->drm_fd, primary-
> >crtc->crtc_id,
>  					  legacy_lut_size, red_lut,
> green_lut, blue_lut), 0);
>  	igt_display_commit(&data->display);
> @@ -752,14 +763,22 @@ static void
>  run_gamma_degamma_tests_for_crtc(data_t *data, igt_crtc_t *crtc,
>  				 bool (*test_t)(data_t*,
> igt_plane_t*))
>  {
> +	bool is_mtk = false;
> +
>  	test_setup(data, crtc);
>  
> +	/* MediaTek requires 10-bit pipeline for accurate (bit true)
> color processing */
> +	if (is_mtk_device(data->drm_fd))
> +		is_mtk = true;
> +

I think using `10bits_mode` or `10bits_depth` is more flexible than
`is_mtk`.

Because other vendors can reuse this code like this:
if (is_mtk_device(data->drm_fd) || is_XXX_device(data->drm_fd))
	10bits_mode = true;

Regards,
Jasdon-JH Lin

>  	/*
> -	 * We assume an 8bits depth per color for degamma/gamma LUTs
> +	 * We assume an 8bits or 10bits depth per color for
> degamma/gamma LUTs
>  	 * for CRC checks with framebuffer references.
>  	 */
> -	data->color_depth = 8;
> -	data->drm_format = DRM_FORMAT_XRGB8888;
> +
> +	/* MediaTek requires 10-bit pipeline for accurate (bit true)
> color processing */
> +	data->color_depth = is_mtk ? 10 : 8;
> +	data->drm_format = is_mtk ? DRM_FORMAT_XRGB2101010 :
> DRM_FORMAT_XRGB8888;
>  	data->mode = igt_output_get_mode(data->output);
>  
>  	igt_require(crtc_output_combo_valid(data, crtc));