[igt-dev] [PATCH i-g-t v2 6/9] tests/kms_color: Add and additional "signed" subtest

[Shankar, Uma]

> -----Original Message-----
> From: igt-dev <igt-dev-bounces at lists.freedesktop.org> On Behalf Of Ville Syrjala
> Sent: Tuesday, April 11, 2023 9:46 PM
> To: igt-dev at lists.freedesktop.org
> Subject: [igt-dev] [PATCH i-g-t v2 6/9] tests/kms_color: Add and additional "signed"
> subtest

Nit: s/and/an

> From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 
> i915 CHV CSC code has a bug where it treats the hardware CSC coefficient as sign-
> magnitude instead of the two's complement value that it is. None of the current
> tests manage to notice this.
> 
> Add a testcase that uses suitable negative coefficients to catch the issue. With CHV's
> s4.12 coefficients the
> -0.25 ends up as -7.75 and the results won't match.

Looks Good to me.
Reviewed-by: Uma Shankar <uma.shankar at intel.com>

> Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> ---
>  tests/kms_color.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/tests/kms_color.c b/tests/kms_color.c index
> d430ee12aaa3..35083122dfbf 100644
> --- a/tests/kms_color.c
> +++ b/tests/kms_color.c
> @@ -959,6 +959,11 @@ run_tests_for_pipe(data_t *data)
>  		{ 0.0, 1.0, 0.0 },
>  		{ 0.0, 0.0, 1.0 },
>  	};
> +	static const color_t colors_cmy[] = {
> +		{ 0.0, 1.0, 1.0 },
> +		{ 1.0, 0.0, 1.0 },
> +		{ 1.0, 1.0, 0.0 }
> +	};
>  	static const struct {
>  		const char *name;
>  		int iter;
> @@ -1040,6 +1045,16 @@ run_tests_for_pipe(data_t *data)
>  		  },
>  		  .desc = "Check the color transformation for 0.75 transparency",
>  		},
> +		{ .name = "ctm-signed",
> +		  .fb_colors = colors_cmy,
> +		  .iter = 3,
> +		  .ctm = {
> +			  -0.25,  0.75,  0.75,
> +			   0.75, -0.25,  0.75,
> +			   0.75,  0.75, -0.25,
> +		  },
> +		  .desc = "Check the color transformation for correct signed
> handling",
> +		},
>  	};
>  	int i;
> 
> --
> 2.39.2