[igt-dev] [PATCH i-g-t 1/1] test/kms_plane_scaling: Don't run scaler clip/clamp with all formats

Juha-Pekka Heikkila juhapekka.heikkila at gmail.com
Wed Mar 24 10:09:42 UTC 2021


Results look all green and there is as expected less time spent on pipes 
b and c.

Reviewed-by: Juha-Pekka Heikkila <juhapekka.heikkila at gmail.com>

/Juha-Pekka


On 23.3.2021 14.29, Mohammed Khajapasha wrote:
> Don't run plane scaler tests with all formats for pipe C & D.
> We probably don't need to test plane scaler tests with each format
> for pipe C & D.
> 
> $ time ./build/tests/kms_plane_scaling
> - real	6m53,143s
> + real	4m39,406s
> 
> Cc: juhapekka.heikkila at gmail.com
> Cc: Mika Kahola <mika.kahola at intel.com>
> Signed-off-by: Mohammed Khajapasha <mohammed.khajapasha at intel.com>
> ---
>   tests/kms_plane_scaling.c | 15 +++++++++++++++
>   1 file changed, 15 insertions(+)
> 
> diff --git a/tests/kms_plane_scaling.c b/tests/kms_plane_scaling.c
> index c63eb428..31f14275 100644
> --- a/tests/kms_plane_scaling.c
> +++ b/tests/kms_plane_scaling.c
> @@ -230,6 +230,18 @@ static bool test_format(data_t *data,
>   	return true;
>   }
>   
> +static bool test_pipe_iteration(data_t *data, enum pipe pipe, int iteration)
> +{
> +	if (!is_i915_device(data->drm_fd) ||
> +	    data->extended)
> +		return true;
> +
> +	if ((pipe > PIPE_B) && (iteration >= 2))
> +		return false;
> +
> +	return true;
> +}
> +
>   static void test_scaler_with_rotation_pipe(data_t *d, enum pipe pipe,
>   					   igt_output_t *output)
>   {
> @@ -532,6 +544,9 @@ test_scaler_with_clipping_clamping_scenario(data_t *d, enum pipe pipe, igt_outpu
>   		unsigned f1 = d->plane1->drm_plane->formats[i];
>   		struct igt_vec tested_formats2;
>   
> +		if (!test_pipe_iteration(d, pipe, i))
> +			continue;
> +
>   		if (!test_format(d, &tested_formats1, f1) ||
>   		    !can_scale(d, f1))
>   			continue;
> 



More information about the igt-dev mailing list