[igt-dev] [PATCH i-g-t V2] tests/kms_plane_multiple: Start continuous CRC capture after commit

[Rodrigo Siqueira]

On 06/28, Anson Jacob wrote:
> From: Victor Lu <victorchengchi.lu at amd.com>
> 
> atomic-pipe-a-tiling-none fails on amdgpu(Raven) because it fails to read the
> CRC when there is no CRC generated (zeroed framebuffer). Move
> igt_pipe_crc_start to be called after the first nonzero FB commit.
> 
> Signed-off-by: Victor Lu <victorchengchi.lu at amd.com>
> Acked-by: Anson Jacob <Anson.Jacob at amd.com>
> Cc: Petri Latvala <petri.latvala at intel.com>
> Cc: Rodrigo Siqueira <Rodrigo.Siqueira at amd.com>
> Cc: Harry Wentland <harry.wentland at amd.com>
> Cc: Mark Yacoub <markyacoub at chromium.org>
> ---
>  tests/kms_plane_multiple.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/kms_plane_multiple.c b/tests/kms_plane_multiple.c
> index 93c6ea7d269d..4424b864b2d6 100644
> --- a/tests/kms_plane_multiple.c
> +++ b/tests/kms_plane_multiple.c
> @@ -288,6 +288,7 @@ test_plane_position_with_output(data_t *data, enum pipe pipe,
>  	igt_plane_t *plane;
>  	int i;
>  	int err, c = 0;
> +	int crc_enabled = 0;
>  	int iterations = opt.iterations < 1 ? 1 : opt.iterations;
>  	bool loop_forever;
>  	char info[256];
> @@ -327,14 +328,16 @@ test_plane_position_with_output(data_t *data, enum pipe pipe,
>  		 igt_output_name(output), kmstest_pipe_name(pipe), c,
>  		 info, opt.seed);
>  
> -	igt_pipe_crc_start(data->pipe_crc);
> -
>  	i = 0;
>  	while (i < iterations || loop_forever) {
>  		/* randomize planes and set up the holes */
>  		prepare_planes(data, pipe, &blue, tiling, c, output);
>  
>  		igt_display_commit2(&data->display, COMMIT_ATOMIC);
> +		if (!crc_enabled) {
> +			igt_pipe_crc_start(data->pipe_crc);
> +			crc_enabled = 1;
> +		}
>  
>  		igt_pipe_crc_get_current(data->display.drm_fd, data->pipe_crc, &crc);
>  
> -- 
> 2.25.1
>

Hi Petri and Hiler,

I was looking at the failure log for this patch [1], and I did not find
any reference to a failure related to kms_plane_multiple, which makes me
think this is a false positive. In this case, is it worth resubmitting
the patch for CI?

1. https://patchwork.freedesktop.org/series/81857/ 

Thanks

-- 
Rodrigo Siqueira
https://siqueira.tech