Re: [igt-dev] [PATCH i-g-t v3] tests/kms_plane_cursor: Handle invalid pipe in big joiner

["Modem, Bhanuprakash" <bhanuprakash.modem@intel.com>]

Hi Nidhi,

On Fri-30-06-2023 01:52 pm, Nidhi Gupta wrote:
> If any pipe is invalid in big joiner currently test throws
> as skip for all the pipes as it is within the dynamic subtest
> block.
> Fixed in IGT by not executing the subtest if valid combo of pipe
> and output not found for big joiner and go on to check for the
> next pipe.
> 
> v2: Fix commit subject and message (Bhanu)
>      Fix nitpicks (Bhanu)
> 
> v3: Have pipe/output valid combo before checking
>      for big joiner for the pipe. (Bhanu)
> 
> Signed-off-by: Nidhi Gupta <nidhi1.gupta@intel.com>
> ---
>   tests/kms_plane_cursor.c | 7 ++++---
>   1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/tests/kms_plane_cursor.c b/tests/kms_plane_cursor.c
> index a7acb18e1..fb290322e 100644
> --- a/tests/kms_plane_cursor.c
> +++ b/tests/kms_plane_cursor.c
> @@ -85,9 +85,6 @@ static void test_init(data_t *data, enum pipe pipe_id, igt_output_t *output)
>   	data->overlay = igt_pipe_get_plane_type(data->pipe, DRM_PLANE_TYPE_OVERLAY);
>   	data->cursor = igt_pipe_get_plane_type(data->pipe, DRM_PLANE_TYPE_CURSOR);
>   
> -	igt_output_set_pipe(data->output, data->pipe_id);
> -	igt_require(i915_pipe_output_combo_valid(display));
> -
>   	igt_info("Using (pipe %s + %s) to run the subtest.\n",
>   		 kmstest_pipe_name(data->pipe_id), igt_output_name(data->output));
>   
> @@ -310,6 +307,10 @@ igt_main
>   							     DRM_PLANE_TYPE_OVERLAY))
>   					continue;
>

Please call igt_display_reset() here before setting the pipe and drop it 
from test_init().

With this change, you can get my R-b.

- Bhanu

> +				igt_output_set_pipe(output, pipe);
> +				if (!i915_pipe_output_combo_valid(&data.display))
> +					continue;
> +
>   				test_init(&data, pipe, output);
>   
>   				for (j = 0; j < ARRAY_SIZE(cursor_sizes); j++) {