[igt-dev] [PATCH i-g-t] tests/i915/kms_fence_pin_leak: Test cleanup

[Modem, Bhanuprakash]

On Wed-07-09-2022 02:47 pm, Ananya Sharma wrote:
> Before running the subtest we need to sanitize the system
> state to default,used igt_display_reset() to sanitize the state.
> 
> Signed-off-by: Ananya Sharma <ananya.sharma at intel.com>
> ---
>   tests/i915/kms_fence_pin_leak.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/i915/kms_fence_pin_leak.c b/tests/i915/kms_fence_pin_leak.c
> index 16eb595f..906b3708 100644
> --- a/tests/i915/kms_fence_pin_leak.c
> +++ b/tests/i915/kms_fence_pin_leak.c
> @@ -124,6 +124,7 @@ static void run_single_test(data_t *data, enum pipe pipe, igt_output_t *output)
>   	struct igt_fb fb[2];
>   	int i;
>   
> +	igt_display_reset(display);
>   	igt_output_set_pipe(output, pipe);
>   
>   	mode = igt_output_get_mode(output);
> @@ -180,8 +181,8 @@ static void run_single_test(data_t *data, enum pipe pipe, igt_output_t *output)
>   	}
>   
>   	igt_plane_set_fb(primary, NULL);
> -	igt_output_set_pipe(output, PIPE_ANY);
> -	igt_display_commit(display);
> +	igt_output_set_pipe(output, PIPE_NONE);
> +	igt_display_commit2(display, display->is_atomic ? COMMIT_ATOMIC : COMMIT_LEGACY);

Why do we need an atomic?

- Bhanu

>   
>   	igt_remove_fb(data->drm_fd, &fb[1]);
>   	igt_remove_fb(data->drm_fd, &fb[0]);