Re: [PATCH i-g-t] tests/intel/kms_cdclk: Skip non-4K outputs instead of skipping the test

[Karthik B S <karthik.b.s@intel.com>]

On 5/1/2026 4:06 PM, Swati Sharma wrote:
> The mode-transition-all-outputs subtest currently skips the entire
> test as soon as any single connected output lacks a >= 4K mode. On
> typical setups (e.g. an eDP panel limited to 2880x1800 alongside
> multiple 4K external displays), this prevents the test from running
> even though enough 4K-capable outputs are available to exercise the
> CDCLK transition.
>
> Replace the igt_require_f() check with a per-output skip: outputs
> without a >= 4K mode are simply omitted from the valid_outputs list
> and the test continues with the remaining ones. The existing
> igt_skip_on_f(count < 2) guard still ensures we only run when at
> least two valid outputs are present.
>
> Co-developed-by: Claude Opus 4.6
> Signed-off-by: Swati Sharma <swati2.sharma@intel.com>
Reviewed-by: Karthik B S <karthik.b.s@intel.com>
> ---
>   tests/intel/kms_cdclk.c | 7 +++++--
>   1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/tests/intel/kms_cdclk.c b/tests/intel/kms_cdclk.c
> index 48bd7276f..070fba400 100644
> --- a/tests/intel/kms_cdclk.c
> +++ b/tests/intel/kms_cdclk.c
> @@ -271,8 +271,11 @@ static void test_mode_transition_on_all_outputs(data_t *data)
>   
>   	for_each_connected_output(display, output) {
>   		mode_highres[count] = *igt_output_get_highres_mode(output);
> -		igt_require_f(is_4k(mode_highres[count]), "Mode >= 4K not found on output %s.\n",
> -			      igt_output_name(output));
> +		if (!is_4k(mode_highres[count])) {
> +			igt_info("Mode >= 4K not found on output %s; skipping\n",
> +				 igt_output_name(output));
> +			continue;
> +		}
>   
>   		mode_lowres[count] = *get_lowres_mode(output);
>