[PATCH i-g-t] tests/kms_bw: Restrict linear-tiling subtest to HDMI outputs and improve skip logic

[Naladala, Ramanaidu]

Hi Pranay,

On 6/17/2025 12:01 PM, Pranay Samala wrote:
> This patch updates the linear-tiling subtest in kms_bw to only use HDMI
> connector. If the platform does not have enough HDMI outputs to match the
> number of requested pipes, the test is skipped with a clear message
> indicating the requirement and the number of HDMI outputs found.
>
> This avoids attempting to force modes on non-HDMI connectors and makes
> the test results clearer and more reliable on systems that don’t have
> enough HDMI connectivity.
>
> Signed-off-by: Pranay Samala <pranay.samala at intel.com>
> ---
>   tests/kms_bw.c | 13 ++++++++++++-
>   1 file changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/tests/kms_bw.c b/tests/kms_bw.c
> index ca05695d3..042edf32f 100644
> --- a/tests/kms_bw.c
> +++ b/tests/kms_bw.c
> @@ -194,7 +194,7 @@ static void run_test_linear_tiling(data_t *data, int pipe, const drmModeModeInfo
>   	igt_crc_t zero, captured[IGT_MAX_PIPES];
>   	int i = 0, num_pipes = 0;
>   	enum pipe p;
> -	int ret;
> +	int ret, hdmi_count = 0;
>   
>   	/* Cannot use igt_display_get_n_pipes() due to fused pipes on i915 where they do
>   	 * not give the numver of valid crtcs and always return IGT_MAX_PIPES */
> @@ -215,6 +215,13 @@ static void run_test_linear_tiling(data_t *data, int pipe, const drmModeModeInfo
>   			continue;
>   		}
>   
> +		if (!physical &&
> +		    output->config.connector->connector_type != DRM_MODE_CONNECTOR_HDMIA &&
> +		    output->config.connector->connector_type != DRM_MODE_CONNECTOR_HDMIB)
> +			continue;
imho: Better to skip the test instead of continue the loop.
Because even though pipe A is not hdmi no point in continue loop to 
check next pipe.
> +
> +		hdmi_count++;
> +
>   		force_output_mode(data, output, mode);
>   
>   		igt_create_color_fb(display->drm_fd, mode->hdisplay,
> @@ -229,6 +236,10 @@ static void run_test_linear_tiling(data_t *data, int pipe, const drmModeModeInfo
>   			 kmstest_pipe_name(i), igt_output_name(output), mode->name);
>   	}
>   
> +	if (!physical && hdmi_count < (pipe + 1))
> +		igt_skip("Skipping test: required %d HDMI outputs, but only found %d HDMI "
> +			 "outputs to force mode\n", pipe + 1, hdmi_count);
> +

if add skip as per 1st comment no need this this skip and hdmi count logic.

>   	ret = igt_display_try_commit_atomic(display,
>   					    DRM_MODE_ATOMIC_ALLOW_MODESET |
>   					    DRM_MODE_ATOMIC_TEST_ONLY,