Re: [PATCH i-g-t 4/5] lib/kms: Make igt_first_crtc() skip if no CRTCs are around

[Jani Nikula <jani.nikula@intel.com>]

On Thu, 05 Mar 2026, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> Everyone using igt_first_crtc() really needs a CRTC, and thus
> we should just skip if none are around.
>
> Most of the callers neglected to even require the non-NULL
> CRTC, but I think all of the do want a CRTC.

*them

Yeah, I think they'd NULL pointer dereference if this returned NULL.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>


>
> Maybe igt_display_require() should even require at least one
> CRTC, but let's do this one small step at a time.
>
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> ---
>  lib/igt_kms.c       | 2 ++
>  tests/nouveau_crc.c | 1 -
>  2 files changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/lib/igt_kms.c b/lib/igt_kms.c
> index 9ebf77b74d04..a9673a9a6565 100644
> --- a/lib/igt_kms.c
> +++ b/lib/igt_kms.c
> @@ -7996,6 +7996,8 @@ igt_crtc_t *igt_first_crtc(igt_display_t *display)
>  	for_each_crtc(display, crtc)
>  		return crtc;
>  
> +	igt_skip("No CRTCs on device\n");
> +
>  	return NULL;
>  }
>  
> diff --git a/tests/nouveau_crc.c b/tests/nouveau_crc.c
> index b71aac45bb79..cad7edc38f20 100644
> --- a/tests/nouveau_crc.c
> +++ b/tests/nouveau_crc.c
> @@ -402,7 +402,6 @@ int igt_main()
>  		     "is reset to its default value after a single capture.");
>  	igt_subtest("ctx-flip-threshold-reset-after-capture") {
>  		data.crtc = igt_first_crtc(&data.display);
> -		igt_require(data.crtc);
>  
>  		run_test(&data, test_ctx_flip_threshold_reset_after_capture);
>  	}

-- 
Jani Nikula, Intel