Re: [igt-dev] [PATCH 2/2] tests/kms_writeback: Test cleanup

[Jessica Zhang <quic_jesszhan@quicinc.com>]

Hi Nidhi,

On 9/8/2022 1:48 AM, Nidhi Gupta wrote:
> Sanitize the system state before starting the subtest.
> 
> Signed-off-by: Nidhi Gupta <nidhi1.gupta@intel.com>
> ---
>   tests/kms_writeback.c | 1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/tests/kms_writeback.c b/tests/kms_writeback.c
> index 2c5421ce..79a5760f 100644
> --- a/tests/kms_writeback.c
> +++ b/tests/kms_writeback.c
> @@ -186,6 +186,7 @@ static int do_writeback_test(igt_output_t *output, uint32_t fb_id,
>   {
>   	int ret;
>   	igt_display_t *display = output->display;
> +	igt_display_reset(display);

I might not have communicated this clearly in the v1 comments, but 
igt_display_reset() will undo all the setup done in the first 
igt_fixture [1] and disable the CRTCs. Furthermore, it would cause the 
commit in do_writeback_test() [2] to fail, since the CRTC would be 
disabled. If you want to call igt_display_reset(), you'll have to re-do 
the setup within that fixture.

Thanks,

Jessica Zhang

[1] 
https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/blob/master/tests/kms_writeback.c#L491

[2] 
https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/blob/master/tests/kms_writeback.c#L198

>   	struct kmstest_connector_config *config = &output->config;
>   
>   	igt_output_set_prop_value(output, IGT_CONNECTOR_CRTC_ID, config->crtc->crtc_id);
> -- 
> 2.36.0
>