[igt-dev] [PATCH i-g-t] tests/kms_cursor_crc: Fix special handling for Intel CHV

[Sharma, Swati2]

LGTM. Please add "Fixes" with gitlab issue# before merging.

Reviewed-by: Swati Sharma <swati2.sharma at intel.com>

On 30-May-23 7:01 PM, Juha-Pekka Heikkila wrote:
> Match software rendering round to what hardware rendering round did
> on Intel CHV.
> 
> Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila at gmail.com>
> ---
>   tests/kms_cursor_crc.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/tests/kms_cursor_crc.c b/tests/kms_cursor_crc.c
> index 3a353a03..61a0d65e 100644
> --- a/tests/kms_cursor_crc.c
> +++ b/tests/kms_cursor_crc.c
> @@ -253,6 +253,10 @@ static void do_single_test(data_t *data, int x, int y, bool hw_test,
>   			igt_assert_crc_equal(hwcrc, &crc_after);
>   		}
>   	} else {
> +		/* If on broken situation on CHV match what hw round did */
> +		if (chv_cursor_broken(data, x) && cursor_visible(data, x, y))
> +			return;
> +
>   		/* Now render the same in software and collect crc */
>   		swbufidx = (data->primary->drm_plane->fb_id ==
>   			    data->primary_fb[SWCOMPARISONBUFFER1].fb_id) ?