[igt-dev] [PATCH i-g-t] kms_plane: Do not leave CRC generation running while creating a format.

Ville Syrjälä ville.syrjala at linux.intel.com
Mon Sep 24 12:41:06 UTC 2018


On Mon, Sep 24, 2018 at 08:44:49AM +0200, Maarten Lankhorst wrote:
> Creating a YUV FB in a high resolution takes too long, and we end up with:
> [drm:drm_crtc_add_crc_entry] *ERROR* Overflow of CRC buffer, userspace reads too slow.
> 
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106885
> Signed-off-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
> ---
>  tests/kms_plane.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/tests/kms_plane.c b/tests/kms_plane.c
> index aa36ddeba31a..d00baeaff72d 100644
> --- a/tests/kms_plane.c
> +++ b/tests/kms_plane.c
> @@ -419,7 +419,7 @@ static void test_format_plane_color(data_t *data, enum pipe pipe,
>  	igt_plane_set_fb(plane, fb);
>  
>  	igt_display_commit2(&data->display, data->display.is_atomic ? COMMIT_ATOMIC : COMMIT_UNIVERSAL);
> -	igt_pipe_crc_get_current(data->display.drm_fd, data->pipe_crc, crc);
> +	igt_pipe_crc_collect_crc(data->pipe_crc, crc);

That's going to slow the test down even more :(

>  
>  	igt_remove_fb(data->drm_fd, &old_fb);
>  }
> @@ -466,7 +466,6 @@ static void test_format_plane(data_t *data, enum pipe pipe,
>  	set_legacy_lut(data, pipe, 0xfc00);
>  
>  	test_init(data, pipe);
> -	igt_pipe_crc_start(data->pipe_crc);

_stop() needs removing too.

>  
>  	igt_info("Testing format " IGT_FORMAT_FMT " on %s.%u\n",
>  		 IGT_FORMAT_ARGS(format),
> -- 
> 2.19.0
> 
> _______________________________________________
> igt-dev mailing list
> igt-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/igt-dev

-- 
Ville Syrjälä
Intel


More information about the igt-dev mailing list