From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga03.intel.com (mga03.intel.com [134.134.136.65]) by gabe.freedesktop.org (Postfix) with ESMTPS id 11A6210E009 for ; Wed, 6 Apr 2022 19:21:00 +0000 (UTC) Date: Wed, 6 Apr 2022 22:20:55 +0300 From: Ville =?iso-8859-1?Q?Syrj=E4l=E4?= To: Juha-Pekka Heikkila Message-ID: References: <20220406191139.31227-1-juhapekka.heikkila@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20220406191139.31227-1-juhapekka.heikkila@gmail.com> Subject: Re: [igt-dev] [PATCH i-g-t] tests/i915/kms_big_fb: move start of crc calculation past first commit List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: igt-dev@lists.freedesktop.org Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" List-ID: On Wed, Apr 06, 2022 at 10:11:39PM +0300, Juha-Pekka Heikkila wrote: > First commit may cause modeset which can on some machines take enough > time to fill up crc buffer causing crc buffer overflow messages. Moved > crc calcualtion start past first commit to avoid these messages. > > I changed small fb not to be cleared with green color, this seem > to take portion of execution time away. > > Signed-off-by: Juha-Pekka Heikkila > --- > tests/i915/kms_big_fb.c | 7 +++---- > 1 file changed, 3 insertions(+), 4 deletions(-) > > diff --git a/tests/i915/kms_big_fb.c b/tests/i915/kms_big_fb.c > index 8724d6069..6c6d674cb 100644 > --- a/tests/i915/kms_big_fb.c > +++ b/tests/i915/kms_big_fb.c > @@ -370,12 +370,12 @@ static bool test_plane(data_t *data) > * rendering pipeline introduces slight differences into > * the result if we try that, and so the crc will not match. > */ > - igt_pipe_crc_start(data->pipe_crc); > copy_pattern(data, small_fb, 0, 0, big_fb, x, y, > small_fb->width, small_fb->height); > > igt_display_commit2(&data->display, data->display.is_atomic ? > COMMIT_ATOMIC : COMMIT_UNIVERSAL); Is is modesetting multiple pipes here? If not then we seem to have a bug somewhere as CRCs should be disabled during most of the modeset sequence. > + igt_pipe_crc_start(data->pipe_crc); > igt_pipe_crc_get_current(data->display.drm_fd, data->pipe_crc, &small_crc); > > igt_plane_set_fb(plane, big_fb); > @@ -420,9 +420,8 @@ static bool test_pipe(data_t *data) > if (igt_rotation_90_or_270(data->rotation)) > igt_swap(width, height); > > - igt_create_color_fb(data->drm_fd, width, height, > - data->format, data->modifier, > - 0, 1, 0, &data->small_fb); > + igt_create_fb(data->drm_fd, width, height, data->format, > + data->modifier, &data->small_fb); I'd split that into a separate patch. > > igt_output_set_pipe(data->output, data->pipe); > > -- > 2.28.0 -- Ville Syrjälä Intel