Hi Laurent, On 15-05-2017 08:05, Laurent Pinchart wrote: > On Monday 15 May 2017 08:47:49 Daniel Vetter wrote: >> On Sun, May 14, 2017 at 02:04:24PM +0300, Laurent Pinchart wrote: >>> On Friday 12 May 2017 17:06:14 Jose Abreu wrote: >>>> On 12-05-2017 10:35, Laurent Pinchart wrote: >>>>> On Tuesday 09 May 2017 18:00:12 Jose Abreu wrote: >>>>>> + if (mode->status == MODE_OK) >>>>>> + mode->status = > drm_mode_validate_connector(connector, >>>>>> > mode); >>>>> I would reverse the arguments order to match the style of the other >>>>> validation functions. >>>> Hmm, I think it makes more sense to pass connector first and then >>>> mode ... >>> I disagree, as this function validates a mode against a pipeline, the same >>> way the other validation functions validate a mode against other >>> parameters, but it's your patch :-) >> Call it drm_connector_validate_mode, because the first argument is >> generally the object we operate on :-) > But the function doesn't validate a mode for a connector, it validates a mode > for a complete pipeline... > Hmm, but note that in the same function there is drm_mode_validate_size() and drm_mode_validate_flag() calls, which take as first argument the mode and then the object to validate (I hadn't seen this). So, maybe leave it as drm_mode_validate_connector() as it takes a connector as argument or change to drm_mode_validate_pipeline() as you said, or even drm_mode_validate_datapath(), drm_mode_validate_videopath(), drm_mode_validate_components() ? Best regards, Jose Miguel Abreu _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
