Hi Russell, On Thursday 06 November 2014 00:54:51 Russell King - ARM Linux wrote: > On Wed, Nov 05, 2014 at 08:47:07PM +0200, Laurent Pinchart wrote: > > (On a side note I believe treating the pitch and size arguments as inputs > > could be a worthwhile extension to the API, but given that we haven't > > rejected incorrect values in the past we're pretty much stuck). > > The bigger picture here is that it is a generic interface to DRM > drivers, and having different behaviour from different drivers is > itself impractical. I totally agree with you, I just believe it would have been nice to be able to extend the API that way, but that's no big deal. Consistency is definitely important. > If nothing breaks through better enforcement of the API, then we > should go with what the API is currently defined to be, rather than > trying to overload it with new features - and then end up in the > situation where we have something trying to use those new features > and no way to know if the driver implements this "bug" or not. -- Regards, Laurent Pinchart _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
