On 11/14/19 6:05 AM, Scott Anderson wrote: >> > > Has any thought been into how this would need to interact with > dmabuf-hints[1]? Without that, it seems like it would be a total > crapshoot for clients to try and use this flag, since they have no idea > what formats+modifers the display controller supports, and instead only > has the list that the GPU supports. > dmabuf-hints would also need to explicitly state that a tranche of > formats+modifiers are supported for this flag.
Well I'm not aware of that hint extension, but I think this wasn't taken into consideration because it isn't the same thing. There's no assurance from the compositor that the buffer will not read/be imported by the GPU. A hint is merely a hint. The compositor can abide or not by that. This flag will explicitly close the client connection if the buffer can't be scanned out when this flag is passed. Regarding the screenshot bit not sure what is the concern here. Are you afraid you won't able to take screenshots? The placeholder graphics is in place only if view to which the buffer is attached can no longer be assigned to a HW plane. It would be nice to know what basic functionality will this break, as I not aware of any. -- Marius Vlad
signature.asc
Description: OpenPGP digital signature
_______________________________________________ wayland-devel mailing list wayland-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/wayland-devel