Sounds good to me. Regards Shashank -----Original Message----- From: Thierry Reding [mailto:thierry.red...@gmail.com] Sent: Thursday, May 15, 2014 12:36 PM To: Sharma, Shashank Cc: Daniel Vetter; intel-gfx; dri-devel; Purushothaman, Vijay A; Mukherjee, Indranil; Shankar, Uma; Jindal, Sonika; Korjani, Vikas Subject: Re: [Intel-gfx] Design review request: DRM color manager
On Thu, May 15, 2014 at 10:52:38AM +0530, Sharma, Shashank wrote: [...] > Please note that the color correction methods changes per platform and > what's valid for one Intel platform may not be valid for other. So the > atomic modeset should have a clear idea of what is supported on which > platforms. I don't think this will be an issue at all. The DRM driver should only expose what's supported on the particular device that it drives. And similarily userspace drivers should enumerate properties to find out which ones are available. Atomic modeset is only the mechanism to make sure it's all applied in one go. But that mechanism is completely generic and can be applied to any properties, therefore no specific knowledge about the available properties will be required in atomic modesetting itself. Thierry
