On Thu, Jan 23, 2014 at 10:05:19AM +0000, Russell King - ARM Linux wrote:
On Thu, Jan 23, 2014 at 11:00:28AM +0100, Daniel Vetter wrote:
On Thu, Jan 23, 2014 at 10:42:02AM +0100, David Herrmann wrote:
If there's ever hardware that truly supports sub-device hotplugging, we can support that by adding a DRM-ClientCap like 3D modes. But as I understand, the current hardware is still a single piece of hardware which just might get probed via different buses and thus is not an atomic entity. So the device is still hotplugged as a whole, we just don't get the events through a single path.
Yeah, if we ever want to allow hotplugging after driver setup that's massive work to do it right. Which is why I want to plug this hole here before someone sneaks in for real ;-)
David did say at the kernel summit that he's not interested at all in hotplugging the components of a DRM system, and his statement appeared to be very clear in that regard that it was never going to happen.
I think we eventually need to be able to hot-plug connectors (and maybe encoders) to support fancy dp1.2 output routing and multiple streams over the same cable. But I agree that hotplugging crtcs is insane and better done by hotplugging an entire drm device with that crtc. In any case not something to worry about right now. -Daniel