On Fri, 2010-05-07 at 14:41 -0700, Eric Anholt wrote: > Applied those two. > > The fact that we're exposing 2 connectors for this situation is bogus in > how the KMS architecture is supposed to work, right? I mean, we've got > 2 "outputs" in this encoder going to one physical connector. The user > should see one connector, and the encoder+connector should sort out how > to program the encoder for that situation.
That sounds more sane to me. I don't know that we've formally taken a stand on that in KMS, but we might as well start. > Would having the same DDC bus for the child devs be a good enough > indication that we should have one SDVO encoder/connector for them and > they get to light up one SDVO encoder to the connector at a time? I can't think of a scenario where "shared DDC bus" wouldn't mean "shared connector". So, yeah. - ajax
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
