On Wed, Nov 27, 2013 at 10:58 AM, Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> wrote: > On Wed, Nov 27, 2013 at 10:37:03AM +0100, Daniel Vetter wrote: >> This reverts commit 3fbd6439e4639ecaeaae6c079e0aa497a1ac3482. >> >> This breaks the boot on my g33 here, right when the initial modeset >> happens. The modeset seems to complete (the external screen lights up >> at the native resolution), but then the box is dead. >> >> Strangely enough when delaying disabling modesetting and manually >> loading the driver later on everything is perfectly fine. I've run out >> of clues, ideas highly welcome. At least the revert here gets my box >> going again. > > Out of curiousity, what does the debug say the two modes being merged are? The old mode somehow has the preferred mode bit whereas the replacement mode doesn't have it. This happens on the sdvo hdmi port. I have no idea why this really matters, since like I've said modesetting works fine anytime afterwards. I'm stumped. -Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
