Hello, > * RandR 1.2 support. > * Several fixes specific the OLPC XO-1. > * Various minor fixes. > > See http://cgit.freedesktop.org/xorg/driver/xf86-video-geode/ for the > latest GIT snapshot. > > Please compare this to our last stable release (2.10.1) and report any > regression that you spot to this list ASAP. If nobody reported any > abnormality by the end of next week, we'll call this our 2.12.0 > release.
I'm quite late, but It appears that the new code now of course has a get_modes xf86OutputFuncsRec, which appears to be called even if GeodeGetDDC has returned FALSE. The call in lx_output_get_modes always calls xf86OutputGetEDID if the output isn't a panel, but it crashes if the GPIO pins are in serial mode instead of DDC. So maybe Jordan has some comments here - maybe RANDR1.2 in server shouldn't be calling get_modes if there is no DDC and it either still does or we aren't signaling that fact back? Also, this is yet another thing that a volunteer could do ;) I fear I won't have time in the short-term future for this, and the common case for me is to have the GPIOs in DDC mode. Regards, Mart Raudsepp _______________________________________________ Xorg-driver-geode mailing list [email protected] http://lists.x.org/mailman/listinfo/xorg-driver-geode
