When I try to execute gnome-keybinding-properties it crashes giving a segmentation fault. gnome-keybinding-properties -d (gnome-keybinding-properties:3626): Gtk-CRITICAL **: gtk_tree_store_get_value: assertion `VALID_ITER (iter, tree_store)' failed (gnome-keybinding-properties:3626): GLib-GObject-WARNING **: gtype.c:3940: type id `0' is invalid (gnome-keybinding-properties:3626): GLib-GObject-WARNING **: can't peek value table for type `<invalid>' which is not currently referenced zsh: segmentation fault gnome-keybinding-properties -d I tried to re-emerge g-c-c but didn't help.
Created attachment 209007 [details] emerge --info
A backtrace of the problem would help http://www.gentoo.org/proj/en/qa/backtraces.xml
Created attachment 209008 [details] backtrace log
I think you are suffering: https://bugzilla.gnome.org/show_bug.cgi?id=591392
Created attachment 210278 [details, diff] 1.patch And attached patch fixes the issue (at least for me) =)
Created attachment 210306 [details] -r1 ebuild
Renamed above patch to gnome-control-center-2.26.0-keybinding-properties-segfault.patch for being applied by ebuild. Thank you @ Pacho Ramos. Patch solves the issue for me too. Added fixed ebuild that applies the patch.
Ops, I think should be closed when patch is applied in cvs. My bad :P
(In reply to comment #8) > Ops, I think should be closed when patch is applied in cvs. > > My bad :P > Exactly ;-) Thanks for testing it
ugh, yeah the original code was obviously prone to problems.
Do you let me commit the fix for this on a new gnome-base/gnome-control-center-2.26.0-r1 release?
(In reply to comment #11) > Do you let me commit the fix for this on a new > gnome-base/gnome-control-center-2.26.0-r1 release? > Feel free to do it and don't forget keywords :)
Fixed in just commited gnome-control-center-2.26.0-r1
This got me too, on a fully stable install. The patch works, and it clearly fixes a simple bug in the original code. Can we stablise -r1?
(In reply to comment #14) > This got me too, on a fully stable install. > The patch works, and it clearly fixes a simple bug in the original code. > > Can we stablise -r1? > This is being handled in bug 308549