On Tue, 2011-01-18 at 13:34 -0500, Archit Baweja wrote: > Hey Francesco, > > If you've compiled the glade source yourself, make sure you've > compiled it with debug information. Usually that involves compiling > with the GCC -g flag. Try the following > > export CFLAGS='-g' ; ./configure
you can also just ./configure --enable-debug to get the debug flags. Cheers, -Tristan > > With that GDB will show you proper names in stack traces instead of > hex values of addresses of functions, routines etc. > > And you're right, what you have is not enough. All that part says is > that the Gtk+ subsystem got a NULL value instead of a valid Gtk+ > object id. So probably some funtion/routine higher up in the call > stack that is faulty. > > Hope that helps, > Archit > > On Tue, Jan 18, 2011 at 12:47 PM, [email protected] > <[email protected]> wrote: > I have a segmentation fault on glade3.7.3 compiled from > source. > Some devel can write please wich steps to do for the right > trace ? > > For now i have just that but i think it is not enough > > (glade-3:31430): GLib-GObject-WARNING **: > /tmp/buildd/glib2.0-2.27.91/./gobject/gtype.c:4197: type id > `0' is invalid > > (glade-3:31430): GLib-GObject-WARNING **: can't peek value > table for > type `<invalid>' which is not currently referenced > Segmentation fault > > > thanks > > -- > > _______________________________________________ > Glade-devel maillist - [email protected] > http://lists.ximian.com/mailman/listinfo/glade-devel > > _______________________________________________ > Glade-devel maillist - [email protected] > http://lists.ximian.com/mailman/listinfo/glade-devel _______________________________________________ Glade-devel maillist - [email protected] http://lists.ximian.com/mailman/listinfo/glade-devel
