On Sat, 27 Nov 2004 16:51:55 +0100, Sven Neumann said: > edward hage <edha@xxxxxxxxx> writes: > > I narrowed the problem down to this program. The following program > > shows a Segmetation Fault on one computer, and works fine for the > > other. > > gdb clearly shows that the crash happens in exit() so it is unlikely > that a GLib function is causing your problem. (Mostly for the archives, and those who have to debug crashes in exit() ;) Note that the most common cause of crashes in exit() is if "somebody" has stomped on the malloc heap badly enough that when exit() tries to flush the stdio buffers, it chases a bad pointer and dies. In other words, the actual bug probably happened long ago and far away, and the program only discovers it when it's trying to wrap things up. It's perfectly possible that incorrect calling of a GLib function could cause such a mangling (most likely by feeding it a pointer to an object that's been malloc()ed on the heap, but isn't in fact a GLib object. It's *much* less likely that an actual GLib bug would be the actual cause of the function.
Attachment:
pgpuUCRqpVdPH.pgp
Description: PGP signature
_______________________________________________ gtk-list@xxxxxxxxx http://mail.gnome.org/mailman/listinfo/gtk-list
