Re: possible bug in gtk

[Owen Taylor <otaylor redhat com>]

Daniel Egger <egger suse de> writes:

> This is a forwarded message from my collegue, maybe someone with more
> insight can comment on this one. I think it might be an gtk-gnutella
> issue since I know it is buggy as hell but I'd like to hear other
> possibilites.

XPending() is never supposed to block. My best guess would be that
the program wasn't actually frozen ... it just receiving and dispatching
a large amount of events for some reason (buggy mouse?), and happened
to be here when it was stopped by the debugger.

My only other guess would be that X got confused because gnutella
was making GTK+ calls in multiple threads or from signal handlers
without locking properly.

Regards,
                                        Owen

> > I let gtk-gnutella run on my home computer. After some
> > time I checked it by ssh from another computer and 
> > it was freezed here:
> > #0  0x403a36ae in select () from /lib/libc.so.6
> > #1  0x402b5d34 in __DTOR_END__ () from /usr/X11R6/lib/libX11.so.6
> > #2  0x4020816a in _XRead () from /usr/X11R6/lib/libX11.so.6
> > #3  0x40207bb0 in _XEventsQueued () from /usr/X11R6/lib/libX11.so.6
> > #4  0x401fc013 in XPending () from /usr/X11R6/lib/libX11.so.6
> > #5  0x40172bd2 in gdk_event_prepare () from /usr/lib/libgdk-1.2.so.0
> > #6  0x401a5e93 in g_main_iterate () from /usr/lib/libglib-1.2.so.0
> > #7  0x401a637c in g_main_run () from /usr/lib/libglib-1.2.so.0
> > #8  0x400bdd2c in gtk_main () from /usr/lib/libgtk-1.2.so.0
> > #9  0x8075e63 in main (argc=1, argv=0xbffff71c) at main.c:298
> > #10 0x402f2c6f in __libc_start_main () from /lib/libc.so.6
> > 
> > When I returned to my computer and moved mouse, it start to 
> > run again.
> > 
> > I think, that bug is in gdk_event_prepare. It calls XPending,
> > which may probably sometimes block (from XPending manpage it is 
> > not clear)