[Bug swing/23558] JComboBox problems on ctwm

[hendrich at informatik dot uni-hamburg dot de]

------- Additional Comments From hendrich at informatik dot uni-hamburg dot de  
2005-09-08 14:28 -------
(In reply to comment #14)
> I'm dropping the priority on this one.  You'll have to narrow the problem down
> by testing on ctwm.

This is what my newly built gdb 6.3 tells me:

java/xxx> gdb /opt/jamvm/bin/jamvm
GNU gdb 6.3
(gdb) break gdk_x_error
...
(gdb) run PR23558 200
Starting program: /opt/jamvm/bin/jamvm PR23558 200
...

Breakpoint 2, gdk_x_error (display=0x8191e58, error=0xbefff1f4)
    at gdkmain-x11.c:505
505     gdkmain-x11.c: No such file or directory.
        in gdkmain-x11.c
(gdb) where
#0  gdk_x_error (display=0x8191e58, error=0xbefff1f4) at gdkmain-x11.c:505
#1  0x48d94b8b in _XError () from /usr/X11R6/lib/libX11.so.6
#2  0x48d92fc6 in _XReply () from /usr/X11R6/lib/libX11.so.6
#3  0x48d79da2 in XGetWindowProperty () from /usr/X11R6/lib/libX11.so.6
#4  0x487c1e8b in window_window_state_cb (widget=0x848e178, event=0x82a0fa4, 
    peer=0x404a0268) at gnu_java_awt_peer_gtk_GtkWindowPeer.c:1722
#5  0x488e97fb in _gtk_marshal_BOOLEAN__BOXED (closure=0x848e178, 
    return_value=0xbefff4a4, n_param_values=2, param_values=0x82a0fa4, 
    invocation_hint=0xbefff4cc, marshal_data=0x0) at gtkmarshalers.c:83
#6  0x48ca7d9b in IA__g_closure_invoke (closure=0x848dde0, 
    return_value=0xbefff4a4, n_param_values=2, param_values=0xbefff5d4, 
    invocation_hint=0xbefff4cc) at gclosure.c:437
#7  0x48cb8870 in signal_emit_unlocked_R (node=0x8428a28, detail=0, 
    instance=0x848e178, emission_return=0xbefff564, 
    instance_and_params=0xbefff5d4) at gsignal.c:2485
#8  0x48cb7d49 in IA__g_signal_emit_valist (instance=0x848e178, signal_id=0, 
    detail=0, var_args=0xbefff764 "|÷ÿ¾È @\b\214÷ÿ¾M²\234HTÞ¨H|÷ÿ¾")
    at gsignal.c:2254
#9  0x48cb819b in IA__g_signal_emit (instance=0x848e178, signal_id=62, 
    detail=0) at gsignal.c:2288
#10 0x489cb606 in gtk_widget_event_internal (widget=0x848e178, event=0x82a0fa4)
    at gtkwidget.c:3616
#11 0x488e71fa in IA__gtk_main_do_event (event=0x82a0fa4) at gtkmain.c:1406
#12 0x48acedff in gdk_event_dispatch (source=0xbefff1f4, callback=0, 
    user_data=0x0) at gdkevents-x11.c:2220
#13 0x48cf9cc1 in g_main_dispatch (context=0x819f230) at gmain.c:1947
#14 0x48cfae09 in IA__g_main_context_dispatch (context=0x819f230)
    at gmain.c:2497
#15 0x48cfb192 in g_main_context_iterate (context=0x819f230, block=1, 
    dispatch=1, self=0x82a0930) at gmain.c:2578
#16 0x48cfb7a0 in IA__g_main_loop_run (loop=0x81e7770) at gmain.c:2782
#17 0x488e6a6b in IA__gtk_main () at gtkmain.c:963
#18 0x487bf33c in Java_gnu_java_awt_peer_gtk_GtkToolkit_gtkMain (
    env=0x8061794, obj=0x401f7a30) at gnu_java_awt_peer_gtk_GtkToolkit.c:331
#19 0x0805d875 in callJNIMethod (env=0x8061794, class=0x401f7a30, 
    sig=0x8070420 "()V", ret_type=0, ostack=0x81e929c, 
    f=0x487bf320 "U\211åS輤þÿ\201Ã\213ê", args=1) at dll_md.c:64
#20 0x0804e931 in callJNIWrapper (class=0x401f7a30, mb=0x816829c, 
    ostack=0xbefff1f4) at dll.c:325
#21 0x08051856 in executeJava () at interp.c:2236
#22 0x0804f557 in executeMethodVaList (ob=Variable "ob" is not available.
) at execute.c:66
#23 0x0804f5f9 in executeMethodArgs (ob=0xbefff1f4, class=0xbefff1f4, 
    mb=0xbefff1f4) at execute.c:38
#24 0x0805b251 in threadStart (arg=0x81e6f40) at thread.c:257
#25 0x4002dc60 in pthread_start_thread () from /lib/libpthread.so.0
#26 0x4002dcdf in pthread_start_thread_event () from /lib/libpthread.so.0
#27 0x4018eb77 in clone () from /lib/libc.so.6


Does this stacktrace look sane, despite the funny characters in
a few of the (varargs) parameters?

If so, I still can't make much of it. Its over ten years now that I last
did some X11 programming, and I never had to bother with signal handlers
and the like.

Anyway, line 1722 in gnu_java_awt_peer_gtk_GtkWindowPeer.c:1722 seems
to query the window manager state:

  XGetWindowProperty (GDK_DISPLAY_XDISPLAY (display),
                      GDK_WINDOW_XID (widget->window),
                      gdk_x11_get_xatom_by_name_for_display (display,           
 
         "_NET_WM_STATE"),
                      0, G_MAXLONG, False, XA_ATOM, &type, &format, &atom_count,
                      &bytes_after, alu.gu_extents);
              
              
Any ideas?

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23558