Re: njamd compatible with gtk 2.2.1?

[Peter Van Wieren <peter vanwieren avlna com>]

Paul

Thanks, I never head of valgrind, but I've tried it.  If I understand 
this right, it looks like a bug somewhere in the Xlibs (thus neither gtk 
nor my code).

Pete



fenwick 1 329% gcc -o example example.c `pkg-config --cflags --libs 
gtk+-2.0 gthread-2.0`
fenwick 0 330% !val
valgrind example
==2709== Memcheck, a memory error detector.
==2709== Copyright (C) 2002-2005, and GNU GPL'd, by Julian Seward et al.
==2709== Using LibVEX rev 1367, a library for dynamic binary translation.
==2709== Copyright (C) 2004-2005, and GNU GPL'd, by OpenWorks LLP.
==2709== Using valgrind-3.0.1, a dynamic binary instrumentation framework.
==2709== Copyright (C) 2000-2005, and GNU GPL'd, by Julian Seward et al.
==2709== For more details, rerun with: -v
==2709==
==2709== Invalid read of size 4
==2709==    at 0x1BF3EBC5: (within /usr/X11R6/lib/libX11.so.6.2)
==2709==    by 0x1BF3EC49: XGetAtomName (in /usr/X11R6/lib/libX11.so.6.2)
==2709==    by 0x1BBBDD7B: gdk_x11_xatom_to_atom_for_display (in 
/usr/lib/libgdk-x11-2.0.so.0.200.1)
==2709==    by 0x1BBB1AFC: (within /usr/lib/libgdk-x11-2.0.so.0.200.1)
==2709==    by 0x1BBB1F35: _gdk_events_queue (in 
/usr/lib/libgdk-x11-2.0.so.0.200.1)
==2709==    by 0x1BBB213C: (within /usr/lib/libgdk-x11-2.0.so.0.200.1)
==2709==    by 0x1BCF5B34: (within /usr/lib/libglib-2.0.so.0.200.1)
==2709==    by 0x1BCF6B77: g_main_context_dispatch (in 
/usr/lib/libglib-2.0.so.0.200.1)
==2709==    by 0x1BCF6E8C: (within /usr/lib/libglib-2.0.so.0.200.1)
==2709==    by 0x1BCF758E: g_main_loop_run (in 
/usr/lib/libglib-2.0.so.0.200.1)
==2709==    by 0x1B9F2F5E: gtk_main (in /usr/lib/libgtk-x11-2.0.so.0.200.1)
==2709==    by 0x8048A00: main (in 
/work/petervw/src/rmake/debug_segv_threads/example)
==2709==  Address 0x1C13F36C is 20 bytes inside a block of size 21 alloc'd
==2709==    at 0x1B8FEA35: malloc (vg_replace_malloc.c:149)
==2709==    by 0x1BF445F0: _XUpdateAtomCache (in 
/usr/X11R6/lib/libX11.so.6.2)
==2709==    by 0x1BF447EF: XInternAtom (in /usr/X11R6/lib/libX11.so.6.2)
==2709==    by 0x1BF60773: XSetWMProtocols (in /usr/X11R6/lib/libX11.so.6.2)
==2709==    by 0x1BBC2169: (within /usr/lib/libgdk-x11-2.0.so.0.200.1)
==2709==    by 0x1BBC26B8: gdk_window_new (in 
/usr/lib/libgdk-x11-2.0.so.0.200.1)
==2709==    by 0x1BAE7A9A: (within /usr/lib/libgtk-x11-2.0.so.0.200.1)
==2709==    by 0x1BCACD76: g_cclosure_marshal_VOID__VOID (in 
/usr/lib/libgobject-2.0.so.0.200.1)
==2709==    by 0x1BC9A246: (within /usr/lib/libgobject-2.0.so.0.200.1)
==2709==    by 0x1BC99ED6: g_closure_invoke (in 
/usr/lib/libgobject-2.0.so.0.200.1)
==2709==    by 0x1BCAC20E: (within /usr/lib/libgobject-2.0.so.0.200.1)
==2709==    by 0x1BCAB9A7: g_signal_emit_valist (in 
/usr/lib/libgobject-2.0.so.0.200.1)
==2709==
==2709== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 86 from 3)
==2709== malloc/free: in use at exit: 596283 bytes in 3117 blocks.
==2709== malloc/free: 8235 allocs, 5118 frees, 968655 bytes allocated.
==2709== For counts of detected errors, rerun with: -v
==2709== searching for pointers to 3117 not-freed blocks.
==2709== checked 1229412 bytes.
==2709==
==2709== LEAK SUMMARY:
==2709==    definitely lost: 0 bytes in 0 blocks.
==2709==      possibly lost: 800 bytes in 20 blocks.
==2709==    still reachable: 595483 bytes in 3097 blocks.
==2709==         suppressed: 0 bytes in 0 blocks.
==2709== Reachable blocks (those to which a pointer was found) are not 
shown.
==2709== To see them, rerun with: --show-reachable=yes

fenwick 1 331% cat example.c

/* gcc -o example example.c `pkg-config --cflags --libs gtk+-2.0 
gthread-2.0` -lnjamd */

#include <gtk/gtk.h>

int main ( int argc , char ** argv , char ** envp )
 {
 GtkWidget *window;
 gtk_init(&argc,&argv);
 window = gtk_window_new(GTK_WINDOW_TOPLEVEL);
 gtk_container_set_border_width(GTK_CONTAINER (window), 10);
 gtk_window_set_title (GTK_WINDOW (window), "njamd");
 gtk_widget_set_size_request (window, 700, 800);
 gtk_widget_show (window);
 gtk_main();
 sleep(10);
 return(0);
 }
fenwick 0 332%


Paul Davis wrote:

> On Fri, 2005-09-30 at 12:55 -0400, Peter Van Wieren wrote:
>  
>
> > Ben,
> >
> > Thanks.  My code cannot have a buffer overrun, for it was simply hello.c 
> > from the tutorial page.  An even more simple version that will still 
> > segfault is as follows.  Can it be that -lnjamd causes a segfault, as 
> > one could not have occured in the code below?
> >    
>
> use valgrind. pray to the god of valgrind before hand, for it is
> munificent in its care and loving for us software engineers, and is
> always good to us.
>
> --p
>
>
>  


--
Peter Van Wieren, Project Engineer, AVL Powertrain Engineering, Inc. USA
734 414 9678 (voice)
734 414 9690 (fax)
email: peter vanwieren avlna com