[Xcb] Re: xcblock.c Assertion

[Paul Moss]

Thanks.  And, yes, I'm using Slackware 10.  I don't recall exactly
when that was released.

It'll take me a little while to upgrade some of that stuff to get you
a better stack trace.  I'll let you know what turns up.

Paul

On 9/15/06, Jamey Sharp <jamey at minilop.net> wrote:
> On Thu, Sep 14, 2006 at 08:53:58PM -0400, Paul Moss wrote:
> > I did a git checkout master; ./autogen.sh ; make ; make install
> > (this was at about 8:30 PM EDT on 9/14)
>
> Thanks for the timestamps. Nothing significant had changed recently in
> either Xlib or XCB before you first reported problems on Monday, though,
> so if this is our bug I guess it isn't especially new.
>
> > GNU gdb 6.1.1
>
> Your gdb is two years and four releases old: it's possible that a newer
> version will be more helpful.
>
> > This GDB was configured as "i486-slackware-linux"...
>
> You're using Slackware, I take it? I don't know if we've had anyone test
> on that distribution before. :-)
>
> > gaim: xcl/xcblock.c:128: _XGetXCBBuffer: Assertion `((int) ((xcb_req)
> > - (dpy->request)) >= 0)' failed.
> > ...
> > #5  0x406ae56c in __assert_fail () from /lib/libc.so.6
> > #6  0x407b4ee0 in _IO_2_1_stdout_ () from /lib/libc.so.6
> > #7  0xbfcbd591 in ?? ()
> > #8  0x407adfe6 in in6addr_loopback () from /lib/libc.so.6
> > #9  0x40638771 in xrmtypes () from /usr/X11R7/lib/libX11.so.6
> > #10 0x00000080 in ?? ()
> > #11 0x406387aa in xrmtypes () from /usr/X11R7/lib/libX11.so.6
> > #12 0x407adfe6 in in6addr_loopback () from /lib/libc.so.6
> > #13 0x406388e0 in xrmtypes () from /usr/X11R7/lib/libX11.so.6
> > #14 0x4000a490 in _dl_map_object_deps () from /lib/ld-linux.so.2
> > Previous frame inner to this frame (corrupt stack?)
> >
> > Does that help any?
>
> You did the right steps, but the stack trace looks completely bogus
> below __assert_fail, and therefore useless. For example,
> in6addr_loopback is read-only data, not a function. Can you upgrade gdb,
> get a version of libc with debugging symbols, and try again?
>
> Sometimes when debugging these sorts of problems, I've had good look
> with studying a trace of the packets. Force the use of TCP: you may have
> to remove "-nolisten tcp" from your X server arguments in gdm, kdm, or
> whatever. Then start up tcpdump to log the packets:
>         sudo tcpdump -w gaim.log -i lo -s 0 tcp portrange 6000-6020
> and, while that's running, run one failing application:
>         DISPLAY=localhost:0 gaim
>
> Then send me or the list the resulting 'gaim.log', or file a bug report
> with the evidence you've got so far -- http://bugs.freedesktop.org/.
>
> I'm either concerned or hopeful that this is some weird interaction
> between Xlib and old versions of other software on your system. But I
> don't have enough information yet.
>
> --Jamey
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.5 (GNU/Linux)
>
> iD8DBQFFCkO5p1aplQ4I9mURAhsoAJ9jqKkOdXDta4mWYYmqAxzDjd751QCfY61I
> 9GaPQ6Cn7sxwJP1LjWO3FV4=
> =0oFS
> -----END PGP SIGNATURE-----
>
>
> _______________________________________________
> Xcb mailing list
> Xcb at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/xcb
>
>