Re: XGammon failing
On Tue, Jul 16, 2002 at 11:08:42AM +1000, Martijn van Oosterhout did blurt:
> On Mon, Jul 15, 2002 at 05:14:31PM -0700, Jeffrey Chimene wrote:
> > Hi:
> > I notice that Woody's XGammon fails fairly consistently (usually when
> > it's about to lose...) The failure message is approximately "floating
> > point error. sorry" i.e., there's not enough information to
> > work out the failure mode.
> > I've examined the diff file, and I don't see anything there that
> > stands out as a potential bug. The source code change involves a
> > slight improvement in a typedef, and the XGammon database is the same
> > size as the previous version. At this point, I suspect the run-time
> > library.
> > I don't believe I've got enough information here to file a useful bug
> > report. What are the next steps to get enough info for the maintainer
> > to use when examining the situation?
> I'd suggest makeing sure that the binary has debugging info (though this is
> not required) and while it's running, attach gdb to it. Tell gdb to continue
> and it should just pop up when the error occurs. From there you should be
> able to do a stack trace to work out which function the error occurs in.
Ok. I don't know why, but somehow I expected that bug reports in this
environment would be different from others.
I'll learn gdb using your advice as a start. Tracking down bugs is
always a good way to learn a new system.
If that doesn't work, I'll rebuild w/ debug enabled; (I assume this
will get me a source code trace) and work the problem from there.
Thanks for your advice,
Microsoft Free By 2003
To UNSUBSCRIBE, email to email@example.com
with a subject of "unsubscribe". Trouble? Contact firstname.lastname@example.org