Re: demo_crash

> Linux, GCC 4.2.2, with both precisions, both Athlon XP (i686) and
> Athlon 64 (x86_64).
>
> I'll try to investigate it as soon as I get home.

Good luck - my biggest hunch right now is that the IEEE float magic
that it's doing is going wrong? that's a total guess though just based
on the fact that it seems quite plain code other than that.


> > I don't agree with reverting back a perfectly good test case when
> > we're not in a build cycle - sweeping it under the carpet is only
> > going to come back and bite us later.
>
> I don't think I proposed that, I was just pointing out that the cause
> for the crash was the SAP space.

sorry - my misunderstanding! :-)


Regards,
Dave