That pretty much settles it. Something in your system is broken,
and it's probably the cpu chip itself. 

Now, I really shouldn't say this, but it does look like your
particular individual failure might actually have a possible SW
workaround, especially with a -current kernel. The -current NetBSD
kernel has an almost complete software IEEE FP interpreter built into
it. (We don't decode instructions that supposedly can't trap, that's the
"almost".)

> From: Werner Backes <werner@bit-1.de>
>
> Hi Ross,
>
> I had to reinstall NetBSD from scratch, because I used the
> disk for my OpenBSD tests, but I'm back now with a running
> NetBSD 1.5.2 system:
>
> > Let's start with your current uname -a,
>
> multia# uname -a
> NetBSD multia 1.5.2 NetBSD 1.5.2 (GENERIC) #37: Sat Aug 18 10:02:07 MEST
> 2001    
> he@albatross.urc.uninett.no:/usr/src/sys/arch/alpha/compile/GENERIC
> alpha
>
> > and also let me know what
> > vintage of userland you are running.
>
> Nothing special, just the plain installation sets
> base,etc,kern and man from
> ftp.netbsd.de:/pub/NetBSD/NetBSD-1.5.2/alpha/binary/sets
>
> If you can help me to decide if my CPU is broken or not I'll be
> very happy...
>                Werner
>