Re: crash building in groff

[der Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>> A quick check in cvs tells that hypot may be using the IEEE version,
>> which will by obvious reasons fail.
> yes, I think it is using the IEEE version, but why should it
> "obviously" fail?

VAX floats and IEEE floats are very different beasts.  Each one is
sign-magnitude-exponent with a base-2 biased exponent, but that's about
where the similarities end.  VAXen have nothing like IEEE quiet NANs
(VAX reserved operands are enough like signaling NANs for that analogy,
while far from perfect, to be reasonable).  VAXen also do not have
infinities, and I have a fuzzy memory saying that negative zero doesn't
work like the IEEE negative zero.  There are probably other differences
I'm forgetting, too.

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                mouse%rodents-montreal.org@localhost
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B