Path: utzoo!utgpu!attcan!uunet!tektronix!orca!tekecs!frip!andrew
From: andrew@frip.gwd.tek.com (Andrew Klossner)
Newsgroups: comp.arch
Subject: Re: MC 88k question
Message-ID: <10206@tekecs.TEK.COM>
Date: 29 Jul 88 00:29:15 GMT
References: <197@quintus.UUCP>
Sender: andrew@tekecs.TEK.COM
Organization: Tektronix, Wilsonville, Oregon
Lines: 14

.> The MC88000 has add and subtract instructions which
.>   - ignore overflow, another set which
.>   - set a carry bit, and another set which
.>   - trap on overflow.
.> But the multiply instruction apparently only comes in a
.>   - quietly truncate to 32 bits
.> flavour.  So what is the recommended way of doing a 32x32 integer
.> multiplication with an overflow check?

Convert to double-precision floating point, multiply, and convert to
integer.  (This was also the way to do it on the National 32016.)

  -=- Andrew Klossner   (decvax!tektronix!tekecs!andrew)       [UUCP]
                        (andrew%tekecs.tek.com@relay.cs.net)   [ARPA]