Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!yale!cmcl2!acf4!tihor
From: tihor@acf4.NYU.EDU (Stephen Tihor)
Newsgroups: comp.arch
Subject: Re: New benchmark (source code)
Message-ID: <12780021@acf4.NYU.EDU>
Date: 10 Jan 90 00:33:00 GMT
References: <2132@sunset.MATH.UCLA.EDU>
Organization: New York University
Lines: 11

Curiously I ran this on a VMS VAX under VAX FORTRAN with full error checking enabled and optimizations disabled and got a reasonable integer overflow in the 
inner loop of INTMUL multiplying 

IMTMUL\I:       1
IMTMUL\J:       1
IMTMUL\K:       1
IMTMUL\M1(1,1): 235517118
IMTMUL\M2(1,1): 891310209

Those do look large for a 32 integer product.  Did I mess something up here?