Path: utzoo!attcan!uunet!samsung!zaphod.mps.ohio-state.edu!rpi!leah!bingvaxu!kym
From: kym@bingvaxu.cc.binghamton.edu (R. Kym Horsell)
Newsgroups: comp.arch
Subject: Re: int x int -> long for * (or is it 32x32->64)
Keywords: arithmetic,arbitrary precision,benchmark,modular arithmetic
Message-ID: <4025@bingvaxu.cc.binghamton.edu>
Date: 15 Sep 90 04:00:30 GMT
References: <3984@bingvaxu.cc.binghamton.edu> <41425@mips.mips.COM> <41497@mips.mips.COM> <69434@sgi.sgi.com>
Reply-To: kym@bingvaxu.cc.binghamton.edu.cc.binghamton.edu (R. Kym Horsell)
Organization: SUNY Binghamton, NY
Lines: 22

In article <69434@sgi.sgi.com> vjs@rhyolite.wpd.sgi.com (Vernon Schryver) writes:
\\\
>Not to disagree, I have many times bemoaned the lack of some kind of fast
>carry indication from the MIPS add instruction while looking for ways to
>make the TCP/IP checksum faster.  If a single instruction could could compute
>	    32-bit + 32-bit -> 32-bit with carry-bit,
>you could substantially better the number instructions in all of the
>algorithms I've found.  The savings of having such a facility, counted in
\\\

I'm afraid the hard-hearted answer has got to be along the lines of:

"What proportion of time is spent in the checksum code, & by what factor
is it increased by not having the convient carry"?

If the answer is a number greater than EPS (your tollerance for such things) 
will you then _personally_ pay me four figures to lay it out in Si?

If the answer is ''no'' I'm afraid you have to stick in the extra 
instructions.

-Kym Horsell