Path: utzoo!mnetor!uunet!husc6!cmcl2!nrl-cmf!ames!oliveb!sun!gorodish!guy
From: guy@gorodish.Sun.COM (Guy Harris)
Newsgroups: comp.arch
Subject: Re: SPARC and multiprocessing
Message-ID: <51428@sun.uucp>
Date: 30 Apr 88 03:33:03 GMT
References: <1521@pt.cs.cmu.edu> <28200135@urbsdc> <4921@bloom-beacon.MIT.EDU> <1680@alliant.Alliant.COM>
Sender: news@sun.uucp
Lines: 19

> 	Sorry, Guy but you misunderstood the purpose of my posting.  I was not
> 	attempting to attack the SPARC.

No, I didn't.  I merely pointed out that you were, as far as I could tell,
making an unwarranted assumption at the start of your discussion; this
assumption unnecessarily colored the rest of your discussion.  David Emberson
of Sun has noted that there is nothing in the *architecture* that demands a
virtual cache.  Current *implementations* may make a virtual cache the best, or
only, way to get maximum performance, but that's a different matter.  If you
were, in fact, referring to the current chips, rather than the SPARC
architecture, I apologize.

>       I am simply attempting to see if anyone has given any thought the
>       the problem of using the SPARC with a virtual cache in a
>       multiprocessor.

David Emberson has also already given an answer to this question; the
answer is "yes, Sun has".  Unfortunately, as he indicated, he's not in a
position to discuss it in detail now.