Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-crg!hoptoad!gnu
From: gnu@hoptoad.uucp (John Gilmore)
Newsgroups: net.micro.68k,net.arch
Subject: Re: 68000 Memory Managment (Bechtolsheim patent)
Message-ID: <1086@hoptoad.uucp>
Date: Mon, 15-Sep-86 01:32:33 EDT
Article-I.D.: hoptoad.1086
Posted: Mon Sep 15 01:32:33 1986
Date-Received: Mon, 15-Sep-86 07:30:00 EDT
References: <508@elmgate.UUCP> <64@mit-prep.ARPA>, <15665@ucbvax.BERKELEY.EDU> <7123@utzoo.UUCP>
Organization: Nebula Consultants in San Francisco
Lines: 28
Xref: mnetor net.micro.68k:1225 net.arch:3047

Andreas Bechtolsheim holds US Patent # 4,527,232 which is on the Sun-2 MMU.

I am pretty sure that the idea of translating virtual addresses while
the row address is being presented to memory, is about as old as DRAMs.
I have never seen a design that hung static RAMs directly in the path
to memory, with no bypass, and with no cacheing or other auto-loading:
they are loaded by software instructions that reference particular
addresses or address spaces.  This idea could be original with Andy.
Another unique thing in his design is having two levels of static RAM
to cut down on the size of the RAMs required, while still having a small,
manageable page size.

Not having the patent in front of me, I don't know whether this is what
it claims as proprietary, or something else.  Typically the claims sections
are written in as broad a language as possible and then fought out in court
if anybody cares.

I don't know if they filed for patent protection in Canada.  I do know that
the patent is NOT owned by Stanford, but by Sun.

I advise MMU designers to get a copy of this patent.  Enforcing the patent
would be a dandy way for Sun to put a competitor out of business if it
ever became worth the bother.

	John Gilmore, Nebula Consultants
-- 
John Gilmore  {sun,ptsfa,lll-crg,ihnp4}!hoptoad!gnu   jgilmore@lll-crg.arpa
		     May the Source be with you!