Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!seismo!mcvax!jack
From: jack@mcvax.UUCP
Newsgroups: comp.arch
Subject: Re: register window machine questions
Message-ID: <7284@boring.mcvax.cwi.nl>
Date: Wed, 25-Feb-87 15:51:53 EST
Article-I.D.: boring.7284
Posted: Wed Feb 25 15:51:53 1987
Date-Received: Fri, 27-Feb-87 23:24:07 EST
References: <4376@columbia.UUCP>
Reply-To: jack@boring.UUCP (Jack Jansen)
Organization: AMOEBA project, CWI, Amsterdam
Lines: 22


Something that I would be very interested in is wether there
is any research done into machines with an unlimited register
file.
What I have in mind is a stackish machine, with an intelligent
stack cache that will delay writes around the stack pointer
(since there's a very good chance that they will never have
to be written at all).

The cache doesn't even have to be that intelligent: data should
be written out as soon as a write to a different address lands
in the same cache entry. Moreover, whenever the stack pointer
is decremented (on routine exit, for instance), you can scratch
all data above it.

One of the advantages of this seems to be that compilers become
simpler (no special cases for routines that run out of registers),
and routines that are recursive at the beginning would benefit
by not having (unnecessary) register-saves.
-- 
	Jack Jansen, jack@cwi.nl (or jack@mcvax.uucp)
	The shell is my oyster.