Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!ucsd!hub.ucsb.edu!tom
From: tom@bears.ucsb.edu (Tom Weinstein)
Newsgroups: comp.arch
Subject: Re: more register flames
Message-ID: <7890@hub.ucsb.edu>
Date: 20 Dec 90 18:38:30 GMT
References: <441@inetg1.arco.com>
Sender: news@hub.ucsb.edu
Reply-To: tom@bears.ucsb.edu
Distribution: usa
Organization: Silicon Graphics Inc.
Lines: 18
In-reply-to: dprrhb@inetg1.Arco.Com's message of 20 Dec 90 15:37:53 GMT

In article <441@inetg1.arco.com>, dprrhb@inetg1.Arco.Com (Reginald H. Beardsley) writes:

>  With respect to all the argument about having large numbers of registers in
> hardware, do any of you read anything more substantial than net news?

>   From "Computer Architecture, A Quantative Approach" by J.L Hennessy and
> D.A. Patterson, p. E-15:
>  "an implementation of SPARC can have as few a 40 physical registers
>   and as many as 520, although most have 128 to 136, so far"

Ack.  This is so completely out of context that I have to say something.
This is taken right out of the middle of a discussion of register
windows.  All this really says is that SPARC can have a variable number
of register windows, each 32 registers wide.

--
He is Bob...eager for fun.         | Tom Weinstein  tom@bears.ucsb.edu
He wears a smile... Everybody run! |                tweinst@polyslo.calpoly.edu