From owner-freebsd-hackers  Thu Dec 12 18:44:48 1996
Return-Path: <owner-hackers>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.4/8.8.4) id SAA05373
          for hackers-outgoing; Thu, 12 Dec 1996 18:44:48 -0800 (PST)
Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1])
          by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id SAA05365
          for <Hackers@FreeBSD.ORG>; Thu, 12 Dec 1996 18:44:38 -0800 (PST)
Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.3/CET-v2.1) with SMTP id CAA12318 for <Hackers@FreeBSD.ORG>; Fri, 13 Dec 1996 02:44:25 GMT
Date: Fri, 13 Dec 1996 11:44:25 +0900 (JST)
From: Michael Hancock <michaelh@cet.co.jp>
To: FreeBSD Hackers <Hackers@FreeBSD.ORG>
Subject: Implementing atomic DCAS on Intel for NBS
Message-ID: <Pine.SV4.3.95.961213112940.12235A-100000@parkplace.cet.co.jp>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

It doesn't look like the P5 has DCAS which is important for NBS because it
wants to atomically manipulate a structure and update it's version
number.

Here's some pseudo code:

int
DCAS(	int *p1, int *p2,
	int old1, int old2
	int new1, int new2)
{
	<begin atomic>	
	if ((*p1 == old1) && (*p2 == old2)) {
		*p1 = new1;
		*p2 = new2;
		return TRUE;
	} else {
		return FALSE;
	}
	<end atomic>
}

Is there an an assembler guru out there who can tell us how many
instructions would be necessary to implement this for a P5? 

Here's a link for an R4000 version:

http://www-dsg.stanford.edu/papers/non-blocking-osdi/node11.html

Regards,


Mike Hancock