From owner-freebsd-smp@FreeBSD.ORG  Fri Apr  4 11:54:10 2003
Return-Path: <owner-freebsd-smp@FreeBSD.ORG>
Delivered-To: freebsd-smp@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP
	id 1C0ED37B408; Fri,  4 Apr 2003 11:54:10 -0800 (PST)
Received: from phoenix.infradead.org (phoenix.infradead.org [195.224.96.167])
	by mx1.FreeBSD.org (Postfix) with ESMTP
	id 9C63443FA3; Fri,  4 Apr 2003 11:54:07 -0800 (PST)
	(envelope-from hch@infradead.org)
Received: from hch by phoenix.infradead.org with local (Exim 4.10)
	id 191XGH-0007jL-00; Fri, 04 Apr 2003 20:54:05 +0100
Date: Fri, 4 Apr 2003 20:54:05 +0100
From: Christoph Hellwig <hch@infradead.org>
To: Garrett Rooney <rooneg@electricjellyfish.net>
Message-ID: <20030404205405.A29698@infradead.org>
References: <XFMail.20030404144550.jhb@FreeBSD.org>
	<3E8DE275.6040000@electricjellyfish.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <3E8DE275.6040000@electricjellyfish.net>;
	from rooneg@electricjellyfish.net on Fri, Apr 04, 2003 at 02:52:21PM -0500
cc: freebsd-smp@FreeBSD.org
cc: John Baldwin <jhb@FreeBSD.org>
Subject: Re: atomic_dec_and_test() in FreeBSD?
X-BeenThere: freebsd-smp@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: FreeBSD SMP implementation group <freebsd-smp.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-smp>,
	<mailto:freebsd-smp-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-smp>
List-Post: <mailto:freebsd-smp@freebsd.org>
List-Help: <mailto:freebsd-smp-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-smp>,
	<mailto:freebsd-smp-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Fri, 04 Apr 2003 19:54:10 -0000

On Fri, Apr 04, 2003 at 02:52:21PM -0500, Garrett Rooney wrote:
> yeah, the default implementation of this function in APR uses mutexes to 
> ensure that the increment is atomic, but craig is looking for a way to 
> implement the 'platform specific faster way to do it without a mutex'.  
> worst case, we can just use the default impl on freebsd 4.x and use 
> atomic_cmpset_int on 5.x.

Just FYI, the i386 atomic.h from Linux doesn't rely on any kernel
features so you could in theory just copy it over.  But I guess the
licensing stand against this.