From owner-freebsd-amd64@FreeBSD.ORG Wed Sep 21 11:45:13 2005 Return-Path: X-Original-To: amd64@FreeBSD.org Delivered-To: freebsd-amd64@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C631516A41F; Wed, 21 Sep 2005 11:45:13 +0000 (GMT) (envelope-from martin@gneto.com) Received: from av12-1-sn2.hy.skanova.net (av12-1-sn2.hy.skanova.net [81.228.8.185]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5E51143D45; Wed, 21 Sep 2005 11:45:12 +0000 (GMT) (envelope-from martin@gneto.com) Received: by av12-1-sn2.hy.skanova.net (Postfix, from userid 502) id 1C5EA3810C; Wed, 21 Sep 2005 13:45:11 +0200 (CEST) Received: from smtp4-2-sn2.hy.skanova.net (smtp4-2-sn2.hy.skanova.net [81.228.8.93]) by av12-1-sn2.hy.skanova.net (Postfix) with ESMTP id 0694737F2D; Wed, 21 Sep 2005 13:45:11 +0200 (CEST) Received: from [192.168.2.30] (h99n2fls34o985.telia.com [213.66.202.99]) by smtp4-2-sn2.hy.skanova.net (Postfix) with ESMTP id E0CF937E52; Wed, 21 Sep 2005 13:45:10 +0200 (CEST) Message-ID: <433147C0.8030900@gneto.com> Date: Wed, 21 Sep 2005 13:45:04 +0200 From: Martin Nilsson User-Agent: Mozilla Thunderbird 1.0.6 (Windows/20050716) X-Accept-Language: sv, en-us, en MIME-Version: 1.0 To: John Baldwin References: <200509201616.22475.jhb@FreeBSD.org> In-Reply-To: <200509201616.22475.jhb@FreeBSD.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: amd64@FreeBSD.org Subject: Re: Patch to use fence instructions X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 21 Sep 2005 11:45:13 -0000 John Baldwin wrote: > This patch changes the atomic operations and bus space barriers to use the x86 > fence instructions. Please test, thanks! > > http://www.FreeBSD.org/~jhb/patches/amd64_fences.patch What kind of performance improvements can we expect with this patch? Is it worthwile to compare performance on dualcore Pentium D with sysbench before and after this patch? Does it affect threads & mutex performance? Sysbench is a benchmark specially made to determine lowlevel performance important for MySQL and be found here: http://sysbench.sourceforge.net/ /Martin