From owner-freebsd-amd64@FreeBSD.ORG Fri Mar 12 11:46:01 2004 Return-Path: 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 802E416A4CE for ; Fri, 12 Mar 2004 11:46:01 -0800 (PST) Received: from host142.ipowerweb.com (host142.ipowerweb.com [66.235.193.61]) by mx1.FreeBSD.org (Postfix) with SMTP id 3F2E943D45 for ; Fri, 12 Mar 2004 11:46:01 -0800 (PST) (envelope-from valour@thejemreport.com) Received: (qmail 60803 invoked from network); 12 Mar 2004 19:45:20 -0000 Received: from unknown (HELO thejemreport.com) (66.67.130.234) by host142.ipowerweb.com with SMTP; 12 Mar 2004 19:45:20 -0000 Message-ID: <40521370.5030806@thejemreport.com> Date: Fri, 12 Mar 2004 14:45:52 -0500 From: Jem Matzan User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.6) Gecko/20040219 X-Accept-Language: en-us, en MIME-Version: 1.0 To: freebsd-amd64@freebsd.org References: <4051A841.9020205@thejemreport.com> <20040312181837.GD56805@funkthat.com> In-Reply-To: <20040312181837.GD56805@funkthat.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: Peer review of AMD64/FreeBSD article X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Mar 2004 19:46:01 -0000 John-Mark Gurney wrote: >Jem Matzan wrote this message on Fri, Mar 12, 2004 at 07:08 -0500: > > >>I've just finished writing this article comparing performance between an >>Athlon64 in 32-bit and 64-bit mode using FreeBSD: >> >>http://www.thejemreport.com/lab64/amd64vsi386.php >> >>(this is a temporary address which will later redirect to the published >>article) >> >> > >A comment on the article, you mention: >Curiously the Pentium4 times are impossible; according to the numbers for -j3 and -j4, it takes longer to execute the utility than it does to complete the entire process. This is either a bug in the time command (in relation to multithreading, perhaps), or something isn't reporting properly. > >You said that you enabled HyperThreading, and THIS is the reason for >the imposible times. Since w/ HT it appears as if there are two cpus >and two processes are running on each of these CPU's, you end up with >a user time that can be 2x real time. > >Hope this helps. > > > Thanks -- I'll change the article to reflect that. I was hoping someone would shed some light on why the user time was off like that... -Jem