From owner-freebsd-stable@FreeBSD.ORG  Tue Apr 13 00:16:23 2010
Return-Path: <owner-freebsd-stable@FreeBSD.ORG>
Delivered-To: freebsd-stable@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id CCBFC106564A;
	Tue, 13 Apr 2010 00:16:23 +0000 (UTC)
	(envelope-from mdpoole@troilus.org)
Received: from na3sys009aog103.obsmtp.com (na3sys009aog103.obsmtp.com
	[74.125.149.71])
	by mx1.freebsd.org (Postfix) with SMTP id 3C7DB8FC1A;
	Tue, 13 Apr 2010 00:16:22 +0000 (UTC)
Received: from source ([74.125.92.27]) by na3sys009aob103.postini.com
	([74.125.148.12]) with SMTP
	ID DSNKS8O31nXVMmfaY0n/+t/55VlIRf/CN1nW@postini.com;
	Mon, 12 Apr 2010 17:16:23 PDT
Received: by qw-out-2122.google.com with SMTP id 5so2165805qwi.7
	for <multiple recipients>; Mon, 12 Apr 2010 17:16:21 -0700 (PDT)
Received: by 10.224.97.137 with SMTP id l9mr1713568qan.222.1271117780259;
	Mon, 12 Apr 2010 17:16:20 -0700 (PDT)
Received: from graviton.troilus.org (pool-72-83-66-147.washdc.east.verizon.net
	[72.83.66.147])
	by mx.google.com with ESMTPS id 23sm3090371qyk.7.2010.04.12.17.16.19
	(version=TLSv1/SSLv3 cipher=RC4-MD5);
	Mon, 12 Apr 2010 17:16:19 -0700 (PDT)
From: Michael Poole <mdpoole@troilus.org>
To: Maho NAKATA <chat95@mac.com>
References: <t2ud763ac661004120231q44e9a4f7z5c0f11a31725deb@mail.gmail.com>
	<h2yea2d4a5b1004120658xba353f17w894d33e08558f3ea@mail.gmail.com>
	<87tyrghiio.fsf@troilus.org>
	<20100413.082856.690091871650385955.chat95@mac.com>
Date: Mon, 12 Apr 2010 20:16:18 -0400
In-Reply-To: <20100413.082856.690091871650385955.chat95@mac.com> (Maho
	NAKATA's message of "Tue, 13 Apr 2010 08:28:56 +0900 (JST)")
Message-ID: <87mxx8gqb1.fsf@troilus.org>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: adrian@freebsd.org, freebsd-stable@freebsd.org
Subject: Re: Only 70% of theoretical peak performance on FreeBSD 8/amd64,
	Corei7 920
X-BeenThere: freebsd-stable@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Production branch of FreeBSD source code <freebsd-stable.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>, 
	<mailto:freebsd-stable-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-stable>
List-Post: <mailto:freebsd-stable@freebsd.org>
List-Help: <mailto:freebsd-stable-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>,
	<mailto:freebsd-stable-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Tue, 13 Apr 2010 00:16:23 -0000

Maho NAKATA writes:

> From: Michael Poole <mdpoole@troilus.org>
> Subject: Re: Only 70% of theoretical peak performance on FreeBSD 8/amd64, Corei7 920
> Date: Mon, 12 Apr 2010 10:06:55 -0400
>
>> Nakata-san's theoretical performance numbers assume 4 to 4.2 operations
>> per core per cycle at the nominal (2.66 GHz, non-TurboBoost) clock rate.
>> (DGEMM is double precision, but I am not familiar enough with scientific
>> computing or with the Nehalem implementation of SSE to know why it is
>> four operations per cycle rather than two -- is it because double
>> precision counts as two FLOPs or is it because of multiple issue?)
>> TurboBoost runs up to 2.93 GHz on this CPU, so it doesn't fit either the
>> theoretical peak performance or the performance discrepancy very well.
>
> Hi Michael,
> I read http://www.intel.com/support/processors/sb/cs-023143.htm
> and TurboBoost on 920 is 2.80GHz.

Ah.  I was looking at http://ark.intel.com/Product.aspx?id=37147 .
Given a 2.80 GHz TurboBoost, the 44.8 GFLOPS theoretical performance
number makes sense.

I think the more important point is that TurboBoost on this CPU gives at
most a 10% speedup, so it cannot explain the 25% performance difference.

Michael