From owner-freebsd-current@FreeBSD.ORG  Sat Feb  4 07:47:31 2006
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
X-Original-To: current@freebsd.org
Delivered-To: freebsd-current@FreeBSD.ORG
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id 4E38716A422
	for <current@freebsd.org>; Sat,  4 Feb 2006 07:47:31 +0000 (GMT)
	(envelope-from ivoras@fer.hr)
Received: from pinus.cc.fer.hr (pinus.cc.fer.hr [161.53.73.18])
	by mx1.FreeBSD.org (Postfix) with ESMTP id B853443D48
	for <current@freebsd.org>; Sat,  4 Feb 2006 07:47:30 +0000 (GMT)
	(envelope-from ivoras@fer.hr)
Received: from [161.53.72.113] (lara.cc.fer.hr [161.53.72.113])
	by pinus.cc.fer.hr (8.12.2/8.12.2) with ESMTP id k147lGFx004254;
	Sat, 4 Feb 2006 08:47:16 +0100 (MET)
Message-ID: <43E45BD8.7070708@fer.hr>
Date: Sat, 04 Feb 2006 08:46:32 +0100
From: Ivan Voras <ivoras@fer.hr>
User-Agent: Mozilla Thunderbird 1.0.6 (X11/20050921)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Andrew Gallatin <gallatin@cs.duke.edu>
References: <17379.56708.421007.613310@grasshopper.cs.duke.edu>
In-Reply-To: <17379.56708.421007.613310@grasshopper.cs.duke.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: current@freebsd.org
Subject: Re: machdep.cpu_idle_hlt and SMP perf?
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
	<freebsd-current.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>, 
	<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Sat, 04 Feb 2006 07:47:31 -0000

Andrew Gallatin wrote:
> Why dooes machdep.cpu_idle_hlt=1 drop my 10GbE network rx
> performance by a considerable amount (7.5Gbs -> 5.5Gbs)?

For what it's worth, here are results of unixbench's "context1" 
benchmark on an old P3 SMP with and without cpu_idle_hlt (6-stable):


x hlt=0
+ hlt=1
+--------------------------------------------------------------------------+
| +     ++   +                    +                   x  xx  x            x|
||_______M___A____________|                          |____M__A_______|     |
+--------------------------------------------------------------------------+
     N           Min           Max        Median           Avg        Stddev
x   5        346203        355234        347787      349229.8     3549.1559
+   5        322710        337269        325631      327663.8     5622.9044
Difference at 95.0% confidence
	-21566 +/- 6857.28
	-6.1753% +/- 1.96354%
	(Student's t, pooled s = 4701.78)