From owner-freebsd-stable@FreeBSD.ORG Thu May 4 09:17:39 2006 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 29EF516A43D for ; Thu, 4 May 2006 09:17:39 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7D52643D6E for ; Thu, 4 May 2006 09:17:35 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id EF5A946D2D; Thu, 4 May 2006 05:17:34 -0400 (EDT) Date: Thu, 4 May 2006 10:17:34 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: Chuck Swiger In-Reply-To: <4458CE13.6060804@mac.com> Message-ID: <20060504101557.Q17611@fledge.watson.org> References: <20060503113955.U1147@ganymede.hub.org> <4458CE13.6060804@mac.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-stable@freebsd.org Subject: Re: Hyperthreading in 6.x ... still frowned upon? X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 04 May 2006 09:17:40 -0000 On Wed, 3 May 2006, Chuck Swiger wrote: >> >> Is it still something that I should disable, and, if so, how in 6.x? > > You should test it for the workloads you have, but most of the time, HT > isn't especially helpful. AMD64 CPUs come in dual-core format rather than > HT-enabled. If you've seen "HT" or "HTT" applied to an AMD system, it's > likely an abbreviation for "HyperTransport" or "HyperTransport Technology". The workloads I've seen the best improvement in performance for HTT have been ones involving a healthy blend of floating point and integer instruction mixes, or ones with a lot of memory stalls. Something worth remembering is that HTT hardware has, in fact, improved since earlier CPUs, and I've seen HTT go from a net loss in some critical workloads to breakeven or win. My recommendation would be to evaluate the performance impact of HTT against your specific workload and see what impact it has, but not be surprised if it doesn't help. Robert N M Watson