From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754047AbXDMPfN (ORCPT ); Fri, 13 Apr 2007 11:35:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754048AbXDMPfN (ORCPT ); Fri, 13 Apr 2007 11:35:13 -0400 Received: from terminus.zytor.com ([192.83.249.54]:52737 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754047AbXDMPfL (ORCPT ); Fri, 13 Apr 2007 11:35:11 -0400 Message-ID: <461FA2EE.5090103@zytor.com> Date: Fri, 13 Apr 2007 08:34:06 -0700 From: "H. Peter Anvin" User-Agent: Thunderbird 1.5.0.10 (X11/20070302) MIME-Version: 1.0 To: Keir Fraser CC: Andi Kleen , Zachary Amsden , Andrew Morton , Virtualization Mailing List , Chris Wright , David Rientjes , Hugh Dickins , Linux Kernel Mailing List Subject: Re: [PATCH 0/4] i386 - pte update optimizations References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Keir Fraser wrote: > On 13/4/07 13:27, "Andi Kleen" wrote: > >>> LOCKed instruction suck really badly on the netburst microarchitecture (like >>> factor of 10x, or not far off). I think it's probably because of their side >>> effect of serialising memory accesses, causing horrible pipeline stalls. >> Unfortunately they tend to be HyperThreaded usually (except for early ones >> and Celerons) and need the LOCK anyways. > > Fair point, although quite a few people disable HT. Note we're talking a UP-only hack. -hpa