From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754483Ab0DSLNL (ORCPT ); Mon, 19 Apr 2010 07:13:11 -0400 Received: from mx1.redhat.com ([209.132.183.28]:12829 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753946Ab0DSLNJ (ORCPT ); Mon, 19 Apr 2010 07:13:09 -0400 Message-ID: <4BCC3AC2.8050601@redhat.com> Date: Mon, 19 Apr 2010 14:13:06 +0300 From: Avi Kivity User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.0.4-1.fc12 Thunderbird/3.0.4 MIME-Version: 1.0 To: Peter Zijlstra CC: Glauber Costa , kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Jeremy Fitzhardinge , Marcelo Tosatti , Zachary Amsden Subject: Re: [PATCH 1/5] Add a global synchronization point for pvclock References: <1271356648-5108-1-git-send-email-glommer@redhat.com> <1271356648-5108-2-git-send-email-glommer@redhat.com> <4BCA026D.3070309@redhat.com> <4BCA02D1.2020608@redhat.com> <1271673836.1674.757.camel@laptop> <4BCC34DF.6030702@redhat.com> <1271674575.1674.793.camel@laptop> In-Reply-To: <1271674575.1674.793.camel@laptop> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 04/19/2010 01:56 PM, Peter Zijlstra wrote: > > >>> Right, do bear in mind that the x86 implementation of atomic64_read() is >>> terrifyingly expensive, it is better to not do that read and simply use >>> the result of the cmpxchg. >>> >>> >>> >> atomic64_read() _is_ cmpxchg64b. Are you thinking of some clever >> implementation for smp i386? >> > > No, what I was suggesting was to rewrite that loop no to need the > initial read but use the cmpxchg result of the previous iteration. > > Something like: > > u64 last = 0; > > /* more stuff */ > > do { > if (ret< last) > return last; > last = cmpxchg64(&last_value, last, ret); > } while (last != ret); > > That only has a single cmpxchg8 in there per loop instead of two > (avoiding the atomic64_read() one). > Still have two cmpxchgs in the common case. The first iteration will fail, fetching last_value, the second will work. It will be better when we have contention, though, so it's worthwhile. -- error compiling committee.c: too many arguments to function