From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932382AbWGYBly (ORCPT ); Mon, 24 Jul 2006 21:41:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751352AbWGYBly (ORCPT ); Mon, 24 Jul 2006 21:41:54 -0400 Received: from colin.muc.de ([193.149.48.1]:40968 "EHLO mail.muc.de") by vger.kernel.org with ESMTP id S1751203AbWGYBlx (ORCPT ); Mon, 24 Jul 2006 21:41:53 -0400 Date: 25 Jul 2006 03:41:51 +0200 Date: Tue, 25 Jul 2006 03:41:51 +0200 From: Andi Kleen To: "H. Peter Anvin" Cc: Rusty Russell , Andrew Morton , Keir Fraser , Jeremy Fitzhardinge , Zachary Amsden , Pratap , Chris Wright , lkml - Kernel Mailing List Subject: Re: [PATCH 6/6] cpuid neatening. Message-ID: <20060725014151.GB91138@muc.de> References: <1153526643.13699.18.camel@localhost.localdomain> <1153526798.13699.23.camel@localhost.localdomain> <1153527194.13699.34.camel@localhost.localdomain> <44C55E8F.2020200@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <44C55E8F.2020200@zytor.com> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jul 24, 2006 at 04:58:07PM -0700, H. Peter Anvin wrote: > Rusty Russell wrote: > >Roll all the cpuid asm into one __cpuid call. It's a little neater, > >and also means only one place to patch for paravirtualization. > > The whole point of those is to avoid the unnecessary write to memory and > pick it back up again. This patch reintroduces that ugliness. Modern gcc should optimize this when it is inlined. If it didn't most abstracted C++ code would be quite unhappy. Also as far as I know there is only a single time critical CPUID in the code and it ignores all output arguments. -Andi