From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762454AbYCDBF1 (ORCPT ); Mon, 3 Mar 2008 20:05:27 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754793AbYCDBFU (ORCPT ); Mon, 3 Mar 2008 20:05:20 -0500 Received: from mga05.intel.com ([192.55.52.89]:12027 "EHLO fmsmga101.fm.intel.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1754154AbYCDBFT (ORCPT ); Mon, 3 Mar 2008 20:05:19 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.25,441,1199692800"; d="scan'208";a="528177268" Message-ID: <47CCA03D.9040007@linux.intel.com> Date: Mon, 03 Mar 2008 17:05:01 -0800 From: Arjan van de Ven User-Agent: Thunderbird 2.0.0.12 (Windows/20080213) MIME-Version: 1.0 To: Linus Torvalds CC: Ingo Molnar , hans.rosenfeld@amd.com, linux-kernel@vger.kernel.org, Thomas Gleixner , "H. Peter Anvin" Subject: Re: bisected boot regression post 2.6.25-rc3.. please revert References: <20080301105646.2c8620d9@laptopd505.fenrus.org> <20080303074620.GC5934@elte.hu> <20080303091304.GA17911@elte.hu> <47CC2A3D.1000307@linux.intel.com> <20080303174009.GA19131@elte.hu> <47CC451A.2060501@linux.intel.com> <47CC74E6.60306@linux.intel.com> In-Reply-To: <47CC74E6.60306@linux.intel.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Arjan van de Ven wrote: > Linus Torvalds wrote: >> >> On Mon, 3 Mar 2008, Arjan van de Ven wrote: >>> interesting observation: if I turn the macros into inlines... the >>> difference >>> goes away. >>> >>> I'm half tempted to just do it as inline period ... any objections ? >> >> Yes, I object. I want to understand why it would matter. If this is a >> compiler bug, it's a really rather bad one. And if it's just some >> stupid bug in our pmd_bad() macro, I still want to know what the >> problem was. >> >> Can you compile both ways and look at what changed at the offending >> site (which is apparently "follow_page()")? >> >> And do you have some odd compiler version? > > it's the F9 gcc, so yeah it's really new. > > I'm staring at the disassembly and it's quite different but > follow_page() is rather large; > trying to make a smaller testcase now sadly a more isolated testcase doesn't show the same behavior yet; so it's some fun interaction ;( more staring at the assembly for me