From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754133AbYL2XSW (ORCPT ); Mon, 29 Dec 2008 18:18:22 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753409AbYL2XSN (ORCPT ); Mon, 29 Dec 2008 18:18:13 -0500 Received: from smtp1.linux-foundation.org ([140.211.169.13]:51463 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753345AbYL2XSN (ORCPT ); Mon, 29 Dec 2008 18:18:13 -0500 Date: Mon, 29 Dec 2008 15:17:31 -0800 From: Andrew Morton To: Nick Piggin Cc: ebiederm@xmission.com, linux-kernel@vger.kernel.org, tglx@linutronix.de, mingo@elte.hu, ijc@hellion.org.uk Subject: Re: early fixmap causes kmap breakage Message-Id: <20081229151731.2a2c5a02.akpm@linux-foundation.org> In-Reply-To: <20081218211543.GB10681@wotan.suse.de> References: <20081218211543.GB10681@wotan.suse.de> X-Mailer: Sylpheed version 2.2.4 (GTK+ 2.8.20; i486-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 18 Dec 2008 22:15:43 +0100 Nick Piggin wrote: > Hi, > > I've debugged a problem where i386+pae systems with more than a few CPUs > blow up at boot in the kmap_atomic code. ping? > The problem is that the kmap_atomic pte pages all need to be contiguous > memory because the pte is calculated via the first kmap pte page + an > offset (so as not to have to walk the page tables every time). > > The fixmap setup code crudely allocates contiguous pte pages, which is fine, > but if it finds an already populated pmd entry, then it will not switch it > to a new, contiguous pte page. So the early fixmap introduces a discontig > page table right in the middle of the kmap atomic fixmaps. > > Commenting out the eaarly fixmap setup in head_32.S gets everything working > properly. What would be the best way to fix this? Could we put the early > fixmap page table in initdata, and then have the fixmap setup proper first > clear its corresponding pmd entry? How come users/testers aren't reporting this?