From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1423367AbXDYSRi (ORCPT ); Wed, 25 Apr 2007 14:17:38 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1423365AbXDYSRi (ORCPT ); Wed, 25 Apr 2007 14:17:38 -0400 Received: from gw.goop.org ([64.81.55.164]:58015 "EHLO mail.goop.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1423359AbXDYSRh (ORCPT ); Wed, 25 Apr 2007 14:17:37 -0400 Message-ID: <462F9B64.3020107@goop.org> Date: Wed, 25 Apr 2007 11:18:12 -0700 From: Jeremy Fitzhardinge User-Agent: Thunderbird 1.5.0.10 (X11/20070302) MIME-Version: 1.0 To: "Eric W. Biederman" CC: "H. Peter Anvin" , Andrew Morton , Andi Kleen , Zachary Amsden , Linux Kernel Mailing List Subject: Re: [PATCH] i386: For debugging, make the initial page table setup less forgiving. References: <200704132149.l3DLnvUY012097@tazenda.hos.anvin.org> <20070425044813.f83dde4e.akpm@linux-foundation.org> <462F73A0.40003@zytor.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Eric W. Biederman wrote: > "H. Peter Anvin" writes: > > >> Andrew Morton wrote: >> >> >>> This patch causes oopses after a minute or so running LTP's >>> >>> ./testcases/bin/growfiles -W gf16 -b -e 1 -i 0 -L 120 -u -g 4090 -T 100 -t >>> >> 408990 -l -C 10 -c 1000 -S 10 -f Lgf02_ >> >>> on everyone's favoutite Vaio, configured with >>> http://userweb.kernel.org/~akpm/config-sony.txt >>> >>> >> *BLINK* >> >> This patch only affects the initial page tables, which should have been >> thrown out *way* long ago at this point. >> > > Yes. I noticed this was happening a few days ago. > I must not have mentioned it loudly enough. > I might have introduced it as part of the paravirt_ops patches. When setting up pagetables under Xen, we need to make sure we preserve the initial mappings Xen put in place (it starts the VM with paging enabled, and a sane initial pagetable). This may have leaked over into native if it doesn't replace an existing entry. Hm, but it should be overwriting small mappings with large ones. Maybe I overlooked that. > I'm halfway through putting together a patchset to address a > bunch of these. > > I haven't yet resolved how I want to allocate the pages for the > identity mapping of the page table yet. I can't use the bootmem > allocate as it exists because that assumes the page is mapped > into the address space already. > Are you going to clash horribly with the paravirt pagetable setup? We should probably coordinate if so. J