From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759861AbZCPSU0 (ORCPT ); Mon, 16 Mar 2009 14:20:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755162AbZCPSUL (ORCPT ); Mon, 16 Mar 2009 14:20:11 -0400 Received: from terminus.zytor.com ([198.137.202.10]:33640 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754297AbZCPSUK (ORCPT ); Mon, 16 Mar 2009 14:20:10 -0400 Message-ID: <49BE9756.6070401@zytor.com> Date: Mon, 16 Mar 2009 11:15:50 -0700 From: "H. Peter Anvin" User-Agent: Thunderbird 2.0.0.19 (X11/20090105) MIME-Version: 1.0 To: Jeremy Fitzhardinge CC: Yinghai Lu , Ingo Molnar , Linux Kernel Mailing List Subject: Re: [crash] Re: Latest brk patchset References: <49BC413B.5020104@zytor.com> <49BC4CAC.202@goop.org> <49BC4DB6.9070403@zytor.com> <49BCA03D.3020605@goop.org> <20090315203802.GA14625@elte.hu> <49BD70EF.7010204@goop.org> <20090315212854.GA23960@elte.hu> <49BD8F15.4020301@goop.org> <20090316085402.GC1062@elte.hu> <49BE7A84.2030503@goop.org> <49BE84D6.3010006@kernel.org> <49BE9248.6050707@goop.org> In-Reply-To: <49BE9248.6050707@goop.org> 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 Jeremy Fitzhardinge wrote: > Yinghai Lu wrote: >> could be max_pfn_mapped change in head_32.S that reduce mapping range >> to _end only. >> > > Yes, I would say you're right. Trimming the mapping to _end only leaves > the space under the kernel avaliable for allocating pagetable from e820 > space. In this case we've got DEBUG_PAGEALLOC enabled, which inhibits > the use of PSE, and BIOS corruption checking on, which eats a chunk of > low memory. In combination, there's only 0x8f000 bytes free below the > kernel, and it needs 0xe1000 to allocate for pagetables. > > Reverting 2bd2753ff46346543ab92e80df9d96366e21baa5 fixes the problem for > me, though it does result in a kernel with a 73MB BSS... > I did kind of think that was a bit fishy, but I somehow assumed you'd taken the consequences into account. Obviously not. The 73 MB brk isn't a problem per se as it will not be used in most systems. This, in fact, is the whole reason for using brk and not just bss: it allows unused brk to be reclaimed at a relatively early stage. -hpa