From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752285Ab1LUGlJ (ORCPT ); Wed, 21 Dec 2011 01:41:09 -0500 Received: from snt0-omc4-s34.snt0.hotmail.com ([65.55.90.237]:9765 "EHLO snt0-omc4-s34.snt0.hotmail.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750733Ab1LUGlI convert rfc822-to-8bit (ORCPT ); Wed, 21 Dec 2011 01:41:08 -0500 X-Greylist: delayed 371 seconds by postgrey-1.27 at vger.kernel.org; Wed, 21 Dec 2011 01:41:07 EST Message-ID: X-Originating-IP: [50.47.182.211] From: Yuhong Bao To: , CC: , , Subject: RE: x86-32 PAE-enabled kernel fails to boot with 64GB of RAM Date: Tue, 20 Dec 2011 22:34:55 -0800 Importance: Normal In-Reply-To: <4EDE5B36.2000203@kernel.org> References: <20111206102628.GC1342@tiehlicka.suse.cz>,<4EDE5B36.2000203@kernel.org> Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT MIME-Version: 1.0 X-OriginalArrivalTime: 21 Dec 2011 06:34:56.0212 (UTC) FILETIME=[A77B4140:01CCBFAA] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > On 12/06/2011 02:26 AM, Michal Hocko wrote: > > > > Quoting from Documentation/vm/highmem.txt: > > " > > The general recommendation is that you don't use more than 8GiB on a > > 32-bit machine - although more might work for you and your workload, > > you're pretty much on your own - don't expect kernel developers to > > really care much if things come apart. > > " > > > > Still, I suspect the failure to cap the memory space... it still > *should* at least limp along. But yes, at some point the wheels really > will just come off the bus. > > Does specifying a kernel"mem=" option (thereby capping the memory > artificially) work? FYI Windows when set to 3G/1G VA split with PAE enabled caps to 16GB of RAM. Yuhong Bao