From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754749AbZHYKYf (ORCPT ); Tue, 25 Aug 2009 06:24:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752223AbZHYKYe (ORCPT ); Tue, 25 Aug 2009 06:24:34 -0400 Received: from e28smtp01.in.ibm.com ([59.145.155.1]:41798 "EHLO e28smtp01.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751050AbZHYKYd (ORCPT ); Tue, 25 Aug 2009 06:24:33 -0400 Date: Tue, 25 Aug 2009 15:54:22 +0530 From: "M. Mohan Kumar" To: Amerigo Wang Cc: Michael Ellerman , linux-kernel@vger.kernel.org, tony.luck@intel.com, linux-ia64@vger.kernel.org, Neil Horman , "Eric W. Biederman" , kamezawa.hiroyu@jp.fujitsu.com, Andi Kleen , akpm@linux-foundation.org, bernhard.walle@gmx.de, Fenghua Yu , Ingo Molnar , Anton Vorontsov Subject: Re: [Patch 6/8] powerpc: add CONFIG_KEXEC_AUTO_RESERVE Message-ID: <20090825102422.GA14591@in.ibm.com> Reply-To: mohan@in.ibm.com References: <20090821065637.4855.32234.sendpatchset@localhost.localdomain> <20090821065739.4855.19179.sendpatchset@localhost.localdomain> <1251121443.10645.56.camel@concordia> <20090824144542.GB11956@in.ibm.com> <4A938698.40302@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4A938698.40302@redhat.com> User-Agent: Mutt/1.5.19 (2009-01-05) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Aug 25, 2009 at 02:37:12PM +0800, Amerigo Wang wrote: > M. Mohan Kumar wrote: >> On Mon, Aug 24, 2009 at 11:44:03PM +1000, Michael Ellerman wrote: >> >>> On Fri, 2009-08-21 at 02:55 -0400, Amerigo Wang wrote: >>> >>> That aside, I don't see how this will be useful in practice, if it only >>> works for memory sizes over 4G? Or are we saying that people with less >>> than 4G don't need crash kernels? If we're not saying that, those users, >>> or those users' distros, still need to do some logic to work out if they >>> have < 4GB of memory and if so pick a crash kernel size. So why can't >>> they pick the size in the > 4GB case also? >>> >> >> True, I wanted to test the patch and when tested on a ppc64 machine which >> has RAM less than 4GB, I have to modify arch_default_crash_size routine to >> return 256MB (I didn't have a PPC64 machine with more than 4GB RAM handy). >> So its better to consider machines with less than 4GB RAM also. >> > > OK, how about 2G on ppc? Is it safe to reserve 256M when I have 2G? I would prefer 2G-4G 128MB. > >> PPC64 crashkernel base is always 32MB. So at least ppc64 code should have >> its own arch_default_crash_base to return 32MB to avoid the kernel warning >> message "Crash kernel location must be 0x2000000" >> > Hmm, good point, how about KDUMP_KERNELBASE? It looks fine for both ppc > and ppc64. Yes, you can use KDUMP_KERNELBASE for arch_default_crash_base Regards, M. Mohan Kumar