From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-gw1.turkuamk.fi (mail-gw1.turkuamk.fi [195.148.208.125]) by ozlabs.org (Postfix) with ESMTP id 144BC67B7E for ; Fri, 1 Sep 2006 18:58:38 +1000 (EST) Message-ID: <44F7F376.4030203@kolumbus.fi> Date: Fri, 01 Sep 2006 11:46:46 +0300 From: =?ISO-8859-1?Q?Mika_Penttil=E4?= MIME-Version: 1.0 To: Mel Gorman Subject: Re: [PATCH 4/6] Have x86_64 use add_active_range() and free_area_init_nodes References: <20060821134518.22179.46355.sendpatchset@skynet.skynet.ie> <20060821134638.22179.44471.sendpatchset@skynet.skynet.ie> <20060831154903.GA7011@skynet.ie> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: akpm@osdl.org, tony.luck@intel.com, linuxppc-dev@ozlabs.org, ak@suse.de, bob.picco@hp.com, Linux Kernel Mailing List , Linux Memory Management List , Keith Mannthey List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , > > Right, it's all very clear now. At some point in the future, I'd like > to visit why SPARSEMEM-based hot-add is not always used but it's a > separate issue. > >> The add areas >> are marked as RESERVED during boot and then later onlined during add. > > That explains the reserve_bootmem_node() > But pages are marked reserved by default. You still have to alloc the bootmem map for the the whole node range, including reserve hot add areas and areas beyond e820-end-of-ram. So all the areas are already reserved, until freed. --Mika From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932339AbWIAIpX (ORCPT ); Fri, 1 Sep 2006 04:45:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932354AbWIAIpW (ORCPT ); Fri, 1 Sep 2006 04:45:22 -0400 Received: from mail-gw1.turkuamk.fi ([195.148.208.125]:40374 "EHLO mail-gw1.turkuamk.fi") by vger.kernel.org with ESMTP id S932339AbWIAIpV (ORCPT ); Fri, 1 Sep 2006 04:45:21 -0400 Message-ID: <44F7F376.4030203@kolumbus.fi> Date: Fri, 01 Sep 2006 11:46:46 +0300 From: =?ISO-8859-1?Q?Mika_Penttil=E4?= User-Agent: Thunderbird 1.5.0.5 (X11/20060808) MIME-Version: 1.0 To: Mel Gorman Cc: Keith Mannthey , akpm@osdl.org, tony.luck@intel.com, Linux Memory Management List , ak@suse.de, bob.picco@hp.com, Linux Kernel Mailing List , linuxppc-dev@ozlabs.org Subject: Re: [PATCH 4/6] Have x86_64 use add_active_range() and free_area_init_nodes References: <20060821134518.22179.46355.sendpatchset@skynet.skynet.ie> <20060821134638.22179.44471.sendpatchset@skynet.skynet.ie> <20060831154903.GA7011@skynet.ie> In-Reply-To: X-MIMETrack: Itemize by SMTP Server on marconi.hallinto.turkuamk.fi/TAMK(Release 6.5.4FP2 HF462|May 23, 2006) at 01.09.2006 11:45:07, Serialize by Router on marconi.hallinto.turkuamk.fi/TAMK(Release 6.5.4FP2 HF462|May 23, 2006) at 01.09.2006 11:45:07, Serialize complete at 01.09.2006 11:45:07, Itemize by SMTP Server on notes.hallinto.turkuamk.fi/TAMK(Release 6.5.4FP2 HF462|May 23, 2006) at 01.09.2006 11:45:07, Serialize by Router on notes.hallinto.turkuamk.fi/TAMK(Release 6.5.4FP2 HF462|May 23, 2006) at 01.09.2006 11:45:19, Serialize complete at 01.09.2006 11:45:19 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > > Right, it's all very clear now. At some point in the future, I'd like > to visit why SPARSEMEM-based hot-add is not always used but it's a > separate issue. > >> The add areas >> are marked as RESERVED during boot and then later onlined during add. > > That explains the reserve_bootmem_node() > But pages are marked reserved by default. You still have to alloc the bootmem map for the the whole node range, including reserve hot add areas and areas beyond e820-end-of-ram. So all the areas are already reserved, until freed. --Mika From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <44F7F376.4030203@kolumbus.fi> Date: Fri, 01 Sep 2006 11:46:46 +0300 From: =?ISO-8859-1?Q?Mika_Penttil=E4?= MIME-Version: 1.0 Subject: Re: [PATCH 4/6] Have x86_64 use add_active_range() and free_area_init_nodes References: <20060821134518.22179.46355.sendpatchset@skynet.skynet.ie> <20060821134638.22179.44471.sendpatchset@skynet.skynet.ie> <20060831154903.GA7011@skynet.ie> In-Reply-To: Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: owner-linux-mm@kvack.org Return-Path: To: Mel Gorman Cc: Keith Mannthey , akpm@osdl.org, tony.luck@intel.com, Linux Memory Management List , ak@suse.de, bob.picco@hp.com, Linux Kernel Mailing List , linuxppc-dev@ozlabs.org List-ID: > > Right, it's all very clear now. At some point in the future, I'd like > to visit why SPARSEMEM-based hot-add is not always used but it's a > separate issue. > >> The add areas >> are marked as RESERVED during boot and then later onlined during add. > > That explains the reserve_bootmem_node() > But pages are marked reserved by default. You still have to alloc the bootmem map for the the whole node range, including reserve hot add areas and areas beyond e820-end-of-ram. So all the areas are already reserved, until freed. --Mika -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org