From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from m7.gw.fujitsu.co.jp ([10.0.50.77]) by fgwmail6.fujitsu.co.jp (Fujitsu Gateway) with ESMTP id k137ZCw8003763 for ; Fri, 3 Feb 2006 16:35:12 +0900 (envelope-from kamezawa.hiroyu@jp.fujitsu.com) Received: from s11.gw.fujitsu.co.jp by m7.gw.fujitsu.co.jp (8.12.10/Fujitsu Domain Master) id k137ZBWg031715 for ; Fri, 3 Feb 2006 16:35:11 +0900 (envelope-from kamezawa.hiroyu@jp.fujitsu.com) Received: from s11.gw.fujitsu.co.jp (s11 [127.0.0.1]) by s11.gw.fujitsu.co.jp (Postfix) with ESMTP id 97B94F8C3A for ; Fri, 3 Feb 2006 16:35:11 +0900 (JST) Received: from fjm506.ms.jp.fujitsu.com (fjm506.ms.jp.fujitsu.com [10.56.99.86]) by s11.gw.fujitsu.co.jp (Postfix) with ESMTP id 528E3F8527 for ; Fri, 3 Feb 2006 16:35:11 +0900 (JST) Received: from [127.0.0.1] (fjmscan502.ms.jp.fujitsu.com [10.56.99.142])by fjm506.ms.jp.fujitsu.com with ESMTP id k137Ytxf002849 for ; Fri, 3 Feb 2006 16:34:57 +0900 Message-ID: <43E307DB.3000903@jp.fujitsu.com> Date: Fri, 03 Feb 2006 16:35:55 +0900 From: KAMEZAWA Hiroyuki MIME-Version: 1.0 Subject: [RFC] pearing off zone from physical memory layout [0/10] Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: linux-mm List-ID: Hi, This series of patches remove members from zone, which depends on physical memory layout, zone_start_pfn, spanned_pages, zone_mem_map against 2.6.16-rc1. By this, zone's meaning will be changed from "a range of memory to be used in a same manner" to "a group of memory to be used in a same manner". Now, the kernel and memmap became sparse if SPARSEMEM=y, but zone is considered as a range of memory. memory-hot-add adds memory to HIGHMEM, but a zone is considered as a range. This means memory layout (after hot add) like this is ok, NORMAL | NORMAL | HIGHMEM | HIGHMEM. but this is insane NORMAL | HIGHMEM | NORMAL | HIGHMEM. (we can do, but insane) IMHO, a zone is an unit of allocation/reclaim of same type of pages. I think that a zone should be defined by its usage/purpose not by physical memory layout. Some codes which wants to walk through all pages in a zone is supported by for_each_page_in_zone() macro. I tested this on my desktop machine a little, but I know this patch needs more work on some arch (ia64 etc...). comments ? -- Kame -- 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