Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759211AbXKLXwY (ORCPT ); Mon, 12 Nov 2007 18:52:24 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751913AbXKLXwQ (ORCPT ); Mon, 12 Nov 2007 18:52:16 -0500 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:42856 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751325AbXKLXwP (ORCPT ); Mon, 12 Nov 2007 18:52:15 -0500 Date: Mon, 12 Nov 2007 15:52:14 -0800 (PST) From: Christoph Lameter X-X-Sender: clameter@schroedinger.engr.sgi.com To: ak@suse.de cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, Mel Gorman , Andy Whitcroft Subject: x86_64: Make sparsemem/vmemmap the default memory model Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2206 Lines: 75 Use sparsemem as the only memory model for UP, SMP and NUMA. Measurements indicate that DISCONTIGMEM has a higher overhead than sparsemem. And FLATMEMs benefits are minimal. So I think its best to simply standardize on sparsemem. Results of page allocator tests (test can be had via git from the slab git tree on git.kernel.org. See the branch tests) Measurements in cycle counts. 1000 allocations were performed and then the average cycle count was calculated. Order FlatMem Discontig SparseMem 0 639 665 641 1 567 647 593 2 679 774 692 3 763 967 781 4 961 1501 962 5 1356 2344 1392 6 2224 3982 2336 7 4869 7225 5074 8 12500 14048 12732 9 27926 28223 28165 10 58578 58714 58682 If this patch is accepted then we can remove the code for discontig and flatmem support from x86_64. (Not sure if I got all the config settings right. Andy?) Signed-off-by: Christoph Lameter --- arch/x86/Kconfig.x86_64 | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) Index: linux-2.6/arch/x86/Kconfig.x86_64 =================================================================== --- linux-2.6.orig/arch/x86/Kconfig.x86_64 2007-11-12 15:17:50.721767735 -0800 +++ linux-2.6/arch/x86/Kconfig.x86_64 2007-11-12 15:21:16.659017509 -0800 @@ -390,28 +390,17 @@ config NUMA_EMU into virtual nodes when booted with "numa=fake=N", where N is the number of nodes. This is only useful for debugging. -config ARCH_DISCONTIGMEM_ENABLE - bool - depends on NUMA - default y - -config ARCH_DISCONTIGMEM_DEFAULT +config ARCH_SPARSEMEM_DEFAULT def_bool y - depends on NUMA config ARCH_SPARSEMEM_ENABLE def_bool y - depends on (NUMA || EXPERIMENTAL) select SPARSEMEM_VMEMMAP_ENABLE config ARCH_MEMORY_PROBE def_bool y depends on MEMORY_HOTPLUG -config ARCH_FLATMEM_ENABLE - def_bool y - depends on !NUMA - source "mm/Kconfig" config MEMORY_HOTPLUG_RESERVE - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/