From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932599AbYDZAQc (ORCPT ); Fri, 25 Apr 2008 20:16:32 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758478AbYDZAPv (ORCPT ); Fri, 25 Apr 2008 20:15:51 -0400 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:55378 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1756714AbYDZAPu (ORCPT ); Fri, 25 Apr 2008 20:15:50 -0400 Message-Id: <20080426001548.829332000@polaris-admin.engr.sgi.com> User-Agent: quilt/0.46-1 Date: Fri, 25 Apr 2008 17:15:48 -0700 From: Mike Travis To: Ingo Molnar Cc: Andrew Morton , Thomas Gleixner , "H. Peter Anvin" , linux-kernel@vger.kernel.org Subject: [PATCH 00/11] x86: cleanup early per cpu variables/accesses v5-folded Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 1/11: Increase the limit of NR_CPUS to 4096 and introduce a boolean called "MAXSMP" which when set (e.g. "allyesconfig"), will set NR_CPUS = 4096 and NODES_SHIFT = 9 (512). Changed max setting for NODES_SHIFT from 15 to 9 to accurately reflect the real limit. 2/11: Introduce a new PER_CPU macro called "EARLY_PER_CPU". This is used by some per_cpu variables that are initialized and accessed before there are per_cpu areas allocated. Add a flag "arch_provides_topology_pointers" that indicates pointers to topology cpumask_t maps are available. Otherwise, use the function returning the cpumask_t value. This is useful if cpumask_t set size is very large to avoid copying data on to/off of the stack. 3/11: Restore the nodenumber field in the x86_64 pda. This field is slightly different than the x86_cpu_to_node_map mainly because it's a static indication of which node the cpu is on while the cpu to node map is a dyanamic mapping that may get reset if the cpu goes offline. This also simplifies the numa_node_id() macro. 4/11: Consolidate node_to_cpumask operations and remove the 256k byte node_to_cpumask_map. This is done by allocating the node_to_cpumask_map array after the number of possible nodes (nr_node_ids) is known. 5/11: Replace usages of MAX_NUMNODES with nr_node_ids in kernel/sched.c, where appropriate. This saves some allocated space as well as many wasted cycles going through node entries that are non-existent. 6/11: Changed some global definitions in drivers/base/cpu.c to static. 7/11: Remove 544k bytes from the kernel by removing the boot_cpu_pda array from the data section and allocating it during startup. 8/11: Increase performance for systems with large count NR_CPUS by limiting the range of the cpumask operators that loop over the bits in a cpumask_t variable. This removes a large amount of wasted cpu cycles. 9/11: Change references from for_each_cpu_mask to for_each_cpu_mask_ptr in all cases for x86_64 and generic code. 10/11: Change references from next_cpu to next_cpu_nr (or for_each_cpu_mask_ptr if applicable), in all cases for x86_64 and generic code. 11/11: Pass reference to cpumask variable in net/sunrpc/svc.c For inclusion into sched-devel/latest tree. Based on: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git + sched-devel/latest .../mingo/linux-2.6-sched-devel.git Signed-off-by: Mike Travis --- v5-folded: Folded in patches 7 - 11 above. Fixed some warnings in NONUMA config build. v4-folded: Folded in Kconfig changes to increase NR_CPU limit to 4096 and add new config option MAXSMP. v3-folded: Folded in follow on "fix" patches to consolidate changes into one place. Includes change to drivers/base/topology.c to fix s390 build error. Includes change to fix preemption warning when numa_node_id is used. checkpatch.pl errors/warnings checked and fixed where possible. v2: remerged PATCH 2/2 with latest x86.git/latest and sched-devel/latest, rebuilt and retested 4k-akpm2, 4k-defconfig, nonuma, & nosmp configs. --