From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758037AbXIQSgk (ORCPT ); Mon, 17 Sep 2007 14:36:40 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757994AbXIQSfO (ORCPT ); Mon, 17 Sep 2007 14:35:14 -0400 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:44998 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1757433AbXIQSfK (ORCPT ); Mon, 17 Sep 2007 14:35:10 -0400 Message-Id: <20070917183507.332345000@sgi.com> User-Agent: quilt/0.46-1 Date: Mon, 17 Sep 2007 11:35:07 -0700 From: travis@sgi.com To: sfr@canb.auug.org.au, Andrew Morton Cc: Andi Kleen , Christoph Lameter , linux-mm@kvack.org, linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org, sparclinux@vger.kernel.org Subject: [PATCH 0/1] ppc64: Convert cpu_sibling_map to a per_cpu data array ppc64 v2 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Stephen Rothwell wrote: > On Mon, 17 Sep 2007 16:28:31 +1000 Stephen Rothwell wrote: >> the topology (on my POWERPC5+ box) is not correct: >> >> cpu0/topology/thread_siblings:0000000f >> cpu1/topology/thread_siblings:0000000f >> cpu2/topology/thread_siblings:0000000f >> cpu3/topology/thread_siblings:0000000f >> >> it used to be: >> >> cpu0/topology/thread_siblings:00000003 >> cpu1/topology/thread_siblings:00000003 >> cpu2/topology/thread_siblings:0000000c >> cpu3/topology/thread_siblings:0000000c > > This would be because we are setting up the cpu_sibling map before we > call setup_per_cpu_areas(). The following patch hopefully should fix this problem. I'm not able to build or test it but the few references to cpu_sibling_map seem to all occur well after setup_per_cpu_areas is called. Thanks Stephen for checking this out! --