From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1761856AbZJNWGf (ORCPT ); Wed, 14 Oct 2009 18:06:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1761048AbZJNWGd (ORCPT ); Wed, 14 Oct 2009 18:06:33 -0400 Received: from mga01.intel.com ([192.55.52.88]:11910 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757834AbZJNWGd (ORCPT ); Wed, 14 Oct 2009 18:06:33 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.44,561,1249282800"; d="scan'208";a="503255892" Message-Id: <20091014214654.423040927@sbs-t61.sc.intel.com> User-Agent: quilt/0.47-1 Date: Wed, 14 Oct 2009 14:46:54 -0700 From: Suresh Siddha To: hpa@zytor.com, mingo@elte.hu, tglx@linutronix.de, arjan@linux.jf.intel.com Cc: linux-kernel@vger.kernel.org, chinang.ma@intel.com Subject: [patch 0/2] x86_64: preserve large page text mappings with CONFIG_DEBUG_RODATA Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Most of the distributions enable CONFIG_DEBUG_RODATA by default, with the assumption that the debug protection checks come with no cost. Our OLTP tests showed 1% improvement (with 2% reduction in system time and 1% increase in iowait idle time) by disabling CONFIG_DEBUG_RODATA on a two socket NHM-EP system. CONFIG_DEBUG_RODATA breaks some of the kernel text/data large page mappings to small pages, thus causing this regression. These two patches address this issue. Experiments with EL5 kernel and mainline show that the first patch in the series (preserve large page mapping for 1st 2MB kernel txt with CONFIG_DEBUG_RODATA) should recover most of the above mentioned regression. We are doing couple of more OLTP tests for the confirmation. thanks, suresh