From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Date: Sun, 11 Jul 2004 08:38:03 +0000 Subject: Re: serious performance regression due to NX patch Message-Id: List-Id: References: <200407100528.i6A5SF8h020094@napali.hpl.hp.com> In-Reply-To: <200407100528.i6A5SF8h020094@napali.hpl.hp.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: davidm@hpl.hp.com Cc: suresh.b.siddha@intel.com, jun.nakajima@intel.com, Andrew Morton , Linus Torvalds , linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org On Fri, 9 Jul 2004, David Mosberger wrote: > The "No eXecute patch for x86" causes a serious exec() performance > degradation on ia64 since it ends up incorrectly turning on the execute > protection on all segments (since most ia64 binaries don't have a > gnu_stack program-header). > > The patch below fixes the problem by turning on VM_EXEC and VM_MAYEXEC > only if VM_DATA_DEFAULT_FLAGS mentions them. Note that on ia64, the > value of VM_DATA_DEFAULT_FLAGS depends on the personality (to support > x86 binaries) and hence I had to move the setting of "def_flags" down a > bit to a point after SET_PERSONALITY() has been done. > > Please apply. ok, agreed. I'll check that it still does the right thing on x86. Ingo From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S266526AbUGKIky (ORCPT ); Sun, 11 Jul 2004 04:40:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S266524AbUGKIkv (ORCPT ); Sun, 11 Jul 2004 04:40:51 -0400 Received: from mx1.redhat.com ([66.187.233.31]:46011 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S266521AbUGKIkn (ORCPT ); Sun, 11 Jul 2004 04:40:43 -0400 Date: Sun, 11 Jul 2004 04:38:03 -0400 (EDT) From: Ingo Molnar X-X-Sender: mingo@devserv.devel.redhat.com To: davidm@hpl.hp.com cc: suresh.b.siddha@intel.com, jun.nakajima@intel.com, Andrew Morton , Linus Torvalds , linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: serious performance regression due to NX patch In-Reply-To: <200407100528.i6A5SF8h020094@napali.hpl.hp.com> Message-ID: References: <200407100528.i6A5SF8h020094@napali.hpl.hp.com> 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 On Fri, 9 Jul 2004, David Mosberger wrote: > The "No eXecute patch for x86" causes a serious exec() performance > degradation on ia64 since it ends up incorrectly turning on the execute > protection on all segments (since most ia64 binaries don't have a > gnu_stack program-header). > > The patch below fixes the problem by turning on VM_EXEC and VM_MAYEXEC > only if VM_DATA_DEFAULT_FLAGS mentions them. Note that on ia64, the > value of VM_DATA_DEFAULT_FLAGS depends on the personality (to support > x86 binaries) and hence I had to move the setting of "def_flags" down a > bit to a point after SET_PERSONALITY() has been done. > > Please apply. ok, agreed. I'll check that it still does the right thing on x86. Ingo