From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932117AbVH3Ddc (ORCPT ); Mon, 29 Aug 2005 23:33:32 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932114AbVH3Ddb (ORCPT ); Mon, 29 Aug 2005 23:33:31 -0400 Received: from omx2-ext.sgi.com ([192.48.171.19]:39908 "EHLO omx2.sgi.com") by vger.kernel.org with ESMTP id S932113AbVH3Ddb (ORCPT ); Mon, 29 Aug 2005 23:33:31 -0400 Date: Mon, 29 Aug 2005 20:28:28 -0700 (PDT) From: Christoph Lameter To: Rusty Lynch cc: Andi Kleen , Rusty Lynch , linux-mm@kvack.org, prasanna@in.ibm.com, linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org, anil.s.keshavamurthy@intel.com Subject: Re: [PATCH] Only process_die notifier in ia64_do_page_fault if KPROBES is configured. In-Reply-To: <20050830001905.GA18279@linux.jf.intel.com> Message-ID: References: <200508262246.j7QMkEoT013490@linux.jf.intel.com> <200508270224.26423.ak@suse.de> <20050830001905.GA18279@linux.jf.intel.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 Mon, 29 Aug 2005, Rusty Lynch wrote: > So, assuming inlining the notifier_call_chain would address Christoph's > conserns, is the following patch something like what you are sugesting? > This would make all the kdebug.h::notify_die() calls use the inline version. Please do not generate any code if the feature cannot ever be used (CONFIG_KPROBES off). With this patch we still have lots of unnecessary code being executed on each page fault.