From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758096AbYHOWp0 (ORCPT ); Fri, 15 Aug 2008 18:45:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752130AbYHOWpM (ORCPT ); Fri, 15 Aug 2008 18:45:12 -0400 Received: from smtp1.linux-foundation.org ([140.211.169.13]:46257 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752111AbYHOWpK (ORCPT ); Fri, 15 Aug 2008 18:45:10 -0400 Date: Fri, 15 Aug 2008 15:44:10 -0700 (PDT) From: Linus Torvalds To: "H. Peter Anvin" cc: Ingo Molnar , Jeremy Fitzhardinge , Steven Rostedt , LKML , Thomas Gleixner , Peter Zijlstra , Andrew Morton , David Miller , Mathieu Desnoyers , Roland McGrath , Ulrich Drepper , Rusty Russell , Gregory Haskins , Arnaldo Carvalho de Melo , "Luis Claudio R. Goncalves" , Clark Williams , srostedt@redhat.com Subject: Re: [PATCH] ftrace: use only 5 byte nops for x86 In-Reply-To: <48A6037B.6030202@zytor.com> Message-ID: References: <48A4C491.4050301@goop.org> <20080815092231.GB22209@elte.hu> <48A6037B.6030202@zytor.com> User-Agent: Alpine 1.10 (LFD 962 2008-03-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 15 Aug 2008, H. Peter Anvin wrote: > Ingo Molnar wrote: > > > > > > > > TODO: Examine the cpuid to determine the nop to use. > > > > > > > Don't think that will help in general. qemu claims its a Pentium II. > > > > ok. I've applied Steve's patch as it's a good bugfix. The extra paranoia > > about faulting on 0x90 seems excessive but doesnt hurt ;-) > > > > I think what we need to do is to test the NOPL instruction and create a > Linux-specific CPUID bit for it. I'll do that and submit to tip:x86/cpu. I also suspect that we'd really be much better off just fixing the generic NOP tables for the 5-byte nop. As far as I could tell, from all the numbers that have been posted, absolutely _none_ show that there is any point at all to the 2-instruction 3/2-byte sequence. So instead of having a magic special ftrace-only thing, why not just do it right, and fix the generic 5-byte nop sequence? Linus