From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754706AbYHSDla (ORCPT ); Mon, 18 Aug 2008 23:41:30 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752612AbYHSDlV (ORCPT ); Mon, 18 Aug 2008 23:41:21 -0400 Received: from tomts22-srv.bellnexxia.net ([209.226.175.184]:59472 "EHLO tomts22-srv.bellnexxia.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752579AbYHSDlV (ORCPT ); Mon, 18 Aug 2008 23:41:21 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AtIFAOfdqUhMRKxB/2dsb2JhbACBYrNhgVg Date: Mon, 18 Aug 2008 23:36:18 -0400 From: Mathieu Desnoyers To: Steven Rostedt Cc: Benjamin Herrenschmidt , Scott Wood , Eran Liberty , linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org, Steven Rostedt , "Paul E. McKenney" Subject: Re: ftrace introduces instability into kernel 2.6.27(-rc2,-rc3) Message-ID: <20080819033618.GA24085@Krystal> References: <48A9901B.1080900@redhat.com> <20080818154746.GA26835@Krystal> <48A9AFA7.8080508@freescale.com> <1219110814.8062.2.camel@pasglop> <1219113549.8062.13.camel@pasglop> <20080819024707.GA22659@Krystal> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.6.21.3-grsec (i686) X-Uptime: 23:35:35 up 75 days, 8:16, 8 users, load average: 0.42, 0.46, 0.43 User-Agent: Mutt/1.5.16 (2007-06-11) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Steven Rostedt (rostedt@goodmis.org) wrote: > > On Mon, 18 Aug 2008, Mathieu Desnoyers wrote: > > > * Steven Rostedt (rostedt@goodmis.org) wrote: > > > > > > On Tue, 19 Aug 2008, Benjamin Herrenschmidt wrote: > > > > > > > > > > > > Hmm, this was originally copied from x86, where we did a cmpxchg, but that > > > > > is probably not needed since all of this is done in kstop_machine. Also, > > > > > only the "get" is needed. If we don't fault there, we wont fault on the > > > > > put (unless we have permissions wrong, and that would be a bug). > > > > > > > > Would it ? How do we make sure the kernel text is mapped writeable ? > > > > > > We map it writeable if FTRACE is enabled. > > > > > > > Argh. See text_poke(). It's there exactly for this purpose on x86. > > Ouch, I just did. text_poke is quite heavy. It would be interesting to see > that performed on 20,000 locations at one time. I could play with it, but > I'm a bit nervous. > It's alread used to modify the LOCK prefixes in alternative.c and did not seem to be too slow for that.. it should therefore be ok. Mathieu > -- Steve -- Mathieu Desnoyers OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68