From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759656AbXKLRFv (ORCPT ); Mon, 12 Nov 2007 12:05:51 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753841AbXKLRFn (ORCPT ); Mon, 12 Nov 2007 12:05:43 -0500 Received: from e35.co.us.ibm.com ([32.97.110.153]:35842 "EHLO e35.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753063AbXKLRFm (ORCPT ); Mon, 12 Nov 2007 12:05:42 -0500 Date: Mon, 12 Nov 2007 09:05:21 -0800 From: "Paul E. McKenney" To: Benjamin Herrenschmidt Cc: Steven Rostedt , linux-kernel@vger.kernel.org, tony@bakeyournoodle.com, paulus@samba.org, dino@in.ibm.com, tytso@us.ibm.com, dvhltc@us.ibm.com, niv@us.ibm.com, antonb@us.ibm.com Subject: Re: [PATCH, RFC] improved hacks to allow -rt to run kernbench on POWER Message-ID: <20071112170521.GA11113@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20071109181028.GA26800@linux.vnet.ibm.com> <1194641524.21340.14.camel@pasglop> <20071109220717.GJ8826@linux.vnet.ibm.com> <1194753541.21340.26.camel@pasglop> <1194814125.6510.14.camel@pasglop> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1194814125.6510.14.camel@pasglop> User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Nov 12, 2007 at 07:48:45AM +1100, Benjamin Herrenschmidt wrote: > > On Sun, 2007-11-11 at 09:45 -0500, Steven Rostedt wrote: > > > Well, I suppose the patch could go in, maybe with some ifdef's > > around > > > the bits in _switch_to, there's little point in doing that on non-rt > > > kernels. > > > > As Nick Piggin already stated, and I'll even state it for the RT > > kernel, > > we do not allow preemption in switch_to. So it is safe to remove those > > "preempt_disable" bits from the patch > > Sure, I know that, I'm not talking about that, I'm talking about the > added code that flush pending batches & save the batch state, since on > non-rt kernel, this is not useful (the batch is only ever active within > a spinlocked section, which cannot be preempted on non-rt). Ah, I need a bit of conditional compilation. Will fix. Thanx, Paul