Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1761219AbYHAVLi (ORCPT ); Fri, 1 Aug 2008 17:11:38 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760774AbYHAVJg (ORCPT ); Fri, 1 Aug 2008 17:09:36 -0400 Received: from e3.ny.us.ibm.com ([32.97.182.143]:39385 "EHLO e3.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760765AbYHAVJe (ORCPT ); Fri, 1 Aug 2008 17:09:34 -0400 Date: Fri, 1 Aug 2008 14:09:13 -0700 From: "Paul E. McKenney" To: Wolfgang Grandegger Cc: Steven Rostedt , Luotao Fu , LKML , RT , Ingo Molnar , Thomas Gleixner Subject: Re: 2.6.24-rc8-rt1: Strange latencies on mpc5200 powerpc - RCU issue? Message-ID: <20080801210913.GC14851@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20080130010716.GB16069@linux.vnet.ibm.com> <47A032E9.4080105@grandegger.com> <20080130102255.GA24890@linux.vnet.ibm.com> <47A0552D.20605@grandegger.com> <20080130105715.GE24890@linux.vnet.ibm.com> <20080130111559.GC20089@pengutronix.de> <486A3EBB.6030905@grandegger.com> <486A573E.60107@grandegger.com> <486B607F.9040903@grandegger.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <486B607F.9040903@grandegger.com> User-Agent: Mutt/1.5.15+20070412 (2007-04-11) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1054 Lines: 27 On Wed, Jul 02, 2008 at 01:03:27PM +0200, Wolfgang Grandegger wrote: > Wolfgang Grandegger wrote: >> Steven Rostedt wrote: >> There seems to be something in trace_mark() keeping latency low: >> http://lxr.linux.no/linux+v2.6.24.4/include/linux/marker.h#L52 >> I will follow your suggestions. > > I removed all "likely" and "unlikely" macros, but the latencies did not > improve. Then I added > > preempt_disable(); > preempt_enable(); > > at the two locations mentioned above, like trace_mark() does, and disabled > CONFIG_RCU_TRACE. That helped to keep the latencies low, and it did for > 2.6.25.8-rt7 as well. As I see it, adding preemption points seem to prevent > high latencies. Interesting. Perhaps a preempt_enable_no_resched() somewhere that should be preempt_enable()? Thanx, Paul -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/