From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754336AbYCQPq7 (ORCPT ); Mon, 17 Mar 2008 11:46:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751581AbYCQPqw (ORCPT ); Mon, 17 Mar 2008 11:46:52 -0400 Received: from gprs189-60.eurotel.cz ([160.218.189.60]:1182 "EHLO spitz.ucw.cz" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751992AbYCQPqv (ORCPT ); Mon, 17 Mar 2008 11:46:51 -0400 Date: Mon, 17 Mar 2008 16:46:39 +0100 From: Pavel Machek To: "Joel K. Greene" Cc: linux-kernel@vger.kernel.org Subject: Re: Timer interrupt stops, causes soft lockup Message-ID: <20080317154639.GA4102@ucw.cz> References: <1205235450.18463.45.camel@linuxdt036.nc.catapult.com> <47D6423D.1E6F.00AA.0@novell.com> <1205255048.18463.60.camel@linuxdt036.nc.catapult.com> <1205260328.18463.67.camel@linuxdt036.nc.catapult.com> <47D68B9D.1E6F.00AA.0@novell.com> <1205265642.18463.82.camel@linuxdt036.nc.catapult.com> <1205322793.20205.9.camel@linuxdt036.nc.catapult.com> <1205326713.20205.27.camel@linuxdt036.nc.catapult.com> <1205332212.20205.33.camel@linuxdt036.nc.catapult.com> <1205510009.29658.83.camel@linuxdt036.nc.catapult.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1205510009.29658.83.camel@linuxdt036.nc.catapult.com> User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi! > Of course when time stops progessing forward at a steady pace all > sorts of Bad Things happen. > > The clock loop is caused by jiffies not incrementing. It seems that wall > time updates itself and then resync's with jiffies every ~4s. > > We verified that all spinlocks ( xtime_lock in particular ) were not > being held, and that the kernel had no reason not to update jiffies. > Other interrupts continue to fire, so we're not stuck in the timer > handler. > > So, we went on to look at the PIC. We threw together a little kernel > module to get some debug info. Nice debugging. > We have over 10 systems demonstrating the problem, many in customer > sites. Does nosmp help? Does noapic/nolapic help? You may want to try testing with HZ=4000... to show the problem faster. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html