From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tim Sander Subject: Re: infinite spin in RT when booting with DHCP on Date: Fri, 3 Feb 2012 00:44:10 +0100 Message-ID: <201202030044.10516.tim01@vlsi.informatik.tu-darmstadt.de> References: <4F292FE0.7090302@digi.com> <20120202201336.GF25594@pengutronix.de> <1328218466.5882.119.camel@gandalf.stny.rr.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-15 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "Uwe =?iso-8859-15?q?Kleine-K=F6nig?=" , Tim Sander , Hector Palacios , linux-rt-users@vger.kernel.org, lclaudio@uudg.org, efault@gmx.de To: Steven Rostedt Return-path: Received: from lnx500.hrz.tu-darmstadt.de ([130.83.156.225]:39890 "EHLO lnx500.hrz.tu-darmstadt.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756743Ab2BBXoZ convert rfc822-to-8bit (ORCPT ); Thu, 2 Feb 2012 18:44:25 -0500 In-Reply-To: <1328218466.5882.119.camel@gandalf.stny.rr.com> Sender: linux-rt-users-owner@vger.kernel.org List-ID: Hi Guys Am Donnerstag 02 Februar 2012, 22:34:26 schrieb Steven Rostedt: > On Thu, 2012-02-02 at 21:13 +0100, Uwe Kleine-K=F6nig wrote: > > enabled the ksoftirq is also running wild with 3.0-rt. This also ha= ppens > > when I fail to reproduce that on a pcm043 machine using 3.0.18-rt34= =2E Can > > you provide me your config. Does the problem occur on every boot fo= r > > you? If not, how often do I need to reboot? >=20 > Note, you also said (on IRC) that you boot off this card via NFS root= =2E > I'm not sure Tim has the same setup, and that may be a big difference= in > how this bug comes about. >=20 > It's a race issue, and it may require enabling the network interface > after the system is up and running. Its indeed a race and it is most of the time only triggered when ubifs = is=20 attached or the ubifs worker thread is active in background. The error = must be=20 there sleeping for at least 2.6.31. We where using 2.6.33-rt (if i reme= mber=20 correctly) and it might occoured once but i wasn't sure back then. Whic= h lead=20 me to the complete wrong direction that ubi was at fault. The bug occur= s only=20 very seldom with static ip addresses configured. So dhcp is needed to=20 reproduce this bug more easily. Then i found out that connman is a good= way to=20 trigger this bug. So to trigger i started dhclient, connman and if it d= idn't=20 show up i did some "ifconfig eth0 up/down" poking. But the first time i= had=20 the hw debugger wired it didn't show up for some reason even when i was= trying=20 to trigger it for an hour or so. @Uwe: I am currently at home and its to late to fire my work computer u= p.=20 You'll get the kconfig tomorrow... ahm later. I will then also test the patch of Steve if i manage that before leavin= g to=20 =46OSDEM :-). Best regards Tim -- To unsubscribe from this list: send the line "unsubscribe linux-rt-user= s" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html