From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934213AbcKQRXy (ORCPT ); Thu, 17 Nov 2016 12:23:54 -0500 Received: from Galois.linutronix.de ([146.0.238.70]:49413 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934008AbcKQRTV (ORCPT ); Thu, 17 Nov 2016 12:19:21 -0500 Date: Thu, 17 Nov 2016 17:31:55 +0100 (CET) From: Thomas Gleixner To: Sebastian Andrzej Siewior cc: Borislav Petkov , Peter Zijlstra , lkml Subject: Re: lockdep splat during suspend In-Reply-To: <20161117162943.ul2gldjt6qhz35dz@linutronix.de> Message-ID: References: <20161117124358.stl7jxepvuhaxv26@pd.tnic> <20161117162943.ul2gldjt6qhz35dz@linutronix.de> User-Agent: Alpine 2.20 (DEB 67 2015-01-07) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 17 Nov 2016, Sebastian Andrzej Siewior wrote: > On 2016-11-17 13:43:58 [+0100], Borislav Petkov wrote: > > Hi dudes, > Hi Boris, > > > this is what I'm seeing on 4.9-rc5+tip/master during suspend. (And yes, > > that printk thing is nuts): > > assuming this happens always, could you please try to revert > 90b14889d2f9 ("kernel/printk: Convert to hotplug state machine") > and see if it still there? I am curious if this is new or old splat. It's a new one and that patch simply missed the fact, that the actions in the original notifier were all w/o FROZEN, but the conversion dropped that. Which probably explains the splat because suspend/resume is where the FROZEN stuff happens. Boris, does the patch below fix the issue for you ? Thanks, tglx 8<----------------------- --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -2194,8 +2194,10 @@ void resume_console(void) */ static int console_cpu_notify(unsigned int cpu) { - console_lock(); - console_unlock(); + if (!cpuhp_tasks_frozen) { + console_lock(); + console_unlock(); + } return 0; }