From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932253Ab0DPOql (ORCPT ); Fri, 16 Apr 2010 10:46:41 -0400 Received: from mx1.redhat.com ([209.132.183.28]:23966 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752416Ab0DPOqk (ORCPT ); Fri, 16 Apr 2010 10:46:40 -0400 Date: Fri, 16 Apr 2010 10:46:24 -0400 From: Don Zickus To: Cyrill Gorcunov Cc: Frederic Weisbecker , mingo@elte.hu, peterz@infradead.org, aris@redhat.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH v2] [watchdog] combine nmi_watchdog and softlockup Message-ID: <20100416144624.GG15159@redhat.com> References: <1271366710-17468-1-git-send-email-dzickus@redhat.com> <20100416014712.GC15570@nowhere> <20100416143232.GA5540@lenovo> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100416143232.GA5540@lenovo> User-Agent: Mutt/1.5.20 (2009-08-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Apr 16, 2010 at 06:32:32PM +0400, Cyrill Gorcunov wrote: > On Fri, Apr 16, 2010 at 03:47:14AM +0200, Frederic Weisbecker wrote: > [...] > > > + > > > +/* Callback function for perf event subsystem */ > > > +void watchdog_overflow_callback(struct perf_event *event, int nmi, > > > + struct perf_sample_data *data, > > > + struct pt_regs *regs) > > > +{ > > > + int this_cpu = smp_processor_id(); > > > + unsigned long touch_ts = per_cpu(watchdog_touch_ts, this_cpu); > > > + > > > + if (touch_ts == 0) { > > > + __touch_watchdog(); > > > + return; > > > + } > > > + > > > + /* check for a hardlockup > > > + * This is done by making sure our timer interrupt > > > + * is incrementing. The timer interrupt should have > > > + * fired multiple times before we overflow'd. If it hasn't > > > + * then this is a good indication the cpu is stuck > > > + */ > > > + if (is_hardlockup(this_cpu)) { > > > + /* only print hardlockups once */ > > > + if (cpumask_test_cpu(this_cpu, to_cpumask(hardlockup_mask))) > > > + return; > > > + > > > + if (hardlockup_panic) > > > + panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu); > > > + else > > > + WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu); > > > + > > > + cpumask_set_cpu(this_cpu, to_cpumask(hardlockup_mask)); > > > > > > > > May be have an arch spin lock there to update your cpu mask safely. > > > > Hmm, this is NMI handler path so from what we protect this per-cpu data? > Do I miss something? /me confused It's not per-cpu which is the problem I believe. If you have multiple cpus dealing with hardlockups than they can overwrite each other's data. Cheers, Don