From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753218AbaBSL54 (ORCPT ); Wed, 19 Feb 2014 06:57:56 -0500 Received: from merlin.infradead.org ([205.233.59.134]:54070 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751245AbaBSL5z (ORCPT ); Wed, 19 Feb 2014 06:57:55 -0500 Date: Wed, 19 Feb 2014 12:57:47 +0100 From: Peter Zijlstra To: Michael wang Cc: paulus@samba.org, Ingo Molnar , acme@ghostprotocols.net, Thomas Gleixner , "H. Peter Anvin" , x86@kernel.org, LKML Subject: Re: [BUG] sched: tip/master show panic while booting/rebooting Message-ID: <20140219115747.GJ15586@twins.programming.kicks-ass.net> References: <5302EBA3.10909@linux.vnet.ibm.com> <20140218112230.GA27965@twins.programming.kicks-ass.net> <5304210B.9060604@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5304210B.9060604@linux.vnet.ibm.com> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Feb 19, 2014 at 11:12:11AM +0800, Michael wang wrote: > On 02/18/2014 07:22 PM, Peter Zijlstra wrote: > > On Tue, Feb 18, 2014 at 01:12:03PM +0800, Michael wang wrote: > >> Hi, Folks > >> > >> Got below panic while testing tip/master on x86 box, it randomly > >> occur while booting or rebooting, any ideas? > > > > The obvious pick would be the relocation of the idle_balance call, but > > I'm not sure how that'll end up stalling. > > I've review the code again, seems like it was caused by the same issue > we discussed (RT/DL enqueued issue), I found that idle_balance() will > return 1 in such cases: > > /* > * While browsing the domains, we released the rq lock. > * A task could have be enqueued in the meantime > */ > if (this_rq->nr_running && !pulled_task) > return 1; > > And if no more fair task to be pulled, we will do balance again and > again in pick_next_task_fair() until panic in NMI... > > The solution we discussed could solve the problem I suppose, it was > already in progress, correct? > > Please let me know if there are any thing I could help with ;-) Great, thanks! Right, I'm waiting for Ingo to appear and commit some patches into tip.