From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============5759142151896836002==" MIME-Version: 1.0 From: Peter Zijlstra To: lkp@lists.01.org Subject: Re: [sched/deadline] kernel BUG at kernel/sched/deadline.c:805! Date: Mon, 16 Feb 2015 15:44:12 +0100 Message-ID: <20150216144412.GS24151@twins.programming.kicks-ass.net> In-Reply-To: <20150216130821.GB5029@twins.programming.kicks-ass.net> List-Id: --===============5759142151896836002== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Mon, Feb 16, 2015 at 02:08:21PM +0100, Peter Zijlstra wrote: > On Mon, Feb 16, 2015 at 03:38:34PM +0300, Kirill Tkhai wrote: > > We shouldn't enqueue migrating tasks. Please, try this one instead ;) > = > Ha, we should amend that task-rq-lock loop for that. See below. > = > I've not yet tested; going to try and reconstruct a .config that > triggers the oops. OK, I had to remove -enable-kvm from Wu's script, otherwise the test would not hit the CBS limit, and then I had to remove the panic on softlockup because well, qemu is so slow you're going to hit that :-) But the good news is that I could reproduce and this patch does indeed seem to solve it. --===============5759142151896836002==-- From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755806AbbBPOoU (ORCPT ); Mon, 16 Feb 2015 09:44:20 -0500 Received: from casper.infradead.org ([85.118.1.10]:55974 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755642AbbBPOoS (ORCPT ); Mon, 16 Feb 2015 09:44:18 -0500 Date: Mon, 16 Feb 2015 15:44:12 +0100 From: Peter Zijlstra To: Kirill Tkhai Cc: Fengguang Wu , Ingo Molnar , LKP , "linux-kernel@vger.kernel.org" , juri.lelli@arm.com Subject: Re: [sched/deadline] kernel BUG at kernel/sched/deadline.c:805! Message-ID: <20150216144412.GS24151@twins.programming.kicks-ass.net> References: <20150216072038.GA17056@wfg-t540p.sh.intel.com> <9846861424087552@web13j.yandex.ru> <1374601424090314@web4j.yandex.ru> <20150216130821.GB5029@twins.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150216130821.GB5029@twins.programming.kicks-ass.net> 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 Mon, Feb 16, 2015 at 02:08:21PM +0100, Peter Zijlstra wrote: > On Mon, Feb 16, 2015 at 03:38:34PM +0300, Kirill Tkhai wrote: > > We shouldn't enqueue migrating tasks. Please, try this one instead ;) > > Ha, we should amend that task-rq-lock loop for that. See below. > > I've not yet tested; going to try and reconstruct a .config that > triggers the oops. OK, I had to remove -enable-kvm from Wu's script, otherwise the test would not hit the CBS limit, and then I had to remove the panic on softlockup because well, qemu is so slow you're going to hit that :-) But the good news is that I could reproduce and this patch does indeed seem to solve it.