From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757454Ab2DYSTL (ORCPT ); Wed, 25 Apr 2012 14:19:11 -0400 Received: from mail-pz0-f51.google.com ([209.85.210.51]:62230 "EHLO mail-pz0-f51.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756340Ab2DYSTK (ORCPT ); Wed, 25 Apr 2012 14:19:10 -0400 Date: Wed, 25 Apr 2012 11:19:05 -0700 From: Tejun Heo To: Jens Axboe Cc: vgoyal@redhat.com, ctalbott@google.com, rni@google.com, linux-kernel@vger.kernel.org, cgroups@vger.kernel.org, containers@lists.linux-foundation.org Subject: Re: [PATCHSET] block: per-queue policy activation, take#2 Message-ID: <20120425181905.GA8989@google.com> References: <1334347895-6268-1-git-send-email-tj@kernel.org> <20120420080943.GG7505@kernel.dk> <4F91503D.2060402@kernel.dk> <20120420171742.GC32324@google.com> <4F91B433.2070209@kernel.dk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4F91B433.2070209@kernel.dk> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Apr 20, 2012 at 09:08:35PM +0200, Jens Axboe wrote: > On 2012-04-20 19:17, Tejun Heo wrote: > > Hello, Jens. > > > > On Fri, Apr 20, 2012 at 02:02:05PM +0200, Jens Axboe wrote: > >> Tejun, everything merged in cleanly. A boot test didn't not go so well, > >> I'm afraid. It complains about stuck RCU. Even with extra RCU > >> debugging/info turned on, it doesn't give my any extra info. > >> > >> Config is attached. I can try and debug this on Monday, for now I've > >> pulled it out of for-next again. > > > > Hmmm... it works fine with the same config here (qemu w/ 4 cpus > > configured). Doesn't stuck RCU usually come from unbalanced RCU > > lock/unlocking? That should trigger context or RCU warnings before > > anything else, weird. Anyways, it would be great if you can provide > > more info on how to reproduce the problem. > > The machine just sat there. I'll get some more info out of it and run a > bisect too. Any progress on this? Thanks. -- tejun