From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S967104AbcCPU30 (ORCPT ); Wed, 16 Mar 2016 16:29:26 -0400 Received: from mail-pf0-f193.google.com ([209.85.192.193]:33838 "EHLO mail-pf0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934790AbcCPU3T (ORCPT ); Wed, 16 Mar 2016 16:29:19 -0400 Date: Wed, 16 Mar 2016 13:29:17 -0700 From: Tejun Heo To: Peter Zijlstra Cc: Li Zefan , Johannes Weiner , cgroups@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com Subject: Re: [PATCH for-4.6] cgroup: ignore css_sets associated with dead cgroups during migration Message-ID: <20160316202917.GD21104@mtj.duckdns.org> References: <20160316004304.GC3447@mtj.duckdns.org> <20160316104527.GG25010@twins.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160316104527.GG25010@twins.programming.kicks-ass.net> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Mar 16, 2016 at 11:45:27AM +0100, Peter Zijlstra wrote: > This doesn't fix the problem that those zombies might actually still > want to use the cgroups they're tied to, as reported here: > > lkml.kernel.org/r/20160314112057.GT6356@twins.programming.kicks-ass.net > > I would strongly suggest to revert 2e91fa7f6d45 wholesale (and mark for > stable) and try again later. For the record, this issue was caused by the cpu controller taking down its css too early during destruction at ->css_offline() and is fixed by switching to ->css_released() instead. Thanks. -- tejun