From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752089Ab2DUFbM (ORCPT ); Sat, 21 Apr 2012 01:31:12 -0400 Received: from cantor2.suse.de ([195.135.220.15]:54380 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750798Ab2DUFbL (ORCPT ); Sat, 21 Apr 2012 01:31:11 -0400 Message-ID: <1334986264.5428.1.camel@marge.simpson.net> Subject: Re: [patch] cgroups: disallow attaching kthreadd From: Mike Galbraith To: Tejun Heo Cc: Peter Zijlstra , Li Zefan , David Rientjes , Paul Menage , LKML , Thomas Gleixner Date: Sat, 21 Apr 2012 07:31:04 +0200 In-Reply-To: <20120420175751.GF32324@google.com> References: <1333475906.7439.7.camel@marge.simpson.net> <20120420175751.GF32324@google.com> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.1 Content-Transfer-Encoding: 7bit Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2012-04-20 at 10:57 -0700, Tejun Heo wrote: > Hello, Mike. > > This is dragging out too long but can you please update the patch so > that both kthreadd and PF_THREAD_BOUND tasks are disallowed from !root > cgroups? cgroups: disallow attaching kthreadd or PF_THREAD_BOUND threads Allowing kthreadd to be moved to a non-root group makes no sense, it being a global resource, and needlessly leads unsuspecting users toward trouble. 1. An RT workqueue worker thread spawned in a task group with no rt_runtime allocated is not schedulable. Simple user error, but harmful to the box. 2. A worker thread which acquires PF_THREAD_BOUND can never leave a cpuset, rendering the cpuset immortal. Save the user some unexpected trouble, just say no. Signed-off-by: Mike Galbraith Acked-by: Peter Zijlstra Acked-by: Thomas Gleixner --- kernel/cgroup.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -60,6 +60,7 @@ #include #include #include /* used in cgroup_attach_proc */ +#include #include @@ -2172,6 +2173,17 @@ static int attach_task_by_pid(struct cgr if (threadgroup) tsk = tsk->group_leader; + + /* + * Workqueue threads may acquire PF_THREAD_BOUND and become + * trapped in a cpuset, or RT worker may be born in a cgroup + * with no rt_runtime allocated. Just say no. + */ + if (tsk == kthreadd_task || (tsk->flags & PF_THREAD_BOUND)) { + ret = -EINVAL; + goto out_unlock_cgroup; + } + get_task_struct(tsk); rcu_read_unlock();