From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Serge E. Hallyn" Subject: Re: 2.6.33-rc4 i686 clone function looping (seems real!) Date: Thu, 21 Jan 2010 14:20:02 -0600 Message-ID: <20100121202002.GA4725@us.ibm.com> References: <1263852243.4745.363.camel@Mercier.safe.ca> <20100119150931.GA7708@us.ibm.com> <1263916851.4745.386.camel@Mercier.safe.ca> <20100121171338.GA16904@us.ibm.com> <1264095325.4283.1198.camel@laptop> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <1264095325.4283.1198.camel@laptop> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Peter Zijlstra Cc: containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org List-Id: containers.vger.kernel.org Quoting Peter Zijlstra (peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org): > On Thu, 2010-01-21 at 11:13 -0600, Serge E. Hallyn wrote: > > The culprit is e2912009fb7b715728311b0d8fe327a1432b3f79 > > sched: Ensure set_task_cpu() is never called on blocked tasks > > > > If you mount both the ns and cpuset cgroups with this patch applied, > > then doing clone with CLONE_NEWPID, CLONE_NEWNET, etc, you get the > > hang. The hang is actually hard enough that alt-sysrq isn't helpful :) > > Still trying to figure out what is going on - Peter, any ideas offhand? > > > Does it replicate on x86_64 ? Hmm, doesn't seem to, no. > /me looks at the NEWPID and NEWNET code > > > Hmm, I have an idea.. does it really need the ns cgroup stuff? To reproduce it? Well, there might be some other way to trigger it, but mounting just cpusets and moving a task into a different cpuset doesn't trigger it... So really it must be something to do with kernel/cgroup.c:cgroup_clone() > Let me try and come up with a debug patch... thanks, -serge