On Fri, 30 Sep 2011 15:30:49 -0700
Andrew Morton <akpm00@gmail.com> wrote:

> On Fri, 30 Sep 2011 16:54:52 +0900
> Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> wrote:
> 
> > In commit:f90d4118, cpuset_wq, a separate workqueue for cpuset, was introduced
> > to avoid a dead lock against cgroup_mutex between async_rebuild_sched_domains()
> > and cgroup_tasks_write().
> > 
> > But check_for_release() has a similar problem:
> > 
> >   check_for_release()
> >     schedule_work(release_agent_work)
> >       cgroup_release_agent()
> >         mutex_lock(&cgroup_mutex)
> > 
> > And I actually see a lockup which seems to be caused by this problem
> > on 2.6.32-131.0.15.el6.x86_64.
> 
> Are you sure the bug is still present in current kernels?  Perhaps
> Tejun's workqueue changes magically made it go away.
> 
Not yet, but I'll check it.

> >
> > ...
> > --- a/include/linux/cgroup.h
> > +++ b/include/linux/cgroup.h
> > @@ -27,6 +27,8 @@ struct css_id;
> > 
> >  extern int cgroup_init_early(void);
> >  extern int cgroup_init(void);
> > +extern void cgroup_wq_init(void);
> > +extern void queue_cgroup_work(struct work_struct *work);
> >  extern void cgroup_lock(void);
> >  extern int cgroup_lock_is_held(void);
> >  extern bool cgroup_lock_live_group(struct cgroup *cgrp);
> 
> Can we spot the odd man out?
> 
> I shall rename queue_cgroup_work() to cgroup_queue_work().
> 
Thank you for fixing it.
I agree cgroup_queue_work() is a better name.

Thanks,
Daisuke Nishimura.