On Wed, Apr 24, 2019 at 06:02:38PM +0200, Oleg Nesterov wrote: > On 04/05, Roman Gushchin wrote: > > > > @@ -5830,6 +5926,12 @@ void cgroup_exit(struct task_struct *tsk) > > spin_lock_irq(&css_set_lock); > > css_set_move_task(tsk, cset, NULL, false); > > cset->nr_tasks--; > > + > > + if (unlikely(cgroup_task_frozen(tsk))) > > + cgroup_freezer_frozen_exit(tsk); > > + else if (unlikely(cgroup_task_freeze(tsk))) > > + cgroup_update_frozen(task_dfl_cgroup(tsk)); > > + Hello, Oleg! A good catch! It's actually a leftover from one of the previous versions of the patchset. In the current iteration I'd replace it with: WARN_ON_ONCE(cgroup_task_frozen(tsk)); just to make sure we're not leaking the frozen bit. Do you agree? Thank you! Roman
