On Thu, 24 May 2018 15:38:39 -0700
"Paul E. McKenney" <paul...@linux.vnet.ibm.com> wrote:
> On Thu, May 24, 2018 at 05:47:52PM -0400, Steven Rostedt wrote:
> > On Wed, 23 May 2018 13:04:58 -0700
> > "Paul E. McKenney" <paul...@linux.vnet.ibm.com> wrote:
> >
> > > Looks plausible to me!
> > >
> >
> > BTW looking at the code:
> >
> > > /* Invoke the callbacks. */
> > > while (list) {
> > > next = list->next;
> > > local_bh_disable();
> > > list->func(list);
> > > local_bh_enable();
> > > list = next;
> > > cond_resched();
> > > }
> > > schedule_timeout_uninterruptible(HZ/10);
> >
> > What's the purpose of this final sleep?
>
> To avoid any possibility of rcu_tasks_kthread() getting into a tight loop.
>
OK, so I'll send out a v4 of what I just sent out ;-)
-- Steve
