On Thu, Oct 31, 2019 at 10:35:22PM +0800, Lai Jiangshan wrote:
> On 2019/10/31 10:10 下午, Paul E. McKenney wrote:
> > On Thu, Oct 31, 2019 at 10:07:59AM +0000, Lai Jiangshan wrote:
> > > Don't need to set ->rcu_read_lock_nesting negative, irq-protected
> > > rcu_preempt_deferred_qs_irqrestore() doesn't expect
> > > ->rcu_read_lock_nesting to be negative to work, it even
> > > doesn't access to ->rcu_read_lock_nesting any more.
> > >
> > > It is true that NMI over rcu_preempt_deferred_qs_irqrestore()
> > > may access to ->rcu_read_lock_nesting, but it is still safe
> > > since rcu_read_unlock_special() can protect itself from NMI.
> >
> > Hmmm... Testing identified the need for this one. But I will wait for
> > your responses on the earlier patches before going any further through
> > this series.
>
> Hmmm... I was wrong, it should be after patch7 to avoid
> the scheduler deadlock.
I was wondering about that. ;-)
Thanx, Paul
> > > Signed-off-by: Lai Jiangshan <laijs@xxxxxxxxxxxxxxxxx>
> > > ---
> > > kernel/rcu/tree_plugin.h | 5 -----
> > > 1 file changed, 5 deletions(-)
> > >
> > > diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
> > > index 82595db04eec..9fe8138ed3c3 100644
> > > --- a/kernel/rcu/tree_plugin.h
> > > +++ b/kernel/rcu/tree_plugin.h
> > > @@ -555,16 +555,11 @@ static bool rcu_preempt_need_deferred_qs(struct task_struct *t)
> > > static void rcu_preempt_deferred_qs(struct task_struct *t)
> > > {
> > > unsigned long flags;
> > > - bool couldrecurse = t->rcu_read_lock_nesting >= 0;
> > > if (!rcu_preempt_need_deferred_qs(t))
> > > return;
> > > - if (couldrecurse)
> > > - t->rcu_read_lock_nesting -= RCU_NEST_BIAS;
> > > local_irq_save(flags);
> > > rcu_preempt_deferred_qs_irqrestore(t, flags);
> > > - if (couldrecurse)
> > > - t->rcu_read_lock_nesting += RCU_NEST_BIAS;
> > > }
> > > /*
> > > --
> > > 2.20.1
> > >
