On Fri, May 16, 2008 at 12:07:52PM -0700, Paul E. McKenney wrote:
> The hlist_del_init_rcu() primitive looks good.
>
> The rest of the RCU code looks fine assuming that "mn->ops->release()"
> either does call_rcu() to defer actual removal, or that the actual
> removal is deferred until after mmu_notifier_release() returns.
Yes, actual removal is deferred until after mmu_notifier_release()
returns.
> Acked-by: Paul E. McKenney <[email protected]>
Thanks for the review Paul! I should also have added your precious
Acked-by to the 1/3 and 3/3 but the important is the ack by email ;)