On Wed, Dec 14, 2022 at 10:51:01AM +0800, Ming Lei wrote: > The pattern of wait_event(percpu_ref_is_zero()) has been used in several > kernel components, and this way actually has the following risk: I'd much rather see those components updated to wait for notification from ->release rather than doing this or update percpu_ref_is_zero() to wait for ->release() to finish. Thanks. -- tejun
