On Fri, Aug 21, 2015 at 10:41:50AM -0700, Mark Brown wrote:
> On Fri, Aug 21, 2015 at 10:30:17AM -0700, Dmitry Torokhov wrote:

> > They are also often wrapped into other objects. For example input
> > device might turn off regulators/clocks in it's close() method, which
> > is called as part of input_unregister_device(). If input device is
> > managed then that state change happens "some time later".

> Right, but my understanding is that the integration with devm with the
> object reference counting was intended to do the right thing.

More specifically: the managed operations are guaranteed to be run in
reverse order so so long as the resources used by the managed input
device are also managed we're supposed to be fine.