> > > panic_blink = i8042_panic_blink; > > > > > > return 0; > > > > > > + err_dev_async_unregister: > > > + dev_async_unregister(&i8042_platform_device->dev); > > > err_free_device: > > > platform_device_put(i8042_platform_device); > > > err_unregister_driver: > > > > I still do not like the interface: manually unregistering seems to be > > ugly & error prone. > > > hmm, yes, we can check device->dev_async and unregister it in driver > core, at device_unregister time. > > then I think we should rename the interfaces to: > device_enable_async_action(dev, type), > and device_disable_async_action(dev) > so that they are more like driver core helper functions. > what do you think? Yes, that would be nicer. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
