WrongState vs InProgress when enabling a modem

[Dan Williams]

On Fri, 2013-09-13 at 08:42 -0700, Ben Chan wrote:
> > Yeah, disable() will actually run mm_iface_modem_wait_for_final_state()
> > so there shouldn't be any need to check the modem state looking for
> > intermediate states.
> >
> 
> Should initialize() and enable() behave similarly, i.e. wait for the final
> state?  Is the early return to allow the UI to be more responsive?

Any UI should be calling any of these methods asynchronously anyway...
but I think that since these steps may take a long time (10, 20 seconds
in some cases? especially if errors occur) they probably were meant to
return immediately and then clients would wait on the state transitions.
But I don't have a problem with the methods waiting for the final state,
that would seem more correct, and shouldn't be a problem given that a
responsive UI wouldn't block until they returned anyway.

Dan