Hi Luca, On Tue, Oct 20, 2026 at 06:51:39PM +0200, Luca Ceresoli wrote: > The drm_dp_aux_unregister() docs don't say it is idempotent, so is it fair > to assume it is and always will?
It is today, as both drm_dp_aux_unregister_devnode() and i2c_del_adapter() return early for an adapter that is not registered, but you are right that this is not a documented guarantee. > I think it would be much cleaner if user drivers took case of calling it at > the appropriate times, not more. > > A flag similar to it6505->extcon_state used in patch 3 could allow that? Agreed, I'll track the registration in a flag in v9, like extcon_registered from patch 1. Cheers Daniel
