> > > + return ret; > > > + > > > + ret = iio_backend_enable(st->back); > > > + if (ret) > > > + return ret; > > > > I'm curious there is no iio_backend_disable() to be done in the exit path? > > > > Ehehe something I have in my mind, yes. I'm just not disabling the core because > it was the same with the previous approach. My goal was to have (more or less) > the same state before vs after introducing the backend. I was thinking in adding > a devm_iio_backend_enable() as a follow up patch and use it in here (or actually > use it for the first axi-dac/dds user as that one will be come from a "clean" > state). > > If you prefer I can already turn iio_backend_enable() -> > devm_iio_backend_enable() and use it in this patch. Might be cleaner to do that. Or add a big fat comment here to say it'll get cleaned up in a follow up patch. > > - Nuno Sá > >
