22 Mar
2015
22 Mar
'15
7:43 p.m.
On Fri, Feb 13, 2015 at 04:43:53AM +0000, Kuninori Morimoto wrote:
- struct list_head unbinded_list;
This should be "unbound". However the whole idea of having this list seems to be a bit problematic - it's moving us back to the old mechanism we used to have where we open coded deferred probe. What would be better would be if we could force the driver to unbind so we go back to the state we were in prior to the card being instantiated. I had a brief look and it looks like device_release_driver() might be what's needed but it's unclear to me if this will do exactly what we want and cause the device to try to rebind. We probably need to ask Greg unless there's an obvious answer I'm missing right now.