Re: repeat... Re: BUG: usb: phy: msm: problem with EPROBE defer and extcon cleanup

["Ivan T. Ivanov" <iivanov@mm-sol.com>]

On Mon, 2015-08-17 at 16:46 -0700, Tim Bird wrote:
> Sorry - I mistyped Ivan's e-mail the first time.
> 
> On 08/17/2015 04:43 PM, Tim Bird wrote:
> > Ivan,
> > 
> > I'm seeing a bug in the phy-msm-usb.c code, when I get an EPROBE defer during probing the driver.
> > The code is statically linked in the kernel, and the dts for my board is specifying an extcon phandle
> > for the ID pin.
> > 
> > Below is a fragment of the kernel output for a boot of this code.  Messages prefixed with TRB
> > are my own debug printks.  Note that the notifier block (at ee23513c) that gets registered
> > with extcon before the first probe failure, ultimately gets used during an ID pin event
> > (status change) by raw_notifier_call_chain().  This notifier block should be unregistered
> > if the probe fails, so that only the second one registered is on the call chain.

Right, I have missed to send one patch, sorry. Please, could you try this one [1].

Regards,
Ivan

[1] https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg957405.html