On Tue, Dec 19, 2017 at 07:14:17AM +0100, Bjørn Mork wrote:
> "SZ Lin (林上智)" <sz....@moxa.com> writes:
> >> Johan Hovold <jo...@kernel.org> writes:
> >> 
> >> >> +static const struct option_blacklist_info yuga_clm920_nc5_blacklist = {
> >> >> +       .reserved = BIT(0) | BIT(1) | BIT(4), };
> >> >
> >> > Do you really need to blacklist the first interface?
> >> 
> >> Good question. Interface #0 does look a lot like a Qualcomm DM/DIAG
> >> function, based on two bulk endpoints, no additional descriptors
> >> and the fact that it is the first interface.  If so, then we do
> >> want a serial driver for it.  There is a basic libqcdm
> >> implementation in ModemManager if you want to test it out.
> >
> >I have confirmed that interface #0 is QCDM/DIAG port in this module,
> >and thus I will remove this from reserved list in next patch.

Thanks to both of you for confirming.

> >Furthermore, interface #1 is ADB port. Should I also remove this from
> >reserved list?
> 
> No. ADB is handled by userspace tools using libusb.  It should not be
> bound to any serial driver, so you will need to blacklist it.  But you
> need to keep the blacklist anyway to include the QCDM/DIAG port
>
> I assume Johan's alternative was to match class/subclass/protocol
> against ff/00/00, which would have worked if you only wanted to include
> interfaces 2 and 3.

Indeed.

Johan
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to