On Tue, Dec 11, 2018 at 02:22:50PM +0100, Jörgen Storvist wrote:
> Den Tue, 11 Dec 2018 09:32:36 +0100
> skrev Re: [PATCH] USB: serial: option: add Fibocom NL668 series:
>
> > On Tue, Dec 11, 2018 at 08:41:24AM +0100, Jörgen Storvist wrote:
> > >
> > > Added USB serial option driver support for Fibocom NL668 series cellular modules.
> > > Reserved USB endpoints 4, 5 and 6 for network + ADB interface.
> > >
> > > Signed-off-by: Jörgen Storvist <jorgen.storvist@xxxxxxxxx>
> > > ---
> > >
> > > Thanks for feedback!
> > > Changes:
> > > Removed name declarations for VID/PID
> > > Added reserved endpoint for ADB interface
> > >
> > > usb-devices
> > > T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 7 Spd=480 MxCh= 0
> > > D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
> > > P: Vendor=1508 ProdID=1001 Rev=03.18
> > > S: Manufacturer=Nodecom NL668 Modem
> > > S: Product=Nodecom NL668-CN Modem
> > > S: SerialNumber=5ced6a52
> > > C: #Ifs= 7 Cfg#= 1 Atr=a0 MxPwr=500mA
> > > I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
> > > I: If#= 1 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
> > > I: If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
> > > I: If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
> > > I: If#= 4 Alt= 0 #EPs= 1 Cls=02(commc) Sub=06 Prot=00 Driver=cdc_ether
> > > I: If#= 5 Alt= 1 #EPs= 2 Cls=0a(data ) Sub=00 Prot=00 Driver=cdc_ether
> > > I: If#= 6 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=42 Prot=01 Driver=(none)
> >
> > Thanks for the update and usb-devices info (you can even put this in the
> > changelog).
> >
> > > drivers/usb/serial/option.c | 2 ++
> > > 1 file changed, 2 insertions(+)
> > >
> > > diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
> > > index e24ff16..ca3e398 100644
> > > --- a/drivers/usb/serial/option.c
> > > +++ b/drivers/usb/serial/option.c
> > > @@ -1941,6 +1941,8 @@ static const struct usb_device_id option_ids[] = {
> > > { USB_DEVICE_AND_INTERFACE_INFO(WETELECOM_VENDOR_ID, WETELECOM_PRODUCT_6802, 0xff, 0xff, 0xff) },
> > > { USB_DEVICE_AND_INTERFACE_INFO(WETELECOM_VENDOR_ID, WETELECOM_PRODUCT_WMD300, 0xff, 0xff, 0xff) },
> > > { USB_DEVICE_AND_INTERFACE_INFO(0x03f0, 0x421d, 0xff, 0xff, 0xff) }, /* HP lt2523 (Novatel E371) */
> > > + { USB_DEVICE(0x1508, 0x1001), /* Fibocom NL668 series */
> > > + .driver_info = RSVD(4) | RSVD(5) | RSVD(6) },
> >
> > Looks like you can use USB_DEVICE_INTERFACE_CLASS() to match on the
> > vendor class instead of blacklisting interface 4 and 5.
> >
> > A more specific match is generally preferred over blacklisting (which
> > adds some overhead).
> >
> > Care to address that in a v3? Remember to include the patch revision in
> > the subject line as well (e.g. "[PATCH v3] USB: ...").
> It seems it would become problematic then if we change USB mode on the
> module to QMI instead of ECM network interface as they share same
> VID/PID value.
>
> Or is there other way to still get endpoints 0-3 bound to option
> driver on class / interface info? Subclass is different on the first
> diagnostical serial interface so USB_DEVICE_AND_INTERFACE_INFO()
> wouldn't be successful then either?
>
> T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 9 Spd=480 MxCh= 0
> D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
> P: Vendor=1508 ProdID=1001 Rev=03.18
> S: Manufacturer=Nodecom NL668 Modem
> S: Product=Nodecom NL668-CN Modem
> S: SerialNumber=5ced6a52
> C: #Ifs= 6 Cfg#= 1 Atr=a0 MxPwr=500mA
> I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option (Serial)
> I: If#= 1 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option (Serial)
> I: If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option (Serial)
> I: If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option (Serial)
> I: If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none) (QMI/RMNET)
> I: If#= 5 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=42 Prot=01 Driver=(none) (ADB)
And you also have ADB as interface five here.
I guess your original patch matching on all interfaces and
blacklisting interfaces 4, 5 and 6 is the best option. An alternative
could be two entries for ff/ff/ff and ff/00/00, but then you'd still
need to blacklist QMI so not that much better.
But please resubmit and include usb-devices for both messages in the
commit message.
Please also base this on my usb-linus branch which already holds some
new ids that prevents this one from applying.
https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial.git/
Thanks,
Johan
