On Tue, Oct 28, 2014 at 11:50:48AM +0100, Johan Hovold wrote:
> On Tue, Oct 28, 2014 at 11:37:15AM +0100, Oliver Neukum wrote:
> > On Tue, 2014-10-28 at 11:17 +0100, Johan Hovold wrote:
> > > On Tue, Oct 28, 2014 at 10:12:08AM +0000, Karl Palsson wrote:
> > > > On Tue, Oct 28, 2014 at 10:20:29AM +0100, Oliver Neukum wrote:
> > > > > Let's drop the warning for modems with unusual capabilities,
> > > > > the associated quirk and blacklist. They made little sense.
> > > > >
> > > > [snip]
> > > > > @@ -1796,11 +1794,6 @@ static const struct usb_device_id acm_ids[] = {
> > > > >
> > > > > /* NOTE: non-Nokia COMM/ACM/0xff is likely MSFT RNDIS... NOT a modem! */
> > > > >
> > > > > - /* Support Lego NXT using pbLua firmware */
> > > > > - { USB_DEVICE(0x0694, 0xff00),
> > > > > - .driver_info = NOT_A_MODEM,
> > > > > - },
> > > > > -
> > > >
> > > > Isn't this dropping the device too, not just the quirk?
> > >
> > > Depending on what its descriptors look like, you may be right. The commit
> > > introducing the quirk:
> > >
> > > ce126644aa10 ("USB: cdc_acm: Silence "It is not a modem." error for pbLua devices")
> > >
> > > looks like if it was just silencing the error message of a device that
> > > was already recognised by the driver, though.
> >
> > Yes, we wouldn't get a warning without probe() being called.
>
> That's obvious.
>
> Karl's point was that if this device is as quirky as his (e.g. has a
> byte-swapped bInterfaceSubClass field), it would no longer be detected
> after your change.
Sorry, I mis-read your reply.
It's a long shot, but someone could have added the device id to the
table (because it was quirky), seen the error message, and then
submitted the id along with the quirk.
Possible, but still unlikely enough to be of no real concern.
We can always add it back if anyone complains.
Johan
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
