Re: [PATCH] pl2303, visor: Match I330 phone more precisely

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Wed, Apr 22, 2015 at 04:20:39PM +0200, Johan Hovold wrote:
> On Wed, Apr 22, 2015 at 02:38:00PM +0200, Greg Kroah-Hartman wrote:
> > On Wed, Apr 22, 2015 at 01:20:38PM +0200, Johan Hovold wrote:
> > > On Wed, Apr 22, 2015 at 12:14:05PM +0200, Jason A. Donenfeld wrote:
> > > > Samsung has just released a portable USB3 SSD, coming in a very small
> > > > and nice form factor. It's USB ID is 04e8:8001, which unfortunately is
> > > > already used by the pl2303 USB serial driver and the Palm Visor driver
> > > > for the Samsung I330 phone cradle. Having pl2303 or visor pick up this
> > > > device ID results in conflicts with the usb-storage driver, which
> > > > handles the newly released portable USB3 SSD.
> > > 
> > > First of all, the device should not be claimed by both pl2303 and visor.
> > > This predates the git, but it looks like the device id should simply be
> > > removed from pl2303. Care to do that as a preparatory patch?
> > 
> > It was added back in 2004 in the 2.4.26 kernel release.  Petr Slansky
> > asked me to add it, and sent a patch as he had the device.
> > 
> > But before that, back in 2003 Adam emailed saying that the visor driver
> > worked with the device, and sent me a patch for it.  I didn't notice the
> > duplicate ids for well over a decade now :(
> >
> > I'd bet this is really a pl2303 device, given that the visor driver was
> > just a "dumb" pipe to the device and the pilot sync tools never cared
> > about baud rates and the like, so odds are the visor entry should be
> > removed.
> 
> My thinking was the other way around; that a dumb, generic device
> (visor) would never work with a more specialised driver (pl2303).
> 
> The opposite could work though as the pl2303 setup commands would simply
> fail, but the bulk in/out layout appears to be the same and should work
> with a dumb device.
> 
> I double checked just now with a pl2303 HX device, and I can't seem to
> get anything through with the default settings (visor or generic
> driver) in loopback.

Ok, fair enough, I wonder how Petr got it to work.  Given that there
were reports of it working just fine for the visor driver a full year
before Petr's patch, odds are he forgot to build that driver into his
kernel.

thanks,

greg k-h