Am Mittwoch, den 17.06.2015, 10:43 -0400 schrieb Alan Stern: > On Tue, 16 Jun 2015, Stefan Koch wrote: > > > > If autoprobing is disabled, all the user has to do is avoid writing to > > > /sys/bus/usb/drivers_probe. > > With my test case that works only with bus_probe_device() becaus it > > checks the autoprobe status and calls then device_attach(). > > > > I have compiled the source once with bus_probe_device() and once with > > device_attach() only. > > > > To conclude: bus_probe_device() was required to ensure that if > > autoprobing was off that no driver was probed without writing it to > > drivers_probe... > > You don't understand what I mean. Your patches should never do _any_ > probing. All probing can be initiated by the user after the interfaces > are authorized. > > If autoprobing is off, the user doesn't have to do anything. Nothing > will be probed. > > If autoprobing is on, the user can tell the kernel to probe the > authorized interfaces by writing to /sys/bus/usb/drivers_probe. > > Alan Stern > You write in another mail: "You could probe all the interfaces whenever any interface is authorized. Or there could be a separate mechanism to initiate probing." The first is the actual approach and this works fine. It is regardless in which order the interfaces for USB-Tethering are authorized. Both works. Before probing *all* interfaces it was needed to authorize interface 1 before interface 0. Also btusb works fine. What do you think? -- 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
