On Fri, 16 Sep 2011, Sebastian Andrzej Siewior wrote: > * Greg KH | 2011-09-09 11:43:54 [-0700]: > > >So we can expect more patches like this adding all of the possible > >permutations of the different gadget devices? > > > >If so, ick, I thought that the work that was done was to prevent this > >from happening, oh well :( > > Would it be possible to come up with a gadget-hub driver which can have > multiple gadget attached? This should get rid of this kind of gadget > drivers, right? Instead of implementing a hub, it might be possible to redo the composite driver in such a way that the individual function drivers are added as separate modules at runtime, rather than all bound together at compile time. The details would be tricky, but I think it could be done. One important point is that somehow you have to inform the composite core when all the function drivers have been loaded and initialized -- otherwise there's no way to tell when it's safe to enable the pull-ups and connect to the host. Alan Stern -- 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
