On Wed, Nov 24, 2010 at 6:50 PM, Filip Aben <f.aben@xxxxxxxxxx> wrote: > This patch just masks one of the symptoms. hso_log_port should never be > called for a net device which is not created in the first place. > > Digging deeper I found this gem: > > if (((port_spec & HSO_PORT_MASK) == HSO_PORT_NETWORK) && > !disable_net) > hso_dev = hso_create_net_device(interface, > port_spec); > else > hso_dev = > hso_create_bulk_serial_device(interface, > port_spec); > > So if we need to make network port but disable_net is set, we just make > a serial device instead on that interface, which is obviously wrong. I'm not saying that there aren't things that need to be fixed with this driver... The driver has been creating this useless serial device since it was accepted back in 2008. Someone suggested simply not creating any device at all for HSO_PORT_NETWORK if disable_net is set about two years ago but this patch was never accepted. Obviously, something like that would be preferred, but my patch could simply be a minimal fix for an oops (and as such could be suitable for stable) in the mean time. Thanks, 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
