On Sun, 29 Aug 2004, David Brownell wrote: > > So all the > > device sees is a port disable (no USB data sent -- I forget whether the > > hub sets the data lines to SE0), followed shortly by a reset, followed by > > re-enumeration. > > Well, re-enumeration in _some_ cases.
The logical-disconnect pathway always does a re-enumeration unless the port status says the device has been unplugged. It even does a debounce, although that's clearly not necessary. It might become necessary, though, once power switching is added into the mix! > Don't we still > need a mode where we can disable a port because > the device is just broken? (SE0 should be a suspend, > as I recall.) If we just want to disable a port, it ought to be enough to call hub_port_disable(). There's no need to disconnect the usb_device structure until the device itself is physically unplugged. I don't think there's any place where the kernel currently wants to do this. If driver binding fails, for instance, we don't do anything at all. Alan Stern ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
