On Tue, 21 Oct 2003, Greg KH wrote: > > Hmm... The interface number fix means making adjustments to a number of > > drivers. Here's the idea: Up to now the USB stack has supported the > > notion, explicitly stated in the USB specification, that interface numbers > > are 0-based array indexes into the list of interfaces supported by a > > configuration. To handle non-compliant devices correctly we need to stop > > supporting that notion -- instead the routine ifnum_to_if() will supply an > > opaque mapping from interface numbers to interfaces. But many drivers use > > the technique of just looking up the bInterfaceNumber element of the > > usb_host_config array. They will all have to be changed to use > > ifnum_to_if() instead. > > I agree. This needs to be fixed. patches to fix this will be accepted, > as well as a patch to the core to allow 1-based interface numbers to > work properly (it's a simple error that lots of firmware developers > probably got wrong.)
Okay, when I have a chance I'll take a look at it. Do you think it will be sufficient to allow 1-based interface numbers, or should we also accept totally-broken arbitrary noncontiguous numbers? > > Is there any way we can set up some sort of temporary tree, so that large > > changes can be shared in the meantime among developers easily via > > BitKeeper? > > Like my usb-2.5 tree? > > Or another one? usb-2.5 is fine with me, since that's the one I have cloned anyhow. I suppose you'll want to change the name at some point, though. Alan Stern ------------------------------------------------------- This SF.net email is sponsored by OSDN developer relations Here's your chance to show off your extensive product knowledge We want to know what you know. Tell us and you have a chance to win $100 http://www.zoomerang.com/survey.zgi?HRPT1X3RYQNC5V4MLNSV3E54 _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
