I guess I don't understand this part of it: the presence of a library or header doesn't mean there is an accessible device, does it? I realize looking at dmesg output is less than optimal, but I'm not sure how else to discover and access devices.
Oliver Rauch wrote: > Hello Paul, > > this is no solution for the backend. May be it could be a way for > configure to find out if USB is available, but I would prefer a test > for an include file or a library. > > Are the other USB backends compiled on systems that do not support USB?
