Re: [Libusb-devel] Patch: libusb/os/linux_usbfs.c: Distingush between usbfs and other kernel mode drivers

[Alan S. <st...@ro...>]

On Wed, 6 Oct 2010, Xiaofan Chen wrote:

> On Wed, Oct 6, 2010 at 4:31 PM, Peter Stuge <pe...@st...> wrote:
> >> Still I am not that convinced that "usbfs" is really a kernel driver,
> >
> > What would convince you?
> 
> If the Linux kernel USB developer like Alan Stern or Greg KH
> say so, I would be convinced.

This probably doesn't matter any more, but for the record, Peter is 
right.  usbfs really is a kernel driver.  A slightly peculiar one, to 
be sure, because it doesn't take any action on its own -- it merely 
relays requests from userspace.  Nevertheless, it is treated just like 
any other kernel driver.

> If there is a kernel driver attached to the USB device (an interface),
> then libusb_claim_interface() should fail. And this is consistent
> with my view that "usbfs" is not a kernel driver in the context
> of libusb.

If another program has claimed an interface, then 
libusb_claim_interface() _will_ fail.

Alan Stern