speedtch usbatm.h,1.4,1.5 usbatm2.c,1.13,1.14

[Roman Kagan]

  Hi Duncan,

On Sat, Feb 12, 2005 at 03:35:00PM +0100, Duncan Sands wrote:
> For me the main problem with getting these changes into the kernel tree
> is that they are not backwards compatible:
> (1) the ioctl for signalling that firmware has been loaded from
> userspace has gone.  It is still possible to load things from userspace,
> by unbinding the kernel driver, loading the firmware, then rebinding it,
> but the old modem_run method will no longer work.

But those ioctls were removed deliberately, weren't they?  If still
there's enough demand to keep using modem_run with it, AFAICS modem_run
can be modified to do ioctl(USBDEVFS_(DIS)CONNECT) at the beginning/end
of firmware loading process, and retain compatibility with older
versions.

> (2) the ATM device is only created once the USB device is fully
> initialized, not straight away like now.  This may cause trouble for
> scripts.

If the device is created, but the firmware isn't loaded yet, any attempt
to open the atm device will fail.  So scripts have to work around this
delay anyway, even with 2.6.10 version.

> What is more, the ATM layer does not yet support sysfs (hotplugging).

Well, the patch has been submitted :)  Waiting for an ACK from Chas.

> > Besides, I think we need to make up a usbatm subtree buildable within
> > the kernel, so that users of 2.6.1[01] could fetch it and replace the
> > original contents of drivers/usb/atm.  I'll add Kconfig and
> > Makefile.kernel to CVS, if you don't mind.
> 
> OK.

Done (the mail notification apparently didn't go through).

Cheers,
  Roman.