On Fri, Sep 19, 2003 at 09:41:59AM -0500, Doug Maxey wrote: > Did the complete dump of /proc/bus/usb that illustrates the device > tree that is giving me fits ever make it to this list? I can resend > if needed.
We can see the problem, but there isn't any workaround. You have two
devices with the exact same descriptors, but which are not exactly the
same. There isn't anything we can do.
We did ask for debug logs to see if we could fix the incompatibility
between the CB protocol and the later-revision device, but you never sent
those.
> It seems that we have another device, in addition to the TEAC floppy
> that has had an unusual_dev entry, this time for the ISC-200. This
> device also appears to be somewhat different from the device matched
> by the original code.
Can you elaborate?
> Did I ever hear a design choice on how to at least workaround this
> issue?
No.
> My current thought is to have a documented config variable that
> enables the code to match the devices that are being shipped by IBM
> on the particular platform. Is this a reasonable solution?
It is reasonable for kernels distributed by IBM. I am highly resistant to
accepting such code into the mainstream kernel. If we can fix the CB
protocol to work with the device, I'd rather do that.
Matt
--
Matthew Dharm Home: [EMAIL PROTECTED]
Maintainer, Linux USB Mass Storage Driver
THEY CASTRATED MY QUAKE BITS! I WANT THEM BACK!!!!
-- Greg
User Friendly, 3/27/1998
pgp00000.pgp
Description: PGP signature
