On Mon, Jun 05, 2000 at 08:28:25PM +0000, Mark Douglas Corner wrote:
> I have an unreleased device that I am attempting to get working under
> linux. It works fine under win98. Also I am having the same problem with
> an Intel USB camera. I know that there are no drivers for these, but as
> you can see that isn't the problem here. I did manage to get a USB
> keyboard to connect correctly.
>
> Fortunately I have a CATC USB Chief analyzer, although I have to note that
> I am somewhat new to USB.
>
> The problem occurs during the GET_DESCRIPTOR transaction. (The second one
> with the proper address add=2) The transaction is not acked. It is sent
> in five groups of three, none of them acked or nacked. There is no
> repsonse from the device at all.
I have seen similar behavior from a USB-Bluetooth module with
usb-uhci. The "set address" is ok, the status phase is naked a few times,
but then successfully completed. About 20ms later, the get_descriptor
fails fifteen times (5*3) without ack/nack.
But now the funny thing: If I simulate a disconnect/connect-cycle of the
device by shorting D+ to ground, the device enumerates without any problems.
:-O
I have tried extending all the waits in hub.c, even tried to simulate
Window's reset behavior, no success. However, I didn't try to disable the full
bandwidth reclamation...
--
Georg Acher, [EMAIL PROTECTED]
http://www.in.tum.de/~acher/
"Oh no, not again !" The bowl of petunias
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
