On Wed, Jun 06, 2001 at 05:30:18PM -0700, David Brownell wrote:
> > > > 75 = EOVERFLOW - never seen this for USB. some grep told me it is ohci
> > > > only and detected by the HC
> > >
> > > I've seen it; EHCI hardware will also report it. Are you sure UHCI won't?
> > > That seems strange. It's just an on-the-wire error, nothing HC-specific
> > > about whether it can happen.
> >
> > Can you describe the nature of this "on-the-wire" error?
>
> Let's see ... linux/Documentation/usb/error-codes.txt says it happens
> when the device returns more data than the maximum packet size, or
> the remaining buffer size. In short, sending more data than the device
> driver told the HCD to expect.
>
> A quick grep suggests that the UHCIs map TD_CTRL_BABBLE
> (sounds like the same error) to -EPIPE, which otherwise means a
> stall ... that seems highly likely to confuse something. A bug, even.
Ack! This is a serious issue.... clear indications of STALL or non-STALL
are critical for usb-storage.
Matt
--
Matthew Dharm Home: [EMAIL PROTECTED]
Maintainer, Linux USB Mass Storage Driver
It's not that hard. No matter what the problem is, tell the customer
to reinstall Windows.
-- Nurse
User Friendly, 3/22/1998
PGP signature
