Re: Null Pointer BUG in uhci_hcd

[Alan Stern]

On Tue, 7 Jul 2009, Michael S. Zick wrote:

> Here is an example of the message paths taken - I have another 27Mbytes of the same
> if this snippet isn't enough.

> Jun 30 10:38:00 cb01 kernel: hub 2-0:1.0: over-current change on port 1
> Jun 30 10:38:02 cb01 kernel: hub 2-0:1.0: connect-debounce failed, port 1 disabled
> Jun 30 10:38:02 cb01 kernel: hub 2-0:1.0: over-current change on port 2
> Jun 30 10:38:02 cb01 kernel: hub 2-0:1.0: over-current change on port 1
> Jun 30 10:38:04 cb01 kernel: hub 2-0:1.0: connect-debounce failed, port 1 disabled
> Jun 30 10:38:04 cb01 kernel: hub 2-0:1.0: over-current change on port 2

This is typical of the errors on bus 2.  ehci-hcd receives
status-change interrupts because of the over-current-change flag.  The
hub driver tries to debounce port 1 (presumably because a device is
connected there) but gives up after two seconds.  Then the cycle
repeats.

> Jun 30 10:38:34 cb01 kernel: hub 1-0:1.0: over-current change on port 1
> Jun 30 10:38:34 cb01 kernel: ehci_hcd 0000:00:10.4: port 1 resume error -19
> Jun 30 10:38:34 cb01 kernel: hub 1-0:1.0: hub_port_status failed (err = -32)
> Jun 30 10:38:34 cb01 kernel: hub 1-0:1.0: connect-debounce failed, port 1 disabled
> Jun 30 10:38:34 cb01 kernel: hub 1-0:1.0: over-current change on port 2
> Jun 30 10:38:34 cb01 kernel: ehci_hcd 0000:00:10.4: port 2 resume error -19
> Jun 30 10:38:34 cb01 kernel: hub 1-0:1.0: hub_port_status failed (err = -32)
> Jun 30 10:38:34 cb01 kernel: hub 1-0:1.0: connect-debounce failed, port 2 disabled

This is typical of the problems on bus 1.  Again we see
over-current-change flags, and this time there's the "resume error
-19".  The only reason that code can appear is if a read returns
0xffffffff, i.e., ~(u32)0.  See handshake() in ehci-hcd.c, and see the
call to handshake in ehci-hub.c shortly before the line containing
"resume error".

It's like I said, your hardware really isn't working properly.

Alan Stern

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/