Re: [linux-usb-devel] Fw: [BUG] USB Storage OOPS and a D state process in 2.6.10

[Srihari Vijayaraghavan <harisri@internode.on.net>]

On Saturday 08 January 2005 14:51, Alan Stern wrote:
> ...
> I see you didn't have CONFIG_USB_STORAGE_DEBUG set when you ran this test.

Actually I did. I do not know what log level the USB debug messages were, but 
the silly syslog did not capture them. Now I have set it to write all kernel 
messages.

> Or else you didn't have your syslog daemon configured to store the
> debugging messages.  If you can get anything unusual to happen with the
> debugging enabled, it would make things much easier to track down.

That is right. I am trying to trigger the bug, and I have ensured that the USB 
debug messages are being written to the log file. But I have not triggered 
the bug yet.

> ...
> > Jan  8 12:35:10 desktop kernel: ehci_hcd 0000:00:10.4: port 5 reset error
> > -110 Jan  8 12:35:10 desktop kernel: hub 1-0:1.0: hub_port_status failed
> > (err = -32)
> > Jan  8 12:35:10 desktop kernel: ehci_hcd 0000:00:10.4: port 5 reset error
> > -110 Jan  8 12:35:10 desktop kernel: hub 1-0:1.0: hub_port_status failed
> > (err = -32)
>
> Those errors are quite unusual.  They seem to indicate there's a problem
> with your computer's USB hardware, as do the numerous spontaneous (?)
> disconnections in your log.  However I'm not familiar with all the details
> of the EHCI driver.  It's faintly possible that plugging the drive into a
> different USB port would work better.

The numerous disconnections (because without debug, I can trigger the D state 
process and oops within a dozen connection/disconnection, but with debug it 
takes 10 times more connection/disconnection to trigger D state process, and 
I do not know how many are needed for oops :-( ) are purposefully generated 
to trigger the bug.

Of course UHCI (USB 1) is not as fast as EHCI (USB 2). Would that not matter 
for CD/DVD burning?

> If you still can't get anything with the debugging option, try running
> without the debugging code and rmmod ehci-hcd before plugging in the
> device.  Maybe if you use a UHCI controller instead the errors won't
> occur.

OK I shall try with UHCI.

Thank you.
Hari