Re: Null Pointer BUG in uhci_hcd

[Michael S. Zick]

On Tue July 7 2009, Alan Stern wrote:
> On Tue, 7 Jul 2009, Michael S. Zick wrote:
> 
> > Will add that to my list - -
> > I need to do that to find/fix that error path message flood loop in ehci anyway.
> 
> What EHCI message flood?
> 

Give me a chance to test what you have already put on my plate.
It isn't very interesting - when the driver shuts down a hub
for "problems found" (I can't be more specific at the moment)
It will try to immediately after resume the hub;
Which leads to it taking the hub out of service;
Which leads to it trying to resume the hub;
etc. etc. etc.
That message flood in the error recovery path.

Set: ignore_oc=1 on the command line and that will stop the flood
at the first message - but only because it hard-locks the kernel with irq's disabled.

On this machine, the "over current" sense is a false indication -
something is not within the expected range.

Thank goodness, no one has asked me to get this kernel certified
for life-support use (not possible, the VIA silicon isn't certified either).  ;)

Mike
> 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/