On Thu, Jan 22, 2009 at 06:28:34PM +0100, Raúl Sánchez Siles wrote:
> Hello:
>
> I think this bug is closely related to [0] but I'm not sure.
>
> Bets are, that again in this case, you're facing an unsupported assembler
> instruction which is not supported by your CPU. The first panic message
> retrieved from kernel was useful in those days to diagnose the cause of the
> panic, but unfortunately I've been unable to find that linux-image package
> version and I couldn't check in the System.map file which function contains the
> crash.
>
> Maybe you can still reproduce it and provide with a fresh panic message and
> either check yourself which function breaks or providing both the crash info
> and the running kernel System.map.
>
> Anyway, I'd suggest reading my latter comment on [0] and taking conclusions
> from there. But I have the impression this time again, it's virtualization
> that's breaking down. I think report should arrive upstream somehow.
Before reporting this upstream, we should test whether it might already
been fixed in the mean time. Can you test, whether a 2.6.30 kernel from
backports.org exhibits the same behaviour?
Cheers,
Moritz