> >How much overhead it places in case of normal operations?
> >If not much, I am willing to apply it.
> The best thing to do in such case is to press SYSRQ+P and to grab lots of
> different EIP addresses. If SYSRQ+P does nothing then it's an hard lockup
> and in such case the NMI oopser will help you.
Ok. I'll try this too.
> So I suggest you to use the IKD patch and to enable only the NMI oopser.
Hmm? I thought that NMI oopser is only applicable to 2.3 kernels..
> If at lockup time you won't see a stack trace, then press SYSRQ+P and
> report us the interesting EIP addresses (the System.map is necessary to
> resolve them later). Thanks.
But I see stack trace even without NMI oopser (from __global_cli)
Or do you want to get trace from second CPU too?
It is in BH handler. And I am suspect that it is in BH handler from serial...
Anyway, thanks.
Bye,
Oleg
-- * Если враг не сдается - его перезагружают! [RAVE *SUCKS* TEAM] [Team $Ui(iDE] [Trans-M Must Die!] [Sepultura Fans Team] [Tolkien forever team]- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/