On Thu, 2005-05-19 at 12:34 -0700, John Klos wrote:
> I had another panic today (7457 G4, netbsd-2, no ALTIVEC option in the 
> kernel config):
[...]
> What's interesting is that I've never seen lost interrupts on the 
> motherboard IDE controller - ever. They're common on PCI IDE cards, but 
> this is new. Is this because the state of the interrupt handlers is all 
> screwy when in the debugger?
> 
> Any ideas about the panic in general?

I don't know about the specific panic you had, but I have seen the
lost interrupts you mention multiple times under i386 (and it was also
in network-related panics).

I think it's because the panic happened in a place where the interrupts
are disabled up to a level where the disk ones are also blocker.  I.e.,
the kernel was at splnet during the panic, thus blocking splbio too.
Am I right?

Cheers,

-- 
Julio M. Merino Vidal <jmmv84@gmail.com>
http://www.livejournal.com/users/jmmv/
The NetBSD Project - http://www.NetBSD.org/