Re: kernel trap 19 with interrupts disabled

On Thu, 10 Jun 2004, Andy Farkas wrote:

> > Don Bowman wrote:
> >
> [... about a different type of NMI]

> My box did the following the other day (its a quad ppro-200 with ecc ram;
> dell 6100/200).  I'd like to know if it is safe to continue running the
> kernel as is, or should it be rebooted?  I pressed 'c' at the debugger
> prompt and nothing bad has happened so far. I've rebooted since.
>
> Jun  6 17:44:47 <kern.crit> hummer kernel: NMI ISA a8, EISA 0
> Jun  6 17:44:47 <kern.crit> hummer kernel: RAM parity error, likely hardware failure.
> Jun  6 17:44:47 <kern.crit> hummer kernel:
> Jun  6 17:44:47 <kern.crit> hummer kernel: Fatal trap 19: non-maskable interrupt trap while in kernel mode
> Jun  6 17:44:47 <kern.crit> hummer kernel: cpuid = 0; apic id = 00
> Jun  6 17:44:47 <kern.crit> hummer kernel: instruction pointer  = 0x8:0xc049cd12

Since the address of the memory with the parity error (if that's really what
caused the NMI) is unknown, there is no way to tell.  I don't know if NMI's
for parity errors are or can be delivered asynchronously, but guess that
they aren't, so the problem might not be near the instruction pointer.

Bruce