Re: [PATCH 5/6] x86, NMI: Allow NMI reason io port (0x61) to be processedon any CPU

From: Cyrill Gorcunov
Date: Mon Feb 28 2011 - 13:48:48 EST

On 02/28/2011 09:37 PM, Don Zickus wrote:
On Sun, Feb 27, 2011 at 02:19:11PM +0300, Cyrill Gorcunov wrote:
On 02/27/2011 04:01 AM, huang ying wrote:

Probably we should put question in another fashion, ie in the fasion of
overall design -- who should be
responsible for handling external nmis, 1) the cpu which apic is configured
to observe such nmis or 2) any cpu?
If we take 1) then no lock is needed and underlied code will report real cpu
number who observed nmi. If
we take 2) then lock is needed but we need a big comment in default_do_nmi
together with probably cpu number
fixed in serr\iochk printk's.

I am OK with both solutions.

Best Regards,
Huang Ying

ok, lets see what others think on this thread

I'm trying to figure out how this affects SGI's systems which currently
enable external NMIs to all cpu's in order to support their nmi button to
dump cpu stacks on a system hang

But feel free to post patches addressing your concerns as I am getting a
little lost in the all the concerns being thrown back and forth.


I was planning to do so today but seems out of time at moment (thought I will
try ;), in particular I thought about dropping lock for a while and restore old
behaviour. *BUT* same time to put some big comment explaining why we do this
(so that Ying's work would not be wasted but rather deffered until proper apic
reconfig implemented).

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at