Re: [PATCH] x86: sysctl to allow panic on IOCK NMI error

[Kurt Garloff]

Ingo,

On Wed, Jul 01, 2009 at 01:10:03PM +0200, Ingo Molnar wrote:
> * Maciej W. Rozycki <macro@xxxxxxxxxxxxxx> wrote:
> 
> > On Tue, 30 Jun 2009, Greg KH wrote:
> > 
> > > >  These days an IOCK NMI typically happens in response to a PCI 
> > > > SERR -- it may be useful to traverse PCI buses to find the 
> > > > offender and dump this information on this occasion too.  The 
> > > > south bridge may have additional status too.
> > > 
> > > Sure, that would be great to have.  Care to make a patch?  :)
> > 
> >  ENOTIME, sorry.  Next year perhaps.  Or a homework project for 
> > one of the newbies. ;)
> 
> You know that this project would kill a newbie, right? :)
> 
> We have no real southbridge drivers on x86 - but we should certainly 
> add some. Also, walking the PCI device tree from NMI context is 
> tricky as the lists there are not NMI safe - we could crash if we 
> happen to get a #IOCK while loading/unloading drivers (which is rare 
> but could happen).

Well -- in case we panic the system anyway this is not necessarily a
big issue (let's print the message before ...) -- if we crash trying
to gather additional info, we'll lose the info. Currently we never have
the info ...

> IMHO it's all very much desired functionality, but highly 
> non-trivial.

Too bad.

Best,
-- 
Kurt Garloff, VP OPS Partner Engineering -- Novell Inc.

Attachment: pgp00000.pgp
Description: PGP signature