lkml.org 
[lkml]   [2009]   [Jul]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] x86: sysctl to allow panic on IOCK NMI error

* Kurt Garloff <garloff@suse.de> wrote:

> Ingo,
>
> On Wed, Jul 01, 2009 at 01:10:03PM +0200, Ingo Molnar wrote:
> > * Maciej W. Rozycki <macro@linux-mips.org> 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 ...

We dont _necessarily_ crash ... The crash/panic is default off and
sysctl driven.

Allowing a crash is not the highest quality of implementation and
i'm somewhat wary of 'allow a little bit of crap' arguments - it's
too similar to the 'little bit pregnant' concept ;-)

Ingo


\
 
 \ /
  Last update: 2009-07-03 11:27    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site