Messages in this thread |  | | | Date | Fri, 3 Jul 2009 09:59:24 +0200 | | From | Ingo Molnar <> | | Subject | Re: [PATCH] x86: sysctl to allow panic on IOCK NMI error |
* Borislav Petkov <borislav.petkov@amd.com> wrote:
> On Thu, Jul 02, 2009 at 09:53:05AM +0200, Ingo Molnar wrote: > > > That shouldn't be a problem if we were about to panic(). For a > > > more sophisticated attempt of recovery -- yes, that would have to > > > be addressed. > > > > We are only panic-ing if the sysctl is set. The diagnostics > > would be useful anyway. The proper approach would be to defer it > > a bit in the non-panic case an read it out from some friendlier > > context - such as the EDAC core. > > Quickly skimming through code shows some functionality is there: > drivers/edac/edac_pci{,_sysfs}.c but doesn't seem to be NMI safe. > CCing some more people.
the 'defer' would mean we put it into a softirq context or so and do it from there - not from NMI context.
But yes, if we are about to panic due to this and want to print out something useful, making EDAC NMI safe would be nice.
Ingo
|  |