Re: kernel 3.10.1 - "NMI received for unknown reason"

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Aug 02, 2013 at 08:24:38AM +0200, Stefan Pietsch wrote:
> On 31.07.2013 11:20, Gleb Natapov wrote:
> > On Wed, Jul 31, 2013 at 11:10:01AM +0200, Stefan Pietsch wrote:
> >> On 30.07.2013 07:31, Gleb Natapov wrote:
> >>
> >>> What happen if you run perf on your host (perf record -a)?
> >>> Do you see same NMI messages?
> >>
> >> It seems that "perf record -a" triggers some delayed NMI messages.
> >> They appear about 20 or 30 minutes after the command. This seems strange.
> > Definitely strange. KVM guest is not running in parallel, correct? 20, 30
> > minutes after perf stopped running or it is running all of the time?
> 
> No, the KVM guest ist not running in parallel. But I'm not able to
> clearly reproduce the NMI messages with "perf record".
> I start "perf record -a" and after some minutes I stop the recording.
> 
> After that it seems NMI messages appear within a random period of time.
> So, I cannot tell what triggers the messages.
When you run KVM with coreduo cpu model it emulates PMU which basically
make is perf front end. If you can reproduce the messages with perf too
it probably means that the problem is not in the KVM itself. If you
disabled NMI watchdog in the guest the messages may go away.
Can you send your guest's dmesg when you boot it with coreduo mode?

--
			Gleb.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux