Re: [Xen-devel] PVops domain 0 crash on NUMA system only Node==1 present (Was: Re: Bug#603632: linux-image-2.6.32-5-xen-amd64: Linux kernel 2.6.32/xen/amd64 booting fine on bare metal, but not as dom0 with Xen 4.0.1 (Dell R410))

[Ian Campbell <Ian.Campbell@xxxxxxxxxxxxx>]

On Fri, 2010-12-03 at 00:27 +0000, Vincent Caron wrote:
> On Thu, 2010-12-02 at 16:12 -0800, Jeremy Fitzhardinge wrote:
> > On 12/02/2010 03:47 PM, Vincent Caron wrote:
> > >   It just happens that your kernel above (2.6.32-27+numa1) boots fine
> > > under hypervisor _when_ passed 'numa=noacpi'. Yeah !
> > >
> > >   I then tried again with Debian Squeeze's latest 2.6.32-28, which
> > > crashes as -27 under hypervisor (and changelog show no xen or
> > > numa-related thingies). Then I added 'numa=noacpi', and it boots fine
> > > too. I got my 8 cores, networking, etc.
> > >
> > >   Enclosed is the dmesg for the latter, Debian, kernel.
> > >
> > >   Is the 'numa=noacpi' a "production acceptable" workaround ?
> > 
> > What about "numa=fake=1"?  I think that should force it to create a
> > single NUMA node.
> 
>   Debian AMD64 2.6.32-28 on hypervisor 4.0.1, option 'numa=fake=1',
> boots fine too on my R410. Dmesg attached.

Thanks.

>   PS: in both "numa=noacpi" and "numa=fake=1" cases where I can actually
> run with a 4.0 hypervisor, I can't reboot. Dmesg ends with 'Restarting
> system.' but the machine never reboots. OTOH reboots works while booting
> native with the same kernel.

That was a known bug in 2.6.32.<something-recent>. AIUI a revert is in
the pipeline for the next 2.6.32.y and Jeremy has a proper fix queued
too, which I presume will make its way to stable in due course.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel