On Mon, Sep 20, 2010 at 2:26 PM, Konrad Rzeszutek Wilk
<konrad.wilk@xxxxxxxxxx> wrote:
> On Mon, Sep 20, 2010 at 11:17:06AM -0700, Bruce Edge wrote:
>> On Mon, Sep 20, 2010 at 9:40 AM, Konrad Rzeszutek Wilk
>> <konrad.wilk@xxxxxxxxxx> wrote:
>> > On Thu, Sep 16, 2010 at 07:13:35AM -0700, Bruce Edge wrote:
>> >> On Thu, Sep 16, 2010 at 12:53 AM, Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
>> >> wrote:
>> >> > On 16/09/2010 00:49, "Bruce Edge" <bruce.edge@xxxxxxxxx> wrote:
>> >> >
>> >> >> (XEN)
>> >> >> (XEN)
>> >> >> (XEN) NMI - MEMORY ERROR
>> >> >
>> >> > PCI parity errors, possibly? It looks rather like you are getting an NMI
>> >> > with fatal consequuences.
>> >> >
>> >> > -- Keir
>> >> >
>> >>
>> >> It seems very odd that a parity error would be so reliably
>> >> recreatable. This also doesn't happen with xen-testing.
>> >
>> > This is irregardless of the PVOPS kernel, right?
>>
>> This does not happen running hvm domU. Only pvops domU kernels have
>> this problem.
>
> Huh? PVOPS DomU kernel can run as HVM or PV. What do you mean by 'HVM DomU'?
>
Sorry, I forget there are so many ways one can run the pvops kernel.
I meant that it didn't happen when I was running an HVM domU using a
different bare-metal kernel. IOW, a non-pvops HVM kernel.
It does happen when I use the pvops kernel in PV mode for the domU.
I have not tried the pvops kernel in hvm mode. I'll try that next.
-Bruce
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|