So I have to wait for xvm 3.3 release?

On Tue, Nov 11, 2008 at 2:41 PM, Lu Baolu <[EMAIL PROTECTED]> wrote:
> Hi,
>
> Domain 0 boots successfully, but when use xvm tools to start the HVM
> guests, the system hangs. The input and output don't work. And the CPU
> fans spin in highest speed.
>
> I am a newbie to xVM debugging. Can anybody give me some hints about
> how to get debugging information in this case? One thing is that I
> can't enter kmdb through press "F1+A", although it works before the
> system hangs.
>
> Thanks,
> Allen
>
> On Tue, Nov 11, 2008 at 2:02 PM, Lu Baolu <[EMAIL PROTECTED]> wrote:
>> Hi David,
>>
>> This xen-unstable snapshot (changeset #17779) doesn't support vt-d2
>> features. After disabling vt-d2 in BIOS setting (vt-d1 still remaining
>> enabled), the domain 0 boots well. Hope the coming 3.3 will fix this
>> problem.
>>
>> vt-d1 is just ok for my current phase 2 work.
>>
>> Thanks,
>> Allen
>>
>>
>> On Fri, Nov 7, 2008 at 10:57 AM, Lu Baolu <[EMAIL PROTECTED]> wrote:
>>> David - I am in vaccation. I will post the debug information the next 
>>> Monday.
>>>
>>> Thanks, Allen
>>>
>>> On Thu, Nov 6, 2008 at 11:40 PM, David Edmondson <[EMAIL PROTECTED]> wrote:
>>>> On Thu, Nov 06, 2008 at 11:16:17PM +0800, Lu Baolu wrote:
>>>>> As you know, I have built a system with the developing xvm source and
>>>>> onnv in matrix-unstable gate, and managed to create HVM guests for
>>>>> Fedora 7, Nevada snv_89 and Windows XP sp2. I did all these with vt-d
>>>>> disabled in BIOS setting. When I turned on vt-d, domain 0 can't do
>>>>> disk io during boot up time. After several failures on reading data
>>>>> from disk, domain 0 panic'ed and rebooted.
>>>>>
>>>>> Is this a known issue that somebody has got it fixed or mine is the
>>>>> first try on a vt-d enabled system?
>>>>
>>>> It's possible that you are the first to run our dom0 on VT-d, as I've
>>>> not tried that myself.
>>>>
>>>> What was the panic?
>>>>
>>>
>>
>
_______________________________________________
xen-discuss mailing list
[email protected]

Reply via email to