| |
|
 |
|
 |
|
|
|
| |
|
|
xen-devel
[Xen-devel] RE: VT-d issues
|
To: |
"Kamala Narasimhan" <Kamala.Narasimhan@xxxxxxxxxx>, "Kay, Allen M" <allen.m.kay@xxxxxxxxx> |
|
Subject: |
[Xen-devel] RE: VT-d issues |
|
From: |
"Han, Weidong" <weidong.han@xxxxxxxxx> |
|
Date: |
Mon, 12 May 2008 10:31:59 +0800 |
|
Cc: |
Ian Pratt <Ian.Pratt@xxxxxxxxxxxxx>, xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>, Keir Fraser <Keir.Fraser@xxxxxxxxxxxxx>, "He, Qing" <qing.he@xxxxxxxxx> |
|
Delivery-date: |
Sun, 11 May 2008 19:32:29 -0700 |
|
Envelope-to: |
www-data@xxxxxxxxxxxxxxxxxx |
|
In-reply-to: |
<0519110A9448BC429B231AD118D72708039A5BC1@xxxxxxxxxxxxxxxxxxxxxx> |
|
List-help: |
<mailto:xen-devel-request@lists.xensource.com?subject=help> |
|
List-id: |
Xen developer discussion <xen-devel.lists.xensource.com> |
|
List-post: |
<mailto:xen-devel@lists.xensource.com> |
|
List-subscribe: |
<http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe> |
|
List-unsubscribe: |
<http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe> |
|
References: |
<0519110A9448BC429B231AD118D72708039A5809@xxxxxxxxxxxxxxxxxxxxxx> <08DF4D958216244799FC84F3514D70F00156920F@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> <0519110A9448BC429B231AD118D72708039A5BC1@xxxxxxxxxxxxxxxxxxxxxx> |
|
Sender: |
xen-devel-bounces@xxxxxxxxxxxxxxxxxxx |
|
Thread-index: |
AciwbZfvcYfvT0dBRY+4ihskqtMIFQASCjLAABb2wVAAsXOsEA== |
|
Thread-topic: |
VT-d issues |
Hi Kamala,
It is caused by the Video ram dirty shadow optimization (cs. 17564)
which wants to touch m2p for passthrough MMIO, while m2p entries for
direct MMIO pages (e.g. 0xffxxx000) don't exist. We will work out a
patch for it later.
Randy (Weidong)
Kamala Narasimhan wrote:
>> Hi Kamala,
>>
>> Our nightly testing didn't encounter issue 1) before, but we didn't
>> test on the same machine you attached. Anyway, we will investigate
>> it. And also, can you try out adding "hap=0" in your hvm.conf?
> Adding "hap=0" to the conf file did not help, unfortunately;
> crashes at the same place. Let me know if I can try something else or
> get any further info you need to help solve this issue.
>
> Thanks,
> Kamala
>
>> For issue 2), I think it's already fixed by c/s 17584.
>>
>> Randy (Weidong)
>>
>> Kamala Narasimhan wrote:
>>> I tried VT-d pass-through with the latest xen-unstable and
>>> encounter the following issue(s) -
>>>
>>> 1) When I pass through a network adapter (doesn't matter which
>>> kind), while I see the device within guest OS (tested with Vista
>>> and XP guest), I encounter a machine reset when the driver for the
>>> passed through network adapter is being installed (either through
>>> plug and play or by manual installation). I have attached
>>> information about the system in question. Has anyone tried
>>> pass-through on a system with the exact configuration provided in
>>> the attachment (including stepping; Note - network pass-through
>>> works on a machine with almost similar configuration except for
>>> stepping # difference)? Also, attached is bug check info if that
>>> is useful. 2) On a different machine, I encounter a bugcheck
>>> similar to the one entered in
> http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=1244
>>> only when I shutdown a domain that pass-through a Intel 82566DM-2
>>> network adapter. As the bug report wasn't specific about the device
>>> that is passed through when the machine reset happens, thought I
>>> would mention in case this helps narrow down the problem.
>>>
>>> Thanks,
>>> Kamala
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
|
|
| |
|
Home