This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


Re[2]: [Xen-devel] VT-D error related to intel IGD on Asus P5Q-EM DO (Q4

To: "Sankaran, Rajesh" <rajesh.sankaran@xxxxxxxxx>
Subject: Re[2]: [Xen-devel] VT-D error related to intel IGD on Asus P5Q-EM DO (Q45 chipset)
From: Sander Eikelenboom <linux@xxxxxxxxxxxxxx>
Date: Fri, 17 Apr 2009 11:29:53 +0200
Cc: "Han, Weidong" <weidong.han@xxxxxxxxx>, "Singh, Ashok P" <ashok.p.singh@xxxxxxxxx>, "Kay, Allen M" <allen.m.kay@xxxxxxxxx>, xen-devel@xxxxxxxxxxxxxxxxxxx
Delivery-date: Fri, 17 Apr 2009 02:30:29 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <5B8ABB4480AEBB4ABFF06629589B20CD61A70536@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
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/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Organization: Eikelenboom IT services
References: <11310048168.20090411001850@xxxxxxxxxxxxxx> <5B8ABB4480AEBB4ABFF06629589B20CD61A70536@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx

Who would be the "platform vendor" ?
The chipset and iGFX are intel, the mobo is Asus and the bios AMI.

If it affects functionality depends on what you what, the iGFX works in
dom0. But passthrough with VT-d doesn't work.

I also tried a pci-e graphic card in the pci-e 16x slot, which disables the
iGFX, but this results in xen disabeling VT-d altogether due to RMRR error.

So yes it affects functionality in the way that it makes VT-D kind of
useless, and that is not why i bought this mobo and proc.

I also wondered if there is a intel-tool which checks if bios adheres to
the VT-d specs/documentation. This way it would be much more simple to
confront the mobo/bios vendors with the problems of there vt-d
implementations, since most vendors (including asus) don't respond well to
problems with opensource implementations. And since a have windows 2008
with hyper-V, nor VMware, nor have plans to buy them to send in a bug
report to asus, i'm kind of stuck.

I somewhere found a link to http://edc.intel.com/Platforms/Xeon-5500/ on
the tab "tools & software" there is a VT-D firmware toolkit, but i don't
have permission to download it. So i can't give that one a try either.

Seems vt-d is a "intel-only" thing due to all the bios related problems
with other vendors :-(



Saturday, April 11, 2009, 12:44:16 AM, you wrote:

> This seem to be a fault from the Intel iGFX device (0:2.0).  The fault
> address is somewhere with no DRAM (ffffff000).
> The likely reason would be that BIOS is not properly initializing the 
> Graphics Translation Table (GTT).
> This can happen if there is an entry in the GTT that BIOS did not zero
> initialize (say it had junk value in it and the valid bit in the entry 
> happened to be present).

> You may report this to the platform vendor.

> Does this block you from using the iGFX device, or this is just a
> transient thing that you see in the Xen log on boot-up, but doesn't affect 
> any functionality ?

> Thanks,
> -Rajesh

> -----Original Message-----
> From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
> [mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Sander Eikelenboom
> Sent: Friday, April 10, 2009 3:19 PM
> To: xen-devel@xxxxxxxxxxxxxxxxxxx
> Subject: [Xen-devel] VT-D error related to intel IGD on Asus P5Q-EM DO (Q45 
> chipset)

> Although in the past few weeks a lot of VT-D related error seems to have
> disappeared I have still one left.

> (XEN) [VT-D]iommu.c:620: iommu_enable_translation: iommu->reg = 
> ffff828bfff57000
> (XEN) [VT-D]iommu.c:620: iommu_enable_translation: iommu->reg = 
> ffff828bfff56000
> (XEN) [VT-D]iommu.c:722: iommu_page_fault: iommu->reg = ffff828bfff56000
> (XEN) [VT-D]iommu.c:691: iommu_fault_status: Fault Overflow
> (XEN) [VT-D]iommu.c:694: iommu_fault_status: Primary Pending Fault
> (XEN) [VT-D]iommu.c:676: iommu_fault:DMA Write: 0:2.0 addr ffffff000
> REASON 5 iommu->reg = ffff828bfff56000
> (XEN) print_vtd_entries: iommu = ffff83007c8dfd20 bdf = 0:2:0 gmfn = ffffff
> (XEN)     root_entry = ffff830079db1000
> (XEN)     root_entry[0] = 7926e001
> (XEN)     context = ffff83007926e000
> (XEN)     context[10] = 101_7acd4001
> (XEN)     l3 = ffff83007acd4000
> (XEN)     l3_index = 3f
> (XEN)     l3[3f] = 0
> (XEN)     l3[3f] not present
> (XEN) [VT-D]iommu.c:620: iommu_enable_translation: iommu->reg = 
> ffff828bfff55000
> (XEN) *** LOADING DOMAIN 0 ***

> Hardware:
> - Asus P5Q-EM DO, with latest bios (0803) and VT-D enabled, IGD (pci 0:2.0)
>   as boot graphics device.
> - Intel Q9400 proc.
> - extra PCI graphics card

> Also attached:
> - xm dmesg  (first part was gone due to length ..)
> - dmesg
> - lspci
> - biosdecode
> - dmidecode
> - acpidump

> Used a fresh copy of xen-unstable and kernel, which seems to

> I hope it's fixable, and not a bios issue.

Best regards,
 Sander                            mailto:linux@xxxxxxxxxxxxxx

Xen-devel mailing list

<Prev in Thread] Current Thread [Next in Thread>