Re: [Xen-users] VT-D RMRR is incorrect

[Steve McIntosh <stephen@xxxxxxxxxxx>]

I've had the same problem trying to run VT-d on an Asus P5Q-EM DO
(Intel Q45 chipset).  Perhaps it's just best to avoid Asus hardware
when it comes to IOMMU virtualization because it appears their boards
may fail to adhere to proper Intel VT-d spec.  Fortunately I had
another motherboard lying around that I swapped it for.

I'm now successfully passing through PCI NICs and USB controllers in
Xen 3.4 on an Intel DQ35JO motherboard with the 1016 bios revision.
(The controller [00:1a.7] takes the 'guestdev=xx:yy.z
reassign_resources' kernel module flag instead of the pciback.hide
because of a memory issue.)  As a result I'm able to use a Creative
USB (Win)Modem Blaster v.92 passed through to a Windows XP HVM.

So, for now, it looks like the only VT-d options are Intel branded
boards.  Personally, I can't wait for the new AMD chipset to come out
with their own IOMMU, that should open up the playing field to more
competition.


On Sun, Jul 26, 2009 at 10:53 AM, srynoname<regtrash@xxxxxxxx> wrote:
>
>
> tmoore wrote:
>>
>>
>> I`ve raised a support case with ASUS to try to get them to fix the BIOS on
>> P6T Deluxe ... I`ll keep this thread updated.
>>
>> The IOMMU is seriously broken on this board, I have tried many workarounds
>> and it always results in failure.
>>
>>
>>
> Hello,
> did you get a reply? Somebody else has something new about the Asus P6T and
> this problem? How about VT-d on windows systems? Thanks!
> --
> View this message in context: 
> http://www.nabble.com/VT-D-RMRR-is-incorrect-tp22005500p24661872.html
> Sent from the Xen - User mailing list archive at Nabble.com.
>
>
> _______________________________________________
> Xen-users mailing list
> Xen-users@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-users
>

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