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


[Xen-users] Re: PCI Passthrough

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Re: PCI Passthrough
From: Christian Tramnitz <chris.ace@xxxxxxx>
Date: Wed, 30 Sep 2009 23:08:04 +0200
Delivery-date: Wed, 30 Sep 2009 14:10:19 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <4AC31E3E0200009900023436@xxxxxxxxxxxxxxxxxxxxx>
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
References: <4AC31E3E0200009900023436@xxxxxxxxxxxxxxxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Thunderbird (Windows/20090812)
Nick Couchman wrote:
A question that I think I probably know the answer to, but I want to verify. Do PCI Passthrough to HVM domUs require VTd or IOMMU? From what I've read in various places around the Internet, the answer is yes, but it's never been explicitly stated. I have a multi-port PCI serial card on a non-VTd-capable machine that I want to passthrough to a Windows domU. PCI passthrough would be the easiest option, except for the lack of VTd. I tried doing something with the IRQ and IOPORTS to pass it through, but I ended up with a device type "Xen Ioports" in my Windows domU and couldn't really get the correct driver installed. Any hints would be appreciated - even if it's to say, "Sorry, you're out of luck - go buy a machine with VTd."



PCI passthrough to HVM domU requires VT-d.

BUT there are experimental patches to passthrough devices to the FIRST HVM domU, i.e. here: http://permalink.gmane.org/gmane.comp.emulators.xen.devel/70672

However this is completely untested and will not just apply and work out of the box. Due to its nature its very unlikely that passthrough to HVM without VT-d will ever make it into Xen at all. So if you're looking for a real solution rather than a quick hack, then I can only tell you "Sorry, you're out of luck - go buy a machine with VTd." ;-)

I just updated the Wiki today with some boards and chipsets known to work (or having problems), have a look here:

Best regards,

Xen-users mailing list

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