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: [Xen-users] Vanishing memory? Whats going on?

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] Vanishing memory? Whats going on?
From: Mark Williamson <mark.williamson@xxxxxxxxxxxx>
Date: Wed, 9 Jul 2008 15:36:02 +0100
Cc: Iustin Pop <iusty@xxxxxxxxx>, bleomycin <diabolical@xxxxxxxxx>
Delivery-date: Wed, 09 Jul 2008 07:36:44 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <20080706044227.GB30892@xxxxxxxxxxxxxxxxx>
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: <18269784.post@xxxxxxxxxxxxxxx> <18298577.post@xxxxxxxxxxxxxxx> <20080706044227.GB30892@xxxxxxxxxxxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: KMail/1.9.9
On Sunday 06 July 2008, Iustin Pop wrote:
> On Sat, Jul 05, 2008 at 08:43:24PM -0700, bleomycin wrote:
> > Thank you very much for the reply. I must not fully understand the point
> > of hvm. Other than being used to run windows, I thought hvm was required
> > to take full advantage of intel/amd's VT technology built into the newer
> > processors? Or is that handled automatically regardless of hvm?
> I think you have been fooled by marketing.
> The VT technology doesn't make virtualization better in general; it's
> however required in order to virtualize Windows and any other OS that
> cannot work in PVM.
> But at any point, PVM is much better than HVM; and VT doesn't give you
> any advantage if the guest runs in PVM. In other words, the fastest way
> to run a guest is PVM mode, and VT support means nothing in that case.
> That is, if I understand the situation correctly. Corrections welcome!

That's pretty much the current situation!

I would add that by combining HVM and PV technologies carefully it might be 
possible to achieve even better performance, potentially with less invasive 
code modifications to the guest!  This isn't done in Xen, though - you either 
use PVM of HVM, with PVM generally being faster.


Push Me Pull You - Distributed SCM tool (http://www.cl.cam.ac.uk/~maw48/pmpu/)

Xen-users mailing list