Xen on top of xen

Xen on top of xen

[Koripella Srinivas]

[-- Attachment #1.1: Type: text/plain, Size: 237 bytes --]

Hi,
Is it possible to run xen in a hvm guest which itself is running on xen?

Thanks




      Office firewalls, cyber cafes, college labs, don't allow you to download CHAT? Click here: http://in.messenger.yahoo.com/webmessengerpromo.php

[-- Attachment #1.2: Type: text/html, Size: 527 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: Xen on top of xen

[pradeep singh rautela]

[-- Attachment #1.1: Type: text/plain, Size: 668 bytes --]

On 5/9/07, Koripella Srinivas <talkwithsrinivas@yahoo.co.in> wrote:
>
> Hi,
> Is it possible to run xen in a hvm guest which itself is running on xen?
>

Yes, you can.

Thanks
~psr

Thanks
>
> ------------------------------
> Office firewalls, cyber cafes, college labs, don't allow you to download
> CHAT? Here's a solution!<http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenger.yahoo.com/webmessengerpromo.php>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>
>


-- 
---
pradeep singh rautela

"Genius is 1% inspiration, and 99% perspiration" - not me :)

[-- Attachment #1.2: Type: text/html, Size: 1705 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: Xen on top of xen

[pradeep singh rautela]

[-- Attachment #1.1: Type: text/plain, Size: 938 bytes --]

On 5/9/07, pradeep singh rautela <rautelap@gmail.com> wrote:
>
>
>
> On 5/9/07, Koripella Srinivas <talkwithsrinivas@yahoo.co.in> wrote:
> >
> > Hi,
> > Is it possible to run xen in a hvm guest which itself is running on xen?
> >
>
> Yes, you can.
>

Just to add only on hvm guests and not  on pv guests.

Thanks
> ~psr
>
> Thanks
> >
> > ------------------------------
> > Office firewalls, cyber cafes, college labs, don't allow you to download
> > CHAT? Here's a solution!<http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenger.yahoo.com/webmessengerpromo.php>
> >
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@lists.xensource.com
> > http://lists.xensource.com/xen-devel
> >
> >
>
>
> --
> ---
> pradeep singh rautela
>
> "Genius is 1% inspiration, and 99% perspiration" - not me :)




-- 
---
pradeep singh rautela

"Genius is 1% inspiration, and 99% perspiration" - not me :)

[-- Attachment #1.2: Type: text/html, Size: 2552 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: Xen on top of xen

[pradeep singh rautela]

[-- Attachment #1.1: Type: text/plain, Size: 702 bytes --]

On 5/9/07, Koripella Srinivas <talkwithsrinivas@yahoo.co.in> wrote:
>
> Hi,
> Is it possible to run xen in a hvm guest which itself is running on xen?
>

Yes , you can run Xen on top of a hvm guest.

Thank you
~psr

Thanks
>
> ------------------------------
> Office firewalls, cyber cafes, college labs, don't allow you to download
> CHAT? Here's a solution!<http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenger.yahoo.com/webmessengerpromo.php>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>
>


-- 
---
pradeep singh rautela

"Genius is 1% inspiration, and 99% perspiration" - not me :)

[-- Attachment #1.2: Type: text/html, Size: 1736 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

RE: Xen on top of xen

[Petersson, Mats]

 

> -----Original Message-----
> From: xen-devel-bounces@lists.xensource.com 
> [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of 
> pradeep singh rautela
> Sent: 09 May 2007 10:11
> To: Koripella Srinivas
> Cc: xen-devel@lists.xensource.com; 
> xen-devel-request@lists.xensource.com
> Subject: Re: [Xen-devel] Xen on top of xen
> 
> 
> 
> On 5/9/07, pradeep singh rautela <rautelap@gmail.com> wrote:
> 
> 
> 
> 	On 5/9/07, Koripella Srinivas < 
> talkwithsrinivas@yahoo.co.in 
> <mailto:talkwithsrinivas@yahoo.co.in> > wrote:
> 
> 		Hi,
> 		Is it possible to run xen in a hvm guest which 
> itself is running on xen?
> 
> 
> 	Yes, you can. 
> 
> 
> Just to add only on hvm guests and not  on pv guests.

And you can't run HVM guests inside the "xen on top of xen", as the HVM
feature is not (in current CPU's) nestable. 

--
Mats
> 
> 
> 
> 	Thanks 
> 	~psr
> 	
> 
> 
> 		
> 		Thanks
> 		
> 
> ________________________________
> 
> 		Office firewalls, cyber cafes, college labs, 
> don't allow you to download CHAT? Here's a solution! 
> <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> r.yahoo.com/webmessengerpromo.php>  
> 
> 		_______________________________________________
> 		Xen-devel mailing list
> 		Xen-devel@lists.xensource.com 
> <mailto:Xen-devel@lists.xensource.com> 
> 		http://lists.xensource.com/xen-devel
> 		
> 		
> 
> 
> 
> 
> 	-- 
> 	---
> 	pradeep singh rautela
> 	
> 	"Genius is 1% inspiration, and 99% perspiration" - not me :) 
> 
> 
> 
> 
> -- 
> ---
> pradeep singh rautela
> 
> "Genius is 1% inspiration, and 99% perspiration" - not me :) 
> 

Re: Xen on top of xen

[Mark Williamson]

Works well for me - I use it to run development versions of Xen (although 
proper testing etc should still be run on "real" hardware to eliminate bugs 
and performance artifacts)

You can't nest HVM guests though, so the Xen-on-Xen will only be able to host 
PV domains.

Cheers,
Mark

On Wednesday 09 May 2007, Koripella Srinivas wrote:
> Hi,
> Is it possible to run xen in a hvm guest which itself is running on xen?
>
> Thanks
>
>
>
>
>       Office firewalls, cyber cafes, college labs, don't allow you to
> download CHAT? Click here:
> http://in.messenger.yahoo.com/webmessengerpromo.php



-- 
Dave: Just a question. What use is a unicyle with no seat?  And no pedals!
Mark: To answer a question with a question: What use is a skateboard?
Dave: Skateboards have wheels.
Mark: My wheel has a wheel!

RE: Xen on top of xen

[Petersson, Mats]

> -----Original Message-----
> From: Koripella Srinivas [mailto:talkwithsrinivas@yahoo.co.in] 
> Sent: 09 May 2007 19:53
> To: Petersson, Mats
> Subject: Re: [Xen-devel] Xen on top of xen
> 
> Can u throw more light on  the non-nestability ?? 

Well, to put it simply, once you set up the SVM/VMX feature, it doesn't
allow another such setup on top of it. It would be technically possible
to support this in a processor designed to do so, but at the moment,
neither of AMD's or Intel's CPU's support this. It is clearly easier to
design a processor that only supports one level of "nesting" of
virtualization[1], and it is also debatable what the REAL benefit of
running nested virtual machines would be - I'm sure if it was a huge
benefit compared to the effort of implementing it[2], at least one of
the two producers of virtualizable x86 processors would have added such
fearures... 

[1] Keeping one set of data for where to return to when the VMExit
happens, for example, is easier than having a "stack" of such. 

[2] On top of the extra effort to design and test such feature, there is
also a potential performance penalty for nesting VM's, since there is
more logic involved. Depending on where that extra logic ends up, it may
be necessary to add a further clock-cycle to some operation. 

--
Mats
> 
> 
> ----- Original Message ----
> From: "Petersson, Mats" <Mats.Petersson@amd.com>
> To: pradeep singh rautela <rautelap@gmail.com>; Koripella 
> Srinivas <talkwithsrinivas@yahoo.co.in>
> Cc: xen-devel@lists.xensource.com; 
> xen-devel-request@lists.xensource.com
> Sent: Wednesday, 9 May, 2007 4:33:57 PM
> Subject: RE: [Xen-devel] Xen on top of xen
> 
> 
> 
> 
> > -----Original Message-----
> > From: xen-devel-bounces@lists.xensource.com 
> > [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of 
> > pradeep singh rautela
> > Sent: 09 May 2007 10:11
> > To: Koripella Srinivas
> > Cc: xen-devel@lists.xensource.com; 
> > xen-devel-request@lists.xensource.com
> > Subject: Re: [Xen-devel] Xen on top of xen
> > 
> > 
> > 
> > On 5/9/07, pradeep singh rautela <rautelap@gmail.com> wrote:
> > 
> > 
> > 
> >     On 5/9/07, Koripella Srinivas < 
> > talkwithsrinivas@yahoo.co.in 
> > <mailto:talkwithsrinivas@yahoo.co.in> > wrote:
> > 
> >         Hi,
> >         Is it possible to run xen in a hvm guest which 
> > itself is running on xen?
> > 
> > 
> >     Yes, you can. 
> > 
> > 
> > Just to add only on hvm guests and not  on pv guests.
> 
> And you can't run HVM guests inside the "xen on top of xen", 
> as the HVM
> feature is not (in current CPU's) nestable. 
> 
> --
> Mats
> > 
> > 
> > 
> >     Thanks 
> >     ~psr
> >     
> > 
> > 
> >         
> >         Thanks
> >         
> > 
> > ________________________________
> > 
> >         Office firewalls, cyber cafes, college labs, 
> > don't allow you to download CHAT? Here's a solution! 
> > <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> > r.yahoo.com/webmessengerpromo.php>  
> > 
> >         _______________________________________________
> >         Xen-devel mailing list
> >         Xen-devel@lists.xensource.com 
> > <mailto:Xen-devel@lists.xensource.com> 
> >         http://lists.xensource.com/xen-devel
> >         
> >         
> > 
> > 
> > 
> > 
> >     -- 
> >     ---
> >     pradeep singh rautela
> >     
> >     "Genius is 1% inspiration, and 99% perspiration" - not me :) 
> > 
> > 
> > 
> > 
> > -- 
> > ---
> > pradeep singh rautela
> > 
> > "Genius is 1% inspiration, and 99% perspiration" - not me :) 
> > 
> 
> 
> 
> 
> 
> ________________________________
> 
> Office firewalls, cyber cafes, college labs, don't allow you 
> to download CHAT? Here's a solution! 
> <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> r.yahoo.com/webmessengerpromo.php>  
> 

Re: Xen on top of xen

[Mark Williamson]

> > Can u throw more light on  the non-nestability ??
>
> Well, to put it simply, once you set up the SVM/VMX feature, it doesn't
> allow another such setup on top of it. It would be technically possible
> to support this in a processor designed to do so, but at the moment,
> neither of AMD's or Intel's CPU's support this.

It could be trapped and emulated in Xen though, right?  The AMD manual was 
pretty clear on all the SVM instructions being trappable.  The VT-x manual I 
had access to when I looked into this was a bit vague in this area.

e.g. a scheme whereby the "outermost" Xen trapped SVM instructions issued by 
the current nested guest and built up a shadow copy of HVM-related data 
structures, then ran the "inner" HVM guest using these?  Future additions to 
SVM implementations (such as nested paging) would reduce the frequency with 
which you have to bounce back into the guests.

> It is clearly easier to 
> design a processor that only supports one level of "nesting" of
> virtualization[1], and it is also debatable what the REAL benefit of
> running nested virtual machines would be - I'm sure if it was a huge
> benefit compared to the effort of implementing it[2], at least one of
> the two producers of virtualizable x86 processors would have added such
> fearures...

Being able to run HVM Xen in HVM Xen could be useful for testing.  Also, from 
a management perspective it would make it possible to subdivide the machine 
between multiple users who each wanted to run a few virtual machines - just 
give them each a Xen in an HVM domain.  Further levels of nesting would be 
overkill, and probably give horrible performance.

Cheers,
Mark

> [1] Keeping one set of data for where to return to when the VMExit
> happens, for example, is easier than having a "stack" of such.
>
> [2] On top of the extra effort to design and test such feature, there is
> also a potential performance penalty for nesting VM's, since there is
> more logic involved. Depending on where that extra logic ends up, it may
> be necessary to add a further clock-cycle to some operation.
>
> --
> Mats
>
> > ----- Original Message ----
> > From: "Petersson, Mats" <Mats.Petersson@amd.com>
> > To: pradeep singh rautela <rautelap@gmail.com>; Koripella
> > Srinivas <talkwithsrinivas@yahoo.co.in>
> > Cc: xen-devel@lists.xensource.com;
> > xen-devel-request@lists.xensource.com
> > Sent: Wednesday, 9 May, 2007 4:33:57 PM
> > Subject: RE: [Xen-devel] Xen on top of xen
> >
> > > -----Original Message-----
> > > From: xen-devel-bounces@lists.xensource.com
> > > [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of
> > > pradeep singh rautela
> > > Sent: 09 May 2007 10:11
> > > To: Koripella Srinivas
> > > Cc: xen-devel@lists.xensource.com;
> > > xen-devel-request@lists.xensource.com
> > > Subject: Re: [Xen-devel] Xen on top of xen
> > >
> > >
> > >
> > > On 5/9/07, pradeep singh rautela <rautelap@gmail.com> wrote:
> > >
> > >
> > >
> > >     On 5/9/07, Koripella Srinivas <
> > > talkwithsrinivas@yahoo.co.in
> > > <mailto:talkwithsrinivas@yahoo.co.in> > wrote:
> > >
> > >         Hi,
> > >         Is it possible to run xen in a hvm guest which
> > > itself is running on xen?
> > >
> > >
> > >     Yes, you can.
> > >
> > >
> > > Just to add only on hvm guests and not  on pv guests.
> >
> > And you can't run HVM guests inside the "xen on top of xen",
> > as the HVM
> > feature is not (in current CPU's) nestable.
> >
> > --
> > Mats
> >
> > >     Thanks
> > >     ~psr
> > >
> > >
> > >
> > >
> > >         Thanks
> > >
> > >
> > > ________________________________
> > >
> > >         Office firewalls, cyber cafes, college labs,
> > > don't allow you to download CHAT? Here's a solution!
> > > <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> > > r.yahoo.com/webmessengerpromo.php>
> > >
> > >         _______________________________________________
> > >         Xen-devel mailing list
> > >         Xen-devel@lists.xensource.com
> > > <mailto:Xen-devel@lists.xensource.com>
> > >         http://lists.xensource.com/xen-devel
> > >
> > >
> > >
> > >
> > >
> > >
> > >     --
> > >     ---
> > >     pradeep singh rautela
> > >
> > >     "Genius is 1% inspiration, and 99% perspiration" - not me :)
> > >
> > >
> > >
> > >
> > > --
> > > ---
> > > pradeep singh rautela
> > >
> > > "Genius is 1% inspiration, and 99% perspiration" - not me :)
> >
> > ________________________________
> >
> > Office firewalls, cyber cafes, college labs, don't allow you
> > to download CHAT? Here's a solution!
> > <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> > r.yahoo.com/webmessengerpromo.php>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel



-- 
Dave: Just a question. What use is a unicyle with no seat?  And no pedals!
Mark: To answer a question with a question: What use is a skateboard?
Dave: Skateboards have wheels.
Mark: My wheel has a wheel!

RE: Xen on top of xen

[Petersson, Mats]

 

> -----Original Message-----
> From: M.A. Williamson [mailto:maw48@hermes.cam.ac.uk] On 
> Behalf Of Mark Williamson
> Sent: 10 May 2007 10:45
> To: xen-devel@lists.xensource.com
> Cc: Petersson, Mats; Koripella Srinivas
> Subject: Re: [Xen-devel] Xen on top of xen
> 
> > > Can u throw more light on  the non-nestability ??
> >
> > Well, to put it simply, once you set up the SVM/VMX 
> feature, it doesn't
> > allow another such setup on top of it. It would be 
> technically possible
> > to support this in a processor designed to do so, but at the moment,
> > neither of AMD's or Intel's CPU's support this.
> 
> It could be trapped and emulated in Xen though, right?  The 
> AMD manual was 
> pretty clear on all the SVM instructions being trappable.  
> The VT-x manual I 
> had access to when I looked into this was a bit vague in this area.

Yes, at least for SVM you could trap and emulate. 
> 
> e.g. a scheme whereby the "outermost" Xen trapped SVM 
> instructions issued by 
> the current nested guest and built up a shadow copy of 
> HVM-related data 
> structures, then ran the "inner" HVM guest using these?  
> Future additions to 
> SVM implementations (such as nested paging) would reduce the 
> frequency with 
> which you have to bounce back into the guests.

I'm not sure if it's worth the effort tho'. It would be relatively slow
(because you'd first get the penalty of the VMExit/VMRun pair, and then
the additional code to emulate the VMExit/VMRun set.

> 
> > It is clearly easier to 
> > design a processor that only supports one level of "nesting" of
> > virtualization[1], and it is also debatable what the REAL benefit of
> > running nested virtual machines would be - I'm sure if it was a huge
> > benefit compared to the effort of implementing it[2], at 
> least one of
> > the two producers of virtualizable x86 processors would 
> have added such
> > fearures...
> 
> Being able to run HVM Xen in HVM Xen could be useful for 
> testing.  Also, from 
> a management perspective it would make it possible to 
> subdivide the machine 
> between multiple users who each wanted to run a few virtual 
> machines - just 
> give them each a Xen in an HVM domain.  Further levels of 
> nesting would be 
> overkill, and probably give horrible performance.

Of those, the management one is the more applicable to our engineering
efforts (i.e. the one that will "sell more processors"), but I'm pretty
sure that the effort is much greater than the gain. 

Do you (or anyone else) know if the PPC version supports nested
virtualization?

--
Mats
> 
> Cheers,
> Mark
> 
> > [1] Keeping one set of data for where to return to when the VMExit
> > happens, for example, is easier than having a "stack" of such.
> >
> > [2] On top of the extra effort to design and test such 
> feature, there is
> > also a potential performance penalty for nesting VM's, 
> since there is
> > more logic involved. Depending on where that extra logic 
> ends up, it may
> > be necessary to add a further clock-cycle to some operation.
> >
> > --
> > Mats
> >
> > > ----- Original Message ----
> > > From: "Petersson, Mats" <Mats.Petersson@amd.com>
> > > To: pradeep singh rautela <rautelap@gmail.com>; Koripella
> > > Srinivas <talkwithsrinivas@yahoo.co.in>
> > > Cc: xen-devel@lists.xensource.com;
> > > xen-devel-request@lists.xensource.com
> > > Sent: Wednesday, 9 May, 2007 4:33:57 PM
> > > Subject: RE: [Xen-devel] Xen on top of xen
> > >
> > > > -----Original Message-----
> > > > From: xen-devel-bounces@lists.xensource.com
> > > > [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of
> > > > pradeep singh rautela
> > > > Sent: 09 May 2007 10:11
> > > > To: Koripella Srinivas
> > > > Cc: xen-devel@lists.xensource.com;
> > > > xen-devel-request@lists.xensource.com
> > > > Subject: Re: [Xen-devel] Xen on top of xen
> > > >
> > > >
> > > >
> > > > On 5/9/07, pradeep singh rautela <rautelap@gmail.com> wrote:
> > > >
> > > >
> > > >
> > > >     On 5/9/07, Koripella Srinivas <
> > > > talkwithsrinivas@yahoo.co.in
> > > > <mailto:talkwithsrinivas@yahoo.co.in> > wrote:
> > > >
> > > >         Hi,
> > > >         Is it possible to run xen in a hvm guest which
> > > > itself is running on xen?
> > > >
> > > >
> > > >     Yes, you can.
> > > >
> > > >
> > > > Just to add only on hvm guests and not  on pv guests.
> > >
> > > And you can't run HVM guests inside the "xen on top of xen",
> > > as the HVM
> > > feature is not (in current CPU's) nestable.
> > >
> > > --
> > > Mats
> > >
> > > >     Thanks
> > > >     ~psr
> > > >
> > > >
> > > >
> > > >
> > > >         Thanks
> > > >
> > > >
> > > > ________________________________
> > > >
> > > >         Office firewalls, cyber cafes, college labs,
> > > > don't allow you to download CHAT? Here's a solution!
> > > > <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> > > > r.yahoo.com/webmessengerpromo.php>
> > > >
> > > >         _______________________________________________
> > > >         Xen-devel mailing list
> > > >         Xen-devel@lists.xensource.com
> > > > <mailto:Xen-devel@lists.xensource.com>
> > > >         http://lists.xensource.com/xen-devel
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >     --
> > > >     ---
> > > >     pradeep singh rautela
> > > >
> > > >     "Genius is 1% inspiration, and 99% perspiration" - not me :)
> > > >
> > > >
> > > >
> > > >
> > > > --
> > > > ---
> > > > pradeep singh rautela
> > > >
> > > > "Genius is 1% inspiration, and 99% perspiration" - not me :)
> > >
> > > ________________________________
> > >
> > > Office firewalls, cyber cafes, college labs, don't allow you
> > > to download CHAT? Here's a solution!
> > > <http://us.rd.yahoo.com/mail/in/ymessenger/*http://in.messenge
> > > r.yahoo.com/webmessengerpromo.php>
> >
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@lists.xensource.com
> > http://lists.xensource.com/xen-devel
> 
> 
> 
> -- 
> Dave: Just a question. What use is a unicyle with no seat?  
> And no pedals!
> Mark: To answer a question with a question: What use is a skateboard?
> Dave: Skateboards have wheels.
> Mark: My wheel has a wheel!
> 
> 
> 

Re: Xen on top of xen

[Mark Williamson]

> > e.g. a scheme whereby the "outermost" Xen trapped SVM
> > instructions issued by
> > the current nested guest and built up a shadow copy of
> > HVM-related data
> > structures, then ran the "inner" HVM guest using these?
> > Future additions to
> > SVM implementations (such as nested paging) would reduce the
> > frequency with
> > which you have to bounce back into the guests.
>
> I'm not sure if it's worth the effort tho'. It would be relatively slow
> (because you'd first get the penalty of the VMExit/VMRun pair, and then
> the additional code to emulate the VMExit/VMRun set.

Mmmm.  The thing that looks to me like it would really make performance nasty 
is jumping in and out of all the levels of nesting in order to do IO 
emulation...

What if IO emulation were being handled by a device model hidden in the 
guest's firmware (has been suggested before for different reasons)?  You'd be 
running the innermost guest context on the host CPU directly, as previously 
suggested, then on an IO exit you could just reflect the IO instruction into 
the device model in that guest.  You potentially wouldn't need to re-enter 
the nested Xen that "owned" the guest until the innermost guest's timeslice 
ran out naturally.

I may well be missing something here, as this ain't my forte ;-)

> > Being able to run HVM Xen in HVM Xen could be useful for
> > testing.  Also, from
> > a management perspective it would make it possible to
> > subdivide the machine
> > between multiple users who each wanted to run a few virtual
> > machines - just
> > give them each a Xen in an HVM domain.  Further levels of
> > nesting would be
> > overkill, and probably give horrible performance.
>
> Of those, the management one is the more applicable to our engineering
> efforts (i.e. the one that will "sell more processors"), but I'm pretty
> sure that the effort is much greater than the gain.

Yep.  Also, the nested management problem could equally be tackled by 
modifying Xen to support some more advanced concept of rights to map memory, 
etc - or even changes at the tools level to make management "look" more 
nested.

Being able to nest Xen is just a /conceptually/ simple way to do this which 
(if it worked) gets us a load of stuff for free (CPU accounting for a domain 
group, atomic suspend/resume for a group, atomic migration for a group, etc.)

> Do you (or anyone else) know if the PPC version supports nested
> virtualization?

I know z/VM supports it (no surprise - z/VM supports all sort of crazy but 
cool things), and it seems like the kind of thing IBM might do - precisely 
for management reasons ;-)  I would guess that the PPC version of Xen doesn't 
support it (yet) but maybe the POWER hypervisor does?  I've cc-ed Hollis 
who'll be able to set the record straight :-)

Cheers,
Mark
-- 
Dave: Just a question. What use is a unicyle with no seat?  And no pedals!
Mark: To answer a question with a question: What use is a skateboard?
Dave: Skateboards have wheels.
Mark: My wheel has a wheel!

Re: Xen on top of xen

[Hollis Blanchard]

On Thu, 2007-05-10 at 11:12 +0100, Mark Williamson wrote:
> > Do you (or anyone else) know if the PPC version supports nested
> > virtualization?
> 
> I know z/VM supports it (no surprise - z/VM supports all sort of crazy but 
> cool things), and it seems like the kind of thing IBM might do - precisely 
> for management reasons ;-)  I would guess that the PPC version of Xen doesn't 
> support it (yet) but maybe the POWER hypervisor does?  I've cc-ed Hollis 
> who'll be able to set the record straight :-)

The hardware extensions defined in the Power Architecture do not support
nested virtualization.

-- 
Hollis Blanchard
IBM Linux Technology Center