All of lore.kernel.org
 help / color / mirror / Atom feed
* Status of nestedhvm (on Intel)
@ 2012-04-16 15:27 Christian Tramnitz
  2012-04-16 15:42 ` Tim Deegan
  2012-04-16 15:44 ` Ian Campbell
  0 siblings, 2 replies; 7+ messages in thread
From: Christian Tramnitz @ 2012-04-16 15:27 UTC (permalink / raw)
  To: xen-devel

May I ask what the current nestedhvm status in Xen 4.1 is? The parameter exists,
the documentation just says that only previously existent cpu features will be
passed through, but I can't seem to get a VM with "vmx" (for testing of ESX or
Hyper-V in a Xen domU) working. "nestedhvm=1" seems just to be ignored...

Is this AMD-only in 4.1? Are additional CPU host features required to make this
work (like EPT, which I do have on this Xeon 5520 box or "Shared EPT" that I
dont have but could be AMD related anyway)? Do I need unstable to make use of it?


Thanks,
   Christian

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of nestedhvm (on Intel)
  2012-04-16 15:27 Status of nestedhvm (on Intel) Christian Tramnitz
@ 2012-04-16 15:42 ` Tim Deegan
  2012-04-16 15:44 ` Ian Campbell
  1 sibling, 0 replies; 7+ messages in thread
From: Tim Deegan @ 2012-04-16 15:42 UTC (permalink / raw)
  To: Christian Tramnitz; +Cc: xen-devel

At 15:27 +0000 on 16 Apr (1334590052), Christian Tramnitz wrote:
> Do I need unstable to make use of it?

Yes, all the nested HVM code is newer than 4.1.

Cheers,

Tim.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of nestedhvm (on Intel)
  2012-04-16 15:27 Status of nestedhvm (on Intel) Christian Tramnitz
  2012-04-16 15:42 ` Tim Deegan
@ 2012-04-16 15:44 ` Ian Campbell
  2012-04-16 19:31   ` Christian Tramnitz
  2012-05-04 15:49   ` Christian Tramnitz
  1 sibling, 2 replies; 7+ messages in thread
From: Ian Campbell @ 2012-04-16 15:44 UTC (permalink / raw)
  To: Christian Tramnitz; +Cc: xen-devel

On Mon, 2012-04-16 at 16:27 +0100, Christian Tramnitz wrote:
> May I ask what the current nestedhvm status in Xen 4.1 is? The parameter exists,
> the documentation just says that only previously existent cpu features will be
> passed through, but I can't seem to get a VM with "vmx" (for testing of ESX or
> Hyper-V in a Xen domU) working. "nestedhvm=1" seems just to be ignored...
> 
> Is this AMD-only in 4.1? Are additional CPU host features required to make this
> work (like EPT, which I do have on this Xeon 5520 box or "Shared EPT" that I
> dont have but could be AMD related anyway)? Do I need unstable to make use of it?

I can't see any support for nested HVM of any sort in 4.1. AFAIK it is
going to be an (experimental) feature in 4.2.

Ian.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of nestedhvm (on Intel)
  2012-04-16 15:44 ` Ian Campbell
@ 2012-04-16 19:31   ` Christian Tramnitz
  2012-05-04 15:49   ` Christian Tramnitz
  1 sibling, 0 replies; 7+ messages in thread
From: Christian Tramnitz @ 2012-04-16 19:31 UTC (permalink / raw)
  To: xen-devel

Ian Campbell <Ian.Campbell <at> citrix.com> writes:

> I can't see any support for nested HVM of any sort in 4.1. AFAIK it is
> going to be an (experimental) feature in 4.2.
> 
> Ian.


My fault, I was looking at config options here
http://www.xen.org/support/generated.html to get used to xl (always using xm
before). I wasn't paying attention to the note above that this is for unstable
only...


Thanks for the hint!


Best regards,
   Christian

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of nestedhvm (on Intel)
  2012-04-16 15:44 ` Ian Campbell
  2012-04-16 19:31   ` Christian Tramnitz
@ 2012-05-04 15:49   ` Christian Tramnitz
  2012-05-04 16:02     ` Jan Beulich
  1 sibling, 1 reply; 7+ messages in thread
From: Christian Tramnitz @ 2012-05-04 15:49 UTC (permalink / raw)
  To: xen-devel

Ian Campbell <Ian.Campbell <at> citrix.com> writes:
> AFAIK it is going to be an (experimental) feature in 4.2.

Short follow-up: I haven't tried with Hyper-V yet, but on current (today's
unstable.hg) Xen 4.2 at least ESXi does not work. VMX flag is visible, but
during load ESXi 5u1 crashes and Xen log shows:

(XEN) vvmx.c:1366:d8 VMX MSR 485 not fully supported yet.
(XEN) hvm.c:1143:d8 All CPUs offline -- powering off.


Does anyone have an idea what MSR 485 does? It is not documented in "Intel 64
and IA-32 Architectures Software Developer's Manual"...



Regards,
   Christian

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of nestedhvm (on Intel)
  2012-05-04 15:49   ` Christian Tramnitz
@ 2012-05-04 16:02     ` Jan Beulich
  2012-05-04 16:12       ` Nakajima, Jun
  0 siblings, 1 reply; 7+ messages in thread
From: Jan Beulich @ 2012-05-04 16:02 UTC (permalink / raw)
  To: Christian Tramnitz; +Cc: xen-devel

>>> On 04.05.12 at 17:49, Christian Tramnitz <chris.ace@gmx.net> wrote:
> Ian Campbell <Ian.Campbell <at> citrix.com> writes:
>> AFAIK it is going to be an (experimental) feature in 4.2.
> 
> Short follow-up: I haven't tried with Hyper-V yet, but on current (today's
> unstable.hg) Xen 4.2 at least ESXi does not work. VMX flag is visible, but
> during load ESXi 5u1 crashes and Xen log shows:
> 
> (XEN) vvmx.c:1366:d8 VMX MSR 485 not fully supported yet.
> (XEN) hvm.c:1143:d8 All CPUs offline -- powering off.
> 
> 
> Does anyone have an idea what MSR 485 does? It is not documented in "Intel 
> 64 and IA-32 Architectures Software Developer's Manual"...

xen/include/asm-x86/msr-index.h says

#define MSR_IA32_VMX_MISC                       0x485

Jan

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Status of nestedhvm (on Intel)
  2012-05-04 16:02     ` Jan Beulich
@ 2012-05-04 16:12       ` Nakajima, Jun
  0 siblings, 0 replies; 7+ messages in thread
From: Nakajima, Jun @ 2012-05-04 16:12 UTC (permalink / raw)
  To: Jan Beulich; +Cc: Christian Tramnitz, xen-devel


[-- Attachment #1.1: Type: text/plain, Size: 1053 bytes --]

And it's documented in A.6 (Appendix A).

A.6 MISCELLANEOUS DATA
The IA32_VMX_MISC MSR (index 485H) consists of the following fields:
...


On Fri, May 4, 2012 at 9:02 AM, Jan Beulich <JBeulich@suse.com> wrote:

> >>> On 04.05.12 at 17:49, Christian Tramnitz <chris.ace@gmx.net> wrote:
> > Ian Campbell <Ian.Campbell <at> citrix.com> writes:
> >> AFAIK it is going to be an (experimental) feature in 4.2.
> >
> > Short follow-up: I haven't tried with Hyper-V yet, but on current
> (today's
> > unstable.hg) Xen 4.2 at least ESXi does not work. VMX flag is visible,
> but
> > during load ESXi 5u1 crashes and Xen log shows:
> >
> > (XEN) vvmx.c:1366:d8 VMX MSR 485 not fully supported yet.
> > (XEN) hvm.c:1143:d8 All CPUs offline -- powering off.
> >
> >
> > Does anyone have an idea what MSR 485 does? It is not documented in
> "Intel
> > 64 and IA-32 Architectures Software Developer's Manual"...
>
> xen/include/asm-x86/msr-index.h says
>
> #define MSR_IA32_VMX_MISC                       0x485
>
> Jan
>
> --
Jun
Intel Open Source Technology Center

[-- Attachment #1.2: Type: text/html, Size: 1723 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

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

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2012-05-04 16:12 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-04-16 15:27 Status of nestedhvm (on Intel) Christian Tramnitz
2012-04-16 15:42 ` Tim Deegan
2012-04-16 15:44 ` Ian Campbell
2012-04-16 19:31   ` Christian Tramnitz
2012-05-04 15:49   ` Christian Tramnitz
2012-05-04 16:02     ` Jan Beulich
2012-05-04 16:12       ` Nakajima, Jun

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.