linux' X86_MSR config option

linux' X86_MSR config option

[Jan Beulich]

.. is out of sync on i386 and x86-64 - should it be allowed on i386, or should
it be suppressed on x86-64? Since Xen fails ill accesses, I would assume
enabling it on i386 is fine.

Jan

Re: linux' X86_MSR config option

[Keir Fraser]

On 30/5/07 10:52, "Jan Beulich" <jbeulich@novell.com> wrote:

> .. is out of sync on i386 and x86-64 - should it be allowed on i386, or should
> it be suppressed on x86-64? Since Xen fails ill accesses, I would assume
> enabling it on i386 is fine.

Is this the MSRs-from-userland option? Should be safe to enable. But
probably fairly pointless?

 -- Keir

Re: linux' X86_MSR config option

[Jan Beulich]

>>> Keir Fraser <keir@xensource.com> 30.05.07 11:53 >>>
>On 30/5/07 10:52, "Jan Beulich" <jbeulich@novell.com> wrote:
>
>> .. is out of sync on i386 and x86-64 - should it be allowed on i386, or should
>> it be suppressed on x86-64? Since Xen fails ill accesses, I would assume
>> enabling it on i386 is fine.
>
>Is this the MSRs-from-userland option? Should be safe to enable.

Yes.

>But probably fairly pointless?

Not so for debugging purposes - I asked because I just now needed it to read
a BIOS-initialized MSR, and it's quite handy if you don't have to reboot into a
native kernel to do just this.

Jan

RE: linux' X86_MSR config option

[Petersson, Mats]

 

> -----Original Message-----
> From: xen-devel-bounces@lists.xensource.com 
> [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of 
> Jan Beulich
> Sent: 30 May 2007 11:03
> To: Keir Fraser
> Cc: xen-devel@lists.xensource.com
> Subject: Re: [Xen-devel] linux' X86_MSR config option
> 
> >>> Keir Fraser <keir@xensource.com> 30.05.07 11:53 >>>
> >On 30/5/07 10:52, "Jan Beulich" <jbeulich@novell.com> wrote:
> >
> >> .. is out of sync on i386 and x86-64 - should it be 
> allowed on i386, or should
> >> it be suppressed on x86-64? Since Xen fails ill accesses, 
> I would assume
> >> enabling it on i386 is fine.
> >
> >Is this the MSRs-from-userland option? Should be safe to enable.
> 
> Yes.
> 
> >But probably fairly pointless?
> 
> Not so for debugging purposes - I asked because I just now 
> needed it to read
> a BIOS-initialized MSR, and it's quite handy if you don't 
> have to reboot into a
> native kernel to do just this.

I'm looking at something similar: A tool that can say if SVM is
enabled/available or not, to be run in Dom0 or before installing Xen.
The same method (directly or indirectly) could also be used by
high-level tools to determine if it's viable to install/create HVM
virtual machines. 

This would certainly be "read only" accesses, as there's nothing much
that can be done if SVM is disabled by BIOS. 

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