[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] svm: allow guest to use EFER.FFXSE and EFER.LMSLE


  • To: Jan Beulich <jbeulich@xxxxxxxxxx>, <xen-devel@xxxxxxxxxxxxxxxxxxx>
  • From: Keir Fraser <keir@xxxxxxxxxxxxx>
  • Date: Fri, 12 Oct 2007 11:08:14 +0100
  • Delivery-date: Fri, 12 Oct 2007 03:08:55 -0700
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: AcgMt80fC7albnirEdyUGwAX8io7RQ==
  • Thread-topic: [Xen-devel] [PATCH] svm: allow guest to use EFER.FFXSE and EFER.LMSLE

On 9/8/07 13:08, "Jan Beulich" <jbeulich@xxxxxxxxxx> wrote:

> (Applies cleanly only on top of the previously sent SVM/LBR patch.)
> 
> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxxxx>

I applied the FFXSE parts but not the LMSLE parts. I don't think it's
reasonable to support an undocumented processor feature, and this one is
particularly worrying:
 * We have to enable it inside Xen in order to detect it. But I have no docs
about what effects that has (even though the time it is enabled for is
admittedly very short).
 * Since the only mentions of EFER.LMSLE that I can find are in
family-specific BIOS and kernel developers' guides, I have no idea whether
the flag is architectural and hence whether that bit in EFER will retain its
meaning in future processor families.

 -- Keir


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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.