[openstack-dev] Host CPU feature exposure

Jiang, Yunhong yunhong.jiang at intel.com
Thu Jan 17 02:27:02 UTC 2013



> -----Original Message-----
> From: Daniel P. Berrange [mailto:berrange at redhat.com]
> Sent: Friday, January 11, 2013 5:28 PM
> To: Jiang, Yunhong
> Cc: OpenStack Development Mailing List; Dugger, Donald D
> Subject: Re: [openstack-dev] Host CPU feature exposure
> 
> On Fri, Jan 11, 2013 at 02:25:52AM +0000, Jiang, Yunhong wrote:
> > >
> > > Again we intentionally don't expose this information because applying
> > > logic based on CPU feature flags is fundamentally non-portable
> > >
> >
> > > We provide an API which allows you to pass in a CPU description (where
> > > a CPU description == a CPU model name + a list of features), and returns
> > > status on whether the host can support that CPU description. This keeps
> > > mgmt applications of the business of doing architecture specific CPU
> > > comparisons.
> >
> >
> > Daniel, can you please elaborate the difference of cpu feature flags
> > and CPU feature? From you description, seems CPU feature flags is x86
> > specifc, while CPU feature is portable. That's a bit confusing to me.
> 
> I didn't imply any difference in those terms.

Ok, I got the impression that they are different from your followed comments that seem a bit conflict to me:

"based on CPU feature flags is fundamentally non-portable" and " CPU description == a CPU model name + a list of features"  The former state that CPU feature flags is architect specific, while the latter states that "CPU model name + features" are architecture in-dependently.



More information about the OpenStack-dev mailing list