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

Re: [Xen-devel] [PATCH] x86/Intel: hide CPUID faulting capability from guests

>>> On 16.09.16 at 11:46, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 16/09/16 07:32, Jan Beulich wrote:
>> We don't currently emulate it, so guests should not be misguided to
>> believe they can (try to) use it.
>> For now, simply return zero to guests for platform MSR reads, and only
>> accept (by discarding) writes of zero. If ever there will be bits we
>> can safely expose to guests, let's handle them by white listing.
>> (As a side note - according to SDM version 059 bit 31 is reserved on
>> all known families.)
>> Reported-by: Kyle Huey <me@xxxxxxxxxxxx>
>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> Acked-by: Andrew Cooper <andrew.cooper3@xxxxxxxxx>, but I would also 
> suggest putting in a d->arch.x86_vendor check against Intel as well.

In general I agree, but then let's please do this for all instances
which currently use the host CPU vendor field at once: There's no
single example of this, and I have to admit that it's also not
immediately clear to me what the best behavior would be in some
of the cases if host and guest vendor disagree.


Xen-devel mailing list



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