Re: [Xen-devel] [PATCH RFC 08/12] xen/x86: setup PVHv2 Dom0 CPUs

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 26.09.16 at 19:05, <roger.pau@xxxxxxxxxx> wrote:
> On Mon, Sep 26, 2016 at 10:19:04AM -0600, Jan Beulich wrote:
>> >>> On 29.07.16 at 18:29, <roger.pau@xxxxxxxxxx> wrote:
>> > Initialize Dom0 BSP/APs and setup the memory and IO permissions.
>> > 
>> > Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
>> > ---
>> > Cc: Jan Beulich <jbeulich@xxxxxxxx>
>> > Cc: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>> > ---
>> > The logic used to setup the CPUID leaves is extremely simplistic (and
>> > probably wrong for hardware different than mine). I'm not sure what's the
>> > best way to deal with this, the code that currently sets the CPUID leaves
>> > for HVM guests lives in libxc, maybe moving it xen/common would be better?
>> 
>> Why don't you just set them from their respective hardware values?
> 
> That's what I'm already doing in this patch, but I'm not sure what should be 
> removed (at least virt extensions should be removed, but I guess there's 
> other stuff that we might want to remove too).

Well, wouldn't the first step be to match current Dom0 behavior?
The reference what to clear here would then be the HVM feature
set and the current behavior of pv_cpuid() for Dom0.

Virt extensions, since you mention them, are particularly something
I'm not sure need clearing (at least mid to long term), since nested
mode could arguably be usable by Dom0.

Jan

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