Re: [PATCH] x86/pti: don't report XenPV as vulnerable

[Juergen Gross]

On 15/06/18 08:16, Jiri Kosina wrote:
> On Fri, 15 Jun 2018, Juergen Gross wrote:
> 
>>>> wrong for 64-bit, too, in case the mitigation is disabled at hypervisor 
>>>> level.
>>>
>>> If that is indeed possible (is it?), then the check we have in 
>>> pti_check_boottime_disable() is wrong as well.
>>
>> No, it isn't. PTI for 32-bit kernels isn't paravirtualized, so it has to
>> be disabled.
> 
> I was talking about this "mitigation disabled at Xen hypervisor level for 
> 64-bit" situation though.
> 

Why? PTI has to be disabled in PV guests as it can't work there due to
missing paravirtualization of the PTI feature (mov to/from %cr3).

The Xen meltdown mitigation ("XPTI") for 64-bit pv guests is primarily
securing the hypervisor against meltdown attacks of the guest. The guest
itself can't do anything in this regard in 64-bit mode, as user and
kernel code are already using different %cr3 values even without PTI.


Juergen