Re: [PATCH][4.15] x86/shadow: suppress "fast fault path" optimization without reserved bits

[Jan Beulich <jbeulich@xxxxxxxx>]

On 25.02.2021 14:20, Ian Jackson wrote:
> Jan Beulich writes ("Re: [PATCH][4.15] x86/shadow: suppress "fast fault path" 
> optimization without reserved bits"):
>> As to 4.15: Without this shadow mode simply won't work on such (new)
>> hardware. Hence something needs to be done anyway. An alternative
>> would be to limit the change to just the guest-no-present entries (to
>> at least allow PV guests to be migrated), and refuse to enable shadow
>> mode for HVM guests on such hardware. (In this case we'd probably
>> better take care of ...
> 
> Thanks for this explanation.
> 
> It sounds like the way you have it in this proposed patch is simpler
> than the alternative.  And that right now it's not a regression, but
> it is needed for running Xen on such newer hardware.

I'm not sure about the "simpler" part.

>> The main risk here is (in particular for the MMIO part of the change
>> I suppose) execution suddenly going a different path, which has been
>> unused / untested (for this specific case) for years.
> 
> That's somewhat concerning.  But I think this only applies to the new
> hardware ?  So it would be risking an XSA but not really risking the
> release very much.

Right - afaict an XSA would also be lurking without us doing anything,
as we'd permit a guest access to pages we didn't mean to hand to it.

> I think therefore:
> 
> Release-Acked-by: Ian Jackson <iwj@xxxxxxxxxxxxxx>

Thanks.

Jan