Re: [Xen-devel] [PATCH v3 2/8] x86/HVM: ignore guest INVD uses

[Jan Beulich <jbeulich@xxxxxxxx>]

On 03.09.2019 12:18, Andrew Cooper wrote:
> On 03/09/2019 10:37, Jan Beulich wrote:
>> The only place we'd expect the insn to be sensibly used is in
>> (virtualization unaware) firmware.
>>
>> Suggested-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
>> ---
>> v3: New.
>>
>> --- a/xen/arch/x86/hvm/emulate.c
>> +++ b/xen/arch/x86/hvm/emulate.c
>> @@ -2210,11 +2210,18 @@ static int hvmemul_cache_op(
>>  
>>          hvmemul_unmap_linear_addr(mapping, addr, 0, hvmemul_ctxt);
>>          /* fall through */
>> -    case x86emul_invd:
>>      case x86emul_wbinvd:
>>      case x86emul_wbnoinvd:
>>          alternative_vcall(hvm_funcs.wbinvd_intercept);
>>          break;
>> +
>> +    case x86emul_invd:
>> +        /*
>> +         * Deliberately ignored: We don't want to issue INVD, and issuing 
>> WBINVD
> 
> I'd phrase this more strongly.  We absolutely must not issue INVD or we
> break cache coherency.
> 
> Ideally with this adjusted, Reviewed-by: Andrew Cooper
> <andrew.cooper3@xxxxxxxxxx>

Thanks, I've replaced "don't want to" with "mustn't".

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel