Re: [PATCH v3] kvm: svm: Use the hardware provided GPA instead of page walk

[Paolo Bonzini]

On 29/11/2016 20:38, Brijesh Singh wrote:
> 
> 
> On 11/29/2016 12:20 PM, Thomas Gleixner wrote:
>> On Tue, 29 Nov 2016, Brijesh Singh wrote:
>>> --- a/arch/x86/kvm/emulate.c
>>> +++ b/arch/x86/kvm/emulate.c
>>> @@ -5483,3 +5483,11 @@ void emulator_writeback_register_cache(struct
>>> x86_emulate_ctxt *ctxt)
>>>  {
>>>      writeback_registers(ctxt);
>>>  }
>>> +
>>> +bool emulator_is_string_op(struct x86_emulate_ctxt *ctxt)
>>> +{
>>> +    if (ctxt->d & String)
>>> +        return true;
>>> +
>>> +    return false;
>>> +}
>>
>> Do we really need a full function call for this simple thing? Just
>> because
>> this horrible CamelCase constant is in emulate.c?
>>
>> What's wrong with moving that thing into a header and make it a trivial
>> inline:
>>
>> #define INS_STRING        (1 << 13)
>>
>> static inline bool emulator_is_string_op(struct x86_emulate_ctxt *ctxt)
>> {
>>     return ctxt->d & INS_STRING;
>> }
>>
>> Hmm?
> 
> One of the recommendation from previous review feedback was to move the
> check inside emulator.c. I am fine with inlining it into kvm_emulate.h
> 
> Hi Paolo,
> 
> Do you want me to spin a new version ?

No, it's not that much of a fast path, so it wouldn't change anything in
practice.

Paolo