> On 15 Oct 2018, at 21:42, Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote:
>
> On 15/10/2018 20:07, Liran Alon wrote:
>> Just out of curiosity, why is this specifically a decision-point for flipping nested=1?
>> Because this is a break of userspace APIs in case of nVMX workloads?
>> What are the conditions for flipping it?
>>
>> If that’s the case, this is also true for the get/set vCPU events IOCTLs.
>> So they need to also be applied before the flip.
>
> Yes, I meant v2 of the whole series. It is not strictly necessary to
> have this before the flip, but we're close enough that it makes sense to
> order that.
>
> Paolo
Kinda weird that this decision seem to easily have been taken for AMD in commit:
4b6e4dca7011 ("KVM: SVM: enable nested svm by default”)
I highly doubt that KVM nSVM is really more stable than KVM nVMX…
But maybe I’m wrong.
-Liran
