On Wed, 7 Apr 2021 22:47:07 +0000 Sean Christopherson wrote:
> On Thu, Apr 08, 2021, Kai Huang wrote:
> > On Wed, 7 Apr 2021 22:35:32 +0000 Sean Christopherson wrote:
> > > On Thu, Apr 08, 2021, Kai Huang wrote:
> > > > On Wed, 7 Apr 2021 22:16:59 +0000 Sean Christopherson wrote:
> > > > > On Thu, Apr 08, 2021, Kai Huang wrote:
> > > > > > +int handle_encls(struct kvm_vcpu *vcpu)
> > > > > > +{
> > > > > > + u32 leaf = (u32)vcpu->arch.regs[VCPU_REGS_RAX];
> > > > >
> > > > > Please use kvm_rax_read(), I've been trying to discourage direct access to the
> > > > > array. Which is ironic because I'm 100% certain I'm to blame for this. :-)
> > > >
> > > > Sure. But I think still, we should convert it to (u32) explicitly, so:
> > > >
> > > > u32 leaf = (u32)kvm_rax_read(vcpu);
> > > >
> > > > ?
> > >
> > > Ya, agreed, it helps document that it's deliberate.
> >
> > Do you have any other comments regarding to other patches? If no I can send
> > another version rather quickly :)
>
> Nope, nothing at this time. Though I'd give folks a few days to review before
> sending the next version, I don't think any of my feedback will affect other
> reviews.
My thinking too, but OK I'll wait for other people's review, plus I'd like to
hear about on how to proceed given the current series has some merge conflicts
with latest kvm/queue, although they are quite straightforward to resolve.
