On Wed, Aug 03, 2022, Maxim Levitsky wrote:
> On Wed, 2022-08-03 at 10:51 -0700, Mingwei Zhang wrote:
> > On Wed, Aug 3, 2022 at 10:18 AM Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote:
> > > On 8/3/22 01:07, Mingwei Zhang wrote:
> > > > + /*
> > > > + * We must first get the vmcs12 pages before checking for interrupts
> > > > + * that might unblock the guest if L1 is using virtual-interrupt
> > > > + * delivery.
> > > > + */
> > > > + if (kvm_check_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu)) {
> > > > + /*
> > > > + * If we have to ask user-space to post-copy a page,
> > > > + * then we have to keep trying to get all of the
> > > > + * VMCS12 pages until we succeed.
> > > > + */
> > > > + if (unlikely(!kvm_x86_ops.nested_ops->get_nested_state_pages(vcpu))) {
> > > > + kvm_make_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu);
> > > > + return 0;
> > > > + }
> > > > + }
> > > > +
> > >
> > > I think request handling (except for KVM_REQ_EVENT) could be more
> > > generically moved from vcpu_enter_guest() to vcpu_run().
> >
> > Yeah, sounds good to me. I can come up with an updated version. At
> > least, I will remove the repeat request here.
>
> Now it all makes sense. I do think that KVM_REQ_GET_NESTED_STATE_PAGES processing
> when the vCPU is halted is indeed missing.
>
> This reminds me that I would be *very* happy to remove the KVM_REQ_GET_NESTED_STATE_PAGES,
> if by any chance there is an agreement to do so upstream.
> This is yet another reason to do so to be honest.
> Just my 0.2 cents of course.
+100
@google folks, what would it take for us to mark KVM_REQ_GET_NESTED_STATE_PAGES
as deprecated in upstream and stop accepting patches/fixes? IIUC, when we eventually
move to userfaultfd, all this goes away, i.e. we do want to ditch this at some point.
