* Paolo Bonzini (pbonz...@redhat.com) wrote:
> Il mer 1 lug 2020, 18:28 Dr. David Alan Gilbert <dgilb...@redhat.com> ha
> scritto:
>
> >
> > > + return (nested_state->format == KVM_STATE_NESTED_FORMAT_SVM &&
> > > + nested_state->size > offsetof(struct kvm_nested_state,
> > data));
> >
> > How does this nested_state->size work? It looks like even if it's 1 byte
> > into 'data' we transmit a whole KVM_STATE_NESTED_SVM_VMCB_SIZE.
> >
>
> The format is documented as having either 0 or KVM_STATE_NESTED_SVM_VMCB_SIZE
> bytes (for vmx instead it can be 0, 4K or 8K).
OK.
Dave
> Paolo
>
>
> > Dave
> >
> > > +}
> > > +
> > > +static const VMStateDescription vmstate_svm_nested_state = {
> > > + .name = "cpu/kvm_nested_state/svm",
> > > + .version_id = 1,
> > > + .minimum_version_id = 1,
> > > + .needed = svm_nested_state_needed,
> > > + .fields = (VMStateField[]) {
> > > + VMSTATE_U64(hdr.svm.vmcb_pa, struct kvm_nested_state),
> > > + VMSTATE_UINT8_ARRAY(data.svm[0].vmcb12,
> > > + struct kvm_nested_state,
> > > + KVM_STATE_NESTED_SVM_VMCB_SIZE),
> > > + VMSTATE_END_OF_LIST()
> > > + }
> > > +};
> > > +
> > > static bool nested_state_needed(void *opaque)
> > > {
> > > X86CPU *cpu = opaque;
> > > CPUX86State *env = &cpu->env;
> > >
> > > return (env->nested_state &&
> > > - vmx_nested_state_needed(env->nested_state));
> > > + (vmx_nested_state_needed(env->nested_state) ||
> > > + svm_nested_state_needed(env->nested_state)));
> > > }
> > >
> > > static int nested_state_post_load(void *opaque, int version_id)
> > > @@ -1139,6 +1166,7 @@ static const VMStateDescription
> > vmstate_kvm_nested_state = {
> > > },
> > > .subsections = (const VMStateDescription*[]) {
> > > &vmstate_vmx_nested_state,
> > > + &vmstate_svm_nested_state,
> > > NULL
> > > }
> > > };
> > > --
> > > 2.26.2
> > >
> > --
> > Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
> >
> >
--
Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
