Re: [PATCH v21 05/23] KVM: arm64: Track the type of VM in kvm_arch

[Suzuki K Poulose]

On 03/10/2026 06:53, Suzuki K Poulose wrote:
> On 02/10/2026 16:29, Marc Zyngier wrote:
> > On Fri, 02 Oct 2026 15:37:00 +0100,
> > Fuad Tabba <tabba@xxxxxxxxxx> wrote:
> > > Hi Marc,
> > >
> > > On Fri, 02 Oct 2026 15:15:06 +0100, Marc Zyngier <maz@xxxxxxxxxx> wrote:
> > > [...]
> > > > Honestly, we introduce the flavor stuff to make it easy to match
> > > > things on a particular VM type in a readable way. So why isn't this
> > > > reading:
> > > >
> > > >          if (vcpu->kvm->arch.vm_flavor != VM_PROTECTED_PKVM)
> > > >                  return;
> > >
> > > I know this is a sketch, but just in case: that's inverted. The sync
> > > only applies to non-protected pKVM VMs (EL2 ignores it for protected
> > > ones), so it would be != VM_PKVM.
> >
> > See what I meant about this stuff being completely intractable? I
> > still have no idea what it means! ;-)
>
> Just to be clear: unprotected_pkvm() != (vm_flavor != VM_PROTECED_PKVM).
> Rather, unprotected_pkvm => (vm_flavor == VM_PKVM).
>
> And the code wanted to bail out early for !unprotected_pkvm(). We can
> stick in "is_protected_kvm_enabled()" for unprotected_pkvm predicate
>
> But, I can drop the helper and use the vm_flavor check.

FWIW: Here is the diff for the above change. If you are happy
with the following, I could fold this in.


diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 90547fbbc8ad7..bcec14c587119 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -772,7 +772,7 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
                kvm_call_hyp_nvhe(__pkvm_vcpu_put);

                /* __pkvm_vcpu_put implies a sync of the state */
-               if (kvm_vm_is_unprotected_pkvm(vcpu->kvm))
+               if (vcpu->kvm->arch.vm_flavor == VM_PKVM)
                        vcpu_set_flag(vcpu, PKVM_HOST_STATE_DIRTY);
        }

@@ -1006,7 +1006,7 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu 
*vcpu)

        if (is_protected_kvm_enabled()) {
                /* Start with the vcpu in a dirty state */
-               if (kvm_vm_is_unprotected_pkvm(vcpu->kvm))
+               if (vcpu->kvm->arch.vm_flavor == VM_PKVM)
                        vcpu_set_flag(vcpu, PKVM_HOST_STATE_DIRTY);
                ret = pkvm_create_hyp_vm(kvm);
                if (ret)
diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
index 384c5d258c7f8..4e16e7d165515 100644
--- a/arch/arm64/kvm/handle_exit.c
+++ b/arch/arm64/kvm/handle_exit.c
@@ -490,7 +490,7 @@ static void handle_exit_pkvm_state(struct kvm_vcpu 
*vcpu, int exception_index)
 {
        int exception_code = ARM_EXCEPTION_CODE(exception_index);

-       if (!kvm_vm_is_unprotected_pkvm(vcpu->kvm))
+       if (!is_protected_kvm_enabled() || vcpu->kvm->arch.vm_flavor != 
VM_PKVM)
                return;

        /*



Cheers
Suzuki>
> Cheers
> Suzuki
>
> >     M.