On Wed, Nov 29, 2017 at 5:30 PM, David Hildenbrand [off-list ref] wrote:

quoted

+++ b/virt/kvm/arm/arm.c

@@ -381,14 +381,11 @@ static void vcpu_power_off(struct kvm_vcpu *vcpu)
 int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
                                  struct kvm_mp_state *mp_state)
 {
-     vcpu_load(vcpu);
-
      if (vcpu->arch.power_off)
              mp_state->mp_state = KVM_MP_STATE_STOPPED;
      else
              mp_state->mp_state = KVM_MP_STATE_RUNNABLE;

-     vcpu_put(vcpu);
      return 0;
 }

Okay, this also makes sense on other architectures. The important thing
is only that we hold the vcpu mutex.

Yes, but as Paolo said, it's better if architecture maintainers do
that themselves.  The risk of me messing things up is way too high
otherwise.

Thanks,
-Christoffer