Paolo Bonzini <pbonzini@xxxxxxxxxx> writes:
> Both Intel and AMD now implement it, so there is no need to check if the
> callback is implemented.
>
> Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
> ---
> arch/x86/kvm/x86.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 59958ce2b681..0492baeb78ab 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -7699,7 +7699,7 @@ static int inject_pending_event(struct kvm_vcpu *vcpu)
> * from L2 to L1 due to pending L1 events which require exit
> * from L2 to L1.
> */
> - if (is_guest_mode(vcpu) && kvm_x86_ops.check_nested_events) {
> + if (is_guest_mode(vcpu)) {
> r = kvm_x86_ops.check_nested_events(vcpu);
> if (r != 0)
> return r;
> @@ -7761,7 +7761,7 @@ static int inject_pending_event(struct kvm_vcpu *vcpu)
> * proposal and current concerns. Perhaps we should be setting
> * KVM_REQ_EVENT only on certain events and not unconditionally?
> */
> - if (is_guest_mode(vcpu) && kvm_x86_ops.check_nested_events) {
> + if (is_guest_mode(vcpu)) {
> r = kvm_x86_ops.check_nested_events(vcpu);
> if (r != 0)
> return r;
> @@ -8527,7 +8527,7 @@ static inline int vcpu_block(struct kvm *kvm, struct kvm_vcpu *vcpu)
>
> static inline bool kvm_vcpu_running(struct kvm_vcpu *vcpu)
> {
> - if (is_guest_mode(vcpu) && kvm_x86_ops.check_nested_events)
> + if (is_guest_mode(vcpu))
> kvm_x86_ops.check_nested_events(vcpu);
>
> return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE &&
While the callback is implemented for both VMX and SVM, it can still be
NULL when !nested (thus patch subject is a bit misleading) but
is_guest_mode() implies this is not the case.
Reviewed-by: Vitaly Kuznetsov <vkuznets@xxxxxxxxxx>
--
Vitaly
