KVM: nSVM: mark vmcb as dirty when forcingly leaving the guest mode

We overwrite most of vmcb fields while doing so, so we must
mark it as dirty.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20210107093854.882483-5-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index a466336aab43adfd879e86178b0d148e55352eb9..a622e63739b4ad07ffb8d74d979eae087f47e30c 100644 (file)
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -754,6 +754,7 @@ void svm_leave_nested(struct vcpu_svm *svm)
                leave_guest_mode(&svm->vcpu);
                copy_vmcb_control_area(&vmcb->control, &hsave->control);
                nested_svm_uninit_mmu_context(&svm->vcpu);
+               vmcb_mark_all_dirty(svm->vmcb);
        }
 
        kvm_clear_request(KVM_REQ_GET_NESTED_STATE_PAGES, &svm->vcpu);