On 03/05/19 11:44, Sean Christopherson wrote:
> Hmm, probably better to check KVM_CAP_NESTED_STATE. If/when support for
> AMD is added it'd be good to automatically pick up whatever testing we can.
> As a bonus it'll test the cap code.
>
Fixed like this:
diff --git a/tools/testing/selftests/kvm/x86_64/vmx_set_nested_state_test.c b/tools/testing/selftests/kvm/x86_64/vmx_set_nested_state_test.c
index 5eea24087d19..61a2163cf9f1 100644
--- a/tools/testing/selftests/kvm/x86_64/vmx_set_nested_state_test.c
+++ b/tools/testing/selftests/kvm/x86_64/vmx_set_nested_state_test.c
@@ -225,6 +225,11 @@ int main(int argc, char *argv[])
struct kvm_nested_state state;
struct kvm_cpuid_entry2 *entry = kvm_get_supported_cpuid_entry(1);
+ if (!kvm_check_cap(KVM_CAP_NESTED_STATE)) {
+ printf("KVM_CAP_NESTED_STATE not available, skipping test\n");
+ exit(KSFT_SKIP);
+ }
+
/*
* AMD currently does not implement set_nested_state, so for now we
* just early out.
Paolo
