On Fri, Nov 30, 2018 at 02:04:32PM -0500, Krish Sadhukhan wrote: > .. as they are used only in nested context. > > Signed-off-by: Krish Sadhukhan <krish.sadhukhan@xxxxxxxxxx> > Reviewed-by: Liran Alon <liran.alon@xxxxxxxxxx> > Reviewed-by: Mihai Carabas <mihai.carabas@xxxxxxxxxx> > Reviewed-by: Mark Kanda <mark.kanda@xxxxxxxxxx> > --- > arch/x86/kvm/vmx.c | 15 ++++++++------- > 1 file changed, 8 insertions(+), 7 deletions(-) > > diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c > index 02edd99..3a74a4c 100644 > --- a/arch/x86/kvm/vmx.c > +++ b/arch/x86/kvm/vmx.c > @@ -13003,7 +13003,8 @@ static int nested_vmx_check_nmi_controls(struct vmcs12 *vmcs12) > return 0; > } > > -static int check_vmentry_prereqs(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) > +static int nested_check_vmentry_prereqs(struct kvm_vcpu *vcpu, > + struct vmcs12 *vmcs12) I think the prefix should be "nested_vmx_" to be consistent with (most) other cases where we prepend nested_. It's helpful to clarify between e.g. nested_cpu...() and nested_vmx...(). For me it gives me a frame of reference, e.g. "oh, this code is querying a property of the nested CPU" vs "oh, this code implements some nested VMX behavior". Sorry for not thinking about this earlier, I feel like I'm serving death by a thousand cuts...