Re: [PATCH 06/14] target/arm/ptw: Pass an ARMSecuritySpace to arm_hcr_el2_eff_secstate()

[Richard Henderson]

On 7/24/23 07:38, Peter Maydell wrote:
> > Does it? HCR_EL2 says "behaves as 0 if EL2 is not enabled in the
> > current Security state". If the current Security state is Root then
> > EL2 isn't enabled (because there's no such thing as EL2 Root), so the
> > function should return 0, shouldn't it?
>
> I guess there's an argument that what the spec really means is
> "the security state described by the current effective value
> of SCR_EL3.{NSE,NS}" (to steal language from the docs of the
> AT operations), though.

Yes, that's how I read it.


r~