Re: [PATCH v19 2/7] firmware: arm_rmm: Check for RMI support at init

[Marc Zyngier <maz@kernel.org>]

On Fri, 25 Sep 2026 16:23:49 +0100,
Suzuki K Poulose <suzuki.poulose@arm.com> wrote:
> 
> On 25/09/2026 11:42, Catalin Marinas wrote:
> > On Thu, Sep 24, 2026 at 02:51:56PM +0100, Suzuki K Poulose wrote:
> >> diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
> >> index 32102c3912fa7..e8b29983b0021 100644
> >> --- a/arch/arm64/kernel/cpufeature.c
> >> +++ b/arch/arm64/kernel/cpufeature.c
> >> @@ -293,6 +293,7 @@ static const struct arm64_ftr_bits ftr_id_aa64isar3[] = {
> >>   static const struct arm64_ftr_bits ftr_id_aa64pfr0[] = {
> >>   	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64PFR0_EL1_CSV3_SHIFT, 4, 0),
> >>   	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64PFR0_EL1_CSV2_SHIFT, 4, 0),
> >> +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64PFR0_EL1_RME_SHIFT, 4, 0),
> > 
> > I think we end up exposing this to (any) guest unless we mask it out in
> > sanitise_id_aa64dfr0_el1(). Do the guests care about the RME version? I
> 
> Valid point. I personally don't think this matters. The FEAT_RME alone
> is not useful or harmful. I could mask it in KVM as a safe bet.

Is there *any* case where the guest (realm or not) should know it is
running on RME enabled HW? I can perfectly imagine that someone would
implement something looking like CCA on non-RME systems.

	M.

-- 
Jazz isn't dead. It just smells funny.