Re: [PATCH 2/2] KVM: arm64: Advertise ID_AA64PFR0_EL1.CSV3=1 if the CPUs are Meltdown-safe

[Marc Zyngier <maz@xxxxxxxxxx>]

Hi David,

On 2020-11-30 15:26, David Brazdil wrote:
> > @@ -1227,9 +1229,16 @@ static int set_id_aa64pfr0_el1(struct kvm_vcpu 
> > *vcpu,
> >  	    (csv2 && arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED))
> >  		return -EINVAL;
> >
> > -	/* We can only differ with CSV2, and anything else is an error */
> > +	/* Same thing for CSV3 */
> > +	csv3 = cpuid_feature_extract_unsigned_field(val, 
> > ID_AA64PFR0_CSV3_SHIFT);
> > +	if (csv3 > 1 ||
> > +	    (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
> > +		return -EINVAL;
> > +
> > +	/* We can only differ with CSV[23], and anything else is an error */
> >  	val ^= read_id_reg(vcpu, rd, false);
> > -	val &= ~(0xFUL << ID_AA64PFR0_CSV2_SHIFT);
> > +	val &= ~((0xFUL << ID_AA64PFR0_CSV2_SHIFT) ||
> > +		 (0xFUL << ID_AA64PFR0_CSV3_SHIFT));
>
> That boolean OR looks like a typo.

It definitely is. Who the hell is writing this code?

Thanks,

        M.
--
Jazz is not dead. It just smells funny...