Re: [RFC PATCH 4/5] target/arm: enable feature ARM_FEATURE_EL2 if EL2 is supported

[Eric Auger <eauger@redhat.com>]

Hi Miguel,

On 2/27/23 17:37, Miguel Luis wrote:
> From: Haibo Xu <haibo.xu@linaro.org>
> 
> KVM_CAP_ARM_EL2 must be supported by the cpu to enable ARM_FEATURE_EL2.
> EL2 bits on ID_AA64PFR0 state unsupported on the value 0b0000.
> 
> Ref: https://lore.kernel.org/qemu-devel/b7c2626e6c720ccc43e57197dff3dac72d613640.1616052890.git.haibo.xu@linaro.org/
> 
> Signed-off-by: Haibo Xu <haibo.xu@linaro.org>
> [Miguel Luis: use of ID_AA64PFR0 for cpu features]
> Signed-off-by: Miguel Luis <miguel.luis@oracle.com>
> ---
>  target/arm/cpu.h   |  2 +-
>  target/arm/kvm64.c | 16 ++++++++++++++++
>  2 files changed, 17 insertions(+), 1 deletion(-)
> 
> diff --git a/target/arm/cpu.h b/target/arm/cpu.h
> index 9aeed3c848..de2a88b43e 100644
> --- a/target/arm/cpu.h
> +++ b/target/arm/cpu.h
> @@ -3961,7 +3961,7 @@ static inline bool isar_feature_aa64_aa32_el1(const ARMISARegisters *id)
>  
>  static inline bool isar_feature_aa64_aa32_el2(const ARMISARegisters *id)
>  {
> -    return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, EL2) >= 2;
> +    return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, EL2) != 0;
>  }
>  
>  static inline bool isar_feature_aa64_ras(const ARMISARegisters *id)
> diff --git a/target/arm/kvm64.c b/target/arm/kvm64.c
> index be8144a2b5..f7ebd731aa 100644
> --- a/target/arm/kvm64.c
> +++ b/target/arm/kvm64.c
> @@ -505,6 +505,7 @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
>       */
>      int fdarray[3];
>      bool sve_supported;
> +    bool el2_supported;
>      bool pmu_supported = false;
>      uint64_t features = 0;
>      int err;
> @@ -535,6 +536,14 @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
>          init.features[0] |= 1 << KVM_ARM_VCPU_SVE;
>      }
>  
> +    /*
> +     * Ask for EL2 if supported.
> +     */
> +    el2_supported = kvm_arm_el2_supported();
> +    if (el2_supported) {
> +        init.features[0] |= 1 << KVM_ARM_VCPU_HAS_EL2;
This doesn't work if your host both supports SVE & NV.

The error output by qemu is not straightforward

qemu-system-aarch64: can't apply global host-arm-cpu.sve=off: Property
'host-arm-cpu.sve' not found

The problem is that we create a scratch VM with a CPU featuring both SVE
and NV and this fails at kernel level, I think on vcpu reset.

The trouble is that we do that even if sve=off at qemu level. So I think
this is a more generic issue related to the way we validate host cpu
features.

Thanks

Eric


> +    }
> +
>      /*
>       * Ask for Pointer Authentication if supported, so that we get
>       * the unsanitized field values for AA64ISAR1_EL1.
> @@ -714,6 +723,10 @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
>      features |= 1ULL << ARM_FEATURE_PMU;
>      features |= 1ULL << ARM_FEATURE_GENERIC_TIMER;
>  
> +    if (el2_supported) {
> +        features |= 1ULL << ARM_FEATURE_EL2;
> +    }
> +
>      ahcf->features = features;
>  
>      return true;
> @@ -881,6 +894,9 @@ int kvm_arch_init_vcpu(CPUState *cs)
>          assert(kvm_arm_sve_supported());
>          cpu->kvm_init_features[0] |= 1 << KVM_ARM_VCPU_SVE;
>      }
> +    if (cpu_isar_feature(aa64_aa32_el2, cpu)) {
> +        cpu->kvm_init_features[0] |= 1 << KVM_ARM_VCPU_HAS_EL2;
> +    }
>      if (cpu_isar_feature(aa64_pauth, cpu)) {
>          cpu->kvm_init_features[0] |= (1 << KVM_ARM_VCPU_PTRAUTH_ADDRESS |
>                                        1 << KVM_ARM_VCPU_PTRAUTH_GENERIC);