On 08/15/2016 11:13 AM, Peter Maydell wrote:
<snip>
>> static inline int arm_feature(CPUARMState *env, int feature)
>> diff --git a/target-arm/cpu64.c b/target-arm/cpu64.c
>> index 1635deb..19e8127 100644
>> --- a/target-arm/cpu64.c
>> +++ b/target-arm/cpu64.c
>> @@ -111,6 +111,9 @@ static void aarch64_a57_initfn(Object *obj)
>> set_feature(&cpu->env, ARM_FEATURE_V8_PMULL);
>> set_feature(&cpu->env, ARM_FEATURE_CRC);
>> set_feature(&cpu->env, ARM_FEATURE_EL3);
>> + if (kvm_enabled()) {
>> + set_feature(&cpu->env, ARM_FEATURE_HOST_PMU);
>> + }
>> cpu->kvm_target = QEMU_KVM_ARM_TARGET_CORTEX_A57;
>> cpu->midr = 0x411fd070;
>> cpu->revidr = 0x00000000;
>> @@ -166,6 +169,9 @@ static void aarch64_a53_initfn(Object *obj)
>> set_feature(&cpu->env, ARM_FEATURE_V8_PMULL);
>> set_feature(&cpu->env, ARM_FEATURE_CRC);
>> set_feature(&cpu->env, ARM_FEATURE_EL3);
>> + if (kvm_enabled()) {
>> + set_feature(&cpu->env, ARM_FEATURE_HOST_PMU);
>> + }
>
> This definitely looks like the wrong place to be checking
> kvm_enabled().
If we don't want to check kvm_enabled() in aarch64_a5x_initfn(), one
viable place is to move it up to cpu.c file, like in
arm_cpu_realizefn(). Do you have any other suggestions?
>
>> cpu->kvm_target = QEMU_KVM_ARM_TARGET_CORTEX_A53;
>> cpu->midr = 0x410fd034;
>> cpu->revidr = 0x00000000;
