Re: [PATCH v3 03/14] KVM: SVM: Detect X2APIC virtualization (x2AVIC) support

[Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>]

Maxim,

On 5/4/22 7:12 PM, Maxim Levitsky wrote:
>> diff --git a/arch/x86/kvm/svm/avic.c b/arch/x86/kvm/svm/avic.c
>> index a8f514212b87..fc3ba6071482 100644
>> --- a/arch/x86/kvm/svm/avic.c
>> +++ b/arch/x86/kvm/svm/avic.c
>> @@ -40,6 +40,15 @@
>>   #define AVIC_GATAG_TO_VMID(x)		((x >> AVIC_VCPU_ID_BITS) & AVIC_VM_ID_MASK)
>>   #define AVIC_GATAG_TO_VCPUID(x)		(x & AVIC_VCPU_ID_MASK)
>>   
>> +enum avic_modes {
>> +	AVIC_MODE_NONE = 0,
>> +	AVIC_MODE_X1,
>> +	AVIC_MODE_X2,
>> +};
>> +
>> +static bool force_avic;
>> +module_param_unsafe(force_avic, bool, 0444);
>> +
>>   /* Note:
>>    * This hash table is used to map VM_ID to a struct kvm_svm,
>>    * when handling AMD IOMMU GALOG notification to schedule in
>> @@ -50,6 +59,7 @@ static DEFINE_HASHTABLE(svm_vm_data_hash, SVM_VM_DATA_HASH_BITS);
>>   static u32 next_vm_id = 0;
>>   static bool next_vm_id_wrapped = 0;
>>   static DEFINE_SPINLOCK(svm_vm_data_hash_lock);
>> +static enum avic_modes avic_mode;
>>   
>>   /*
>>    * This is a wrapper of struct amd_iommu_ir_data.
>> @@ -1077,3 +1087,33 @@ void avic_vcpu_unblocking(struct kvm_vcpu *vcpu)
>>   
>>   	avic_vcpu_load(vcpu);
>>   }
>> +
>> +/*
>> + * Note:
>> + * - The module param avic enable both xAPIC and x2APIC mode.
>> + * - Hypervisor can support both xAVIC and x2AVIC in the same guest.
>> + * - The mode can be switched at run-time.
>> + */
>> +bool avic_hardware_setup(struct kvm_x86_ops *x86_ops)
>> +{
>> +	if (!npt_enabled)
>> +		return false;
>> +
>> +	if (boot_cpu_has(X86_FEATURE_AVIC)) {
>> +		avic_mode = AVIC_MODE_X1;
>> +		pr_info("AVIC enabled\n");
>> +	} else if (force_avic) {
>> +		pr_warn("AVIC is not supported in CPUID but force enabled");
>> +		pr_warn("Your system might crash and burn");
> I think in this case avic_mode should also be set to AVIC_MODE_X1
> (Hopefully this won't be needed for systems that have x2avic enabled)

You are right. My appology :(

Actually, x2AVIC depends on both CPUID bits (i.e. X86_FEATURE_AVIC and X86_FEATURE_X2AVIC).
If the force_avic option is only applicable to only the X86_FEATURE_AVIC bit, we would
need to check for the following condition before enabling x2AVIC support in the driver:

   if ((X86_FEATURE_AVIC | avic_force) & X86_FEATURE_X2AVIC)
	avic_mode = AVIC_MODE_X2

Regards,
Suravee