Re: [RFCv2 PATCH 06/12] KVM: SVM: Do not update logical APIC ID table when in x2APIC mode

[Maxim Levitsky]

On Tue, 2022-03-08 at 10:39 -0600, Suravee Suthikulpanit wrote:
> In X2APIC mode the Logical Destination Register is read-only,
> which provides a fixed mapping between the logical and physical
> APIC IDs. Therefore, there is no Logical APIC ID table in X2AVIC
> and the processor uses the X2APIC ID in the backing page to create
> a vCPU’s logical ID. Also, when x2AVIC is activated, a guest write
> to the x2APIC LDR register would result in #GP injection into
> the guest by the hardware.
> 
> Therefore, add logic to check x2APIC mode before updating logical
> APIC ID table.
> 
> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@xxxxxxx>
> ---
>  arch/x86/kvm/svm/avic.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kvm/svm/avic.c b/arch/x86/kvm/svm/avic.c
> index 5329b93dc4cd..4d7a8743196e 100644
> --- a/arch/x86/kvm/svm/avic.c
> +++ b/arch/x86/kvm/svm/avic.c
> @@ -406,6 +406,10 @@ static int avic_ldr_write(struct kvm_vcpu *vcpu, u8 g_physical_id, u32 ldr)
>  	bool flat;
>  	u32 *entry, new_entry;
>  
> +	/* Note: x2AVIC does not use logical APIC ID table */
> +	if (apic_x2apic_mode(vcpu->arch.apic))
> +		return 0;
> +
>  	flat = kvm_lapic_get_reg(vcpu->arch.apic, APIC_DFR) == APIC_DFR_FLAT;
>  	entry = avic_get_logical_id_entry(vcpu, ldr, flat);
>  	if (!entry)
> @@ -424,8 +428,13 @@ static void avic_invalidate_logical_id_entry(struct kvm_vcpu *vcpu)
>  {
>  	struct vcpu_svm *svm = to_svm(vcpu);
>  	bool flat = svm->dfr_reg == APIC_DFR_FLAT;
> -	u32 *entry = avic_get_logical_id_entry(vcpu, svm->ldr_reg, flat);
> +	u32 *entry;
> +
> +	/* Note: x2AVIC does not use logical APIC ID table */
> +	if (apic_x2apic_mode(vcpu->arch.apic))
> +		return;
>  
> +	entry = avic_get_logical_id_entry(vcpu, svm->ldr_reg, flat);
>  	if (entry)
>  		clear_bit(AVIC_LOGICAL_ID_ENTRY_VALID_BIT, (unsigned long *)entry);
>  }

Reviewed-by: Maxim Levitsky <mlevitsk@xxxxxxxxxx>

Best regards,
	Maxim Levitsky