Re: [PATCH] KVM: arm64: vgic-v4: Initialize GICv4.1 even in the absence of a virtual ITS

[Marc Zyngier <maz@kernel.org>]

Hi Zenghui,

On 2020-04-30 12:24, Zenghui Yu wrote:
> Hi Marc,
> 
> On 2020/4/25 17:44, Marc Zyngier wrote:
>> KVM now expects to be able to use HW-accelerated delivery of vSGIs
>> as soon as the guest has enabled thm. Unfortunately, we only
> them
>> initialize the GICv4 context if we have a virtual ITS exposed to
>> the guest.
>> 
>> Fix it by always initializing the GICv4.1 context if it is
>> available on the host.
>> 
>> Fixes: 2291ff2f2a56 ("KVM: arm64: GICv4.1: Plumb SGI implementation 
>> selection in the distributor")
>> Signed-off-by: Marc Zyngier <maz@kernel.org>
>> ---
>>   virt/kvm/arm/vgic/vgic-init.c    | 9 ++++++++-
>>   virt/kvm/arm/vgic/vgic-mmio-v3.c | 3 ++-
>>   2 files changed, 10 insertions(+), 2 deletions(-)
>> 
>> diff --git a/virt/kvm/arm/vgic/vgic-init.c 
>> b/virt/kvm/arm/vgic/vgic-init.c
>> index a963b9d766b73..8e6f350c3bcd1 100644
>> --- a/virt/kvm/arm/vgic/vgic-init.c
>> +++ b/virt/kvm/arm/vgic/vgic-init.c
>> @@ -294,8 +294,15 @@ int vgic_init(struct kvm *kvm)
>>   		}
>>   	}
>>   -	if (vgic_has_its(kvm)) {
>> +	if (vgic_has_its(kvm))
>>   		vgic_lpi_translation_cache_init(kvm);
>> +
>> +	/*
>> +	 * If we have GICv4.1 enabled, unconditionnaly request enable the
>> +	 * v4 support so that we get HW-accelerated vSGIs. Otherwise, only
>> +	 * enable it if we present a virtual ITS to the guest.
>> +	 */
>> +	if (vgic_supports_direct_msis(kvm)) {
>>   		ret = vgic_v4_init(kvm);
>>   		if (ret)
>>   			goto out;
>> diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c 
>> b/virt/kvm/arm/vgic/vgic-mmio-v3.c
>> index e72dcc4542475..26b11dcd45524 100644
>> --- a/virt/kvm/arm/vgic/vgic-mmio-v3.c
>> +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c
>> @@ -50,7 +50,8 @@ bool vgic_has_its(struct kvm *kvm)
>>     bool vgic_supports_direct_msis(struct kvm *kvm)
>>   {
>> -	return kvm_vgic_global_state.has_gicv4 && vgic_has_its(kvm);
>> +	return (kvm_vgic_global_state.has_gicv4_1 ||
>> +		(kvm_vgic_global_state.has_gicv4 && vgic_has_its(kvm)));
>>   }
> 
> Not related to this patch, but I think that the function name can be
> improved a bit after this change. It now indicates whether the vGIC
> supports direct MSIs injection *or* direct SGIs injection, not just
> MSIs. And if vgic_has_its() is false, we don't even support MSIs.

Yes, I noticed that too. But in the spirit of keeping the change minimal
and avoid later conflicts with potential fixes, I decided against 
changing
it right now.

> The fix itself looks correct to me,
> 
> Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>

Thanks,

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