Re: [PATCH v3 04/14] KVM: s390: device attribute to set AP interpretive execution

[Halil Pasic <pasic@linux.vnet.ibm.com>]

On 03/20/2018 06:58 PM, Tony Krowiak wrote:
> I spoke with Christian this morning and he made a suggestion which I think would provide the best solution here.
> This is my proposal:
> 1. Get rid of the KVM_S390_VM_CRYPTO_INTERPRET_AP device attribute and return to setting ECA.28 from the
>    mdev device open callback.
> 2. Since there may be vcpus online at the time the mdev device open is called, we must first take all running vcpus out of
>    SIE and block them. Christian suggested the kvm_s390_vcpu_block_all(struct kvm *kvm) function will do the trick. So I
>    propose introducing a function like the following to be called during mdev open:

There is one thing you missed, otherwise I'm *very* satisfied with this
proposal.

What you have missed IMHO is vcpu hottplug. So IMHO you should keep
kvm->arch.crypto.apie, and update it accordingly ...


> 
>     int kvm_ap_set_interpretive_exec(struct kvm *kvm, bool enable)
>     {
>         int i;
>         struct kvm_vcpu *vcpu;
> 
>         if (!test_kvm_cpu_feat(kvm, KVM_S390_VM_CPU_FEAT_AP))
>             return -EOPNOTSUPP;
> 
>         mutex_lock(&kvm->lock);
> 
>         kvm_s390_vcpu_block_all(kvm);

... let's say here.

> 
>         kvm_for_each_vcpu(i, vcpu, kvm) {

And here you can call kvm_s390_vcpu_crypto_setup(vcpu) (the changes to
this function will be required for hotplug) if you like

>             if (enable)
>                 vcpu->arch.sie_block->eca |= ECA_APIE;
>             else
>                 vcpu->arch.sie_block->eca &= ~ECA_APIE;

or keep this stuff, it does not really matter to me.

>         }
> 
>         kvm_s390_vcpu_unblock_all(kvm);
> 
>         mutex_unlock(&kvm->lock);
> 
>         return 0;
>     }
> 
>    This interface allows us to set ECA.28 even if vcpus are running

I tend to agree. I will give it a proper review when this gets more
formal (e.g. v4 (preferably) or patches to be fixed up to this series).

Please don't forget to revisit the discussion on kvm_s390_vm_set_crypto:
if the mechanism there isn't right for ECA.28 I think you should tell
us why it's OK for the other attributes if it's OK. If it is not then
I guess you will want to do a stand alone patch for that.