Re: [RFC PATCH v1 09/10] KVM: s390: Add capability for storage key extension of MEM_OP IOCTL

[Christian Borntraeger]

Am 18.01.22 um 10:52 schrieb Janis Schoetterl-Glausch:
> Availability of the KVM_CAP_S390_MEM_OP_SKEY capability signals that:
> * The vcpu MEM_OP IOCTL supports storage key checking.
> * The vm MEM_OP IOCTL exists.
>
> Signed-off-by: Janis Schoetterl-Glausch <scgl@xxxxxxxxxxxxx>
> ---
> Maybe this should be redesigned. As is, the capability mixes
> support of storage keys for the vcpu ioctl with the availability
> of the vm ioctl (which always supports storage keys).
>
> We could have two capabilities, one to indicate the availability
> of the vm memop and another used to derive the available functionality.
> Janosch suggested that the second capability indicates the availability
> of a "query" memop operation.

I think one capability covering both changes is totally fine as long as we document
that in api.rst.

>   arch/s390/kvm/kvm-s390.c | 1 +
>   include/uapi/linux/kvm.h | 1 +
>   2 files changed, 2 insertions(+)
>
> diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
> index ab07389fb4d9..3c6517ad43a3 100644
> --- a/arch/s390/kvm/kvm-s390.c
> +++ b/arch/s390/kvm/kvm-s390.c
> @@ -565,6 +565,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
>   	case KVM_CAP_S390_VCPU_RESETS:
>   	case KVM_CAP_SET_GUEST_DEBUG:
>   	case KVM_CAP_S390_DIAG318:
> +	case KVM_CAP_S390_MEM_OP_SKEY:
>   		r = 1;
>   		break;
>   	case KVM_CAP_SET_GUEST_DEBUG2:
> diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
> index dd04170287fd..1bb38efd1156 100644
> --- a/include/uapi/linux/kvm.h
> +++ b/include/uapi/linux/kvm.h
> @@ -1134,6 +1134,7 @@ struct kvm_ppc_resize_hpt {
>   #define KVM_CAP_EXIT_ON_EMULATION_FAILURE 204
>   #define KVM_CAP_ARM_MTE 205
>   #define KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 206
> +#define KVM_CAP_S390_MEM_OP_SKEY 209
>   
>   #ifdef KVM_CAP_IRQ_ROUTING
>