Re: [PATCH v5 19/21] KVM: s390: add KVM_S390_ZPCI_OP to manage guest zPCI devices

[Pierre Morel <pmorel@linux.ibm.com>]

On 4/4/22 19:43, Matthew Rosato wrote:
> The KVM_S390_ZPCI_OP ioctl provides a mechanism for managing
> hardware-assisted virtualization features for s390X zPCI passthrough.
> Add the first 2 operations, which can be used to enable/disable
> the specified device for Adapter Event Notification interpretation.
> 
> Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
> ---
>   Documentation/virt/kvm/api.rst | 45 +++++++++++++++++++++++
>   arch/s390/kvm/kvm-s390.c       | 23 ++++++++++++
>   arch/s390/kvm/pci.c            | 65 ++++++++++++++++++++++++++++++++++
>   arch/s390/kvm/pci.h            |  2 ++
>   include/uapi/linux/kvm.h       | 31 ++++++++++++++++
>   5 files changed, 166 insertions(+)
> 
> diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
> index d13fa6600467..474f502c2ea0 100644
> --- a/Documentation/virt/kvm/api.rst
> +++ b/Documentation/virt/kvm/api.rst
> @@ -5645,6 +5645,51 @@ enabled with ``arch_prctl()``, but this may change in the future.
>   The offsets of the state save areas in struct kvm_xsave follow the contents
>   of CPUID leaf 0xD on the host.
>   
> +4.135 KVM_S390_ZPCI_OP
> +--------------------
> +
> +:Capability: KVM_CAP_S390_ZPCI_OP
> +:Architectures: s390
> +:Type: vcpu ioctl
> +:Parameters: struct kvm_s390_zpci_op (in)
> +:Returns: 0 on success, <0 on error
> +
> +Used to manage hardware-assisted virtualization features for zPCI devices.
> +
> +Parameters are specified via the following structure::
> +
> +  struct kvm_s390_zpci_op {
> +	/* in */
> +	__u32 fh;		/* target device */
> +	__u8  op;		/* operation to perform */
> +	__u8  pad[3];
> +	union {
> +		/* for KVM_S390_ZPCIOP_REG_AEN */
> +		struct {
> +			__u64 ibv;	/* Guest addr of interrupt bit vector */
> +			__u64 sb;	/* Guest addr of summary bit */
> +			__u32 flags;
> +			__u32 noi;	/* Number of interrupts */
> +			__u8 isc;	/* Guest interrupt subclass */
> +			__u8 sbo;	/* Offset of guest summary bit vector */
> +			__u16 pad;
> +		} reg_aen;
> +		__u64 reserved[8];
> +	} u;
> +  };
> +
> +The type of operation is specified in the "op" field.
> +KVM_S390_ZPCIOP_REG_AEN is used to register the VM for adapter event
> +notification interpretation, which will allow firmware delivery of adapter
> +events directly to the vm, with KVM providing a backup delivery mechanism;
> +KVM_S390_ZPCIOP_DEREG_AEN is used to subsequently disable interpretation of
> +adapter event notifications.
> +
> +The target zPCI function must also be specified via the "fh" field.  For the
> +KVM_S390_ZPCIOP_REG_AEN operation, additional information to establish firmware
> +delivery must be provided via the "reg_aen" struct.
> +
> +The "reserved" field is meant for future extensions.
>   
>   5. The kvm_run structure
>   ========================
> diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
> index 704d85214f4f..65a53e22f686 100644
> --- a/arch/s390/kvm/kvm-s390.c
> +++ b/arch/s390/kvm/kvm-s390.c
> @@ -616,6 +616,12 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
>   	case KVM_CAP_S390_PROTECTED:
>   		r = is_prot_virt_host();
>   		break;
> +	case KVM_CAP_S390_ZPCI_OP:
> +		if (kvm_s390_pci_interp_allowed())
> +			r = 1;
> +		else
> +			r = 0;
> +		break;
>   	default:
>   		r = 0;
>   	}
> @@ -2621,6 +2627,23 @@ long kvm_arch_vm_ioctl(struct file *filp,
>   			r = -EFAULT;
>   		break;
>   	}
> +	case KVM_S390_ZPCI_OP: {
> +		struct kvm_s390_zpci_op args;
> +
> +		r = -EINVAL;
> +		if (!IS_ENABLED(CONFIG_VFIO_PCI))
> +			break;
> +		if (copy_from_user(&args, argp, sizeof(args))) {
> +			r = -EFAULT;
> +			break;
> +		}
> +		r = kvm_s390_pci_zpci_op(kvm, &args);
> +		if (r)
> +			break;
> +		if (copy_to_user(argp, &args, sizeof(args)))
> +			r = -EFAULT;
> +		break;
> +	}
>   	default:
>   		r = -ENOTTY;
>   	}
> diff --git a/arch/s390/kvm/pci.c b/arch/s390/kvm/pci.c
> index 66565f5f3f43..d941657bcb39 100644
> --- a/arch/s390/kvm/pci.c
> +++ b/arch/s390/kvm/pci.c
> @@ -588,6 +588,71 @@ void kvm_s390_pci_clear_list(struct kvm *kvm)
>   		unregister_kvm(kzdev->zdev);
>   }
>   
> +static struct kvm_zdev *get_kzdev_by_fh(struct kvm *kvm, u32 fh)
> +{
> +	struct kvm_zdev *kzdev, *retval = NULL;
> +
> +	spin_lock(&kvm->arch.kzdev_list_lock);
> +	list_for_each_entry(kzdev, &kvm->arch.kzdev_list, entry) {
> +		if (kzdev->zdev->fh == fh) {
> +			retval = kzdev;
> +			break;
> +		}
> +	}
> +	spin_unlock(&kvm->arch.kzdev_list_lock);
> +
> +	return retval;
> +}
> +
> +static int kvm_s390_pci_zpci_reg_aen(struct zpci_dev *zdev,
> +				     struct kvm_s390_zpci_op *args)
> +{
> +	struct zpci_fib fib = {};
> +
> +	fib.fmt0.aibv = args->u.reg_aen.ibv;
> +	fib.fmt0.isc = args->u.reg_aen.isc;
> +	fib.fmt0.noi = args->u.reg_aen.noi;
> +	if (args->u.reg_aen.sb != 0) {
> +		fib.fmt0.aisb = args->u.reg_aen.sb;
> +		fib.fmt0.aisbo = args->u.reg_aen.sbo;
> +		fib.fmt0.sum = 1;
> +	} else {
> +		fib.fmt0.aisb = 0;
> +		fib.fmt0.aisbo = 0;
> +		fib.fmt0.sum = 0;
> +	}
> +
> +	if (args->u.reg_aen.flags & KVM_S390_ZPCIOP_REGAEN_HOST)
> +		return kvm_s390_pci_aif_enable(zdev, &fib, true);
> +	else
> +		return kvm_s390_pci_aif_enable(zdev, &fib, false);
> +}
> +
> +int kvm_s390_pci_zpci_op(struct kvm *kvm, struct kvm_s390_zpci_op *args)
> +{
> +	struct kvm_zdev *kzdev;
> +	struct zpci_dev *zdev;
> +	int r;
> +
> +	kzdev = get_kzdev_by_fh(kvm, args->fh);
> +	if (!kzdev)
> +		return -ENODEV;
> +	zdev = kzdev->zdev;
> +
> +	switch (args->op) {
> +	case KVM_S390_ZPCIOP_REG_AEN:
> +		r = kvm_s390_pci_zpci_reg_aen(zdev, args);
> +		break;
> +	case KVM_S390_ZPCIOP_DEREG_AEN:
> +		r = kvm_s390_pci_aif_disable(zdev, false);
> +		break;
> +	default:
> +		r = -EINVAL;
> +	}
> +
> +	return r;
> +}
> +
>   int kvm_s390_pci_init(void)
>   {
>   	aift = kzalloc(sizeof(struct zpci_aift), GFP_KERNEL);
> diff --git a/arch/s390/kvm/pci.h b/arch/s390/kvm/pci.h
> index cb5ec3208923..b053b50c0904 100644
> --- a/arch/s390/kvm/pci.h
> +++ b/arch/s390/kvm/pci.h
> @@ -59,6 +59,8 @@ void kvm_s390_pci_aen_exit(void);
>   void kvm_s390_pci_init_list(struct kvm *kvm);
>   void kvm_s390_pci_clear_list(struct kvm *kvm);
>   
> +int kvm_s390_pci_zpci_op(struct kvm *kvm, struct kvm_s390_zpci_op *args);
> +
>   int kvm_s390_pci_init(void);
>   
>   static inline bool kvm_s390_pci_interp_allowed(void)
> diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
> index 91a6fe4e02c0..014adb30d42c 100644
> --- a/include/uapi/linux/kvm.h
> +++ b/include/uapi/linux/kvm.h
> @@ -1144,6 +1144,7 @@ struct kvm_ppc_resize_hpt {
>   #define KVM_CAP_S390_MEM_OP_EXTENSION 211
>   #define KVM_CAP_PMU_CAPABILITY 212
>   #define KVM_CAP_DISABLE_QUIRKS2 213
> +#define KVM_CAP_S390_ZPCI_OP 214
>   
>   #ifdef KVM_CAP_IRQ_ROUTING
>   
> @@ -2060,4 +2061,34 @@ struct kvm_stats_desc {
>   /* Available with KVM_CAP_XSAVE2 */
>   #define KVM_GET_XSAVE2		  _IOR(KVMIO,  0xcf, struct kvm_xsave)
>   
> +/* Available with KVM_CAP_S390_ZPCI_OP */
> +#define KVM_S390_ZPCI_OP	  _IOW(KVMIO,  0xd0, struct kvm_s390_zpci_op)
> +
> +struct kvm_s390_zpci_op {
> +	/* in */
> +	__u32 fh;		/* target device */
> +	__u8  op;		/* operation to perform */
> +	__u8  pad[3];
> +	union {
> +		/* for KVM_S390_ZPCIOP_REG_AEN */
> +		struct {
> +			__u64 ibv;	/* Guest addr of interrupt bit vector */
> +			__u64 sb;	/* Guest addr of summary bit */
> +			__u32 flags;
> +			__u32 noi;	/* Number of interrupts */
> +			__u8 isc;	/* Guest interrupt subclass */
> +			__u8 sbo;	/* Offset of guest summary bit vector */
> +			__u16 pad;
> +		} reg_aen;
> +		__u64 reserved[8];
> +	} u;
> +};
> +
> +/* types for kvm_s390_zpci_op->op */
> +#define KVM_S390_ZPCIOP_REG_AEN		0
> +#define KVM_S390_ZPCIOP_DEREG_AEN	1
> +
> +/* flags for kvm_s390_zpci_op->u.reg_aen.flags */
> +#define KVM_S390_ZPCIOP_REGAEN_HOST	(1 << 0)
> +
>   #endif /* __LINUX_KVM_H */
> 


The purpose is to setup the IRQ for the zPCI device wouldn't it be more 
logical to go through VFIO_DEVICE_SET_IRQS and there jump to the zPCI 
handling, like it is done for the capabilities, instead to go directly 
through KVM?
We would also spare the research for the real FH in KVM.






-- 
Pierre Morel
IBM Lab Boeblingen