vdpa/mlx5: support device features provisioning

This patch implements features provisioning for mlx5_vdpa.

1) Validate the provisioned features are a subset of the parent
    features.
2) Clearing features that are not wanted by userspace.

For example:

    # vdpa mgmtdev show
    pci/0000:41:04.2:
      supported_classes net
      max_supported_vqs 65
      dev_features CSUM GUEST_CSUM MTU MAC HOST_TSO4 HOST_TSO6 STATUS CTRL_VQ CTRL_VLAN MQ CTRL_MAC_ADDR VERSION_1 ACCESS_PLATFORM

1) Provision vDPA device with all features derived from the parent

    # vdpa dev add name vdpa1 mgmtdev pci/0000:41:04.2
    # vdpa dev config show
    vdpa1: mac e4:11:c6:d3:45:f0 link up link_announce false max_vq_pairs 1 mtu 1500
      negotiated_features CSUM GUEST_CSUM MTU HOST_TSO4 HOST_TSO6 STATUS CTRL_VQ CTRL_VLAN MQ CTRL_MAC_ADDR VERSION_1 ACCESS_PLATFORM

2) Provision vDPA device with a subset of parent features

    # vdpa dev add name vdpa1 mgmtdev pci/0000:41:04.2 device_features 0x300020000
    # vdpa dev config show
    vdpa1:
      negotiated_features CTRL_VQ VERSION_1 ACCESS_PLATFORM

Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
Reviewed-by: Eli Cohen <elic@nvidia.com>
Message-Id: <1675725124-7375-7-git-send-email-si-wei.liu@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

Author: siwliu-kernel

drivers/vdpa/mlx5/net/mlx5_vnet.c

@@ -2209,6 +2209,7 @@ static u64 get_supported_features(struct mlx5_core_dev *mdev)
 	mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_STATUS);
 	mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_MTU);
 	mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_CTRL_VLAN);
+	mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_MAC);
 
 	return mlx_vdpa_features;
 }
@@ -3099,6 +3100,7 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
 	struct mlx5_vdpa_dev *mvdev;
 	struct mlx5_vdpa_net *ndev;
 	struct mlx5_core_dev *mdev;
+	u64 device_features;
 	u32 max_vqs;
 	u16 mtu;
 	int err;
@@ -3107,6 +3109,24 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
 		return -ENOSPC;
 
 	mdev = mgtdev->madev->mdev;
+	device_features = mgtdev->mgtdev.supported_features;
+	if (add_config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) {
+		if (add_config->device_features & ~device_features) {
+			dev_warn(mdev->device,
+				 "The provisioned features 0x%llx are not supported by this device with features 0x%llx\n",
+				 add_config->device_features, device_features);
+			return -EINVAL;
+		}
+		device_features &= add_config->device_features;
+	}
+	if (!(device_features & BIT_ULL(VIRTIO_F_VERSION_1) &&
+	      device_features & BIT_ULL(VIRTIO_F_ACCESS_PLATFORM))) {
+		dev_warn(mdev->device,
+			 "Must provision minimum features 0x%llx for this device",
+			 BIT_ULL(VIRTIO_F_VERSION_1) | BIT_ULL(VIRTIO_F_ACCESS_PLATFORM));
+		return -EOPNOTSUPP;
+	}
+
 	if (!(MLX5_CAP_DEV_VDPA_EMULATION(mdev, virtio_queue_type) &
 	    MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT)) {
 		dev_warn(mdev->device, "missing support for split virtqueues\n");
@@ -3135,7 +3155,6 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
 	if (IS_ERR(ndev))
 		return PTR_ERR(ndev);
 
-	ndev->mvdev.mlx_features = mgtdev->mgtdev.supported_features;
 	ndev->mvdev.max_vqs = max_vqs;
 	mvdev = &ndev->mvdev;
 	mvdev->mdev = mdev;
@@ -3157,15 +3176,15 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
 			goto err_alloc;
 	}
 
-	if (ndev->mvdev.mlx_features & BIT_ULL(VIRTIO_NET_F_MTU)) {
+	if (device_features & BIT_ULL(VIRTIO_NET_F_MTU)) {
 		err = query_mtu(mdev, &mtu);
 		if (err)
 			goto err_alloc;
 
 		ndev->config.mtu = cpu_to_mlx5vdpa16(mvdev, mtu);
 	}
 
-	if (ndev->mvdev.mlx_features & BIT_ULL(VIRTIO_NET_F_STATUS)) {
+	if (device_features & BIT_ULL(VIRTIO_NET_F_STATUS)) {
 		if (get_link_state(mvdev))
 			ndev->config.status |= cpu_to_mlx5vdpa16(mvdev, VIRTIO_NET_S_LINK_UP);
 		else
@@ -3174,7 +3193,9 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
 
 	if (add_config->mask & (1 << VDPA_ATTR_DEV_NET_CFG_MACADDR)) {
 		memcpy(ndev->config.mac, add_config->net.mac, ETH_ALEN);
-	} else {
+	/* No bother setting mac address in config if not going to provision _F_MAC */
+	} else if ((add_config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) == 0 ||
+		   device_features & BIT_ULL(VIRTIO_NET_F_MAC)) {
 		err = mlx5_query_nic_vport_mac_address(mdev, 0, 0, config->mac);
 		if (err)
 			goto err_alloc;
@@ -3185,11 +3206,26 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
 		err = mlx5_mpfs_add_mac(pfmdev, config->mac);
 		if (err)
 			goto err_alloc;
-
-		ndev->mvdev.mlx_features |= BIT_ULL(VIRTIO_NET_F_MAC);
+	} else if ((add_config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) == 0) {
+		/*
+		 * We used to clear _F_MAC feature bit if seeing
+		 * zero mac address when device features are not
+		 * specifically provisioned. Keep the behaviour
+		 * so old scripts do not break.
+		 */
+		device_features &= ~BIT_ULL(VIRTIO_NET_F_MAC);
+	} else if (device_features & BIT_ULL(VIRTIO_NET_F_MAC)) {
+		/* Don't provision zero mac address for _F_MAC */
+		mlx5_vdpa_warn(&ndev->mvdev,
+			       "No mac address provisioned?\n");
+		err = -EINVAL;
+		goto err_alloc;
 	}
 
-	config->max_virtqueue_pairs = cpu_to_mlx5vdpa16(mvdev, max_vqs / 2);
+	if (device_features & BIT_ULL(VIRTIO_NET_F_MQ))
+		config->max_virtqueue_pairs = cpu_to_mlx5vdpa16(mvdev, max_vqs / 2);
+
+	ndev->mvdev.mlx_features = device_features;
 	mvdev->vdev.dma_dev = &mdev->pdev->dev;
 	err = mlx5_vdpa_alloc_resources(&ndev->mvdev);
 	if (err)
@@ -3289,7 +3325,8 @@ static int mlx5v_probe(struct auxiliary_device *adev,
 	mgtdev->mgtdev.id_table = id_table;
 	mgtdev->mgtdev.config_attr_mask = BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MACADDR) |
 					  BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MAX_VQP) |
-					  BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MTU);
+					  BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MTU) |
+					  BIT_ULL(VDPA_ATTR_DEV_FEATURES);
 	mgtdev->mgtdev.max_supported_vqs =
 		MLX5_CAP_DEV_VDPA_EMULATION(mdev, max_num_virtio_queues) + 1;
 	mgtdev->mgtdev.supported_features = get_supported_features(mdev);