RDMA: Introduce and use GID attr helper to read RoCE L2 fields

Instead of RoCE drivers figuring out vlan, smac fields while working on
QP/AH, provide a helper routine to read the L2 fields such as vlan_id and
source mac address.

This moves logic from mlx5 driver to core for wider usage for RoCE ports.

This is a preparation patch to allow detaching netdev in subsequent patch.

Signed-off-by: Parav Pandit <[e-mail]>
Signed-off-by: Leon Romanovsky <[e-mail]>
Signed-off-by: Jason Gunthorpe <[e-mail]>

drivers/infiniband/core/cache.c

···
 	return ndev;
 }
 
+static int get_lower_dev_vlan(struct net_device *lower_dev, void *data)
+{
+	u16 *vlan_id = data;
+
+	if (is_vlan_dev(lower_dev))
+		*vlan_id = vlan_dev_vlan_id(lower_dev);
+
+	/* We are interested only in first level vlan device, so
+	 * always return 1 to stop iterating over next level devices.
+	 */
+	return 1;
+}
+
+/**
+ * rdma_read_gid_l2_fields - Read the vlan ID and source MAC address
+ *			     of a GID entry.
+ *
+ * @attr:	GID attribute pointer whose L2 fields to be read
+ * @vlan_id:	Pointer to vlan id to fill up if the GID entry has
+ *		vlan id. It is optional.
+ * @smac:	Pointer to smac to fill up for a GID entry. It is optional.
+ *
+ * rdma_read_gid_l2_fields() returns 0 on success and returns vlan id
+ * (if gid entry has vlan) and source MAC, or returns error.
+ */
+int rdma_read_gid_l2_fields(const struct ib_gid_attr *attr,
+			    u16 *vlan_id, u8 *smac)
+{
+	struct net_device *ndev;
+
+	ndev = attr->ndev;
+	if (!ndev)
+		return -EINVAL;
+
+	if (smac)
+		ether_addr_copy(smac, ndev->dev_addr);
+	if (vlan_id) {
+		*vlan_id = 0xffff;
+		if (is_vlan_dev(ndev)) {
+			*vlan_id = vlan_dev_vlan_id(ndev);
+		} else {
+			/* If the netdev is upper device and if it's lower
+			 * device is vlan device, consider vlan id of the
+			 * the lower vlan device for this gid entry.
+			 */
+			rcu_read_lock();
+			netdev_walk_all_lower_dev_rcu(attr->ndev,
+					get_lower_dev_vlan, vlan_id);
+			rcu_read_unlock();
+		}
+	}
+	return 0;
+}
+EXPORT_SYMBOL(rdma_read_gid_l2_fields);
+
 static int config_non_roce_gid_cache(struct ib_device *device,
 				     u8 port, int gid_tbl_len)
 {

drivers/infiniband/hw/bnxt_re/ib_verbs.c

···
 	struct bnxt_re_dev *rdev = to_bnxt_re_dev(attr->device, ibdev);
 	struct bnxt_qplib_sgid_tbl *sgid_tbl = &rdev->qplib_res.sgid_tbl;
 
-	if ((attr->ndev) && is_vlan_dev(attr->ndev))
-		vlan_id = vlan_dev_vlan_id(attr->ndev);
+	rc = rdma_read_gid_l2_fields(attr, &vlan_id, NULL);
+	if (rc)
+		return rc;
 
 	rc = bnxt_qplib_add_sgid(sgid_tbl, (struct bnxt_qplib_gid *)&attr->gid,
 				 rdev->qplib_res.netdev->dev_addr,
···
 				qp_attr->ah_attr.roce.dmac);
 
 		sgid_attr = qp_attr->ah_attr.grh.sgid_attr;
-		memcpy(qp->qplib_qp.smac, sgid_attr->ndev->dev_addr,
-		       ETH_ALEN);
+		rc = rdma_read_gid_l2_fields(sgid_attr, NULL,
+					     &qp->qplib_qp.smac[0]);
+		if (rc)
+			return rc;
+
 		nw_type = rdma_gid_attr_network_type(sgid_attr);
 		switch (nw_type) {
 		case RDMA_NETWORK_IPV4:
···
 
 	memset(&qp->qp1_hdr, 0, sizeof(qp->qp1_hdr));
 
-	if (is_vlan_dev(sgid_attr->ndev))
-		vlan_id = vlan_dev_vlan_id(sgid_attr->ndev);
+	rc = rdma_read_gid_l2_fields(sgid_attr, &vlan_id, NULL);
+	if (rc)
+		return rc;
+
 	/* Get network header type for this GID */
 	nw_type = rdma_gid_attr_network_type(sgid_attr);
 	switch (nw_type) {

drivers/infiniband/hw/hns/hns_roce_ah.c

···
 	u16 vlan_tag = 0xffff;
 	const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr);
 	bool vlan_en = false;
+	int ret;
+
+	gid_attr = ah_attr->grh.sgid_attr;
+	ret = rdma_read_gid_l2_fields(gid_attr, &vlan_tag, NULL);
+	if (ret)
+		return ret;
 
 	/* Get mac address */
 	memcpy(ah->av.mac, ah_attr->roce.dmac, ETH_ALEN);
 
-	gid_attr = ah_attr->grh.sgid_attr;
-	if (is_vlan_dev(gid_attr->ndev)) {
-		vlan_tag = vlan_dev_vlan_id(gid_attr->ndev);
+	if (vlan_tag < VLAN_CFI_MASK) {
 		vlan_en = true;
-	}
-
-	if (vlan_tag < 0x1000)
 		vlan_tag |= (rdma_ah_get_sl(ah_attr) &
 			     HNS_ROCE_VLAN_SL_BIT_MASK) <<
 			     HNS_ROCE_VLAN_SL_SHIFT;
+	}
 
 	ah->av.port_pd = cpu_to_le32(to_hr_pd(ibah->pd)->pdn |
 				     (rdma_ah_get_port_num(ah_attr) <<

drivers/infiniband/hw/hns/hns_roce_hw_v2.c

···
 #include <linux/types.h>
 #include <net/addrconf.h>
 #include <rdma/ib_addr.h>
+#include <rdma/ib_cache.h>
 #include <rdma/ib_umem.h>
 #include <rdma/uverbs_ioctl.h>
 
···
 
 		if (is_roce_protocol) {
 			gid_attr = attr->ah_attr.grh.sgid_attr;
-			vlan = rdma_vlan_dev_vlan_id(gid_attr->ndev);
+			ret = rdma_read_gid_l2_fields(gid_attr, &vlan, NULL);
+			if (ret)
+				goto out;
 		}
 
-		if (is_vlan_dev(gid_attr->ndev)) {
+		if (vlan < VLAN_CFI_MASK) {
 			roce_set_bit(context->byte_76_srqn_op_en,
 				     V2_QPC_BYTE_76_RQ_VLAN_EN_S, 1);
 			roce_set_bit(qpc_mask->byte_76_srqn_op_en,

drivers/infiniband/hw/mlx4/ah.c

···
 	 */
 	gid_attr = ah_attr->grh.sgid_attr;
 	if (gid_attr) {
-		if (is_vlan_dev(gid_attr->ndev))
-			vlan_tag = vlan_dev_vlan_id(gid_attr->ndev);
-		memcpy(ah->av.eth.s_mac, gid_attr->ndev->dev_addr, ETH_ALEN);
+		ret = rdma_read_gid_l2_fields(gid_attr, &vlan_tag,
+					      &ah->av.eth.s_mac[0]);
+		if (ret)
+			return ret;
+
 		ret = mlx4_ib_gid_index_to_real_index(ibdev, gid_attr);
 		if (ret < 0)
 			return ret;

drivers/infiniband/hw/mlx4/qp.c

···
 
 		if (is_eth) {
 			gid_attr = attr->ah_attr.grh.sgid_attr;
-			vlan = rdma_vlan_dev_vlan_id(gid_attr->ndev);
-			memcpy(smac, gid_attr->ndev->dev_addr, ETH_ALEN);
+			err = rdma_read_gid_l2_fields(gid_attr, &vlan,
+						      &smac[0]);
+			if (err)
+				goto out;
 		}
 
 		if (mlx4_set_path(dev, attr, attr_mask, qp, &context->pri_path,

drivers/infiniband/hw/mlx5/main.c

···
 	return err;
 }
 
-struct mlx5_ib_vlan_info {
-	u16 vlan_id;
-	bool vlan;
-};
-
-static int get_lower_dev_vlan(struct net_device *lower_dev, void *data)
-{
-	struct mlx5_ib_vlan_info *vlan_info = data;
-
-	if (is_vlan_dev(lower_dev)) {
-		vlan_info->vlan = true;
-		vlan_info->vlan_id = vlan_dev_vlan_id(lower_dev);
-	}
-	/* We are interested only in first level vlan device, so
-	 * always return 1 to stop iterating over next level devices.
-	 */
-	return 1;
-}
-
 static int set_roce_addr(struct mlx5_ib_dev *dev, u8 port_num,
 			 unsigned int index, const union ib_gid *gid,
 			 const struct ib_gid_attr *attr)
 {
 	enum ib_gid_type gid_type = IB_GID_TYPE_IB;
-	struct mlx5_ib_vlan_info vlan_info = { };
+	u16 vlan_id = 0xffff;
 	u8 roce_version = 0;
 	u8 roce_l3_type = 0;
 	u8 mac[ETH_ALEN];
+	int ret;
 
 	if (gid) {
 		gid_type = attr->gid_type;
-		ether_addr_copy(mac, attr->ndev->dev_addr);
-
-		if (is_vlan_dev(attr->ndev)) {
-			vlan_info.vlan = true;
-			vlan_info.vlan_id = vlan_dev_vlan_id(attr->ndev);
-		} else {
-			/* If the netdev is upper device and if it's lower
-			 * lower device is vlan device, consider vlan id of
-			 * the lower vlan device for this gid entry.
-			 */
-			rcu_read_lock();
-			netdev_walk_all_lower_dev_rcu(attr->ndev,
-					get_lower_dev_vlan, &vlan_info);
-			rcu_read_unlock();
-		}
+		ret = rdma_read_gid_l2_fields(attr, &vlan_id, &mac[0]);
+		if (ret)
+			return ret;
 	}
 
 	switch (gid_type) {
···
 
 	return mlx5_core_roce_gid_set(dev->mdev, index, roce_version,
 				      roce_l3_type, gid->raw, mac,
-				      vlan_info.vlan, vlan_info.vlan_id,
+				      vlan_id < VLAN_CFI_MASK, vlan_id,
 				      port_num);
 }
 

drivers/infiniband/hw/ocrdma/ocrdma_ah.c

···
 	if (atomic_cmpxchg(&dev->update_sl, 1, 0))
 		ocrdma_init_service_level(dev);
 
+	sgid_attr = attr->grh.sgid_attr;
+	status = rdma_read_gid_l2_fields(sgid_attr, &vlan_tag, NULL);
+	if (status)
+		return status;
+
 	status = ocrdma_alloc_av(dev, ah);
 	if (status)
 		goto av_err;
-
-	sgid_attr = attr->grh.sgid_attr;
-	if (is_vlan_dev(sgid_attr->ndev))
-		vlan_tag = vlan_dev_vlan_id(sgid_attr->ndev);
 
 	/* Get network header type for this GID */
 	ah->hdr_type = rdma_gid_attr_network_type(sgid_attr);

drivers/infiniband/hw/ocrdma/ocrdma_hw.c

···
 	int status;
 	struct rdma_ah_attr *ah_attr = &attrs->ah_attr;
 	const struct ib_gid_attr *sgid_attr;
-	u32 vlan_id = 0xFFFF;
+	u16 vlan_id = 0xFFFF;
 	u8 mac_addr[6], hdr_type;
 	union {
 		struct sockaddr     _sockaddr;
···
 	       sizeof(cmd->params.dgid));
 
 	sgid_attr = ah_attr->grh.sgid_attr;
-	vlan_id = rdma_vlan_dev_vlan_id(sgid_attr->ndev);
-	memcpy(mac_addr, sgid_attr->ndev->dev_addr, ETH_ALEN);
+	status = rdma_read_gid_l2_fields(sgid_attr, &vlan_id, &mac_addr[0]);
+	if (status)
+		return status;
 
 	qp->sgid_idx = grh->sgid_index;
 	memcpy(&cmd->params.sgid[0], &sgid_attr->gid.raw[0],

drivers/infiniband/hw/qedr/qedr_roce_cm.c

···
 	bool has_udp = false;
 	int i;
 
+	rc = rdma_read_gid_l2_fields(sgid_attr, &vlan_id, NULL);
+	if (rc)
+		return rc;
+
+	if (vlan_id < VLAN_CFI_MASK)
+		has_vlan = true;
+
 	send_size = 0;
 	for (i = 0; i < swr->num_sge; ++i)
 		send_size += swr->sg_list[i].length;
-
-	vlan_id = rdma_vlan_dev_vlan_id(sgid_attr->ndev);
-	if (vlan_id < VLAN_CFI_MASK)
-		has_vlan = true;
 
 	has_udp = (sgid_attr->gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP);
 	if (!has_udp) {

drivers/infiniband/hw/qedr/verbs.c

···
 	enum rdma_network_type nw_type;
 	const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr);
 	u32 ipv4_addr;
+	int ret;
 	int i;
 
 	gid_attr = grh->sgid_attr;
-	qp_params->vlan_id = rdma_vlan_dev_vlan_id(gid_attr->ndev);
+	ret = rdma_read_gid_l2_fields(gid_attr, &qp_params->vlan_id, NULL);
+	if (ret)
+		return ret;
 
 	nw_type = rdma_gid_attr_network_type(gid_attr);
 	switch (nw_type) {

include/rdma/ib_cache.h

···
 		       void *),
 	void *context);
 
+int rdma_read_gid_l2_fields(const struct ib_gid_attr *attr,
+			    u16 *vlan_id, u8 *smac);
+
 /**
  * ib_get_cached_pkey - Returns a cached PKey table entry
  * @device: The device to query.