IB/mlx4: Add VLAN support for IBoE

This patch allows IBoE traffic to be encapsulated in 802.1Q tagged
VLAN frames.  The VLAN tag is encoded in the GID and derived from it
by a simple computation.

The netdev notifier callback is modified to catch VLAN device
addition/removal and the port's GID table is updated to reflect the
change, so that for each netdevice there is an entry in the GID table.
When the port's GID table is exhausted, GID entries will not be added.
Only children of the main interfaces can add to the GID table; if a
VLAN interface is added on another VLAN interface (e.g. "vconfig add
eth2.6 8"), then that interfaces will not add an entry to the GID
table.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

6 files changed, 191 insertions, 27 deletions

diff --git a/drivers/infiniband/hw/mlx4/ah.c b/drivers/infiniband/hw/mlx4/ah.c
index 3bf3544..4b8f9c4 100644
--- a/drivers/infiniband/hw/mlx4/ah.c
+++ b/drivers/infiniband/hw/mlx4/ah.c
@@ -31,6 +31,7 @@
  */
 
 #include <rdma/ib_addr.h>
+#include <rdma/ib_cache.h>
 
 #include <linux/slab.h>
 #include <linux/inet.h>
@@ -91,17 +92,26 @@ static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr
 {
 	struct mlx4_ib_dev *ibdev = to_mdev(pd->device);
 	struct mlx4_dev *dev = ibdev->dev;
+	union ib_gid sgid;
 	u8 mac[6];
 	int err;
 	int is_mcast;
+	u16 vlan_tag;
 
 	err = mlx4_ib_resolve_grh(ibdev, ah_attr, mac, &is_mcast, ah_attr->port_num);
 	if (err)
 		return ERR_PTR(err);
 
 	memcpy(ah->av.eth.mac, mac, 6);
+	err = ib_get_cached_gid(pd->device, ah_attr->port_num, ah_attr->grh.sgid_index, &sgid);
+	if (err)
+		return ERR_PTR(err);
+	vlan_tag = rdma_get_vlan_id(&sgid);
+	if (vlan_tag < 0x1000)
+		vlan_tag |= (ah_attr->sl & 7) << 13;
 	ah->av.eth.port_pd = cpu_to_be32(to_mpd(pd)->pdn | (ah_attr->port_num << 24));
 	ah->av.eth.gid_index = ah_attr->grh.sgid_index;
+	ah->av.eth.vlan = cpu_to_be16(vlan_tag);
 	if (ah_attr->static_rate) {
 		ah->av.eth.stat_rate = ah_attr->static_rate + MLX4_STAT_RATE_OFFSET;
 		while (ah->av.eth.stat_rate > IB_RATE_2_5_GBPS + MLX4_STAT_RATE_OFFSET &&
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index e65db73..8736bd8 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -38,6 +38,7 @@
 #include <linux/netdevice.h>
 #include <linux/inetdevice.h>
 #include <linux/rtnetlink.h>
+#include <linux/if_vlan.h>
 
 #include <rdma/ib_smi.h>
 #include <rdma/ib_user_verbs.h>
@@ -79,6 +80,8 @@ static void init_query_mad(struct ib_smp *mad)
 	mad->method	   = IB_MGMT_METHOD_GET;
 }
 
+static union ib_gid zgid;
+
 static int mlx4_ib_query_device(struct ib_device *ibdev,
 				struct ib_device_attr *props)
 {
@@ -755,12 +758,17 @@ static struct device_attribute *mlx4_class_attributes[] = {
 	&dev_attr_board_id
 };
 
-static void mlx4_addrconf_ifid_eui48(u8 *eui, struct net_device *dev)
+static void mlx4_addrconf_ifid_eui48(u8 *eui, u16 vlan_id, struct net_device *dev)
 {
 	memcpy(eui, dev->dev_addr, 3);
 	memcpy(eui + 5, dev->dev_addr + 3, 3);
-	eui[3] = 0xFF;
-	eui[4] = 0xFE;
+	if (vlan_id < 0x1000) {
+		eui[3] = vlan_id >> 8;
+		eui[4] = vlan_id & 0xff;
+	} else {
+		eui[3] = 0xff;
+		eui[4] = 0xfe;
+	}
 	eui[0] ^= 2;
 }
 
@@ -802,28 +810,93 @@ static int update_ipv6_gids(struct mlx4_ib_dev *dev, int port, int clear)
 {
 	struct net_device *ndev = dev->iboe.netdevs[port - 1];
 	struct update_gid_work *work;
+	struct net_device *tmp;
+	int i;
+	u8 *hits;
+	int ret;
+	union ib_gid gid;
+	int free;
+	int found;
+	int need_update = 0;
+	u16 vid;
 
 	work = kzalloc(sizeof *work, GFP_ATOMIC);
 	if (!work)
 		return -ENOMEM;
 
-	if (!clear) {
-		mlx4_addrconf_ifid_eui48(&work->gids[0].raw[8], ndev);
-		work->gids[0].global.subnet_prefix = cpu_to_be64(0xfe80000000000000LL);
+	hits = kzalloc(128, GFP_ATOMIC);
+	if (!hits) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	read_lock(&dev_base_lock);
+	for_each_netdev(&init_net, tmp) {
+		if (ndev && (tmp == ndev || rdma_vlan_dev_real_dev(tmp) == ndev)) {
+			gid.global.subnet_prefix = cpu_to_be64(0xfe80000000000000LL);
+			vid = rdma_vlan_dev_vlan_id(tmp);
+			mlx4_addrconf_ifid_eui48(&gid.raw[8], vid, ndev);
+			found = 0;
+			free = -1;
+			for (i = 0; i < 128; ++i) {
+				if (free < 0 &&
+				    !memcmp(&dev->iboe.gid_table[port - 1][i], &zgid, sizeof zgid))
+					free = i;
+				if (!memcmp(&dev->iboe.gid_table[port - 1][i], &gid, sizeof gid)) {
+					hits[i] = 1;
+					found = 1;
+					break;
+				}
+			}
+
+			if (!found) {
+				if (tmp == ndev &&
+				    (memcmp(&dev->iboe.gid_table[port - 1][0],
+					    &gid, sizeof gid) ||
+				     !memcmp(&dev->iboe.gid_table[port - 1][0],
+					     &zgid, sizeof gid))) {
+					dev->iboe.gid_table[port - 1][0] = gid;
+					++need_update;
+					hits[0] = 1;
+				} else if (free >= 0) {
+					dev->iboe.gid_table[port - 1][free] = gid;
+					hits[free] = 1;
+					++need_update;
+				}
+			}
+		}
 	}
+	read_unlock(&dev_base_lock);
+
+	for (i = 0; i < 128; ++i)
+		if (!hits[i]) {
+			if (memcmp(&dev->iboe.gid_table[port - 1][i], &zgid, sizeof zgid))
+				++need_update;
+			dev->iboe.gid_table[port - 1][i] = zgid;
+		}
 
-	INIT_WORK(&work->work, update_gids_task);
-	work->port = port;
-	work->dev = dev;
-	queue_work(wq, &work->work);
+	if (need_update) {
+		memcpy(work->gids, dev->iboe.gid_table[port - 1], sizeof work->gids);
+		INIT_WORK(&work->work, update_gids_task);
+		work->port = port;
+		work->dev = dev;
+		queue_work(wq, &work->work);
+	} else
+		kfree(work);
 
+	kfree(hits);
 	return 0;
+
+out:
+	kfree(work);
+	return ret;
 }
 
 static void handle_en_event(struct mlx4_ib_dev *dev, int port, unsigned long event)
 {
 	switch (event) {
 	case NETDEV_UP:
+	case NETDEV_CHANGEADDR:
 		update_ipv6_gids(dev, port, 0);
 		break;
 
@@ -871,9 +944,11 @@ static int mlx4_ib_netdev_event(struct notifier_block *this, unsigned long event
 		}
 	}
 
-	if (dev == iboe->netdevs[0])
+	if (dev == iboe->netdevs[0] ||
+	    (iboe->netdevs[0] && rdma_vlan_dev_real_dev(dev) == iboe->netdevs[0]))
 		handle_en_event(ibdev, 1, event);
-	else if (dev == iboe->netdevs[1])
+	else if (dev == iboe->netdevs[1]
+		 || (iboe->netdevs[1] && rdma_vlan_dev_real_dev(dev) == iboe->netdevs[1]))
 		handle_en_event(ibdev, 2, event);
 
 	spin_unlock(&iboe->lock);
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 2696484..9a7794a 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -37,6 +37,7 @@
 
 #include <rdma/ib_cache.h>
 #include <rdma/ib_pack.h>
+#include <rdma/ib_addr.h>
 
 #include <linux/mlx4/qp.h>
 
@@ -57,10 +58,11 @@ enum {
 enum {
 	/*
 	 * Largest possible UD header: send with GRH and immediate
-	 * data plus 14 bytes for an Ethernet header.  (LRH would only
-	 * use 8 bytes, so Ethernet is the biggest case)
+	 * data plus 18 bytes for an Ethernet header with VLAN/802.1Q
+	 * tag.  (LRH would only use 8 bytes, so Ethernet is the
+	 * biggest case)
 	 */
-	MLX4_IB_UD_HEADER_SIZE		= 78,
+	MLX4_IB_UD_HEADER_SIZE		= 82,
 	MLX4_IB_LSO_HEADER_SPARE	= 128,
 };
 
@@ -879,6 +881,8 @@ static int mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah,
 		IB_LINK_LAYER_ETHERNET;
 	u8 mac[6];
 	int is_mcast;
+	u16 vlan_tag;
+	int vidx;
 
 	path->grh_mylmc     = ah->src_path_bits & 0x7f;
 	path->rlid	    = cpu_to_be16(ah->dlid);
@@ -907,10 +911,10 @@ static int mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah,
 		memcpy(path->rgid, ah->grh.dgid.raw, 16);
 	}
 
-	path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE |
-		((port - 1) << 6) | ((ah->sl & 0xf) << 2);
-
 	if (is_eth) {
+		path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE |
+			((port - 1) << 6) | ((ah->sl & 7) << 3) | ((ah->sl & 8) >> 1);
+
 		if (!(ah->ah_flags & IB_AH_GRH))
 			return -1;
 
@@ -922,7 +926,18 @@ static int mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah,
 		path->ackto = MLX4_IB_LINK_TYPE_ETH;
 		/* use index 0 into MAC table for IBoE */
 		path->grh_mylmc &= 0x80;
-	}
+
+		vlan_tag = rdma_get_vlan_id(&dev->iboe.gid_table[port - 1][ah->grh.sgid_index]);
+		if (vlan_tag < 0x1000) {
+			if (mlx4_find_cached_vlan(dev->dev, port, vlan_tag, &vidx))
+				return -ENOENT;
+
+			path->vlan_index = vidx;
+			path->fl = 1 << 6;
+		}
+	} else
+		path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE |
+			((port - 1) << 6) | ((ah->sl & 0xf) << 2);
 
 	return 0;
 }
@@ -1277,13 +1292,16 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_send_wr *wr,
 	struct mlx4_wqe_mlx_seg *mlx = wqe;
 	struct mlx4_wqe_inline_seg *inl = wqe + sizeof *mlx;
 	struct mlx4_ib_ah *ah = to_mah(wr->wr.ud.ah);
+	union ib_gid sgid;
 	u16 pkey;
 	int send_size;
 	int header_size;
 	int spc;
 	int i;
 	int is_eth;
+	int is_vlan = 0;
 	int is_grh;
+	u16 vlan;
 
 	send_size = 0;
 	for (i = 0; i < wr->num_sge; ++i)
@@ -1291,7 +1309,13 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_send_wr *wr,
 
 	is_eth = rdma_port_get_link_layer(sqp->qp.ibqp.device, sqp->qp.port) == IB_LINK_LAYER_ETHERNET;
 	is_grh = mlx4_ib_ah_grh_present(ah);
-	ib_ud_header_init(send_size, !is_eth, is_eth, 0, is_grh, 0, &sqp->ud_header);
+	if (is_eth) {
+		ib_get_cached_gid(ib_dev, be32_to_cpu(ah->av.ib.port_pd) >> 24,
+				  ah->av.ib.gid_index, &sgid);
+		vlan = rdma_get_vlan_id(&sgid);
+		is_vlan = vlan < 0x1000;
+	}
+	ib_ud_header_init(send_size, !is_eth, is_eth, is_vlan, is_grh, 0, &sqp->ud_header);
 
 	if (!is_eth) {
 		sqp->ud_header.lrh.service_level =
@@ -1345,7 +1369,15 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_send_wr *wr,
 		memcpy(sqp->ud_header.eth.smac_h, smac, 6);
 		if (!memcmp(sqp->ud_header.eth.smac_h, sqp->ud_header.eth.dmac_h, 6))
 			mlx->flags |= cpu_to_be32(MLX4_WQE_CTRL_FORCE_LOOPBACK);
-		sqp->ud_header.eth.type = cpu_to_be16(MLX4_IB_IBOE_ETHERTYPE);
+		if (!is_vlan) {
+			sqp->ud_header.eth.type = cpu_to_be16(MLX4_IB_IBOE_ETHERTYPE);
+		} else {
+			u16 pcp;
+
+			sqp->ud_header.vlan.type = cpu_to_be16(MLX4_IB_IBOE_ETHERTYPE);
+			pcp = (be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 27 & 3) << 13;
+			sqp->ud_header.vlan.tag = cpu_to_be16(vlan | pcp);
+		}
 	} else {
 		sqp->ud_header.lrh.virtual_lane    = !sqp->qp.ibqp.qp_num ? 15 : 0;
 		if (sqp->ud_header.lrh.destination_lid == IB_LID_PERMISSIVE)
@@ -1507,13 +1539,14 @@ static void set_masked_atomic_seg(struct mlx4_wqe_masked_atomic_seg *aseg,
 }
 
 static void set_datagram_seg(struct mlx4_wqe_datagram_seg *dseg,
-			     struct ib_send_wr *wr)
+			     struct ib_send_wr *wr, __be16 *vlan)
 {
 	memcpy(dseg->av, &to_mah(wr->wr.ud.ah)->av, sizeof (struct mlx4_av));
 	dseg->dqpn = cpu_to_be32(wr->wr.ud.remote_qpn);
 	dseg->qkey = cpu_to_be32(wr->wr.ud.remote_qkey);
 	dseg->vlan = to_mah(wr->wr.ud.ah)->av.eth.vlan;
 	memcpy(dseg->mac, to_mah(wr->wr.ud.ah)->av.eth.mac, 6);
+	*vlan = dseg->vlan;
 }
 
 static void set_mlx_icrc_seg(void *dseg)
@@ -1616,6 +1649,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 	__be32 uninitialized_var(lso_hdr_sz);
 	__be32 blh;
 	int i;
+	__be16 vlan = cpu_to_be16(0xffff);
 
 	spin_lock_irqsave(&qp->sq.lock, flags);
 
@@ -1719,7 +1753,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			break;
 
 		case IB_QPT_UD:
-			set_datagram_seg(wqe, wr);
+			set_datagram_seg(wqe, wr, &vlan);
 			wqe  += sizeof (struct mlx4_wqe_datagram_seg);
 			size += sizeof (struct mlx4_wqe_datagram_seg) / 16;
 
@@ -1797,6 +1831,11 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 		ctrl->owner_opcode = mlx4_ib_opcode[wr->opcode] |
 			(ind & qp->sq.wqe_cnt ? cpu_to_be32(1 << 31) : 0) | blh;
 
+		if (be16_to_cpu(vlan) < 0x1000) {
+			ctrl->ins_vlan = 1 << 6;
+			ctrl->vlan_tag = vlan;
+		}
+
 		stamp = ind + qp->sq_spare_wqes;
 		ind += DIV_ROUND_UP(size * 16, 1U << qp->sq.wqe_shift);
 
@@ -1946,17 +1985,27 @@ static int to_ib_qp_access_flags(int mlx4_flags)
 	return ib_flags;
 }
 
-static void to_ib_ah_attr(struct mlx4_dev *dev, struct ib_ah_attr *ib_ah_attr,
+static void to_ib_ah_attr(struct mlx4_ib_dev *ibdev, struct ib_ah_attr *ib_ah_attr,
 				struct mlx4_qp_path *path)
 {
+	struct mlx4_dev *dev = ibdev->dev;
+	int is_eth;
+
 	memset(ib_ah_attr, 0, sizeof *ib_ah_attr);
 	ib_ah_attr->port_num	  = path->sched_queue & 0x40 ? 2 : 1;
 
 	if (ib_ah_attr->port_num == 0 || ib_ah_attr->port_num > dev->caps.num_ports)
 		return;
 
+	is_eth = rdma_port_get_link_layer(&ibdev->ib_dev, ib_ah_attr->port_num) ==
+		IB_LINK_LAYER_ETHERNET;
+	if (is_eth)
+		ib_ah_attr->sl = ((path->sched_queue >> 3) & 0x7) |
+		((path->sched_queue & 4) << 1);
+	else
+		ib_ah_attr->sl = (path->sched_queue >> 2) & 0xf;
+
 	ib_ah_attr->dlid	  = be16_to_cpu(path->rlid);
-	ib_ah_attr->sl		  = (path->sched_queue >> 2) & 0xf;
 	ib_ah_attr->src_path_bits = path->grh_mylmc & 0x7f;
 	ib_ah_attr->static_rate   = path->static_rate ? path->static_rate - 5 : 0;
 	ib_ah_attr->ah_flags      = (path->grh_mylmc & (1 << 7)) ? IB_AH_GRH : 0;
@@ -2009,8 +2058,8 @@ int mlx4_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr
 		to_ib_qp_access_flags(be32_to_cpu(context.params2));
 
 	if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) {
-		to_ib_ah_attr(dev->dev, &qp_attr->ah_attr, &context.pri_path);
-		to_ib_ah_attr(dev->dev, &qp_attr->alt_ah_attr, &context.alt_path);
+		to_ib_ah_attr(dev, &qp_attr->ah_attr, &context.pri_path);
+		to_ib_ah_attr(dev, &qp_attr->alt_ah_attr, &context.alt_path);
 		qp_attr->alt_pkey_index = context.alt_path.pkey_index & 0x7f;
 		qp_attr->alt_port_num	= qp_attr->alt_ah_attr.port_num;
 	}
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index a0d8a26..9a87c4f 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -69,6 +69,7 @@ static void mlx4_en_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
 	int err;
+	int idx;
 
 	if (!priv->vlgrp)
 		return;
@@ -83,7 +84,10 @@ static void mlx4_en_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
 		if (err)
 			en_err(priv, "Failed configuring VLAN filter\n");
 	}
+	if (mlx4_register_vlan(mdev->dev, priv->port, vid, &idx))
+		en_err(priv, "failed adding vlan %d\n", vid);
 	mutex_unlock(&mdev->state_lock);
+
 }
 
 static void mlx4_en_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
@@ -91,6 +95,7 @@ static void mlx4_en_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
 	int err;
+	int idx;
 
 	if (!priv->vlgrp)
 		return;
@@ -101,6 +106,11 @@ static void mlx4_en_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
 
 	/* Remove VID from port VLAN filter */
 	mutex_lock(&mdev->state_lock);
+	if (!mlx4_find_cached_vlan(mdev->dev, priv->port, vid, &idx))
+		mlx4_unregister_vlan(mdev->dev, priv->port, idx);
+	else
+		en_err(priv, "could not find vid %d in cache\n", vid);
+
 	if (mdev->device_up && priv->port_up) {
 		err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, priv->vlgrp);
 		if (err)
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 4492109..dab5eaf 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -463,6 +463,7 @@ struct mlx4_en_priv {
 	char *mc_addrs;
 	int mc_addrs_cnt;
 	struct mlx4_en_stat_out_mbox hw_stats;
+	int vids[128];
 };
 
 
diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c
index 606aa58..56371ef 100644
--- a/drivers/net/mlx4/port.c
+++ b/drivers/net/mlx4/port.c
@@ -182,6 +182,25 @@ static int mlx4_set_port_vlan_table(struct mlx4_dev *dev, u8 port,
 	return err;
 }
 
+int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx)
+{
+	struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table;
+	int i;
+
+	for (i = 0; i < MLX4_MAX_VLAN_NUM; ++i) {
+		if (table->refs[i] &&
+		    (vid == (MLX4_VLAN_MASK &
+			      be32_to_cpu(table->entries[i])))) {
+			/* VLAN already registered, increase reference count */
+			*idx = i;
+			return 0;
+		}
+	}
+
+	return -ENOENT;
+}
+EXPORT_SYMBOL_GPL(mlx4_find_cached_vlan);
+
 int mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index)
 {
 	struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table;