From patchwork Thu Jun 10 19:08:11 2010
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
X-Patchwork-Submitter: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
X-Patchwork-Id: 105443
Received: from vger.kernel.org (vger.kernel.org [209.132.180.67])
	by demeter.kernel.org (8.14.3/8.14.3) with ESMTP id o5AJ7a6q017747
	for <patchwork-linux-rdma@patchwork.kernel.org>;
	Thu, 10 Jun 2010 19:08:17 GMT
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1759722Ab0FJTIQ (ORCPT
	<rfc822;patchwork-linux-rdma@patchwork.kernel.org>);
	Thu, 10 Jun 2010 15:08:16 -0400
Received: from mail.mellanox.co.il ([194.90.237.43]:48229 "EHLO
	mellanox.co.il"
	rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP
	id S1759699Ab0FJTIP (ORCPT <rfc822;linux-rdma@vger.kernel.org>);
	Thu, 10 Jun 2010 15:08:15 -0400
Received: from Internal Mail-Server by MTLPINE1 (envelope-from
	yevgenyp@mellanox.co.il) with SMTP; 10 Jun 2010 22:09:03 +0300
Received: from vnc8.lab.mtl.com ([10.4.45.8]) by mtlexch01.mtl.com with
	Microsoft SMTPSVC(6.0.3790.3959); Thu, 10 Jun 2010 22:08:12 +0300
Message-ID: <4C11381B.90101@mellanox.co.il>
Date: Thu, 10 Jun 2010 22:08:11 +0300
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
User-Agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9.1.9)
	Gecko/20100317 Thunderbird/3.0.4
MIME-Version: 1.0
To: Roland Dreier <rdreier@cisco.com>
CC: linux-rdma@vger.kernel.org
Subject: [PATCH 17/19 V4] mlx4: Using mcg tables for Ethernet Unicast
	steering.
X-OriginalArrivalTime: 10 Jun 2010 19:08:12.0149 (UTC)
	FILETIME=[4571B250:01CB08D0]
X-TM-AS-Product-Ver: SMEX-8.0.0.1181-6.000.1038-17438.001
X-TM-AS-Result: No--24.182500-8.000000-31
X-TM-AS-User-Approved-Sender: No
X-TM-AS-User-Blocked-Sender: No
Sender: linux-rdma-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-rdma.vger.kernel.org>
X-Mailing-List: linux-rdma@vger.kernel.org
X-Greylist: IP, sender and recipient auto-whitelisted, not delayed by
	milter-greylist-4.2.3 (demeter.kernel.org [140.211.167.41]);
	Thu, 10 Jun 2010 19:08:17 +0000 (UTC)


diff --git a/drivers/net/mlx4/cmd.c b/drivers/net/mlx4/cmd.c
index 660d001..7efa85f 100644
--- a/drivers/net/mlx4/cmd.c
+++ b/drivers/net/mlx4/cmd.c
@@ -514,7 +514,7 @@ static int mlx4_RESOURCE_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vh
 		switch (vhcr->op) {
 		case MLX4_CMD_ALLOC_RES:
 			ret = mlx4_register_mac(dev, vhcr->op_modifier,
-						vhcr->in_param, (int *) &vhcr->out_param);
+						vhcr->in_param, (int *) &vhcr->out_param, 1);
 			vhcr->errno = ret;
 			break;
 		case MLX4_CMD_FREE_RES:
@@ -937,6 +937,14 @@ static struct mlx4_cmd_info {
 		.wrapper = mlx4_MCAST_wrapper
 	},
 	{
+		.opcode = MLX4_CMD_PROMISC,
+		.has_inbox = false,
+		.has_outbox = false,
+		.out_is_imm = false,
+		.verify = NULL,
+		.wrapper = mlx4_PROMISC_wrapper
+	},
+	{
 		.opcode = MLX4_CMD_DIAG_RPRT,
 		.has_inbox = false,
 		.has_outbox = true,
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index d171945..6dd47e8 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -240,8 +240,12 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 			priv->flags |= MLX4_EN_FLAG_PROMISC;
 
 			/* Enable promiscouos mode */
-			err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
-						     priv->base_qpn, 1);
+			if (!mdev->dev->caps.vep_uc_steering)
+				err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
+							     priv->base_qpn, 1);
+			else
+				err = mlx4_unicast_promisc_add(mdev->dev, priv->base_qpn,
+							       priv->port - 1);
 			if (err)
 				en_err(priv, "Failed enabling "
 					     "promiscous mode\n");
@@ -253,6 +257,15 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 				en_err(priv, "Failed disabling "
 					     "multicast filter\n");
 
+			/* Add the default qp number as multicast promisc */
+			if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
+				err = mlx4_multicast_promisc_add(mdev->dev, priv->base_qpn,
+								 priv->port - 1);
+				if (err)
+					en_err(priv, "Failed entering multicast promisc mode\n");
+				priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
+			}
+
 			/* Disable port VLAN filter */
 			err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, NULL);
 			if (err)
@@ -271,11 +284,24 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 		priv->flags &= ~MLX4_EN_FLAG_PROMISC;
 
 		/* Disable promiscouos mode */
-		err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
-					     priv->base_qpn, 0);
+		if (!mdev->dev->caps.vep_uc_steering)
+			err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
+						     priv->base_qpn, 0);
+		else
+			err = mlx4_unicast_promisc_remove(mdev->dev, priv->base_qpn,
+							  priv->port - 1);
 		if (err)
 			en_err(priv, "Failed disabling promiscous mode\n");
 
+		/* Disable Multicast promisc */
+		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
+			err = mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
+							    priv->port - 1);
+			if (err)
+				en_err(priv, "Failed disabling multicast promiscous mode\n");
+			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
+		}
+
 		/* Enable port VLAN filter */
 		err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, priv->vlgrp);
 		if (err)
@@ -288,9 +314,27 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 					  0, MLX4_MCAST_DISABLE);
 		if (err)
 			en_err(priv, "Failed disabling multicast filter\n");
+
+		/* Add the default qp number as multicast promisc */
+		if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
+			err = mlx4_multicast_promisc_add(mdev->dev, priv->base_qpn,
+							 priv->port - 1);
+			if (err)
+				en_err(priv, "Failed entering multicast promisc mode\n");
+			priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
+		}
 	} else {
 		int i;
 
+		/* Disable Multicast promisc */
+		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
+			err = mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
+							    priv->port - 1);
+			if (err)
+				en_err(priv, "Failed disabling multicast promiscous mode\n");
+			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
+		}
+
 		err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
 					  0, MLX4_MCAST_DISABLE);
 		if (err)
@@ -328,6 +372,7 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 					  0, MLX4_MCAST_ENABLE);
 		if (err)
 			en_err(priv, "Failed enabling multicast filter\n");
+
 	}
 out:
 	mutex_unlock(&mdev->state_lock);
@@ -615,7 +660,7 @@ int mlx4_en_start_port(struct net_device *dev)
 	/* Set port mac number */
 	en_dbg(DRV, priv, "Setting mac for port %d\n", priv->port);
 	err = mlx4_register_mac(mdev->dev, priv->port,
-				priv->mac, &priv->base_qpn);
+				priv->mac, &priv->base_qpn, 0);
 	if (err) {
 		en_err(priv, "Failed setting port mac\n");
 		goto cq_err;
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 773de63..92b2132 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -854,6 +854,7 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 #define	 INIT_HCA_MC_BASE_OFFSET	 (INIT_HCA_MCAST_OFFSET + 0x00)
 #define	 INIT_HCA_LOG_MC_ENTRY_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x12)
 #define	 INIT_HCA_LOG_MC_HASH_SZ_OFFSET	 (INIT_HCA_MCAST_OFFSET + 0x16)
+#define  INIT_HCA_UC_STEERING_OFFSET	 (INIT_HCA_MCAST_OFFSET + 0x18)
 #define	 INIT_HCA_LOG_MC_TABLE_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x1b)
 #define INIT_HCA_TPT_OFFSET		 0x0f0
 #define	 INIT_HCA_DMPT_BASE_OFFSET	 (INIT_HCA_TPT_OFFSET + 0x00)
@@ -914,6 +915,8 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 	MLX4_PUT(inbox, param->mc_base,		INIT_HCA_MC_BASE_OFFSET);
 	MLX4_PUT(inbox, param->log_mc_entry_sz, INIT_HCA_LOG_MC_ENTRY_SZ_OFFSET);
 	MLX4_PUT(inbox, param->log_mc_hash_sz,  INIT_HCA_LOG_MC_HASH_SZ_OFFSET);
+	if (dev->caps.vep_mc_steering)
+		MLX4_PUT(inbox, (u8) (1 << 3),	INIT_HCA_UC_STEERING_OFFSET);
 	MLX4_PUT(inbox, param->log_mc_table_sz, INIT_HCA_LOG_MC_TABLE_SZ_OFFSET);
 
 	/* TPT attributes */
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index b1eb69e..f7fed9a 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -1180,6 +1180,7 @@ static int mlx4_init_port_info(struct mlx4_dev *dev, int port)
 	info->dev = dev;
 	info->port = port;
 	if (!mlx4_is_slave(dev)) {
+		INIT_RADIX_TREE(&info->mac_tree, GFP_KERNEL);
 		mlx4_init_mac_table(dev, &info->mac_table);
 		mlx4_init_vlan_table(dev, &info->vlan_table);
 		info->base_qpn = dev->caps.reserved_qps_base[MLX4_QP_REGION_ETH_ADDR] +
@@ -1209,6 +1210,58 @@ static void mlx4_cleanup_port_info(struct mlx4_port_info *info)
 	device_remove_file(&info->dev->pdev->dev, &info->port_attr);
 }
 
+static int mlx4_init_steering(struct mlx4_dev *dev)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	int num_entries = max(dev->caps.num_ports, dev->caps.pf_num);
+	int i, j;
+
+	priv->steer = kzalloc(sizeof(struct mlx4_steer) * num_entries, GFP_KERNEL);
+	if (!priv->steer)
+		return -ENOMEM;
+
+	for (i = 0; i < num_entries; i++) {
+		for (j = 0; j < MLX4_NUM_STEERS; j++) {
+			INIT_LIST_HEAD(&priv->steer[i].promisc_qps[j]);
+			INIT_LIST_HEAD(&priv->steer[i].steer_entries[j]);
+		}
+	}
+	return 0;
+}
+
+static void mlx4_clear_steering(struct mlx4_dev *dev)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct mlx4_steer_index *entry, *tmp_entry;
+	struct mlx4_promisc_qp *pqp, *tmp_pqp;
+	int num_entries = max(dev->caps.num_ports, dev->caps.pf_num);
+	int i, j;
+
+	for (i = 0; i < num_entries; i++) {
+		for (j = 0; j < MLX4_NUM_STEERS; j++) {
+			list_for_each_entry_safe(pqp, tmp_pqp,
+						 &priv->steer[i].promisc_qps[j],
+						 list) {
+				list_del(&pqp->list);
+				kfree(pqp);
+			}
+			list_for_each_entry_safe(entry, tmp_entry,
+						 &priv->steer[i].steer_entries[j],
+						 list) {
+				list_del(&entry->list);
+				list_for_each_entry_safe(pqp, tmp_pqp,
+							 &entry->duplicates,
+							 list) {
+					list_del(&pqp->list);
+					kfree(pqp);
+				}
+				kfree(entry);
+			}
+		}
+	}
+	kfree(priv->steer);
+}
+
 static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct mlx4_priv *priv;
@@ -1386,6 +1439,12 @@ slave_start:
 		goto err_free_eq;
 	}
 
+	if (!mlx4_is_slave(dev)) {
+		err = mlx4_init_steering(dev);
+		if (err)
+			goto err_free_eq;
+	}
+
 	err = mlx4_setup_hca(dev);
 	if (err == -EBUSY && (dev->flags & MLX4_FLAG_MSI_X) && !mlx4_is_slave(dev)) {
 		dev->flags &= ~MLX4_FLAG_MSI_X;
@@ -1394,7 +1453,7 @@ slave_start:
 	}
 
 	if (err)
-		goto err_free_eq;
+		goto err_steer;
 
 	for (port = 1; port <= dev->caps.num_ports; port++) {
 		err = mlx4_init_port_info(dev, port);
@@ -1434,6 +1493,10 @@ err_port:
 	mlx4_cleanup_pd_table(dev);
 	mlx4_cleanup_uar_table(dev);
 
+err_steer:
+	if (!mlx4_is_slave(dev))
+		mlx4_clear_steering(dev);
+
 err_free_eq:
 	mlx4_free_eq_table(dev);
 
@@ -1510,6 +1573,8 @@ static void mlx4_remove_one(struct pci_dev *pdev)
 		iounmap(priv->kar);
 		mlx4_uar_free(dev, &priv->driver_uar);
 		mlx4_cleanup_uar_table(dev);
+		if (!mlx4_is_slave(dev))
+			mlx4_clear_steering(dev);
 		mlx4_free_eq_table(dev);
 		mlx4_close_hca(dev);
 		if (mlx4_is_mfunc(dev))
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index e3dc486..e1dcd86 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -50,28 +50,35 @@ struct mlx4_mgm {
 
 static const u8 zero_gid[16];	/* automatically initialized to 0 */
 
-static int mlx4_READ_MCG(struct mlx4_dev *dev, int index,
-			 struct mlx4_cmd_mailbox *mailbox)
+static int mlx4_READ_ENTRY(struct mlx4_dev *dev, int index,
+			   struct mlx4_cmd_mailbox *mailbox)
 {
-	return mlx4_cmd_box(dev, 0, mailbox->dma, index, 0, MLX4_CMD_READ_MCG,
-			    MLX4_CMD_TIME_CLASS_A);
+	return mlx4_cmd_box(dev, 0, mailbox->dma, index, 0,
+			    MLX4_CMD_READ_MCG, MLX4_CMD_TIME_CLASS_A);
 }
 
-static int mlx4_WRITE_MCG(struct mlx4_dev *dev, int index,
-			  struct mlx4_cmd_mailbox *mailbox)
+static int mlx4_WRITE_ENTRY(struct mlx4_dev *dev, int index,
+			    struct mlx4_cmd_mailbox *mailbox)
 {
-	return mlx4_cmd(dev, mailbox->dma, index, 0, MLX4_CMD_WRITE_MCG,
-			MLX4_CMD_TIME_CLASS_A);
+	return mlx4_cmd(dev, mailbox->dma, index, 0,
+			MLX4_CMD_WRITE_MCG, MLX4_CMD_TIME_CLASS_A);
 }
 
-static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
-			  u16 *hash, u8 op_mod)
+static int mlx4_WRITE_PROMISC(struct mlx4_dev *dev,
+			      struct mlx4_cmd_mailbox *mailbox)
+{
+	return mlx4_cmd(dev, mailbox->dma, 0, 0x2,
+			MLX4_CMD_WRITE_MCG, MLX4_CMD_TIME_CLASS_A);
+}
+
+static int mlx4_GID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
+			 u16 *hash, u8 op_mod)
 {
 	u64 imm;
 	int err;
 
-	err = mlx4_cmd_imm(dev, mailbox->dma, &imm, 0, op_mod, MLX4_CMD_MGID_HASH,
-			   MLX4_CMD_TIME_CLASS_A);
+	err = mlx4_cmd_imm(dev, mailbox->dma, &imm, 0, op_mod,
+			   MLX4_CMD_MGID_HASH, MLX4_CMD_TIME_CLASS_A);
 
 	if (!err)
 		*hash = imm;
@@ -80,6 +87,425 @@ static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox
 }
 
 /*
+ * Helper functions to manage multifunction steering data structures.
+ * Used only for Ethernet steering.
+ */
+
+static struct mlx4_promisc_qp *get_promisc_qp(struct mlx4_dev *dev, u8 pf_num,
+					      enum mlx4_steer_type steer,
+					      u32 qpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_promisc_qp *pqp;
+
+	list_for_each_entry(pqp, &s_steer->promisc_qps[steer], list) {
+		if (pqp->qpn == qpn)
+			return pqp;
+	}
+	/* not found */
+	return NULL;
+}
+
+/*
+ * Add new entry to steering data structure.
+ * All promisc QPs should be added as well
+ */
+static int new_steering_entry(struct mlx4_dev *dev, u8 pf_num,
+			      enum mlx4_steer_type steer,
+			      unsigned int index, u32 qpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_mgm *mgm;
+	u32 members_count;
+	struct mlx4_steer_index *new_entry;
+	struct mlx4_promisc_qp *pqp;
+	struct mlx4_promisc_qp *dqp;
+	u32 prot;
+	int err;
+
+	new_entry = kzalloc(sizeof *new_entry, GFP_KERNEL);
+	if (!new_entry)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&new_entry->duplicates);
+	new_entry->index = index;
+	list_add_tail(&new_entry->list, &s_steer->steer_entries[steer]);
+
+	/* If the given qpn is also a promisc qp,
+	 * it should be inserted to duplicates list
+	 */
+	pqp = get_promisc_qp(dev, pf_num, steer, qpn);
+	if (pqp) {
+		dqp = kmalloc(sizeof *dqp, GFP_KERNEL);
+		if (!dqp) {
+			err = -ENOMEM;
+			goto out_alloc;
+		}
+		dqp->qpn = qpn;
+		list_add_tail(&pqp->list, &new_entry->duplicates);
+	}
+
+	/* if no promisc qps for this vep, we are done */
+	if (list_empty(&s_steer->promisc_qps[steer]))
+		return 0;
+
+	/* now need to add all the promisc qps to the new
+	 * steering entry, as they should also receive the packets
+	 * destined to this address */
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox)) {
+		err = -ENOMEM;
+		goto out_alloc;
+	}
+	mgm = mailbox->buf;
+
+	err = mlx4_READ_ENTRY(dev, index, mailbox);
+	if (err)
+		goto out_mailbox;
+
+	members_count = be32_to_cpu(mgm->members_count) & 0xffffff;
+	prot = be32_to_cpu(mgm->members_count) >> 30;
+	list_for_each_entry(pqp, &s_steer->promisc_qps[steer], list) {
+		/* don't add already existing qpn */
+		if (pqp->qpn == qpn)
+			continue;
+		if (members_count == MLX4_QP_PER_MGM) {
+			/* out of space */
+			err = -ENOMEM;
+			goto out_mailbox;
+		}
+
+		/* add the qpn */
+		mgm->qp[members_count++] = cpu_to_be32(pqp->qpn & MGM_QPN_MASK);
+	}
+	/* update the qps count and update the entry with all the promisc qps*/
+	mgm->members_count = cpu_to_be32(members_count | (prot << 30));
+	err = mlx4_WRITE_ENTRY(dev, index, mailbox);
+
+out_mailbox:
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	if (!err)
+		return 0;
+out_alloc:
+	if (dqp) {
+		list_del(&dqp->list);
+		kfree(&dqp);
+	}
+	list_del(&new_entry->list);
+	kfree(new_entry);
+	return err;
+}
+
+/* update the data structures with existing steering entry */
+static int existing_steering_entry(struct mlx4_dev *dev, u8 pf_num,
+				   enum mlx4_steer_type steer,
+				   unsigned int index, u32 qpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_steer_index *tmp_entry, *entry = NULL;
+	struct mlx4_promisc_qp *pqp;
+	struct mlx4_promisc_qp *dqp;
+
+	pqp = get_promisc_qp(dev, pf_num, steer, qpn);
+	if (!pqp)
+		return 0; /* nothing to do */
+
+	list_for_each_entry(tmp_entry, &s_steer->steer_entries[steer], list) {
+		if (tmp_entry->index == entry->index) {
+			entry = tmp_entry;
+			break;
+		}
+	}
+	if (unlikely(!entry)) {
+		mlx4_warn(dev, "Steering entry at index %x is not registered\n", index);
+		return -EINVAL;
+	}
+
+	/* the given qpn is listed as a promisc qpn
+	 * we need to add it as a duplicate to this entry
+	 * for future refernce */
+	list_for_each_entry(dqp, &entry->duplicates, list) {
+		if (dqp->qpn == pqp->qpn)
+			return 0; /* qp is already duplicated */
+	}
+
+	/* add the qp as a duplicate on this index */
+	dqp = kmalloc(sizeof *dqp, GFP_KERNEL);
+	if (!dqp)
+		return -ENOMEM;
+	dqp->qpn = qpn;
+	list_add_tail(&pqp->list, &entry->duplicates);
+
+	return 0;
+}
+
+/* Check whether a qpn is a duplicate on steering entry
+ * If so, it should not be removed from mgm */
+static bool check_duplicate_entry(struct mlx4_dev *dev, u8 pf_num,
+				  enum mlx4_steer_type steer,
+				  unsigned int index, u32 qpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_steer_index *tmp_entry, *entry = NULL;
+	struct mlx4_promisc_qp *dqp, *tmp_dqp;
+
+	/* if qp is not promisc, it cannot be duplicated */
+	if (!get_promisc_qp(dev, pf_num, steer, qpn))
+		return false;
+
+	/* The qp is promisc qp so it is a duplicate on this index
+	 * Find the index entry, and remove the duplicate */
+	list_for_each_entry(tmp_entry, &s_steer->steer_entries[steer], list) {
+		if (tmp_entry->index == index) {
+			entry = tmp_entry;
+			break;
+		}
+	}
+	if (unlikely(!entry)) {
+		mlx4_warn(dev, "Steering entry for index %x is not registered\n", index);
+		return false;
+	}
+	list_for_each_entry_safe(dqp, tmp_dqp, &entry->duplicates, list) {
+		if (dqp->qpn == qpn) {
+			list_del(&dqp->list);
+			kfree(dqp);
+		}
+	}
+	return true;
+}
+
+/* I a steering entry contains only promisc QPs, it can be removed. */
+static bool can_remove_steering_entry(struct mlx4_dev *dev, u8 pf_num,
+				      enum mlx4_steer_type steer,
+				      unsigned int index, u32 tqpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_mgm *mgm;
+	struct mlx4_steer_index *entry = NULL, *tmp_entry;
+	u32 qpn;
+	u32 members_count;
+	bool ret = false;
+	int i;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return false;
+	mgm = mailbox->buf;
+
+	if (mlx4_READ_ENTRY(dev, index, mailbox))
+		goto out;
+	members_count = be32_to_cpu(mgm->members_count) & 0xffffff;
+	for (i = 0;  i < members_count; i++) {
+		qpn = be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK;
+		if (!get_promisc_qp(dev, pf_num, steer, qpn) && qpn != tqpn) {
+			/* the qp is not promisc, the entry can't be removed */
+			goto out;
+		}
+	}
+	 /* All the qps currently registered for this entry are promiscuous,
+	  * it can be removed */
+	ret = true;
+	list_for_each_entry_safe(entry, tmp_entry, &s_steer->steer_entries[steer], list) {
+		if (entry->index == index) {
+			list_del(&entry->list);
+			kfree(entry);
+		}
+	}
+
+out:
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return ret;
+}
+
+
+static int add_promisc_qp(struct mlx4_dev *dev, u8 pf_num,
+			  enum mlx4_steer_type steer, u32 qpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_mgm *mgm;
+	struct mlx4_steer_index *entry;
+	struct mlx4_promisc_qp *pqp;
+	struct mlx4_promisc_qp *dqp;
+	u32 members_count;
+	u32 prot;
+	int i;
+	bool found;
+	int last_index;
+	int err;
+
+	if (get_promisc_qp(dev, pf_num, steer, qpn))
+		return 0; /* Noting to do, already exists */
+
+	pqp = kmalloc(sizeof *pqp, GFP_KERNEL);
+	if (!pqp)
+		return -ENOMEM;
+	pqp->qpn = qpn;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox)) {
+		err = -ENOMEM;
+		goto out_alloc;
+	}
+	mgm = mailbox->buf;
+
+	/* the promisc qp needs to be added for each one of the steering
+	 * entries, if it already exists, needs to be added as a duplicate
+	 * for this entry */
+	list_for_each_entry(entry, &s_steer->steer_entries[steer], list) {
+		err = mlx4_READ_ENTRY(dev, entry->index, mailbox);
+		if (err)
+			goto out_mailbox;
+
+		members_count = be32_to_cpu(mgm->members_count) & 0xffffff;
+		prot = be32_to_cpu(mgm->members_count) >> 30;
+		found = false;
+		for (i = 0; i < members_count; i++) {
+			if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qpn) {
+				/* Entry already exists, add to duplicates */
+				dqp = kmalloc(sizeof *dqp, GFP_KERNEL);
+				if (!dqp)
+					goto out_mailbox;
+				dqp->qpn = qpn;
+				list_add_tail(&dqp->list, &entry->duplicates);
+				found = true;
+			}
+		}
+		if (!found) {
+			/* Need to add the qpn to mgm */
+			if (members_count == MLX4_QP_PER_MGM) {
+				/* entry is full */
+				err = -ENOMEM;
+				goto out_mailbox;
+			}
+			mgm->qp[members_count++] = cpu_to_be32(qpn & MGM_QPN_MASK);
+			mgm->members_count = cpu_to_be32(members_count | (prot << 30));
+			err = mlx4_WRITE_ENTRY(dev, entry->index, mailbox);
+			if (err)
+				goto out_mailbox;
+		}
+		last_index = entry->index;
+	}
+
+	/* add the new qpn to list of promisc qps */
+	list_add_tail(&pqp->list, &s_steer->promisc_qps[steer]);
+	/* now need to add all the promisc qps to default entry */
+	memset(mgm, 0, sizeof *mgm);
+	mgm->gid[7] = pf_num << 4;
+	mgm->next_gid_index = cpu_to_be32(1 << 4);
+	members_count = 0;
+	list_for_each_entry(dqp, &s_steer->promisc_qps[steer], list)
+		mgm->qp[members_count++] = cpu_to_be32(dqp->qpn & MGM_QPN_MASK);
+	mgm->members_count = cpu_to_be32(members_count | MLX4_PROT_ETH << 30);
+
+	err = mlx4_WRITE_PROMISC(dev, mailbox);
+	if (err)
+		goto out_list;
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return 0;
+
+out_list:
+	list_del(&pqp->list);
+out_mailbox:
+	/* TODO: undo partial addition of promisc qps */
+	mlx4_free_cmd_mailbox(dev, mailbox);
+out_alloc:
+	kfree(pqp);
+	return err;
+}
+
+static int remove_promisc_qp(struct mlx4_dev *dev, u8 pf_num,
+			     enum mlx4_steer_type steer, u32 qpn)
+{
+	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num];
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_mgm *mgm;
+	struct mlx4_steer_index *entry;
+	struct mlx4_promisc_qp *pqp;
+	struct mlx4_promisc_qp *dqp;
+	u32 members_count;
+	bool found;
+	bool back_to_list = false;
+	int loc, i;
+	int err;
+
+	pqp = get_promisc_qp(dev, pf_num, steer, qpn);
+	if (unlikely(!pqp)) {
+		mlx4_warn(dev, "QP %x is not promiscuous QP\n", qpn);
+		/* nothing to do */
+		return 0;
+	}
+
+	/*remove from list of promisc qps */
+	list_del(&pqp->list);
+	kfree(pqp);
+
+	/* set the default entry not to include the removed one */
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox)) {
+		err = -ENOMEM;
+		back_to_list = true;
+		goto out_list;
+	}
+	mgm = mailbox->buf;
+	mgm->gid[7] = pf_num << 4;
+	mgm->next_gid_index = cpu_to_be32(1 << 4);
+	members_count = 0;
+	list_for_each_entry(dqp, &s_steer->promisc_qps[steer], list)
+		mgm->qp[members_count++] = cpu_to_be32(dqp->qpn & MGM_QPN_MASK);
+	mgm->members_count = cpu_to_be32(members_count | MLX4_PROT_ETH << 30);
+
+	err = mlx4_WRITE_PROMISC(dev,mailbox);
+	if (err)
+		goto out_mailbox;
+
+	/* remove the qp from all the steering entries*/
+	list_for_each_entry(entry, &s_steer->steer_entries[steer], list) {
+		found = false;
+		list_for_each_entry(dqp, &entry->duplicates, list) {
+			if (dqp->qpn == qpn) {
+				found = true;
+				break;
+			}
+		}
+		if (found) {
+			/* a duplicate, no need to change the mgm,
+			 * only update the duplicates list */
+			list_del(&dqp->list);
+			kfree(dqp);
+		} else {
+			err = mlx4_READ_ENTRY(dev, entry->index, mailbox);
+				if (err)
+					goto out_mailbox;
+			members_count = be32_to_cpu(mgm->members_count) & 0xffffff;
+			for (loc = -1, i = 0; i < members_count; ++i)
+				if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qpn)
+					loc = i;
+
+			mgm->members_count = cpu_to_be32(--members_count |
+							 (MLX4_PROT_ETH << 30));
+			mgm->qp[loc] = mgm->qp[i - 1];
+			mgm->qp[i - 1] = 0;
+
+			err = mlx4_WRITE_ENTRY(dev, entry->index, mailbox);
+				if (err)
+					goto out_mailbox;
+		}
+
+	}
+
+out_mailbox:
+	mlx4_free_cmd_mailbox(dev, mailbox);
+out_list:
+	if (back_to_list)
+		list_add_tail(&pqp->list, &s_steer->promisc_qps[steer]);
+	return err;
+}
+
+/*
  * Caller must hold MCG table semaphore.  gid and mgm parameters must
  * be properly aligned for command interface.
  *
@@ -94,10 +520,11 @@ static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox
  * If no AMGM exists for given gid, *index = -1, *prev = index of last
  * entry in hash chain and *mgm holds end of hash chain.
  */
-static int find_mgm(struct mlx4_dev *dev,
-		    u8 *gid, enum mlx4_protocol prot,
-		    struct mlx4_cmd_mailbox *mgm_mailbox,
-		    u16 *hash, int *prev, int *index)
+static int find_entry(struct mlx4_dev *dev,
+		      u8 *gid, enum mlx4_protocol prot,
+		      enum mlx4_steer_type steer,
+		      struct mlx4_cmd_mailbox *mgm_mailbox,
+		      u16 *hash, int *prev, int *index)
 {
 	struct mlx4_cmd_mailbox *mailbox;
 	struct mlx4_mgm *mgm = mgm_mailbox->buf;
@@ -112,7 +539,7 @@ static int find_mgm(struct mlx4_dev *dev,
 
 	memcpy(mgid, gid, 16);
 
-	err = mlx4_MGID_HASH(dev, mailbox, hash, op_mod);
+	err = mlx4_GID_HASH(dev, mailbox, hash, op_mod);
 	mlx4_free_cmd_mailbox(dev, mailbox);
 	if (err)
 		return err;
@@ -124,7 +551,7 @@ static int find_mgm(struct mlx4_dev *dev,
 	*prev  = -1;
 
 	do {
-		err = mlx4_READ_MCG(dev, *index, mgm_mailbox);
+		err = mlx4_READ_ENTRY(dev, *index, mgm_mailbox);
 		if (err)
 			return err;
 
@@ -148,10 +575,9 @@ static int find_mgm(struct mlx4_dev *dev,
 	return err;
 }
 
-static int mlx4_multicast_attach_common(struct mlx4_dev *dev,
-					struct mlx4_qp *qp, u8 gid[16],
-					int block_mcast_loopback,
-					enum mlx4_protocol prot)
+int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  int block_mcast_loopback, enum mlx4_protocol prot,
+			  enum mlx4_steer_type steer)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_cmd_mailbox *mailbox;
@@ -162,6 +588,8 @@ static int mlx4_multicast_attach_common(struct mlx4_dev *dev,
 	int link = 0;
 	int i;
 	int err;
+	u8 pf_num = gid[7] >> 4;
+	u8 new_entry = 0;
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox))
@@ -170,13 +598,15 @@ static int mlx4_multicast_attach_common(struct mlx4_dev *dev,
 
 	mutex_lock(&priv->mcg_table.mutex);
 
-	err = find_mgm(dev, gid, prot, mailbox, &hash, &prev, &index);
+	err = find_entry(dev, gid, prot, steer, mailbox, &hash, &prev, &index);
 	if (err)
 		goto out;
 
 	if (index != -1) {
-		if (!memcmp(mgm->gid, zero_gid, 16))
+		if (!memcmp(mgm->gid, zero_gid, 16)) {
+			new_entry = 1;
 			memcpy(mgm->gid, gid, 16);
+		}
 	} else {
 		link = 1;
 
@@ -215,25 +645,32 @@ static int mlx4_multicast_attach_common(struct mlx4_dev *dev,
 	mgm->members_count       = cpu_to_be32(members_count | ((u32) prot << 30));
 	mgm->next_gid_index	 = cpu_to_be32(!!(dev->caps.vep_mc_steering) << 4);
 
-	err = mlx4_WRITE_MCG(dev, index, mailbox);
+	err = mlx4_WRITE_ENTRY(dev, index, mailbox);
 	if (err)
 		goto out;
 
 	if (!link)
 		goto out;
 
-	err = mlx4_READ_MCG(dev, prev, mailbox);
+	err = mlx4_READ_ENTRY(dev, prev, mailbox);
 	if (err)
 		goto out;
 
 	mgm->next_gid_index = cpu_to_be32((index << 6) |
 					  (!!(dev->caps.vep_mc_steering) << 4));
 
-	err = mlx4_WRITE_MCG(dev, prev, mailbox);
+	err = mlx4_WRITE_ENTRY(dev, prev, mailbox);
 	if (err)
 		goto out;
 
 out:
+	if (!err && prot == MLX4_PROT_ETH) {
+		/* manage the steering entry for promisc mode */
+		if (new_entry)
+			new_steering_entry(dev, pf_num, steer, index, qp->qpn);
+		else
+			existing_steering_entry(dev, pf_num, steer, index, qp->qpn);
+	}
 	if (err && link && index != -1) {
 		if (index < dev->caps.num_mgms)
 			mlx4_warn(dev, "Got AMGM index %d < %d",
@@ -248,9 +685,9 @@ out:
 	return err;
 }
 
-static int mlx4_multicast_detach_common(struct mlx4_dev *dev,
-					struct mlx4_qp *qp, u8 gid[16],
-					enum mlx4_protocol prot)
+int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp,
+				 u8 gid[16], enum mlx4_protocol prot,
+				 enum mlx4_steer_type steer)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_cmd_mailbox *mailbox;
@@ -269,7 +706,7 @@ static int mlx4_multicast_detach_common(struct mlx4_dev *dev,
 
 	mutex_lock(&priv->mcg_table.mutex);
 
-	err = find_mgm(dev, gid, prot, mailbox, &hash, &prev, &index);
+	err = find_entry(dev, gid, prot, steer, mailbox, &hash, &prev, &index);
 	if (err)
 		goto out;
 
@@ -279,6 +716,11 @@ static int mlx4_multicast_detach_common(struct mlx4_dev *dev,
 		goto out;
 	}
 
+	/* if this qp is also a promisc qp, it shouldn't be removed */
+	if (prot == MLX4_PROT_ETH &&
+	    check_duplicate_entry(dev, pf_num, steer, index, qp->qpn))
+		goto out;
+
 	members_count = be32_to_cpu(mgm->members_count);
 	for (loc = -1, i = 0; i < members_count; ++i)
 		if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn)
@@ -295,8 +737,8 @@ static int mlx4_multicast_detach_common(struct mlx4_dev *dev,
 	mgm->qp[loc]       = mgm->qp[i - 1];
 	mgm->qp[i - 1]     = 0;
 
-	if (i != 1) {
-		err = mlx4_WRITE_MCG(dev, index, mailbox);
+	if (!can_remove_steering_entry(dev, pf_num, steer, index, qp->qpn) && i != 1) {
+		err = mlx4_WRITE_ENTRY(dev, index, mailbox);
 		goto out;
 	}
 
@@ -304,16 +746,13 @@ static int mlx4_multicast_detach_common(struct mlx4_dev *dev,
 		/* Remove entry from MGM */
 		int amgm_index = be32_to_cpu(mgm->next_gid_index) >> 6;
 		if (amgm_index) {
-			err = mlx4_READ_MCG(dev, amgm_index, mailbox);
+			err = mlx4_READ_ENTRY(dev, amgm_index, mailbox);
 			if (err)
 				goto out;
-		} else {
+		} else
 			memset(mgm->gid, 0, 16);
-			if (prot == MLX4_PROT_ETH)
-				mgm->gid[7] = pf_num << 4;
-		}
 
-		err = mlx4_WRITE_MCG(dev, index, mailbox);
+		err = mlx4_WRITE_ENTRY(dev, index, mailbox);
 		if (err)
 			goto out;
 
@@ -328,13 +767,13 @@ static int mlx4_multicast_detach_common(struct mlx4_dev *dev,
 	} else {
 		/* Remove entry from AMGM */
 		int cur_next_index = be32_to_cpu(mgm->next_gid_index);
-		err = mlx4_READ_MCG(dev, prev, mailbox);
+		err = mlx4_READ_ENTRY(dev, prev, mailbox);
 		if (err)
 			goto out;
 
 		mgm->next_gid_index = cpu_to_be32(cur_next_index);
 
-		err = mlx4_WRITE_MCG(dev, prev, mailbox);
+		err = mlx4_WRITE_ENTRY(dev, prev, mailbox);
 		if (err)
 			goto out;
 
@@ -360,16 +799,17 @@ int mlx4_MCAST_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr,
 	struct mlx4_qp qp; /* dummy for calling attach/detach */
 	u8 *gid = inbox->buf;
 	enum mlx4_protocol prot = (vhcr->in_modifier >> 28) & 0x7;
+	u8 pf_num = mlx4_priv(dev)->mfunc.master.slave_state[slave].pf_num;
 
 	if (prot == MLX4_PROT_ETH)
-		gid[7] |= slave << 4;
+		gid[7] |= (pf_num << 4 | MLX4_MC_STEER << 1);
 
 	qp.qpn = vhcr->in_modifier & 0xffffff;
 	if (vhcr->op_modifier)
-		return mlx4_multicast_attach_common(dev, &qp, gid,
-					     vhcr->in_modifier >> 31, prot);
+		return mlx4_qp_attach_common(dev, &qp, gid, vhcr->in_modifier >> 31,
+					     prot, MLX4_MC_STEER);
 	else
-		return mlx4_multicast_detach_common(dev, &qp, gid, prot);
+		return mlx4_qp_detach_common(dev, &qp, gid, prot, MLX4_MC_STEER);
 }
 
 static int mlx4_MCAST(struct mlx4_dev *dev, struct mlx4_qp *qp,
@@ -410,10 +850,11 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 		return mlx4_MCAST(dev, qp, gid, 1, block_mcast_loopback, prot);
 
 	if (mlx4_is_master(dev) && prot == MLX4_PROT_ETH)
-		gid[7] |= dev->caps.function << 4;
+		gid[7] |= (dev->caps.function << 4 | MLX4_MC_STEER << 1);
 
-	return mlx4_multicast_attach_common(dev, qp, gid,
-					    block_mcast_loopback, prot);
+	return mlx4_qp_attach_common(dev, qp, gid,
+				     block_mcast_loopback, prot,
+				     MLX4_MC_STEER);
 }
 EXPORT_SYMBOL_GPL(mlx4_multicast_attach);
 
@@ -427,12 +868,82 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 		return mlx4_MCAST(dev, qp, gid, 0, 0, prot);
 
 	if (mlx4_is_master(dev) && prot == MLX4_PROT_ETH)
-		gid[7] |= dev->caps.function << 4;
+		gid[7] |= (dev->caps.function << 4 | MLX4_MC_STEER << 1);
 
-	return mlx4_multicast_detach_common(dev, qp, gid, prot);
+	return mlx4_qp_detach_common(dev, qp, gid, prot, MLX4_MC_STEER);
 }
 EXPORT_SYMBOL_GPL(mlx4_multicast_detach);
 
+int mlx4_PROMISC_wrapper(struct mlx4_dev *dev, int slave,
+			 struct mlx4_vhcr *vhcr,
+			 struct mlx4_cmd_mailbox *inbox,
+			 struct mlx4_cmd_mailbox *outbox)
+{
+	u8 pf_num = mlx4_priv(dev)->mfunc.master.slave_state[slave].pf_num;
+	u32 qpn = (u32) vhcr->in_param & 0xffffffff;
+	u8 port = vhcr->in_param >> 63;
+	enum mlx4_steer_type steer = vhcr->in_modifier;
+	if (vhcr->op_modifier)
+		return add_promisc_qp(dev, pf_num | port, steer, qpn);
+	else
+		return remove_promisc_qp(dev, pf_num | port, steer, qpn);
+}
+
+static int mlx4_PROMISC(struct mlx4_dev *dev, u32 qpn,
+			enum mlx4_steer_type steer, u8 add, u8 port)
+{
+	return mlx4_cmd(dev, (u64) qpn | (u64) port << 63, (u32) steer, add,
+			MLX4_CMD_PROMISC, MLX4_CMD_TIME_CLASS_A);
+}
+
+int mlx4_multicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port)
+{
+	if (!dev->caps.vep_mc_steering)
+		return 0;
+
+	if (mlx4_is_slave(dev))
+		return mlx4_PROMISC(dev, qpn, MLX4_MC_STEER, 1, port);
+
+	return add_promisc_qp(dev, dev->caps.function | port, MLX4_MC_STEER, qpn);
+}
+EXPORT_SYMBOL_GPL(mlx4_multicast_promisc_add);
+
+int mlx4_multicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port)
+{
+	if (!dev->caps.vep_mc_steering)
+		return 0;
+
+	if (mlx4_is_slave(dev))
+		return mlx4_PROMISC(dev, qpn, MLX4_MC_STEER, 0, port);
+
+	return remove_promisc_qp(dev, dev->caps.function | port, MLX4_MC_STEER, qpn);
+}
+EXPORT_SYMBOL_GPL(mlx4_multicast_promisc_remove);
+
+int mlx4_unicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port)
+{
+	if (!dev->caps.vep_mc_steering)
+		return 0;
+
+	if (mlx4_is_slave(dev))
+		return mlx4_PROMISC(dev, qpn, MLX4_UC_STEER, 1, port);
+
+	return add_promisc_qp(dev, dev->caps.function | port, MLX4_UC_STEER, qpn);
+}
+EXPORT_SYMBOL_GPL(mlx4_unicast_promisc_add);
+
+int mlx4_unicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port)
+{
+	if (!dev->caps.vep_mc_steering)
+		return 0;
+
+	if (mlx4_is_slave(dev))
+		return mlx4_PROMISC(dev, qpn, MLX4_UC_STEER, 0, port);
+
+	return remove_promisc_qp(dev, dev->caps.function | port, MLX4_UC_STEER, qpn);
+}
+EXPORT_SYMBOL_GPL(mlx4_unicast_promisc_remove);
+
 int mlx4_init_mcg_table(struct mlx4_dev *dev)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index d722467..8530032 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -234,6 +234,17 @@ struct mlx4_vlan_fltr {
 	__be32 entry[VLAN_FLTR_SIZE];
 };
 
+struct mlx4_promisc_qp {
+	struct list_head list;
+	u32 qpn;
+};
+
+struct mlx4_steer_index {
+	struct list_head list;
+	unsigned int index;
+	struct list_head duplicates;
+};
+
 struct mlx4_slave_state {
 	u8 comm_toggle;
 	u8 last_cmd;
@@ -425,6 +436,9 @@ struct mlx4_set_port_rqp_calc_context {
 	__be32 mcast;
 };
 
+struct mlx4_mac_entry {
+	u64 mac;
+};
 
 struct mlx4_port_info {
 	struct mlx4_dev	       *dev;
@@ -433,6 +447,7 @@ struct mlx4_port_info {
 	struct device_attribute port_attr;
 	enum mlx4_port_type	tmp_type;
 	struct mlx4_mac_table	mac_table;
+	struct radix_tree_root	mac_tree;
 	struct mlx4_vlan_table	vlan_table;
 	int			base_qpn;
 };
@@ -444,6 +459,11 @@ struct mlx4_sense {
 	struct delayed_work	sense_poll;
 };
 
+struct mlx4_steer {
+	struct list_head promisc_qps[MLX4_NUM_STEERS];
+	struct list_head steer_entries[MLX4_NUM_STEERS];
+};
+
 struct mlx4_priv {
 	struct mlx4_dev		dev;
 
@@ -476,6 +496,7 @@ struct mlx4_priv {
 	struct mlx4_port_info	port[MLX4_MAX_PORTS + 1];
 	struct mlx4_sense       sense;
 	struct mutex		port_mutex;
+	struct mlx4_steer	*steer;
 };
 
 static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
@@ -610,6 +631,15 @@ int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps);
 int mlx4_MCAST_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr,
 						     struct mlx4_cmd_mailbox *inbox,
 						     struct mlx4_cmd_mailbox *outbox);
+int mlx4_PROMISC_wrapper(struct mlx4_dev *dev, int slave,
+			 struct mlx4_vhcr *vhcr,
+			 struct mlx4_cmd_mailbox *inbox,
+			 struct mlx4_cmd_mailbox *outbox);
+int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  enum mlx4_protocol prot, enum mlx4_steer_type steer);
+int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  int block_mcast_loopback, enum mlx4_protocol prot,
+			  enum mlx4_steer_type steer);
 int mlx4_SET_MCAST_FLTR_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr,
 				struct mlx4_cmd_mailbox *inbox,
 				struct mlx4_cmd_mailbox *outbox);
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index e0ce6c5..cc1a6f0 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -480,6 +480,7 @@ struct mlx4_en_priv {
 	struct mlx4_en_rss_map rss_map;
 	u32 flags;
 #define MLX4_EN_FLAG_PROMISC	0x1
+#define MLX4_EN_FLAG_MC_PROMISC	0x2
 	u32 tx_ring_num;
 	u32 rx_ring_num;
 	u32 rx_skb_size;
diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c
index 2437720..939efef 100644
--- a/drivers/net/mlx4/port.c
+++ b/drivers/net/mlx4/port.c
@@ -90,11 +90,62 @@ static int mlx4_set_port_mac_table(struct mlx4_dev *dev, u8 port,
 	return err;
 }
 
-int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn)
+static int mlx4_uc_steer_add(struct mlx4_dev *dev, u8 port,
+			     u64 mac, int *qpn, u8 reserve)
+{
+	struct mlx4_qp qp;
+	u8 pf_num;
+	u8 gid[16] = {0};
+	int err;
+
+	if (reserve) {
+		err = mlx4_qp_reserve_range(dev, 1, 1, qpn);
+		if (err) {
+			mlx4_err(dev, "Failed to reserve qp for mac registration\n");
+			return err;
+		}
+	}
+	qp.qpn = *qpn;
+
+	pf_num = ((u8) (mac >> 48)) | (port - 1);
+	mac &= 0xffffffffffffULL;
+	mac = cpu_to_be64(mac << 16);
+	memcpy(&gid[10], &mac, ETH_ALEN);
+	gid[7] = pf_num << 4 | MLX4_UC_STEER << 1;
+
+	err = mlx4_qp_attach_common(dev, &qp, gid, 0,
+				    MLX4_PROT_ETH, MLX4_UC_STEER);
+	if (err && reserve)
+		mlx4_qp_release_range(dev, *qpn, 1);
+
+	return err;
+}
+
+static void mlx4_uc_steer_release(struct mlx4_dev *dev, u8 port,
+				  u64 mac, int qpn, u8 free)
+{
+	struct mlx4_qp qp;
+	u8 pf_num;
+	u8 gid[16] = {0};
+
+	qp.qpn = qpn;
+	pf_num = ((u8) (mac >> 48)) | (port - 1);
+	mac &= 0xffffffffffffULL;
+	mac = cpu_to_be64(mac << 16);
+	memcpy(&gid[10], &mac, ETH_ALEN);
+	gid[7] = pf_num << 4 | MLX4_UC_STEER << 1;
+
+	mlx4_qp_detach_common(dev, &qp, gid, MLX4_PROT_ETH, MLX4_UC_STEER);
+	if (free)
+		mlx4_qp_release_range(dev, qpn, 1);
+}
+
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn, u8 wrap)
 {
 	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
 	struct mlx4_mac_table *table = &info->mac_table;
 	u64 out_param;
+	struct mlx4_mac_entry *entry;
 	int i, err = 0;
 	int free = -1;
 
@@ -104,9 +155,25 @@ int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn)
 		if (!err)
 			*qpn = out_param;
 		return err;
-	} else
+	} else if (!wrap)
 		mac |= (u64) (dev->caps.function) << 48;
 
+	if (dev->caps.vep_uc_steering) {
+		err = mlx4_uc_steer_add(dev, port, mac, qpn, 1);
+		if (!err) {
+			entry = kmalloc(sizeof *entry, GFP_KERNEL);
+			if (!entry) {
+				mlx4_uc_steer_release(dev, port, mac, *qpn, 1);
+				return -ENOMEM;
+			}
+			entry->mac = mac;
+			err = radix_tree_insert(&info->mac_tree, *qpn, entry);
+			if (err)
+				mlx4_uc_steer_release(dev, port, mac, *qpn, 1);
+		}
+		return err;
+	}
+
 	mlx4_dbg(dev, "Registering MAC: 0x%llx\n", (unsigned long long) mac);
 	mutex_lock(&table->mutex);
 	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
@@ -121,6 +188,7 @@ int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn)
 			goto out;
 		}
 	}
+
 	mlx4_dbg(dev, "Free MAC index is %d\n", free);
 
 	if (table->total == table->max) {
@@ -164,12 +232,22 @@ void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn)
 	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
 	struct mlx4_mac_table *table = &info->mac_table;
 	int index = qpn - info->base_qpn;
+	struct mlx4_mac_entry *entry;
 
 	if (mlx4_is_slave(dev)) {
 		mlx4_cmd(dev, qpn, RES_MAC, port,
 			 MLX4_CMD_FREE_RES, MLX4_CMD_TIME_CLASS_A);
 		return;
 	}
+	if (dev->caps.vep_uc_steering) {
+		entry = radix_tree_lookup(&info->mac_tree, qpn);
+		if (entry) {
+			mlx4_uc_steer_release(dev, port, entry->mac, qpn, 1);
+			radix_tree_delete(&info->mac_tree, qpn);
+			kfree(entry);
+		}
+		return;
+	}
 
 	mutex_lock(&table->mutex);
 
@@ -189,6 +267,7 @@ int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac)
 	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
 	struct mlx4_mac_table *table = &info->mac_table;
 	int index = qpn - info->base_qpn;
+	struct mlx4_mac_entry *entry;
 	int err;
 
 	if (mlx4_is_slave(dev)) {
@@ -197,6 +276,14 @@ int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac)
 		return err;
 	}
 
+	if (dev->caps.vep_uc_steering) {
+		entry = radix_tree_lookup(&info->mac_tree, qpn);
+		if (!entry)
+			return -EINVAL;
+		mlx4_uc_steer_release(dev, port, entry->mac, qpn, 0);
+		return mlx4_uc_steer_add(dev, port, entry->mac, &qpn, 0);
+	}
+
 	mutex_lock(&table->mutex);
 
 	err = validate_index(dev, table, index);
diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h
index 236e03f..9225791 100644
--- a/include/linux/mlx4/cmd.h
+++ b/include/linux/mlx4/cmd.h
@@ -132,6 +132,7 @@ enum {
 	MLX4_CMD_GET_EVENT	 = 0xf03,
 	MLX4_CMD_QUERY_SLAVE_CAP = 0xf04,
 	MLX4_CMD_MCAST_ATTACH    = 0xf05,
+	MLX4_CMD_PROMISC         = 0xf07,
 
 	/* debug commands */
 	MLX4_CMD_QUERY_DEBUG_MSG = 0x2a,
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index b999e39..c03a176 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -179,6 +179,12 @@ enum mlx4_special_vlan_idx {
 	MLX4_VLAN_REGULAR
 };
 
+enum mlx4_steer_type {
+	MLX4_UC_STEER = 0,
+	MLX4_MC_STEER,
+	MLX4_NUM_STEERS
+};
+
 enum {
 	MLX4_NUM_FEXCH          = 64 * 1024,
 };
@@ -505,9 +511,13 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			  int block_mcast_loopback, enum mlx4_protocol prot);
 int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			  enum mlx4_protocol prot);
+int mlx4_multicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_multicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_unicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_unicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port);
 int mlx4_SET_MCAST_FLTR(struct mlx4_dev *dev, u8 port, u64 mac, u64 clear, u8 mode);
 
-int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn);
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn, u8 wrap);
 void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn);
 int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac);