Merge tag 'rdma-for-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband

Pull InfiniBand/RDMA changes from Roland Dreier:
 - Updates to the qib low-level driver
 - First chunk of changes for SR-IOV support for mlx4 IB
 - RDMA CM support for IPv6-only binding
 - Other misc cleanups and fixes

Fix up some add-add conflicts in include/linux/mlx4/device.h and
drivers/net/ethernet/mellanox/mlx4/main.c

* tag 'rdma-for-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (30 commits)
  IB/qib: checkpatch fixes
  IB/qib: Add congestion control agent implementation
  IB/qib: Reduce sdma_lock contention
  IB/qib: Fix an incorrect log message
  IB/qib: Fix QP RCU sparse warnings
  mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them
  mlx4_core: Allow guests to have IB ports
  mlx4_core: Implement mechanism for reserved Q_Keys
  net/mlx4_core: Free ICM table in case of error
  IB/cm: Destroy idr as part of the module init error flow
  mlx4_core: Remove double function declarations
  IB/mlx4: Fill the masked_atomic_cap attribute in query device
  IB/mthca: Fill in sq_sig_type in query QP
  IB/mthca: Warning about event for non-existent QPs should show event type
  IB/qib: Fix sparse RCU warnings in qib_keys.c
  net/mlx4_core: Initialize IB port capabilities for all slaves
  mlx4: Use port management change event instead of smp_snoop
  IB/qib: RCU locking for MR validation
  IB/qib: Avoid returning EBUSY from MR deregister
  IB/qib: Fix UC MR refs for immediate operations
  ...

9 files changed, 206 insertions, 130 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c
index 69ba5727048..a52922ed85c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c
@@ -131,7 +131,7 @@ static void *mlx4_en_get_netdev(struct mlx4_dev *dev, void *ctx, u8 port)
 }
 
 static void mlx4_en_event(struct mlx4_dev *dev, void *endev_ptr,
-			  enum mlx4_dev_event event, int port)
+			  enum mlx4_dev_event event, unsigned long port)
 {
 	struct mlx4_en_dev *mdev = (struct mlx4_en_dev *) endev_ptr;
 	struct mlx4_en_priv *priv;
@@ -156,7 +156,8 @@ static void mlx4_en_event(struct mlx4_dev *dev, void *endev_ptr,
 		if (port < 1 || port > dev->caps.num_ports ||
 		    !mdev->pndev[port])
 			return;
-		mlx4_warn(mdev, "Unhandled event %d for port %d\n", event, port);
+		mlx4_warn(mdev, "Unhandled event %d for port %d\n", event,
+			  (int) port);
 	}
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index cd48337cbfc..99a04648fab 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -83,6 +83,15 @@ enum {
 			       (1ull << MLX4_EVENT_TYPE_FLR_EVENT)	    | \
 			       (1ull << MLX4_EVENT_TYPE_FATAL_WARNING))
 
+static u64 get_async_ev_mask(struct mlx4_dev *dev)
+{
+	u64 async_ev_mask = MLX4_ASYNC_EVENT_MASK;
+	if (dev->caps.flags & MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV)
+		async_ev_mask |= (1ull << MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT);
+
+	return async_ev_mask;
+}
+
 static void eq_set_ci(struct mlx4_eq *eq, int req_not)
 {
 	__raw_writel((__force u32) cpu_to_be32((eq->cons_index & 0xffffff) |
@@ -474,6 +483,11 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 
 			break;
 
+		case MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT:
+			mlx4_dispatch_event(dev, MLX4_DEV_EVENT_PORT_MGMT_CHANGE,
+					    (unsigned long) eqe);
+			break;
+
 		case MLX4_EVENT_TYPE_EEC_CATAS_ERROR:
 		case MLX4_EVENT_TYPE_ECC_DETECT:
 		default:
@@ -957,7 +971,7 @@ int mlx4_init_eq_table(struct mlx4_dev *dev)
 		priv->eq_table.have_irq = 1;
 	}
 
-	err = mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+	err = mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0,
 			  priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
 	if (err)
 		mlx4_warn(dev, "MAP_EQ for async EQ %d failed (%d)\n",
@@ -997,7 +1011,7 @@ void mlx4_cleanup_eq_table(struct mlx4_dev *dev)
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	int i;
 
-	mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 1,
+	mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 1,
 		    priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
 
 	mlx4_free_irqs(dev);
@@ -1041,7 +1055,7 @@ int mlx4_test_interrupts(struct mlx4_dev *dev)
 		mlx4_cmd_use_polling(dev);
 
 		/* Map the new eq to handle all asyncronous events */
-		err = mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+		err = mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0,
 				  priv->eq_table.eq[i].eqn);
 		if (err) {
 			mlx4_warn(dev, "Failed mapping eq for interrupt test\n");
@@ -1055,7 +1069,7 @@ int mlx4_test_interrupts(struct mlx4_dev *dev)
 	}
 
 	/* Return to default */
-	mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+	mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0,
 		    priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 1d70657058a..c6964848732 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -109,6 +109,7 @@ static void dump_dev_cap_flags(struct mlx4_dev *dev, u64 flags)
 		[41] = "Unicast VEP steering support",
 		[42] = "Multicast VEP steering support",
 		[48] = "Counters support",
+		[59] = "Port management change event support",
 	};
 	int i;
 
@@ -174,6 +175,7 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_FLAGS_OFFSET		0x0
 #define QUERY_FUNC_CAP_NUM_PORTS_OFFSET		0x1
 #define QUERY_FUNC_CAP_PF_BHVR_OFFSET		0x4
+#define QUERY_FUNC_CAP_FMR_OFFSET		0x8
 #define QUERY_FUNC_CAP_QP_QUOTA_OFFSET		0x10
 #define QUERY_FUNC_CAP_CQ_QUOTA_OFFSET		0x14
 #define QUERY_FUNC_CAP_SRQ_QUOTA_OFFSET		0x18
@@ -183,25 +185,44 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_MAX_EQ_OFFSET		0x2c
 #define QUERY_FUNC_CAP_RESERVED_EQ_OFFSET	0X30
 
+#define QUERY_FUNC_CAP_FMR_FLAG			0x80
+#define QUERY_FUNC_CAP_FLAG_RDMA		0x40
+#define QUERY_FUNC_CAP_FLAG_ETH			0x80
+
+/* when opcode modifier = 1 */
 #define QUERY_FUNC_CAP_PHYS_PORT_OFFSET		0x3
+#define QUERY_FUNC_CAP_RDMA_PROPS_OFFSET	0x8
 #define QUERY_FUNC_CAP_ETH_PROPS_OFFSET		0xc
 
+#define QUERY_FUNC_CAP_ETH_PROPS_FORCE_MAC	0x40
+#define QUERY_FUNC_CAP_ETH_PROPS_FORCE_VLAN	0x80
+
+#define QUERY_FUNC_CAP_RDMA_PROPS_FORCE_PHY_WQE_GID 0x80
+
 	if (vhcr->op_modifier == 1) {
 		field = vhcr->in_modifier;
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
 
-		field = 0; /* ensure fvl bit is not set */
+		field = 0;
+		/* ensure force vlan and force mac bits are not set */
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
+		/* ensure that phy_wqe_gid bit is not set */
+		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_RDMA_PROPS_OFFSET);
+
 	} else if (vhcr->op_modifier == 0) {
-		field = 1 << 7; /* enable only ethernet interface */
+		/* enable rdma and ethernet interfaces */
+		field = (QUERY_FUNC_CAP_FLAG_ETH | QUERY_FUNC_CAP_FLAG_RDMA);
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS_OFFSET);
 
 		field = dev->caps.num_ports;
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_NUM_PORTS_OFFSET);
 
-		size = 0; /* no PF behavious is set for now */
+		size = 0; /* no PF behaviour is set for now */
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_PF_BHVR_OFFSET);
 
+		field = 0; /* protected FMR support not available as yet */
+		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FMR_OFFSET);
+
 		size = dev->caps.num_qps;
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP_QUOTA_OFFSET);
 
@@ -254,11 +275,12 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, struct mlx4_func_cap *func_cap)
 	outbox = mailbox->buf;
 
 	MLX4_GET(field, outbox, QUERY_FUNC_CAP_FLAGS_OFFSET);
-	if (!(field & (1 << 7))) {
-		mlx4_err(dev, "The host doesn't support eth interface\n");
+	if (!(field & (QUERY_FUNC_CAP_FLAG_ETH | QUERY_FUNC_CAP_FLAG_RDMA))) {
+		mlx4_err(dev, "The host supports neither eth nor rdma interfaces\n");
 		err = -EPROTONOSUPPORT;
 		goto out;
 	}
+	func_cap->flags = field;
 
 	MLX4_GET(field, outbox, QUERY_FUNC_CAP_NUM_PORTS_OFFSET);
 	func_cap->num_ports = field;
@@ -297,17 +319,27 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, struct mlx4_func_cap *func_cap)
 		if (err)
 			goto out;
 
-		MLX4_GET(field, outbox, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
-		if (field & (1 << 7)) {
-			mlx4_err(dev, "VLAN is enforced on this port\n");
-			err = -EPROTONOSUPPORT;
-			goto out;
-		}
+		if (dev->caps.port_type[i] == MLX4_PORT_TYPE_ETH) {
+			MLX4_GET(field, outbox, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
+			if (field & QUERY_FUNC_CAP_ETH_PROPS_FORCE_VLAN) {
+				mlx4_err(dev, "VLAN is enforced on this port\n");
+				err = -EPROTONOSUPPORT;
+				goto out;
+			}
 
-		if (field & (1 << 6)) {
-			mlx4_err(dev, "Force mac is enabled on this port\n");
-			err = -EPROTONOSUPPORT;
-			goto out;
+			if (field & QUERY_FUNC_CAP_ETH_PROPS_FORCE_MAC) {
+				mlx4_err(dev, "Force mac is enabled on this port\n");
+				err = -EPROTONOSUPPORT;
+				goto out;
+			}
+		} else if (dev->caps.port_type[i] == MLX4_PORT_TYPE_IB) {
+			MLX4_GET(field, outbox, QUERY_FUNC_CAP_RDMA_PROPS_OFFSET);
+			if (field & QUERY_FUNC_CAP_RDMA_PROPS_FORCE_PHY_WQE_GID) {
+				mlx4_err(dev, "phy_wqe_gid is "
+					 "enforced on this ib port\n");
+				err = -EPROTONOSUPPORT;
+				goto out;
+			}
 		}
 
 		MLX4_GET(field, outbox, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
@@ -707,14 +739,12 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 {
 	u64 def_mac;
 	u8 port_type;
+	u16 short_field;
 	int err;
 
-#define MLX4_PORT_SUPPORT_IB		(1 << 0)
-#define MLX4_PORT_SUGGEST_TYPE		(1 << 3)
-#define MLX4_PORT_DEFAULT_SENSE		(1 << 4)
-#define MLX4_VF_PORT_ETH_ONLY_MASK	(0xff & ~MLX4_PORT_SUPPORT_IB & \
-					 ~MLX4_PORT_SUGGEST_TYPE & \
-					 ~MLX4_PORT_DEFAULT_SENSE)
+#define MLX4_VF_PORT_NO_LINK_SENSE_MASK	0xE0
+#define QUERY_PORT_CUR_MAX_PKEY_OFFSET	0x0c
+#define QUERY_PORT_CUR_MAX_GID_OFFSET	0x0e
 
 	err = mlx4_cmd_box(dev, 0, outbox->dma, vhcr->in_modifier, 0,
 			   MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B,
@@ -730,20 +760,58 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 		MLX4_GET(port_type, outbox->buf,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
 
-		/* Allow only Eth port, no link sensing allowed */
-		port_type &= MLX4_VF_PORT_ETH_ONLY_MASK;
-
-		/* check eth is enabled for this port */
-		if (!(port_type & 2))
-			mlx4_dbg(dev, "QUERY PORT: eth not supported by host");
+		/* No link sensing allowed */
+		port_type &= MLX4_VF_PORT_NO_LINK_SENSE_MASK;
+		/* set port type to currently operating port type */
+		port_type |= (dev->caps.port_type[vhcr->in_modifier] & 0x3);
 
 		MLX4_PUT(outbox->buf, port_type,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
+
+		short_field = 1; /* slave max gids */
+		MLX4_PUT(outbox->buf, short_field,
+			 QUERY_PORT_CUR_MAX_GID_OFFSET);
+
+		short_field = dev->caps.pkey_table_len[vhcr->in_modifier];
+		MLX4_PUT(outbox->buf, short_field,
+			 QUERY_PORT_CUR_MAX_PKEY_OFFSET);
 	}
 
 	return err;
 }
 
+int mlx4_get_slave_pkey_gid_tbl_len(struct mlx4_dev *dev, u8 port,
+				    int *gid_tbl_len, int *pkey_tbl_len)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	u32			*outbox;
+	u16			field;
+	int			err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	err =  mlx4_cmd_box(dev, 0, mailbox->dma, port, 0,
+			    MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B,
+			    MLX4_CMD_WRAPPED);
+	if (err)
+		goto out;
+
+	outbox = mailbox->buf;
+
+	MLX4_GET(field, outbox, QUERY_PORT_CUR_MAX_GID_OFFSET);
+	*gid_tbl_len = field;
+
+	MLX4_GET(field, outbox, QUERY_PORT_CUR_MAX_PKEY_OFFSET);
+	*pkey_tbl_len = field;
+
+out:
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+EXPORT_SYMBOL(mlx4_get_slave_pkey_gid_tbl_len);
+
 int mlx4_map_cmd(struct mlx4_dev *dev, u16 op, struct mlx4_icm *icm, u64 virt)
 {
 	struct mlx4_cmd_mailbox *mailbox;
@@ -890,11 +958,12 @@ int mlx4_QUERY_FW(struct mlx4_dev *dev)
 		((fw_ver & 0xffff0000ull) >> 16) |
 		((fw_ver & 0x0000ffffull) << 16);
 
+	MLX4_GET(lg, outbox, QUERY_FW_PPF_ID);
+	dev->caps.function = lg;
+
 	if (mlx4_is_slave(dev))
 		goto out;
 
-	MLX4_GET(lg, outbox, QUERY_FW_PPF_ID);
-	dev->caps.function = lg;
 
 	MLX4_GET(cmd_if_rev, outbox, QUERY_FW_CMD_IF_REV_OFFSET);
 	if (cmd_if_rev < MLX4_COMMAND_INTERFACE_MIN_REV ||
@@ -975,9 +1044,12 @@ int mlx4_QUERY_FW_wrapper(struct mlx4_dev *dev, int slave,
 	if (err)
 		return err;
 
-	/* for slaves, zero out everything except FW version */
+	/* for slaves, set pci PPF ID to invalid and zero out everything
+	 * else except FW version */
 	outbuf[0] = outbuf[1] = 0;
 	memset(&outbuf[8], 0, QUERY_FW_OUT_SIZE - 8);
+	outbuf[QUERY_FW_PPF_ID] = MLX4_INVALID_SLAVE_ID;
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.c b/drivers/net/ethernet/mellanox/mlx4/icm.c
index a9ade1c3cad..88b7b3e75ab 100644
--- a/drivers/net/ethernet/mellanox/mlx4/icm.c
+++ b/drivers/net/ethernet/mellanox/mlx4/icm.c
@@ -413,6 +413,8 @@ err:
 			mlx4_free_icm(dev, table->icm[i], use_coherent);
 		}
 
+	kfree(table->icm);
+
 	return -ENOMEM;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.h b/drivers/net/ethernet/mellanox/mlx4/icm.h
index b10c07a1dc1..19e4efc0b34 100644
--- a/drivers/net/ethernet/mellanox/mlx4/icm.h
+++ b/drivers/net/ethernet/mellanox/mlx4/icm.h
@@ -81,13 +81,7 @@ int mlx4_init_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table,
 			u64 virt, int obj_size,	int nobj, int reserved,
 			int use_lowmem, int use_coherent);
 void mlx4_cleanup_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table);
-int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, int obj);
-void mlx4_table_put(struct mlx4_dev *dev, struct mlx4_icm_table *table, int obj);
 void *mlx4_table_find(struct mlx4_icm_table *table, int obj, dma_addr_t *dma_handle);
-int mlx4_table_get_range(struct mlx4_dev *dev, struct mlx4_icm_table *table,
-			 int start, int end);
-void mlx4_table_put_range(struct mlx4_dev *dev, struct mlx4_icm_table *table,
-			  int start, int end);
 
 static inline void mlx4_icm_first(struct mlx4_icm *icm,
 				  struct mlx4_icm_iter *iter)
diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c
index b4e9f6f5cc0..116895ac8b3 100644
--- a/drivers/net/ethernet/mellanox/mlx4/intf.c
+++ b/drivers/net/ethernet/mellanox/mlx4/intf.c
@@ -115,7 +115,8 @@ void mlx4_unregister_interface(struct mlx4_interface *intf)
 }
 EXPORT_SYMBOL_GPL(mlx4_unregister_interface);
 
-void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int port)
+void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type,
+			 unsigned long param)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_device_context *dev_ctx;
@@ -125,7 +126,7 @@ void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int por
 
 	list_for_each_entry(dev_ctx, &priv->ctx_list, list)
 		if (dev_ctx->intf->event)
-			dev_ctx->intf->event(dev, dev_ctx->context, type, port);
+			dev_ctx->intf->event(dev, dev_ctx->context, type, param);
 
 	spin_unlock_irqrestore(&priv->ctx_lock, flags);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 42645166bae..e8f8ebb4ae6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -218,6 +218,10 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		dev->caps.vl_cap[i]	    = dev_cap->max_vl[i];
 		dev->caps.ib_mtu_cap[i]	    = dev_cap->ib_mtu[i];
+		dev->phys_caps.gid_phys_table_len[i]  = dev_cap->max_gids[i];
+		dev->phys_caps.pkey_phys_table_len[i] = dev_cap->max_pkeys[i];
+		/* set gid and pkey table operating lengths by default
+		 * to non-sriov values */
 		dev->caps.gid_table_len[i]  = dev_cap->max_gids[i];
 		dev->caps.pkey_table_len[i] = dev_cap->max_pkeys[i];
 		dev->caps.port_width_cap[i] = dev_cap->max_port_width[i];
@@ -312,29 +316,19 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			/* if only ETH is supported - assign ETH */
 			if (dev->caps.supported_type[i] == MLX4_PORT_TYPE_ETH)
 				dev->caps.port_type[i] = MLX4_PORT_TYPE_ETH;
-			/* if only IB is supported,
-			 * assign IB only if SRIOV is off*/
+			/* if only IB is supported, assign IB */
 			else if (dev->caps.supported_type[i] ==
-				 MLX4_PORT_TYPE_IB) {
-				if (dev->flags & MLX4_FLAG_SRIOV)
-					dev->caps.port_type[i] =
-						MLX4_PORT_TYPE_NONE;
-				else
-					dev->caps.port_type[i] =
-						MLX4_PORT_TYPE_IB;
-			/* if IB and ETH are supported,
-			 * first of all check if SRIOV is on */
-			} else if (dev->flags & MLX4_FLAG_SRIOV)
-				dev->caps.port_type[i] = MLX4_PORT_TYPE_ETH;
+				 MLX4_PORT_TYPE_IB)
+				dev->caps.port_type[i] = MLX4_PORT_TYPE_IB;
 			else {
-				/* In non-SRIOV mode, we set the port type
-				 * according to user selection of port type,
-				 * if usere selected none, take the FW hint */
-				if (port_type_array[i-1] == MLX4_PORT_TYPE_NONE)
+				/* if IB and ETH are supported, we set the port
+				 * type according to user selection of port type;
+				 * if user selected none, take the FW hint */
+				if (port_type_array[i - 1] == MLX4_PORT_TYPE_NONE)
 					dev->caps.port_type[i] = dev->caps.suggested_type[i] ?
 						MLX4_PORT_TYPE_ETH : MLX4_PORT_TYPE_IB;
 				else
-					dev->caps.port_type[i] = port_type_array[i-1];
+					dev->caps.port_type[i] = port_type_array[i - 1];
 			}
 		}
 		/*
@@ -415,6 +409,23 @@ static int mlx4_how_many_lives_vf(struct mlx4_dev *dev)
 	return ret;
 }
 
+int mlx4_get_parav_qkey(struct mlx4_dev *dev, u32 qpn, u32 *qkey)
+{
+	u32 qk = MLX4_RESERVED_QKEY_BASE;
+	if (qpn >= dev->caps.base_tunnel_sqpn + 8 * MLX4_MFUNC_MAX ||
+	    qpn < dev->caps.sqp_start)
+		return -EINVAL;
+
+	if (qpn >= dev->caps.base_tunnel_sqpn)
+		/* tunnel qp */
+		qk += qpn - dev->caps.base_tunnel_sqpn;
+	else
+		qk += qpn - dev->caps.sqp_start;
+	*qkey = qk;
+	return 0;
+}
+EXPORT_SYMBOL(mlx4_get_parav_qkey);
+
 int mlx4_is_slave_active(struct mlx4_dev *dev, int slave)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
@@ -515,8 +526,13 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 		return -ENODEV;
 	}
 
-	for (i = 1; i <= dev->caps.num_ports; ++i)
+	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		dev->caps.port_mask[i] = dev->caps.port_type[i];
+		if (mlx4_get_slave_pkey_gid_tbl_len(dev, i,
+						    &dev->caps.gid_table_len[i],
+						    &dev->caps.pkey_table_len[i]))
+			return -ENODEV;
+	}
 
 	if (dev->caps.uar_page_size * (dev->caps.num_uars -
 				       dev->caps.reserved_uars) >
@@ -553,7 +569,7 @@ int mlx4_change_port_types(struct mlx4_dev *dev,
 		for (port = 1; port <= dev->caps.num_ports; port++) {
 			mlx4_CLOSE_PORT(dev, port);
 			dev->caps.port_type[port] = port_types[port - 1];
-			err = mlx4_SET_PORT(dev, port);
+			err = mlx4_SET_PORT(dev, port, -1);
 			if (err) {
 				mlx4_err(dev, "Failed to set port %d, "
 					      "aborting\n", port);
@@ -739,7 +755,7 @@ static ssize_t set_port_ib_mtu(struct device *dev,
 	mlx4_unregister_device(mdev);
 	for (port = 1; port <= mdev->caps.num_ports; port++) {
 		mlx4_CLOSE_PORT(mdev, port);
-		err = mlx4_SET_PORT(mdev, port);
+		err = mlx4_SET_PORT(mdev, port, -1);
 		if (err) {
 			mlx4_err(mdev, "Failed to set port %d, "
 				      "aborting\n", port);
@@ -1192,6 +1208,17 @@ err:
 	return -EIO;
 }
 
+static void mlx4_parav_master_pf_caps(struct mlx4_dev *dev)
+{
+	int i;
+
+	for (i = 1; i <= dev->caps.num_ports; i++) {
+		dev->caps.gid_table_len[i] = 1;
+		dev->caps.pkey_table_len[i] =
+			dev->phys_caps.pkey_phys_table_len[i] - 1;
+	}
+}
+
 static int mlx4_init_hca(struct mlx4_dev *dev)
 {
 	struct mlx4_priv	  *priv = mlx4_priv(dev);
@@ -1231,6 +1258,9 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 			goto err_stop_fw;
 		}
 
+		if (mlx4_is_master(dev))
+			mlx4_parav_master_pf_caps(dev);
+
 		priv->fs_hash_mode = MLX4_FS_L2_HASH;
 
 		switch (priv->fs_hash_mode) {
@@ -1522,12 +1552,24 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 					  "with caps = 0\n", port, err);
 			dev->caps.ib_port_def_cap[port] = ib_port_default_caps;
 
+			/* initialize per-slave default ib port capabilities */
+			if (mlx4_is_master(dev)) {
+				int i;
+				for (i = 0; i < dev->num_slaves; i++) {
+					if (i == mlx4_master_func_num(dev))
+						continue;
+					priv->mfunc.master.slave_state[i].ib_cap_mask[port] =
+							ib_port_default_caps;
+				}
+			}
+
 			if (mlx4_is_mfunc(dev))
 				dev->caps.port_ib_mtu[port] = IB_MTU_2048;
 			else
 				dev->caps.port_ib_mtu[port] = IB_MTU_4096;
 
-			err = mlx4_SET_PORT(dev, port);
+			err = mlx4_SET_PORT(dev, port, mlx4_is_master(dev) ?
+					    dev->caps.pkey_table_len[port] : -1);
 			if (err) {
 				mlx4_err(dev, "Failed to set port %d, aborting\n",
 					port);
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index d2c436b10fb..59ebc033963 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -351,66 +351,6 @@ struct mlx4_srq_context {
 	__be64			db_rec_addr;
 };
 
-struct mlx4_eqe {
-	u8			reserved1;
-	u8			type;
-	u8			reserved2;
-	u8			subtype;
-	union {
-		u32		raw[6];
-		struct {
-			__be32	cqn;
-		} __packed comp;
-		struct {
-			u16	reserved1;
-			__be16	token;
-			u32	reserved2;
-			u8	reserved3[3];
-			u8	status;
-			__be64	out_param;
-		} __packed cmd;
-		struct {
-			__be32	qpn;
-		} __packed qp;
-		struct {
-			__be32	srqn;
-		} __packed srq;
-		struct {
-			__be32	cqn;
-			u32	reserved1;
-			u8	reserved2[3];
-			u8	syndrome;
-		} __packed cq_err;
-		struct {
-			u32	reserved1[2];
-			__be32	port;
-		} __packed port_change;
-		struct {
-			#define COMM_CHANNEL_BIT_ARRAY_SIZE	4
-			u32 reserved;
-			u32 bit_vec[COMM_CHANNEL_BIT_ARRAY_SIZE];
-		} __packed comm_channel_arm;
-		struct {
-			u8	port;
-			u8	reserved[3];
-			__be64	mac;
-		} __packed mac_update;
-		struct {
-			u8	port;
-		} __packed sw_event;
-		struct {
-			__be32	slave_id;
-		} __packed flr_event;
-		struct {
-			__be16  current_temperature;
-			__be16  warning_threshold;
-		} __packed warming;
-	}			event;
-	u8			slave_id;
-	u8			reserved3[2];
-	u8			owner;
-} __packed;
-
 struct mlx4_eq {
 	struct mlx4_dev	       *dev;
 	void __iomem	       *doorbell;
@@ -902,7 +842,8 @@ void mlx4_catas_init(void);
 int mlx4_restart_one(struct pci_dev *pdev);
 int mlx4_register_device(struct mlx4_dev *dev);
 void mlx4_unregister_device(struct mlx4_dev *dev);
-void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int port);
+void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type,
+			 unsigned long param);
 
 struct mlx4_dev_cap;
 struct mlx4_init_hca_param;
@@ -1043,7 +984,7 @@ int mlx4_change_port_types(struct mlx4_dev *dev,
 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table);
 void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table);
 
-int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port);
+int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz);
 /* resource tracker functions*/
 int mlx4_get_slave_from_resource_id(struct mlx4_dev *dev,
 				    enum mlx4_resource resource_type,
@@ -1086,6 +1027,8 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			    struct mlx4_cmd_info *cmd);
 int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps);
 
+int mlx4_get_slave_pkey_gid_tbl_len(struct mlx4_dev *dev, u8 port,
+				    int *gid_tbl_len, int *pkey_tbl_len);
 
 int mlx4_QP_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 			   struct mlx4_vhcr *vhcr,
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index 028833ffc56..e36dd0f2fa7 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -775,14 +775,15 @@ int mlx4_SET_PORT_wrapper(struct mlx4_dev *dev, int slave,
 enum {
 	MLX4_SET_PORT_VL_CAP	 = 4, /* bits 7:4 */
 	MLX4_SET_PORT_MTU_CAP	 = 12, /* bits 15:12 */
+	MLX4_CHANGE_PORT_PKEY_TBL_SZ = 20,
 	MLX4_CHANGE_PORT_VL_CAP	 = 21,
 	MLX4_CHANGE_PORT_MTU_CAP = 22,
 };
 
-int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port)
+int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz)
 {
 	struct mlx4_cmd_mailbox *mailbox;
-	int err, vl_cap;
+	int err, vl_cap, pkey_tbl_flag = 0;
 
 	if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH)
 		return 0;
@@ -795,11 +796,17 @@ int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port)
 
 	((__be32 *) mailbox->buf)[1] = dev->caps.ib_port_def_cap[port];
 
+	if (pkey_tbl_sz >= 0 && mlx4_is_master(dev)) {
+		pkey_tbl_flag = 1;
+		((__be16 *) mailbox->buf)[20] = cpu_to_be16(pkey_tbl_sz);
+	}
+
 	/* IB VL CAP enum isn't used by the firmware, just numerical values */
 	for (vl_cap = 8; vl_cap >= 1; vl_cap >>= 1) {
 		((__be32 *) mailbox->buf)[0] = cpu_to_be32(
 			(1 << MLX4_CHANGE_PORT_MTU_CAP) |
 			(1 << MLX4_CHANGE_PORT_VL_CAP)  |
+			(pkey_tbl_flag << MLX4_CHANGE_PORT_PKEY_TBL_SZ) |
 			(dev->caps.port_ib_mtu[port] << MLX4_SET_PORT_MTU_CAP) |
 			(vl_cap << MLX4_SET_PORT_VL_CAP));
 		err = mlx4_cmd(dev, mailbox->dma, port, 0, MLX4_CMD_SET_PORT,