Merge branches 'cma', 'cxgb4', 'flowsteer', 'ipoib', 'misc', 'mlx4', 'mlx5',...

Merge branches 'cma', 'cxgb4', 'flowsteer', 'ipoib', 'misc', 'mlx4', 'mlx5', 'ocrdma', 'qib', 'srp' and 'usnic' into for-next

Documentation/scsi/scsi_transport_srp/Makefile

+all: rport_state_diagram.svg rport_state_diagram.png
+
+rport_state_diagram.svg: rport_state_diagram.dot
+	dot -Tsvg -o $@ $<
+
+rport_state_diagram.png: rport_state_diagram.dot
+	dot -Tpng -o $@ $<

Documentation/scsi/scsi_transport_srp/rport_state_diagram.dot

+digraph srp_initiator {
+	node [shape = doublecircle]; running lost;
+	node [shape = circle];
+
+	{
+	  rank = min;
+	  running_rta	[ label = "running;\nreconnect\ntimer\nactive" ];
+	};
+	running		[ label = "running;\nreconnect\ntimer\nstopped" ];
+	blocked;
+	failfast	[ label = "fail I/O\nfast" ];
+	lost;
+
+	running -> running_rta		[ label = "fast_io_fail_tmo = off and\ndev_loss_tmo = off;\nsrp_start_tl_fail_timers()" ];
+	running_rta -> running		[ label = "fast_io_fail_tmo = off and\ndev_loss_tmo = off;\nreconnecting succeeded" ];
+	running -> blocked		[ label = "fast_io_fail_tmo >= 0 or\ndev_loss_tmo >= 0;\nsrp_start_tl_fail_timers()" ];
+	running -> failfast		[ label = "fast_io_fail_tmo = off and\ndev_loss_tmo = off;\nreconnecting failed\n" ];
+	blocked -> failfast		[ label = "fast_io_fail_tmo\nexpired or\nreconnecting\nfailed" ];
+	blocked -> lost			[ label = "dev_loss_tmo\nexpired or\nsrp_stop_rport_timers()" ];
+	failfast -> lost		[ label = "dev_loss_tmo\nexpired or\nsrp_stop_rport_timers()" ];
+	blocked -> running		[ label = "reconnecting\nsucceeded" ];
+	failfast -> failfast		[ label = "reconnecting\nfailed" ];
+	failfast -> running		[ label = "reconnecting\nsucceeded" ];
+	running -> lost			[ label = "srp_stop_rport_timers()" ];
+	running_rta -> lost		[ label = "srp_stop_rport_timers()" ];
+}

MAINTAINERS

@@ -2158,6 +2158,11 @@ M:	Nishank Trivedi <nistrive@cisco.com>
 S:	Supported
 F:	drivers/net/ethernet/cisco/enic/
 
+CISCO VIC LOW LATENCY NIC DRIVER
+M:      Upinder Malhi <umalhi@cisco.com>
+S:      Supported
+F:      drivers/infiniband/hw/usnic
+
 CIRRUS LOGIC EP93XX ETHERNET DRIVER
 M:	Hartley Sweeten <hsweeten@visionengravers.com>
 L:	netdev@vger.kernel.org
@@ -7468,7 +7473,7 @@ S:	Maintained
 F:	drivers/scsi/sr*
 
 SCSI RDMA PROTOCOL (SRP) INITIATOR
-M:	David Dillow <dillowda@ornl.gov>
+M:	Bart Van Assche <bvanassche@acm.org>
 L:	linux-rdma@vger.kernel.org
 S:	Supported
 W:	http://www.openfabrics.org

drivers/infiniband/Kconfig

@@ -53,6 +53,7 @@ source "drivers/infiniband/hw/mlx4/Kconfig"
 source "drivers/infiniband/hw/mlx5/Kconfig"
 source "drivers/infiniband/hw/nes/Kconfig"
 source "drivers/infiniband/hw/ocrdma/Kconfig"
+source "drivers/infiniband/hw/usnic/Kconfig"
 
 source "drivers/infiniband/ulp/ipoib/Kconfig"
 

drivers/infiniband/Makefile

@@ -10,6 +10,7 @@ obj-$(CONFIG_MLX4_INFINIBAND)		+= hw/mlx4/
 obj-$(CONFIG_MLX5_INFINIBAND)		+= hw/mlx5/
 obj-$(CONFIG_INFINIBAND_NES)		+= hw/nes/
 obj-$(CONFIG_INFINIBAND_OCRDMA)		+= hw/ocrdma/
+obj-$(CONFIG_INFINIBAND_USNIC)		+= hw/usnic/
 obj-$(CONFIG_INFINIBAND_IPOIB)		+= ulp/ipoib/
 obj-$(CONFIG_INFINIBAND_SRP)		+= ulp/srp/
 obj-$(CONFIG_INFINIBAND_SRPT)		+= ulp/srpt/

drivers/infiniband/core/iwcm.c

@@ -334,7 +334,6 @@ static void destroy_cm_id(struct iw_cm_id *cm_id)
 {
 	struct iwcm_id_private *cm_id_priv;
 	unsigned long flags;
-	int ret;
 
 	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
 	/*
@@ -350,7 +349,7 @@ static void destroy_cm_id(struct iw_cm_id *cm_id)
 		cm_id_priv->state = IW_CM_STATE_DESTROYING;
 		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
 		/* destroy the listening endpoint */
-		ret = cm_id->device->iwcm->destroy_listen(cm_id);
+		cm_id->device->iwcm->destroy_listen(cm_id);
 		spin_lock_irqsave(&cm_id_priv->lock, flags);
 		break;
 	case IW_CM_STATE_ESTABLISHED:

drivers/infiniband/core/sysfs.c

@@ -613,6 +613,7 @@ static ssize_t show_node_type(struct device *device,
 	case RDMA_NODE_IB_CA:	  return sprintf(buf, "%d: CA\n", dev->node_type);
 	case RDMA_NODE_RNIC:	  return sprintf(buf, "%d: RNIC\n", dev->node_type);
 	case RDMA_NODE_USNIC:	  return sprintf(buf, "%d: usNIC\n", dev->node_type);
+	case RDMA_NODE_USNIC_UDP: return sprintf(buf, "%d: usNIC UDP\n", dev->node_type);
 	case RDMA_NODE_IB_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type);
 	case RDMA_NODE_IB_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type);
 	default:		  return sprintf(buf, "%d: <unknown>\n", dev->node_type);

drivers/infiniband/core/verbs.c

@@ -116,6 +116,8 @@ rdma_node_get_transport(enum rdma_node_type node_type)
 		return RDMA_TRANSPORT_IWARP;
 	case RDMA_NODE_USNIC:
 		return RDMA_TRANSPORT_USNIC;
+	case RDMA_NODE_USNIC_UDP:
+		return RDMA_TRANSPORT_USNIC_UDP;
 	default:
 		BUG();
 		return 0;
@@ -133,6 +135,7 @@ enum rdma_link_layer rdma_port_get_link_layer(struct ib_device *device, u8 port_
 		return IB_LINK_LAYER_INFINIBAND;
 	case RDMA_TRANSPORT_IWARP:
 	case RDMA_TRANSPORT_USNIC:
+	case RDMA_TRANSPORT_USNIC_UDP:
 		return IB_LINK_LAYER_ETHERNET;
 	default:
 		return IB_LINK_LAYER_UNSPECIFIED;

drivers/infiniband/hw/amso1100/c2_intr.c

@@ -169,7 +169,8 @@ static void handle_vq(struct c2_dev *c2dev, u32 mq_index)
 		 * We should never get here, as the adapter should
 		 * never send us a reply that we're not expecting.
 		 */
-		vq_repbuf_free(c2dev, host_msg);
+		if (reply_msg != NULL)
+			vq_repbuf_free(c2dev, host_msg);
 		pr_debug("handle_vq: UNEXPECTEDLY got NULL req\n");
 		return;
 	}

drivers/infiniband/hw/cxgb4/mem.c

@@ -76,7 +76,7 @@ static int _c4iw_write_mem_dma_aligned(struct c4iw_rdev *rdev, u32 addr,
 	INIT_ULPTX_WR(req, wr_len, 0, 0);
 	req->wr.wr_hi = cpu_to_be32(FW_WR_OP(FW_ULPTX_WR) |
 			(wait ? FW_WR_COMPL(1) : 0));
-	req->wr.wr_lo = wait ? (__force __be64)&wr_wait : 0;
+	req->wr.wr_lo = wait ? (__force __be64)(unsigned long) &wr_wait : 0L;
 	req->wr.wr_mid = cpu_to_be32(FW_WR_LEN16(DIV_ROUND_UP(wr_len, 16)));
 	req->cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE));
 	req->cmd |= cpu_to_be32(V_T5_ULP_MEMIO_ORDER(1));

drivers/infiniband/hw/mlx4/main.c

@@ -55,6 +55,7 @@
 #define DRV_RELDATE	"April 4, 2008"
 
 #define MLX4_IB_FLOW_MAX_PRIO 0xFFF
+#define MLX4_IB_FLOW_QPN_MASK 0xFFFFFF
 
 MODULE_AUTHOR("Roland Dreier");
 MODULE_DESCRIPTION("Mellanox ConnectX HCA InfiniBand driver");
@@ -92,21 +93,27 @@ static union ib_gid zgid;
 
 static int check_flow_steering_support(struct mlx4_dev *dev)
 {
+	int eth_num_ports = 0;
 	int ib_num_ports = 0;
-	int i;
-
-	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB)
-		ib_num_ports++;
 
-	if (dev->caps.steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED) {
-		if (ib_num_ports || mlx4_is_mfunc(dev)) {
-			pr_warn("Device managed flow steering is unavailable "
-				"for IB ports or in multifunction env.\n");
-			return 0;
+	int dmfs = dev->caps.steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED;
+
+	if (dmfs) {
+		int i;
+		mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH)
+			eth_num_ports++;
+		mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB)
+			ib_num_ports++;
+		dmfs &= (!ib_num_ports ||
+			 (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_DMFS_IPOIB)) &&
+			(!eth_num_ports ||
+			 (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_FS_EN));
+		if (ib_num_ports && mlx4_is_mfunc(dev)) {
+			pr_warn("Device managed flow steering is unavailable for IB port in multifunction env.\n");
+			dmfs = 0;
 		}
-		return 1;
 	}
-	return 0;
+	return dmfs;
 }
 
 static int mlx4_ib_query_device(struct ib_device *ibdev,
@@ -165,7 +172,7 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 			props->device_cap_flags |= IB_DEVICE_MEM_WINDOW_TYPE_2B;
 		else
 			props->device_cap_flags |= IB_DEVICE_MEM_WINDOW_TYPE_2A;
-	if (check_flow_steering_support(dev->dev))
+	if (dev->steering_support ==  MLX4_STEERING_MODE_DEVICE_MANAGED)
 		props->device_cap_flags |= IB_DEVICE_MANAGED_FLOW_STEERING;
 	}
 
@@ -819,6 +826,7 @@ struct mlx4_ib_steering {
 };
 
 static int parse_flow_attr(struct mlx4_dev *dev,
+			   u32 qp_num,
 			   union ib_flow_spec *ib_spec,
 			   struct _rule_hw *mlx4_spec)
 {
@@ -834,6 +842,14 @@ static int parse_flow_attr(struct mlx4_dev *dev,
 		mlx4_spec->eth.vlan_tag = ib_spec->eth.val.vlan_tag;
 		mlx4_spec->eth.vlan_tag_msk = ib_spec->eth.mask.vlan_tag;
 		break;
+	case IB_FLOW_SPEC_IB:
+		type = MLX4_NET_TRANS_RULE_ID_IB;
+		mlx4_spec->ib.l3_qpn =
+			cpu_to_be32(qp_num);
+		mlx4_spec->ib.qpn_mask =
+			cpu_to_be32(MLX4_IB_FLOW_QPN_MASK);
+		break;
+
 
 	case IB_FLOW_SPEC_IPV4:
 		type = MLX4_NET_TRANS_RULE_ID_IPV4;
@@ -865,6 +881,115 @@ static int parse_flow_attr(struct mlx4_dev *dev,
 	return mlx4_hw_rule_sz(dev, type);
 }
 
+struct default_rules {
+	__u32 mandatory_fields[IB_FLOW_SPEC_SUPPORT_LAYERS];
+	__u32 mandatory_not_fields[IB_FLOW_SPEC_SUPPORT_LAYERS];
+	__u32 rules_create_list[IB_FLOW_SPEC_SUPPORT_LAYERS];
+	__u8  link_layer;
+};
+static const struct default_rules default_table[] = {
+	{
+		.mandatory_fields = {IB_FLOW_SPEC_IPV4},
+		.mandatory_not_fields = {IB_FLOW_SPEC_ETH},
+		.rules_create_list = {IB_FLOW_SPEC_IB},
+		.link_layer = IB_LINK_LAYER_INFINIBAND
+	}
+};
+
+static int __mlx4_ib_default_rules_match(struct ib_qp *qp,
+					 struct ib_flow_attr *flow_attr)
+{
+	int i, j, k;
+	void *ib_flow;
+	const struct default_rules *pdefault_rules = default_table;
+	u8 link_layer = rdma_port_get_link_layer(qp->device, flow_attr->port);
+
+	for (i = 0; i < sizeof(default_table)/sizeof(default_table[0]); i++,
+	     pdefault_rules++) {
+		__u32 field_types[IB_FLOW_SPEC_SUPPORT_LAYERS];
+		memset(&field_types, 0, sizeof(field_types));
+
+		if (link_layer != pdefault_rules->link_layer)
+			continue;
+
+		ib_flow = flow_attr + 1;
+		/* we assume the specs are sorted */
+		for (j = 0, k = 0; k < IB_FLOW_SPEC_SUPPORT_LAYERS &&
+		     j < flow_attr->num_of_specs; k++) {
+			union ib_flow_spec *current_flow =
+				(union ib_flow_spec *)ib_flow;
+
+			/* same layer but different type */
+			if (((current_flow->type & IB_FLOW_SPEC_LAYER_MASK) ==
+			     (pdefault_rules->mandatory_fields[k] &
+			      IB_FLOW_SPEC_LAYER_MASK)) &&
+			    (current_flow->type !=
+			     pdefault_rules->mandatory_fields[k]))
+				goto out;
+
+			/* same layer, try match next one */
+			if (current_flow->type ==
+			    pdefault_rules->mandatory_fields[k]) {
+				j++;
+				ib_flow +=
+					((union ib_flow_spec *)ib_flow)->size;
+			}
+		}
+
+		ib_flow = flow_attr + 1;
+		for (j = 0; j < flow_attr->num_of_specs;
+		     j++, ib_flow += ((union ib_flow_spec *)ib_flow)->size)
+			for (k = 0; k < IB_FLOW_SPEC_SUPPORT_LAYERS; k++)
+				/* same layer and same type */
+				if (((union ib_flow_spec *)ib_flow)->type ==
+				    pdefault_rules->mandatory_not_fields[k])
+					goto out;
+
+		return i;
+	}
+out:
+	return -1;
+}
+
+static int __mlx4_ib_create_default_rules(
+		struct mlx4_ib_dev *mdev,
+		struct ib_qp *qp,
+		const struct default_rules *pdefault_rules,
+		struct _rule_hw *mlx4_spec) {
+	int size = 0;
+	int i;
+
+	for (i = 0; i < sizeof(pdefault_rules->rules_create_list)/
+			sizeof(pdefault_rules->rules_create_list[0]); i++) {
+		int ret;
+		union ib_flow_spec ib_spec;
+		switch (pdefault_rules->rules_create_list[i]) {
+		case 0:
+			/* no rule */
+			continue;
+		case IB_FLOW_SPEC_IB:
+			ib_spec.type = IB_FLOW_SPEC_IB;
+			ib_spec.size = sizeof(struct ib_flow_spec_ib);
+
+			break;
+		default:
+			/* invalid rule */
+			return -EINVAL;
+		}
+		/* We must put empty rule, qpn is being ignored */
+		ret = parse_flow_attr(mdev->dev, 0, &ib_spec,
+				      mlx4_spec);
+		if (ret < 0) {
+			pr_info("invalid parsing\n");
+			return -EINVAL;
+		}
+
+		mlx4_spec = (void *)mlx4_spec + ret;
+		size += ret;
+	}
+	return size;
+}
+
 static int __mlx4_ib_create_flow(struct ib_qp *qp, struct ib_flow_attr *flow_attr,
 			  int domain,
 			  enum mlx4_net_trans_promisc_mode flow_type,
@@ -876,6 +1001,7 @@ static int __mlx4_ib_create_flow(struct ib_qp *qp, struct ib_flow_attr *flow_att
 	struct mlx4_ib_dev *mdev = to_mdev(qp->device);
 	struct mlx4_cmd_mailbox *mailbox;
 	struct mlx4_net_trans_rule_hw_ctrl *ctrl;
+	int default_flow;
 
 	static const u16 __mlx4_domain[] = {
 		[IB_FLOW_DOMAIN_USER] = MLX4_DOMAIN_UVERBS,
@@ -910,8 +1036,21 @@ static int __mlx4_ib_create_flow(struct ib_qp *qp, struct ib_flow_attr *flow_att
 
 	ib_flow = flow_attr + 1;
 	size += sizeof(struct mlx4_net_trans_rule_hw_ctrl);
+	/* Add default flows */
+	default_flow = __mlx4_ib_default_rules_match(qp, flow_attr);
+	if (default_flow >= 0) {
+		ret = __mlx4_ib_create_default_rules(
+				mdev, qp, default_table + default_flow,
+				mailbox->buf + size);
+		if (ret < 0) {
+			mlx4_free_cmd_mailbox(mdev->dev, mailbox);
+			return -EINVAL;
+		}
+		size += ret;
+	}
 	for (i = 0; i < flow_attr->num_of_specs; i++) {
-		ret = parse_flow_attr(mdev->dev, ib_flow, mailbox->buf + size);
+		ret = parse_flow_attr(mdev->dev, qp->qp_num, ib_flow,
+				      mailbox->buf + size);
 		if (ret < 0) {
 			mlx4_free_cmd_mailbox(mdev->dev, mailbox);
 			return -EINVAL;
@@ -1682,6 +1821,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	}
 
 	if (check_flow_steering_support(dev)) {
+		ibdev->steering_support = MLX4_STEERING_MODE_DEVICE_MANAGED;
 		ibdev->ib_dev.create_flow	= mlx4_ib_create_flow;
 		ibdev->ib_dev.destroy_flow	= mlx4_ib_destroy_flow;
 
@@ -1710,8 +1850,35 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	spin_lock_init(&ibdev->sm_lock);
 	mutex_init(&ibdev->cap_mask_mutex);
 
+	if (ibdev->steering_support == MLX4_STEERING_MODE_DEVICE_MANAGED) {
+		ibdev->steer_qpn_count = MLX4_IB_UC_MAX_NUM_QPS;
+		err = mlx4_qp_reserve_range(dev, ibdev->steer_qpn_count,
+					    MLX4_IB_UC_STEER_QPN_ALIGN,
+					    &ibdev->steer_qpn_base);
+		if (err)
+			goto err_counter;
+
+		ibdev->ib_uc_qpns_bitmap =
+			kmalloc(BITS_TO_LONGS(ibdev->steer_qpn_count) *
+				sizeof(long),
+				GFP_KERNEL);
+		if (!ibdev->ib_uc_qpns_bitmap) {
+			dev_err(&dev->pdev->dev, "bit map alloc failed\n");
+			goto err_steer_qp_release;
+		}
+
+		bitmap_zero(ibdev->ib_uc_qpns_bitmap, ibdev->steer_qpn_count);
+
+		err = mlx4_FLOW_STEERING_IB_UC_QP_RANGE(
+				dev, ibdev->steer_qpn_base,
+				ibdev->steer_qpn_base +
+				ibdev->steer_qpn_count - 1);
+		if (err)
+			goto err_steer_free_bitmap;
+	}
+
 	if (ib_register_device(&ibdev->ib_dev, NULL))
-		goto err_counter;
+		goto err_steer_free_bitmap;
 
 	if (mlx4_ib_mad_init(ibdev))
 		goto err_reg;
@@ -1762,6 +1929,13 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 err_reg:
 	ib_unregister_device(&ibdev->ib_dev);
 
+err_steer_free_bitmap:
+	kfree(ibdev->ib_uc_qpns_bitmap);
+
+err_steer_qp_release:
+	if (ibdev->steering_support == MLX4_STEERING_MODE_DEVICE_MANAGED)
+		mlx4_qp_release_range(dev, ibdev->steer_qpn_base,
+				      ibdev->steer_qpn_count);
 err_counter:
 	for (; i; --i)
 		if (ibdev->counters[i - 1] != -1)
@@ -1782,6 +1956,69 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	return NULL;
 }
 
+int mlx4_ib_steer_qp_alloc(struct mlx4_ib_dev *dev, int count, int *qpn)
+{
+	int offset;
+
+	WARN_ON(!dev->ib_uc_qpns_bitmap);
+
+	offset = bitmap_find_free_region(dev->ib_uc_qpns_bitmap,
+					 dev->steer_qpn_count,
+					 get_count_order(count));
+	if (offset < 0)
+		return offset;
+
+	*qpn = dev->steer_qpn_base + offset;
+	return 0;
+}
+
+void mlx4_ib_steer_qp_free(struct mlx4_ib_dev *dev, u32 qpn, int count)
+{
+	if (!qpn ||
+	    dev->steering_support != MLX4_STEERING_MODE_DEVICE_MANAGED)
+		return;
+
+	BUG_ON(qpn < dev->steer_qpn_base);
+
+	bitmap_release_region(dev->ib_uc_qpns_bitmap,
+			      qpn - dev->steer_qpn_base,
+			      get_count_order(count));
+}
+
+int mlx4_ib_steer_qp_reg(struct mlx4_ib_dev *mdev, struct mlx4_ib_qp *mqp,
+			 int is_attach)
+{
+	int err;
+	size_t flow_size;
+	struct ib_flow_attr *flow = NULL;
+	struct ib_flow_spec_ib *ib_spec;
+
+	if (is_attach) {
+		flow_size = sizeof(struct ib_flow_attr) +
+			    sizeof(struct ib_flow_spec_ib);
+		flow = kzalloc(flow_size, GFP_KERNEL);
+		if (!flow)
+			return -ENOMEM;
+		flow->port = mqp->port;
+		flow->num_of_specs = 1;
+		flow->size = flow_size;
+		ib_spec = (struct ib_flow_spec_ib *)(flow + 1);
+		ib_spec->type = IB_FLOW_SPEC_IB;
+		ib_spec->size = sizeof(struct ib_flow_spec_ib);
+		/* Add an empty rule for IB L2 */
+		memset(&ib_spec->mask, 0, sizeof(ib_spec->mask));
+
+		err = __mlx4_ib_create_flow(&mqp->ibqp, flow,
+					    IB_FLOW_DOMAIN_NIC,
+					    MLX4_FS_REGULAR,
+					    &mqp->reg_id);
+	} else {
+		err = __mlx4_ib_destroy_flow(mdev->dev, mqp->reg_id);
+	}
+	kfree(flow);
+	return err;
+}
+
 static void mlx4_ib_remove(struct mlx4_dev *dev, void *ibdev_ptr)
 {
 	struct mlx4_ib_dev *ibdev = ibdev_ptr;
@@ -1795,6 +2032,13 @@ static void mlx4_ib_remove(struct mlx4_dev *dev, void *ibdev_ptr)
 			pr_warn("failure unregistering notifier\n");
 		ibdev->iboe.nb.notifier_call = NULL;
 	}
+
+	if (ibdev->steering_support == MLX4_STEERING_MODE_DEVICE_MANAGED) {
+		mlx4_qp_release_range(dev, ibdev->steer_qpn_base,
+				      ibdev->steer_qpn_count);
+		kfree(ibdev->ib_uc_qpns_bitmap);
+	}
+
 	iounmap(ibdev->uar_map);
 	for (p = 0; p < ibdev->num_ports; ++p)
 		if (ibdev->counters[p] != -1)

drivers/infiniband/hw/mlx4/mlx4_ib.h

@@ -68,6 +68,8 @@ enum {
 /*module param to indicate if SM assigns the alias_GUID*/
 extern int mlx4_ib_sm_guid_assign;
 
+#define MLX4_IB_UC_STEER_QPN_ALIGN 1
+#define MLX4_IB_UC_MAX_NUM_QPS     256
 struct mlx4_ib_ucontext {
 	struct ib_ucontext	ibucontext;
 	struct mlx4_uar		uar;
@@ -153,6 +155,7 @@ struct mlx4_ib_wq {
 enum mlx4_ib_qp_flags {
 	MLX4_IB_QP_LSO = IB_QP_CREATE_IPOIB_UD_LSO,
 	MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK = IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK,
+	MLX4_IB_QP_NETIF = IB_QP_CREATE_NETIF_QP,
 	MLX4_IB_SRIOV_TUNNEL_QP = 1 << 30,
 	MLX4_IB_SRIOV_SQP = 1 << 31,
 };
@@ -270,6 +273,7 @@ struct mlx4_ib_qp {
 	struct list_head	gid_list;
 	struct list_head	steering_rules;
 	struct mlx4_ib_buf	*sqp_proxy_rcv;
+	u64			reg_id;
 
 };
 
@@ -494,6 +498,10 @@ struct mlx4_ib_dev {
 	struct kobject	       *dev_ports_parent[MLX4_MFUNC_MAX];
 	struct mlx4_ib_iov_port	iov_ports[MLX4_MAX_PORTS];
 	struct pkey_mgt		pkeys;
+	unsigned long *ib_uc_qpns_bitmap;
+	int steer_qpn_count;
+	int steer_qpn_base;
+	int steering_support;
 };
 
 struct ib_event_work {
@@ -752,5 +760,9 @@ void mlx4_ib_device_unregister_sysfs(struct mlx4_ib_dev *device);
 
 __be64 mlx4_ib_gen_node_guid(void);
 
+int mlx4_ib_steer_qp_alloc(struct mlx4_ib_dev *dev, int count, int *qpn);
+void mlx4_ib_steer_qp_free(struct mlx4_ib_dev *dev, u32 qpn, int count);
+int mlx4_ib_steer_qp_reg(struct mlx4_ib_dev *mdev, struct mlx4_ib_qp *mqp,
+			 int is_attach);
 
 #endif /* MLX4_IB_H */

drivers/infiniband/hw/mlx4/qp.c

@@ -716,6 +716,14 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO)
 			qp->flags |= MLX4_IB_QP_LSO;
 
+		if (init_attr->create_flags & IB_QP_CREATE_NETIF_QP) {
+			if (dev->steering_support ==
+			    MLX4_STEERING_MODE_DEVICE_MANAGED)
+				qp->flags |= MLX4_IB_QP_NETIF;
+			else
+				goto err;
+		}
+
 		err = set_kernel_sq_size(dev, &init_attr->cap, qp_type, qp);
 		if (err)
 			goto err;
@@ -765,7 +773,11 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		if (init_attr->qp_type == IB_QPT_RAW_PACKET)
 			err = mlx4_qp_reserve_range(dev->dev, 1, 1 << 8, &qpn);
 		else
-			err = mlx4_qp_reserve_range(dev->dev, 1, 1, &qpn);
+			if (qp->flags & MLX4_IB_QP_NETIF)
+				err = mlx4_ib_steer_qp_alloc(dev, 1, &qpn);
+			else
+				err = mlx4_qp_reserve_range(dev->dev, 1, 1,
+							    &qpn);
 		if (err)
 			goto err_proxy;
 	}
@@ -790,8 +802,12 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 	return 0;
 
 err_qpn:
-	if (!sqpn)
-		mlx4_qp_release_range(dev->dev, qpn, 1);
+	if (!sqpn) {
+		if (qp->flags & MLX4_IB_QP_NETIF)
+			mlx4_ib_steer_qp_free(dev, qpn, 1);
+		else
+			mlx4_qp_release_range(dev->dev, qpn, 1);
+	}
 err_proxy:
 	if (qp->mlx4_ib_qp_type == MLX4_IB_QPT_PROXY_GSI)
 		free_proxy_bufs(pd->device, qp);
@@ -932,8 +948,12 @@ static void destroy_qp_common(struct mlx4_ib_dev *dev, struct mlx4_ib_qp *qp,
 
 	mlx4_qp_free(dev->dev, &qp->mqp);
 
-	if (!is_sqp(dev, qp) && !is_tunnel_qp(dev, qp))
-		mlx4_qp_release_range(dev->dev, qp->mqp.qpn, 1);
+	if (!is_sqp(dev, qp) && !is_tunnel_qp(dev, qp)) {
+		if (qp->flags & MLX4_IB_QP_NETIF)
+			mlx4_ib_steer_qp_free(dev, qp->mqp.qpn, 1);
+		else
+			mlx4_qp_release_range(dev->dev, qp->mqp.qpn, 1);
+	}
 
 	mlx4_mtt_cleanup(dev->dev, &qp->mtt);
 
@@ -987,9 +1007,16 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	 */
 	if (init_attr->create_flags & ~(MLX4_IB_QP_LSO |
 					MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK |
-					MLX4_IB_SRIOV_TUNNEL_QP | MLX4_IB_SRIOV_SQP))
+					MLX4_IB_SRIOV_TUNNEL_QP |
+					MLX4_IB_SRIOV_SQP |
+					MLX4_IB_QP_NETIF))
 		return ERR_PTR(-EINVAL);
 
+	if (init_attr->create_flags & IB_QP_CREATE_NETIF_QP) {
+		if (init_attr->qp_type != IB_QPT_UD)
+			return ERR_PTR(-EINVAL);
+	}
+
 	if (init_attr->create_flags &&
 	    (udata ||
 	     ((init_attr->create_flags & ~MLX4_IB_SRIOV_SQP) &&
@@ -1235,6 +1262,7 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
 	struct mlx4_qp_context *context;
 	enum mlx4_qp_optpar optpar = 0;
 	int sqd_event;
+	int steer_qp = 0;
 	int err = -EINVAL;
 
 	context = kzalloc(sizeof *context, GFP_KERNEL);
@@ -1319,6 +1347,11 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
 			optpar |= MLX4_QP_OPTPAR_COUNTER_INDEX;
 		} else
 			context->pri_path.counter_index = 0xff;
+
+		if (qp->flags & MLX4_IB_QP_NETIF) {
+			mlx4_ib_steer_qp_reg(dev, qp, 1);
+			steer_qp = 1;
+		}
 	}
 
 	if (attr_mask & IB_QP_PKEY_INDEX) {
@@ -1547,9 +1580,14 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
 		qp->sq_next_wqe = 0;
 		if (qp->rq.wqe_cnt)
 			*qp->db.db  = 0;
+
+		if (qp->flags & MLX4_IB_QP_NETIF)
+			mlx4_ib_steer_qp_reg(dev, qp, 0);
 	}
 
 out:
+	if (err && steer_qp)
+		mlx4_ib_steer_qp_reg(dev, qp, 0);
 	kfree(context);
 	return err;
 }
@@ -2762,6 +2800,9 @@ int mlx4_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr
 	if (qp->flags & MLX4_IB_QP_LSO)
 		qp_init_attr->create_flags |= IB_QP_CREATE_IPOIB_UD_LSO;
 
+	if (qp->flags & MLX4_IB_QP_NETIF)
+		qp_init_attr->create_flags |= IB_QP_CREATE_NETIF_QP;
+
 	qp_init_attr->sq_sig_type =
 		qp->sq_signal_bits == cpu_to_be32(MLX4_WQE_CTRL_CQ_UPDATE) ?
 		IB_SIGNAL_ALL_WR : IB_SIGNAL_REQ_WR;

drivers/infiniband/hw/mlx4/sysfs.c

@@ -582,8 +582,10 @@ static int add_port(struct mlx4_ib_dev *dev, int port_num, int slave)
 	p->pkey_group.attrs =
 		alloc_group_attrs(show_port_pkey, store_port_pkey,
 				  dev->dev->caps.pkey_table_len[port_num]);
-	if (!p->pkey_group.attrs)
+	if (!p->pkey_group.attrs) {
+		ret = -ENOMEM;
 		goto err_alloc;
+	}
 
 	ret = sysfs_create_group(&p->kobj, &p->pkey_group);
 	if (ret)
@@ -591,8 +593,10 @@ static int add_port(struct mlx4_ib_dev *dev, int port_num, int slave)
 
 	p->gid_group.name  = "gid_idx";
 	p->gid_group.attrs = alloc_group_attrs(show_port_gid_idx, NULL, 1);
-	if (!p->gid_group.attrs)
+	if (!p->gid_group.attrs) {
+		ret = -ENOMEM;
 		goto err_free_pkey;
+	}
 
 	ret = sysfs_create_group(&p->kobj, &p->gid_group);
 	if (ret)

drivers/infiniband/hw/mlx5/cq.c

@@ -73,14 +73,24 @@ static void *get_cqe(struct mlx5_ib_cq *cq, int n)
 	return get_cqe_from_buf(&cq->buf, n, cq->mcq.cqe_sz);
 }
 
+static u8 sw_ownership_bit(int n, int nent)
+{
+	return (n & nent) ? 1 : 0;
+}
+
 static void *get_sw_cqe(struct mlx5_ib_cq *cq, int n)
 {
 	void *cqe = get_cqe(cq, n & cq->ibcq.cqe);
 	struct mlx5_cqe64 *cqe64;
 
 	cqe64 = (cq->mcq.cqe_sz == 64) ? cqe : cqe + 64;
-	return ((cqe64->op_own & MLX5_CQE_OWNER_MASK) ^
-		!!(n & (cq->ibcq.cqe + 1))) ? NULL : cqe;
+
+	if (likely((cqe64->op_own) >> 4 != MLX5_CQE_INVALID) &&
+	    !((cqe64->op_own & MLX5_CQE_OWNER_MASK) ^ !!(n & (cq->ibcq.cqe + 1)))) {
+		return cqe;
+	} else {
+		return NULL;
+	}
 }
 
 static void *next_cqe_sw(struct mlx5_ib_cq *cq)
@@ -351,6 +361,11 @@ static void handle_atomics(struct mlx5_ib_qp *qp, struct mlx5_cqe64 *cqe64,
 	qp->sq.last_poll = tail;
 }
 
+static void free_cq_buf(struct mlx5_ib_dev *dev, struct mlx5_ib_cq_buf *buf)
+{
+	mlx5_buf_free(&dev->mdev, &buf->buf);
+}
+
 static int mlx5_poll_one(struct mlx5_ib_cq *cq,
 			 struct mlx5_ib_qp **cur_qp,
 			 struct ib_wc *wc)
@@ -366,6 +381,7 @@ static int mlx5_poll_one(struct mlx5_ib_cq *cq,
 	void *cqe;
 	int idx;
 
+repoll:
 	cqe = next_cqe_sw(cq);
 	if (!cqe)
 		return -EAGAIN;
@@ -379,7 +395,18 @@ static int mlx5_poll_one(struct mlx5_ib_cq *cq,
 	 */
 	rmb();
 
-	/* TBD: resize CQ */
+	opcode = cqe64->op_own >> 4;
+	if (unlikely(opcode == MLX5_CQE_RESIZE_CQ)) {
+		if (likely(cq->resize_buf)) {
+			free_cq_buf(dev, &cq->buf);
+			cq->buf = *cq->resize_buf;
+			kfree(cq->resize_buf);
+			cq->resize_buf = NULL;
+			goto repoll;
+		} else {
+			mlx5_ib_warn(dev, "unexpected resize cqe\n");
+		}
+	}
 
 	qpn = ntohl(cqe64->sop_drop_qpn) & 0xffffff;
 	if (!*cur_qp || (qpn != (*cur_qp)->ibqp.qp_num)) {
@@ -398,7 +425,6 @@ static int mlx5_poll_one(struct mlx5_ib_cq *cq,
 	}
 
 	wc->qp  = &(*cur_qp)->ibqp;
-	opcode = cqe64->op_own >> 4;
 	switch (opcode) {
 	case MLX5_CQE_REQ:
 		wq = &(*cur_qp)->sq;
@@ -503,15 +529,11 @@ static int alloc_cq_buf(struct mlx5_ib_dev *dev, struct mlx5_ib_cq_buf *buf,
 		return err;
 
 	buf->cqe_size = cqe_size;
+	buf->nent = nent;
 
 	return 0;
 }
 
-static void free_cq_buf(struct mlx5_ib_dev *dev, struct mlx5_ib_cq_buf *buf)
-{
-	mlx5_buf_free(&dev->mdev, &buf->buf);
-}
-
 static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 			  struct ib_ucontext *context, struct mlx5_ib_cq *cq,
 			  int entries, struct mlx5_create_cq_mbox_in **cqb,
@@ -576,16 +598,16 @@ static void destroy_cq_user(struct mlx5_ib_cq *cq, struct ib_ucontext *context)
 	ib_umem_release(cq->buf.umem);
 }
 
-static void init_cq_buf(struct mlx5_ib_cq *cq, int nent)
+static void init_cq_buf(struct mlx5_ib_cq *cq, struct mlx5_ib_cq_buf *buf)
 {
 	int i;
 	void *cqe;
 	struct mlx5_cqe64 *cqe64;
 
-	for (i = 0; i < nent; i++) {
-		cqe = get_cqe(cq, i);
-		cqe64 = (cq->buf.cqe_size == 64) ? cqe : cqe + 64;
-		cqe64->op_own = 0xf1;
+	for (i = 0; i < buf->nent; i++) {
+		cqe = get_cqe_from_buf(buf, i, buf->cqe_size);
+		cqe64 = buf->cqe_size == 64 ? cqe : cqe + 64;
+		cqe64->op_own = MLX5_CQE_INVALID << 4;
 	}
 }
 
@@ -610,7 +632,7 @@ static int create_cq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq,
 	if (err)
 		goto err_db;
 
-	init_cq_buf(cq, entries);
+	init_cq_buf(cq, &cq->buf);
 
 	*inlen = sizeof(**cqb) + sizeof(*(*cqb)->pas) * cq->buf.buf.npages;
 	*cqb = mlx5_vzalloc(*inlen);
@@ -818,12 +840,266 @@ void mlx5_ib_cq_clean(struct mlx5_ib_cq *cq, u32 qpn, struct mlx5_ib_srq *srq)
 
 int mlx5_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period)
 {
-	return -ENOSYS;
+	struct mlx5_modify_cq_mbox_in *in;
+	struct mlx5_ib_dev *dev = to_mdev(cq->device);
+	struct mlx5_ib_cq *mcq = to_mcq(cq);
+	int err;
+	u32 fsel;
+
+	if (!(dev->mdev.caps.flags & MLX5_DEV_CAP_FLAG_CQ_MODER))
+		return -ENOSYS;
+
+	in = kzalloc(sizeof(*in), GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	in->cqn = cpu_to_be32(mcq->mcq.cqn);
+	fsel = (MLX5_CQ_MODIFY_PERIOD | MLX5_CQ_MODIFY_COUNT);
+	in->ctx.cq_period = cpu_to_be16(cq_period);
+	in->ctx.cq_max_count = cpu_to_be16(cq_count);
+	in->field_select = cpu_to_be32(fsel);
+	err = mlx5_core_modify_cq(&dev->mdev, &mcq->mcq, in, sizeof(*in));
+	kfree(in);
+
+	if (err)
+		mlx5_ib_warn(dev, "modify cq 0x%x failed\n", mcq->mcq.cqn);
+
+	return err;
+}
+
+static int resize_user(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq,
+		       int entries, struct ib_udata *udata, int *npas,
+		       int *page_shift, int *cqe_size)
+{
+	struct mlx5_ib_resize_cq ucmd;
+	struct ib_umem *umem;
+	int err;
+	int npages;
+	struct ib_ucontext *context = cq->buf.umem->context;
+
+	err = ib_copy_from_udata(&ucmd, udata, sizeof(ucmd));
+	if (err)
+		return err;
+
+	if (ucmd.reserved0 || ucmd.reserved1)
+		return -EINVAL;
+
+	umem = ib_umem_get(context, ucmd.buf_addr, entries * ucmd.cqe_size,
+			   IB_ACCESS_LOCAL_WRITE, 1);
+	if (IS_ERR(umem)) {
+		err = PTR_ERR(umem);
+		return err;
+	}
+
+	mlx5_ib_cont_pages(umem, ucmd.buf_addr, &npages, page_shift,
+			   npas, NULL);
+
+	cq->resize_umem = umem;
+	*cqe_size = ucmd.cqe_size;
+
+	return 0;
+}
+
+static void un_resize_user(struct mlx5_ib_cq *cq)
+{
+	ib_umem_release(cq->resize_umem);
+}
+
+static int resize_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq,
+			 int entries, int cqe_size)
+{
+	int err;
+
+	cq->resize_buf = kzalloc(sizeof(*cq->resize_buf), GFP_KERNEL);
+	if (!cq->resize_buf)
+		return -ENOMEM;
+
+	err = alloc_cq_buf(dev, cq->resize_buf, entries, cqe_size);
+	if (err)
+		goto ex;
+
+	init_cq_buf(cq, cq->resize_buf);
+
+	return 0;
+
+ex:
+	kfree(cq->resize_buf);
+	return err;
+}
+
+static void un_resize_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq)
+{
+	free_cq_buf(dev, cq->resize_buf);
+	cq->resize_buf = NULL;
+}
+
+static int copy_resize_cqes(struct mlx5_ib_cq *cq)
+{
+	struct mlx5_ib_dev *dev = to_mdev(cq->ibcq.device);
+	struct mlx5_cqe64 *scqe64;
+	struct mlx5_cqe64 *dcqe64;
+	void *start_cqe;
+	void *scqe;
+	void *dcqe;
+	int ssize;
+	int dsize;
+	int i;
+	u8 sw_own;
+
+	ssize = cq->buf.cqe_size;
+	dsize = cq->resize_buf->cqe_size;
+	if (ssize != dsize) {
+		mlx5_ib_warn(dev, "resize from different cqe size is not supported\n");
+		return -EINVAL;
+	}
+
+	i = cq->mcq.cons_index;
+	scqe = get_sw_cqe(cq, i);
+	scqe64 = ssize == 64 ? scqe : scqe + 64;
+	start_cqe = scqe;
+	if (!scqe) {
+		mlx5_ib_warn(dev, "expected cqe in sw ownership\n");
+		return -EINVAL;
+	}
+
+	while ((scqe64->op_own >> 4) != MLX5_CQE_RESIZE_CQ) {
+		dcqe = get_cqe_from_buf(cq->resize_buf,
+					(i + 1) & (cq->resize_buf->nent),
+					dsize);
+		dcqe64 = dsize == 64 ? dcqe : dcqe + 64;
+		sw_own = sw_ownership_bit(i + 1, cq->resize_buf->nent);
+		memcpy(dcqe, scqe, dsize);
+		dcqe64->op_own = (dcqe64->op_own & ~MLX5_CQE_OWNER_MASK) | sw_own;
+
+		++i;
+		scqe = get_sw_cqe(cq, i);
+		scqe64 = ssize == 64 ? scqe : scqe + 64;
+		if (!scqe) {
+			mlx5_ib_warn(dev, "expected cqe in sw ownership\n");
+			return -EINVAL;
+		}
+
+		if (scqe == start_cqe) {
+			pr_warn("resize CQ failed to get resize CQE, CQN 0x%x\n",
+				cq->mcq.cqn);
+			return -ENOMEM;
+		}
+	}
+	++cq->mcq.cons_index;
+	return 0;
 }
 
 int mlx5_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 {
-	return -ENOSYS;
+	struct mlx5_ib_dev *dev = to_mdev(ibcq->device);
+	struct mlx5_ib_cq *cq = to_mcq(ibcq);
+	struct mlx5_modify_cq_mbox_in *in;
+	int err;
+	int npas;
+	int page_shift;
+	int inlen;
+	int uninitialized_var(cqe_size);
+	unsigned long flags;
+
+	if (!(dev->mdev.caps.flags & MLX5_DEV_CAP_FLAG_RESIZE_CQ)) {
+		pr_info("Firmware does not support resize CQ\n");
+		return -ENOSYS;
+	}
+
+	if (entries < 1)
+		return -EINVAL;
+
+	entries = roundup_pow_of_two(entries + 1);
+	if (entries > dev->mdev.caps.max_cqes + 1)
+		return -EINVAL;
+
+	if (entries == ibcq->cqe + 1)
+		return 0;
+
+	mutex_lock(&cq->resize_mutex);
+	if (udata) {
+		err = resize_user(dev, cq, entries, udata, &npas, &page_shift,
+				  &cqe_size);
+	} else {
+		cqe_size = 64;
+		err = resize_kernel(dev, cq, entries, cqe_size);
+		if (!err) {
+			npas = cq->resize_buf->buf.npages;
+			page_shift = cq->resize_buf->buf.page_shift;
+		}
+	}
+
+	if (err)
+		goto ex;
+
+	inlen = sizeof(*in) + npas * sizeof(in->pas[0]);
+	in = mlx5_vzalloc(inlen);
+	if (!in) {
+		err = -ENOMEM;
+		goto ex_resize;
+	}
+
+	if (udata)
+		mlx5_ib_populate_pas(dev, cq->resize_umem, page_shift,
+				     in->pas, 0);
+	else
+		mlx5_fill_page_array(&cq->resize_buf->buf, in->pas);
+
+	in->field_select = cpu_to_be32(MLX5_MODIFY_CQ_MASK_LOG_SIZE  |
+				       MLX5_MODIFY_CQ_MASK_PG_OFFSET |
+				       MLX5_MODIFY_CQ_MASK_PG_SIZE);
+	in->ctx.log_pg_sz = page_shift - MLX5_ADAPTER_PAGE_SHIFT;
+	in->ctx.cqe_sz_flags = cqe_sz_to_mlx_sz(cqe_size) << 5;
+	in->ctx.page_offset = 0;
+	in->ctx.log_sz_usr_page = cpu_to_be32(ilog2(entries) << 24);
+	in->hdr.opmod = cpu_to_be16(MLX5_CQ_OPMOD_RESIZE);
+	in->cqn = cpu_to_be32(cq->mcq.cqn);
+
+	err = mlx5_core_modify_cq(&dev->mdev, &cq->mcq, in, inlen);
+	if (err)
+		goto ex_alloc;
+
+	if (udata) {
+		cq->ibcq.cqe = entries - 1;
+		ib_umem_release(cq->buf.umem);
+		cq->buf.umem = cq->resize_umem;
+		cq->resize_umem = NULL;
+	} else {
+		struct mlx5_ib_cq_buf tbuf;
+		int resized = 0;
+
+		spin_lock_irqsave(&cq->lock, flags);
+		if (cq->resize_buf) {
+			err = copy_resize_cqes(cq);
+			if (!err) {
+				tbuf = cq->buf;
+				cq->buf = *cq->resize_buf;
+				kfree(cq->resize_buf);
+				cq->resize_buf = NULL;
+				resized = 1;
+			}
+		}
+		cq->ibcq.cqe = entries - 1;
+		spin_unlock_irqrestore(&cq->lock, flags);
+		if (resized)
+			free_cq_buf(dev, &tbuf);
+	}
+	mutex_unlock(&cq->resize_mutex);
+
+	mlx5_vfree(in);
+	return 0;
+
+ex_alloc:
+	mlx5_vfree(in);
+
+ex_resize:
+	if (udata)
+		un_resize_user(cq);
+	else
+		un_resize_kernel(dev, cq);
+ex:
+	mutex_unlock(&cq->resize_mutex);
+	return err;
 }
 
 int mlx5_ib_get_cqe_size(struct mlx5_ib_dev *dev, struct ib_cq *ibcq)

drivers/infiniband/hw/mlx5/main.c

@@ -541,6 +541,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	struct mlx5_ib_ucontext *context;
 	struct mlx5_uuar_info *uuari;
 	struct mlx5_uar *uars;
+	int gross_uuars;
 	int num_uars;
 	int uuarn;
 	int err;
@@ -559,11 +560,13 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	if (req.total_num_uuars == 0)
 		return ERR_PTR(-EINVAL);
 
-	req.total_num_uuars = ALIGN(req.total_num_uuars, MLX5_BF_REGS_PER_PAGE);
+	req.total_num_uuars = ALIGN(req.total_num_uuars,
+				    MLX5_NON_FP_BF_REGS_PER_PAGE);
 	if (req.num_low_latency_uuars > req.total_num_uuars - 1)
 		return ERR_PTR(-EINVAL);
 
-	num_uars = req.total_num_uuars / MLX5_BF_REGS_PER_PAGE;
+	num_uars = req.total_num_uuars / MLX5_NON_FP_BF_REGS_PER_PAGE;
+	gross_uuars = num_uars * MLX5_BF_REGS_PER_PAGE;
 	resp.qp_tab_size      = 1 << dev->mdev.caps.log_max_qp;
 	resp.bf_reg_size      = dev->mdev.caps.bf_reg_size;
 	resp.cache_line_size  = L1_CACHE_BYTES;
@@ -585,7 +588,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 		goto out_ctx;
 	}
 
-	uuari->bitmap = kcalloc(BITS_TO_LONGS(req.total_num_uuars),
+	uuari->bitmap = kcalloc(BITS_TO_LONGS(gross_uuars),
 				sizeof(*uuari->bitmap),
 				GFP_KERNEL);
 	if (!uuari->bitmap) {
@@ -595,13 +598,13 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	/*
 	 * clear all fast path uuars
 	 */
-	for (i = 0; i < req.total_num_uuars; i++) {
+	for (i = 0; i < gross_uuars; i++) {
 		uuarn = i & 3;
 		if (uuarn == 2 || uuarn == 3)
 			set_bit(i, uuari->bitmap);
 	}
 
-	uuari->count = kcalloc(req.total_num_uuars, sizeof(*uuari->count), GFP_KERNEL);
+	uuari->count = kcalloc(gross_uuars, sizeof(*uuari->count), GFP_KERNEL);
 	if (!uuari->count) {
 		err = -ENOMEM;
 		goto out_bitmap;

drivers/infiniband/hw/mlx5/mlx5_ib.h

@@ -195,6 +195,7 @@ struct mlx5_ib_cq_buf {
 	struct mlx5_buf		buf;
 	struct ib_umem		*umem;
 	int			cqe_size;
+	int			nent;
 };
 
 enum mlx5_ib_qp_flags {
@@ -220,7 +221,7 @@ struct mlx5_ib_cq {
 	/* protect resize cq
 	 */
 	struct mutex		resize_mutex;
-	struct mlx5_ib_cq_resize *resize_buf;
+	struct mlx5_ib_cq_buf  *resize_buf;
 	struct ib_umem	       *resize_umem;
 	int			cqe_size;
 };
@@ -264,7 +265,6 @@ struct mlx5_ib_mr {
 	enum ib_wc_status	status;
 	struct mlx5_ib_dev     *dev;
 	struct mlx5_create_mkey_mbox_out out;
-	unsigned long		start;
 };
 
 struct mlx5_ib_fast_reg_page_list {

drivers/infiniband/hw/mlx5/mr.c

@@ -146,7 +146,6 @@ static int add_keys(struct mlx5_ib_dev *dev, int c, int num)
 		spin_lock_irq(&ent->lock);
 		ent->pending++;
 		spin_unlock_irq(&ent->lock);
-		mr->start = jiffies;
 		err = mlx5_core_create_mkey(&dev->mdev, &mr->mmr, in,
 					    sizeof(*in), reg_mr_callback,
 					    mr, &mr->out);

drivers/infiniband/hw/mlx5/qp.c

@@ -340,14 +340,57 @@ static int qp_has_rq(struct ib_qp_init_attr *attr)
 	return 1;
 }
 
+static int first_med_uuar(void)
+{
+	return 1;
+}
+
+static int next_uuar(int n)
+{
+	n++;
+
+	while (((n % 4) & 2))
+		n++;
+
+	return n;
+}
+
+static int num_med_uuar(struct mlx5_uuar_info *uuari)
+{
+	int n;
+
+	n = uuari->num_uars * MLX5_NON_FP_BF_REGS_PER_PAGE -
+		uuari->num_low_latency_uuars - 1;
+
+	return n >= 0 ? n : 0;
+}
+
+static int max_uuari(struct mlx5_uuar_info *uuari)
+{
+	return uuari->num_uars * 4;
+}
+
+static int first_hi_uuar(struct mlx5_uuar_info *uuari)
+{
+	int med;
+	int i;
+	int t;
+
+	med = num_med_uuar(uuari);
+	for (t = 0, i = first_med_uuar();; i = next_uuar(i)) {
+		t++;
+		if (t == med)
+			return next_uuar(i);
+	}
+
+	return 0;
+}
+
 static int alloc_high_class_uuar(struct mlx5_uuar_info *uuari)
 {
-	int nuuars = uuari->num_uars * MLX5_BF_REGS_PER_PAGE;
-	int start_uuar;
 	int i;
 
-	start_uuar = nuuars - uuari->num_low_latency_uuars;
-	for (i = start_uuar; i < nuuars; i++) {
+	for (i = first_hi_uuar(uuari); i < max_uuari(uuari); i = next_uuar(i)) {
 		if (!test_bit(i, uuari->bitmap)) {
 			set_bit(i, uuari->bitmap);
 			uuari->count[i]++;
@@ -360,19 +403,10 @@ static int alloc_high_class_uuar(struct mlx5_uuar_info *uuari)
 
 static int alloc_med_class_uuar(struct mlx5_uuar_info *uuari)
 {
-	int nuuars = uuari->num_uars * MLX5_BF_REGS_PER_PAGE;
-	int minidx = 1;
-	int uuarn;
-	int end;
+	int minidx = first_med_uuar();
 	int i;
 
-	end = nuuars - uuari->num_low_latency_uuars;
-
-	for (i = 1; i < end; i++) {
-		uuarn = i & 3;
-		if (uuarn == 2 || uuarn == 3)
-			continue;
-
+	for (i = first_med_uuar(); i < first_hi_uuar(uuari); i = next_uuar(i)) {
 		if (uuari->count[i] < uuari->count[minidx])
 			minidx = i;
 	}
@@ -489,12 +523,12 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 {
 	struct mlx5_ib_ucontext *context;
 	struct mlx5_ib_create_qp ucmd;
-	int page_shift;
+	int page_shift = 0;
 	int uar_index;
 	int npages;
-	u32 offset;
+	u32 offset = 0;
 	int uuarn;
-	int ncont;
+	int ncont = 0;
 	int err;
 
 	err = ib_copy_from_udata(&ucmd, udata, sizeof(ucmd));
@@ -510,11 +544,16 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_HIGH);
 	if (uuarn < 0) {
 		mlx5_ib_dbg(dev, "failed to allocate low latency UUAR\n");
-		mlx5_ib_dbg(dev, "reverting to high latency\n");
-		uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_LOW);
+		mlx5_ib_dbg(dev, "reverting to medium latency\n");
+		uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_MEDIUM);
 		if (uuarn < 0) {
-			mlx5_ib_dbg(dev, "uuar allocation failed\n");
-			return uuarn;
+			mlx5_ib_dbg(dev, "failed to allocate medium latency UUAR\n");
+			mlx5_ib_dbg(dev, "reverting to high latency\n");
+			uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_LOW);
+			if (uuarn < 0) {
+				mlx5_ib_warn(dev, "uuar allocation failed\n");
+				return uuarn;
+			}
 		}
 	}
 
@@ -525,23 +564,29 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	if (err)
 		goto err_uuar;
 
-	qp->umem = ib_umem_get(pd->uobject->context, ucmd.buf_addr,
-			       qp->buf_size, 0, 0);
-	if (IS_ERR(qp->umem)) {
-		mlx5_ib_dbg(dev, "umem_get failed\n");
-		err = PTR_ERR(qp->umem);
-		goto err_uuar;
+	if (ucmd.buf_addr && qp->buf_size) {
+		qp->umem = ib_umem_get(pd->uobject->context, ucmd.buf_addr,
+				       qp->buf_size, 0, 0);
+		if (IS_ERR(qp->umem)) {
+			mlx5_ib_dbg(dev, "umem_get failed\n");
+			err = PTR_ERR(qp->umem);
+			goto err_uuar;
+		}
+	} else {
+		qp->umem = NULL;
 	}
 
-	mlx5_ib_cont_pages(qp->umem, ucmd.buf_addr, &npages, &page_shift,
-			   &ncont, NULL);
-	err = mlx5_ib_get_buf_offset(ucmd.buf_addr, page_shift, &offset);
-	if (err) {
-		mlx5_ib_warn(dev, "bad offset\n");
-		goto err_umem;
+	if (qp->umem) {
+		mlx5_ib_cont_pages(qp->umem, ucmd.buf_addr, &npages, &page_shift,
+				   &ncont, NULL);
+		err = mlx5_ib_get_buf_offset(ucmd.buf_addr, page_shift, &offset);
+		if (err) {
+			mlx5_ib_warn(dev, "bad offset\n");
+			goto err_umem;
+		}
+		mlx5_ib_dbg(dev, "addr 0x%llx, size %d, npages %d, page_shift %d, ncont %d, offset %d\n",
+			    ucmd.buf_addr, qp->buf_size, npages, page_shift, ncont, offset);
 	}
-	mlx5_ib_dbg(dev, "addr 0x%llx, size %d, npages %d, page_shift %d, ncont %d, offset %d\n",
-		    ucmd.buf_addr, qp->buf_size, npages, page_shift, ncont, offset);
 
 	*inlen = sizeof(**in) + sizeof(*(*in)->pas) * ncont;
 	*in = mlx5_vzalloc(*inlen);
@@ -549,7 +594,8 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 		err = -ENOMEM;
 		goto err_umem;
 	}
-	mlx5_ib_populate_pas(dev, qp->umem, page_shift, (*in)->pas, 0);
+	if (qp->umem)
+		mlx5_ib_populate_pas(dev, qp->umem, page_shift, (*in)->pas, 0);
 	(*in)->ctx.log_pg_sz_remote_qpn =
 		cpu_to_be32((page_shift - MLX5_ADAPTER_PAGE_SHIFT) << 24);
 	(*in)->ctx.params2 = cpu_to_be32(offset << 6);
@@ -580,7 +626,8 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	mlx5_vfree(*in);
 
 err_umem:
-	ib_umem_release(qp->umem);
+	if (qp->umem)
+		ib_umem_release(qp->umem);
 
 err_uuar:
 	free_uuar(&context->uuari, uuarn);
@@ -593,7 +640,8 @@ static void destroy_qp_user(struct ib_pd *pd, struct mlx5_ib_qp *qp)
 
 	context = to_mucontext(pd->uobject->context);
 	mlx5_ib_db_unmap_user(context, &qp->db);
-	ib_umem_release(qp->umem);
+	if (qp->umem)
+		ib_umem_release(qp->umem);
 	free_uuar(&context->uuari, qp->uuarn);
 }
 
@@ -2212,6 +2260,10 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 
 		qp->db.db[MLX5_SND_DBR] = cpu_to_be32(qp->sq.cur_post);
 
+		/* Make sure doorbell record is visible to the HCA before
+		 * we hit doorbell */
+		wmb();
+
 		if (bf->need_lock)
 			spin_lock(&bf->lock);
 

drivers/infiniband/hw/mlx5/user.h

@@ -93,6 +93,9 @@ struct mlx5_ib_create_cq_resp {
 
 struct mlx5_ib_resize_cq {
 	__u64	buf_addr;
+	__u16	cqe_size;
+	__u16	reserved0;
+	__u32	reserved1;
 };
 
 struct mlx5_ib_create_srq {

drivers/infiniband/hw/nes/nes_cm.c

@@ -1354,8 +1354,7 @@ static int nes_addr_resolve_neigh(struct nes_vnic *nesvnic, u32 dst_ip, int arpi
 				  neigh->ha, ntohl(rt->rt_gateway));
 
 			if (arpindex >= 0) {
-				if (!memcmp(nesadapter->arp_table[arpindex].mac_addr,
-					    neigh->ha, ETH_ALEN)) {
+				if (ether_addr_equal(nesadapter->arp_table[arpindex].mac_addr, neigh->ha)) {
 					/* Mac address same as in nes_arp_table */
 					goto out;
 				}

drivers/infiniband/hw/ocrdma/ocrdma_ah.c

@@ -84,6 +84,7 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah,
 	memcpy((u8 *)ah->av + eth_sz, &grh, sizeof(struct ocrdma_grh));
 	if (vlan_enabled)
 		ah->av->valid |= OCRDMA_AV_VLAN_VALID;
+	ah->av->valid = cpu_to_le32(ah->av->valid);
 	return status;
 }
 

drivers/infiniband/hw/ocrdma/ocrdma_sli.h

@@ -31,7 +31,7 @@
 #define Bit(_b) (1 << (_b))
 
 #define OCRDMA_GEN1_FAMILY	0xB
-#define OCRDMA_GEN2_FAMILY	0x2
+#define OCRDMA_GEN2_FAMILY	0x0F
 
 #define OCRDMA_SUBSYS_ROCE 10
 enum {
@@ -1694,7 +1694,7 @@ struct ocrdma_grh {
 	u16	rsvd;
 } __packed;
 
-#define OCRDMA_AV_VALID		Bit(0)
+#define OCRDMA_AV_VALID		Bit(7)
 #define OCRDMA_AV_VLAN_VALID	Bit(1)
 
 struct ocrdma_av {

drivers/infiniband/hw/qib/qib_ud.c

@@ -57,13 +57,20 @@ static void qib_ud_loopback(struct qib_qp *sqp, struct qib_swqe *swqe)
 	struct qib_sge *sge;
 	struct ib_wc wc;
 	u32 length;
+	enum ib_qp_type sqptype, dqptype;
 
 	qp = qib_lookup_qpn(ibp, swqe->wr.wr.ud.remote_qpn);
 	if (!qp) {
 		ibp->n_pkt_drops++;
 		return;
 	}
-	if (qp->ibqp.qp_type != sqp->ibqp.qp_type ||
+
+	sqptype = sqp->ibqp.qp_type == IB_QPT_GSI ?
+			IB_QPT_UD : sqp->ibqp.qp_type;
+	dqptype = qp->ibqp.qp_type == IB_QPT_GSI ?
+			IB_QPT_UD : qp->ibqp.qp_type;
+
+	if (dqptype != sqptype ||
 	    !(ib_qib_state_ops[qp->state] & QIB_PROCESS_RECV_OK)) {
 		ibp->n_pkt_drops++;
 		goto drop;

drivers/infiniband/hw/usnic/Kconfig

+config INFINIBAND_USNIC
+	tristate "Verbs support for Cisco VIC"
+	depends on NETDEVICES && ETHERNET && INET && PCI && INTEL_IOMMU
+	select ENIC
+	select NET_VENDOR_CISCO
+	select PCI_IOV
+	select INFINIBAND_USER_ACCESS
+	---help---
+	  This is a low-level driver for Cisco's Virtual Interface
+	  Cards (VICs), including the VIC 1240 and 1280 cards.

drivers/infiniband/hw/usnic/Makefile

+ccflags-y := -Idrivers/net/ethernet/cisco/enic
+
+obj-$(CONFIG_INFINIBAND_USNIC)+= usnic_verbs.o
+
+usnic_verbs-y=\
+usnic_fwd.o \
+usnic_transport.o \
+usnic_uiom.o \
+usnic_uiom_interval_tree.o \
+usnic_vnic.o \
+usnic_ib_main.o \
+usnic_ib_qp_grp.o \
+usnic_ib_sysfs.o \
+usnic_ib_verbs.o \
+usnic_debugfs.o \

drivers/infiniband/hw/usnic/usnic.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_H_
+#define USNIC_H_
+
+#define DRV_NAME	"usnic_verbs"
+
+#define PCI_DEVICE_ID_CISCO_VIC_USPACE_NIC	0x00cf	/* User space NIC */
+
+#define DRV_VERSION    "1.0.3"
+#define DRV_RELDATE    "December 19, 2013"
+
+#endif /* USNIC_H_ */

drivers/infiniband/hw/usnic/usnic_abi.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+
+#ifndef USNIC_ABI_H
+#define USNIC_ABI_H
+
+/* ABI between userspace and kernel */
+#define USNIC_UVERBS_ABI_VERSION	4
+
+#define USNIC_QP_GRP_MAX_WQS		8
+#define USNIC_QP_GRP_MAX_RQS		8
+#define USNIC_QP_GRP_MAX_CQS		16
+
+enum usnic_transport_type {
+	USNIC_TRANSPORT_UNKNOWN		= 0,
+	USNIC_TRANSPORT_ROCE_CUSTOM	= 1,
+	USNIC_TRANSPORT_IPV4_UDP	= 2,
+	USNIC_TRANSPORT_MAX		= 3,
+};
+
+struct usnic_transport_spec {
+	enum usnic_transport_type	trans_type;
+	union {
+		struct {
+			uint16_t	port_num;
+		} usnic_roce;
+		struct {
+			uint32_t	sock_fd;
+		} udp;
+	};
+};
+
+struct usnic_ib_create_qp_cmd {
+	struct usnic_transport_spec	spec;
+};
+
+/*TODO: Future - usnic_modify_qp needs to pass in generic filters */
+struct usnic_ib_create_qp_resp {
+	u32				vfid;
+	u32				qp_grp_id;
+	u64				bar_bus_addr;
+	u32				bar_len;
+/*
+ * WQ, RQ, CQ are explicity specified bc exposing a generic resources inteface
+ * expands the scope of ABI to many files.
+ */
+	u32				wq_cnt;
+	u32				rq_cnt;
+	u32				cq_cnt;
+	u32				wq_idx[USNIC_QP_GRP_MAX_WQS];
+	u32				rq_idx[USNIC_QP_GRP_MAX_RQS];
+	u32				cq_idx[USNIC_QP_GRP_MAX_CQS];
+	u32				transport;
+	u32				reserved[9];
+};
+
+#endif /* USNIC_ABI_H */

drivers/infiniband/hw/usnic/usnic_common_pkt_hdr.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_CMN_PKT_HDR_H
+#define USNIC_CMN_PKT_HDR_H
+
+#define USNIC_ROCE_ETHERTYPE		(0x8915)
+#define USNIC_ROCE_GRH_VER              (8)
+#define USNIC_PROTO_VER                 (1)
+#define USNIC_ROCE_GRH_VER_SHIFT        (4)
+
+#endif /* USNIC_COMMON_PKT_HDR_H */

drivers/infiniband/hw/usnic/usnic_common_util.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_CMN_UTIL_H
+#define USNIC_CMN_UTIL_H
+
+static inline void
+usnic_mac_to_gid(const char *const mac, char *raw_gid)
+{
+	raw_gid[0] = 0xfe;
+	raw_gid[1] = 0x80;
+	memset(&raw_gid[2], 0, 6);
+	raw_gid[8] = mac[0]^2;
+	raw_gid[9] = mac[1];
+	raw_gid[10] = mac[2];
+	raw_gid[11] = 0xff;
+	raw_gid[12] = 0xfe;
+	raw_gid[13] = mac[3];
+	raw_gid[14] = mac[4];
+	raw_gid[15] = mac[5];
+}
+
+static inline void
+usnic_mac_ip_to_gid(const char *const mac, const __be32 inaddr, char *raw_gid)
+{
+	raw_gid[0] = 0xfe;
+	raw_gid[1] = 0x80;
+	memset(&raw_gid[2], 0, 2);
+	memcpy(&raw_gid[4], &inaddr, 4);
+	raw_gid[8] = mac[0]^2;
+	raw_gid[9] = mac[1];
+	raw_gid[10] = mac[2];
+	raw_gid[11] = 0xff;
+	raw_gid[12] = 0xfe;
+	raw_gid[13] = mac[3];
+	raw_gid[14] = mac[4];
+	raw_gid[15] = mac[5];
+}
+
+static inline void
+usnic_write_gid_if_id_from_mac(char *mac, char *raw_gid)
+{
+	raw_gid[8] = mac[0]^2;
+	raw_gid[9] = mac[1];
+	raw_gid[10] = mac[2];
+	raw_gid[11] = 0xff;
+	raw_gid[12] = 0xfe;
+	raw_gid[13] = mac[3];
+	raw_gid[14] = mac[4];
+	raw_gid[15] = mac[5];
+}
+
+#endif /* USNIC_COMMON_UTIL_H */

drivers/infiniband/hw/usnic/usnic_debugfs.c

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#include <linux/debugfs.h>
+#include <linux/module.h>
+
+#include "usnic.h"
+#include "usnic_log.h"
+#include "usnic_debugfs.h"
+#include "usnic_ib_qp_grp.h"
+#include "usnic_transport.h"
+
+static struct dentry *debugfs_root;
+static struct dentry *flows_dentry;
+
+static ssize_t usnic_debugfs_buildinfo_read(struct file *f, char __user *data,
+						size_t count, loff_t *ppos)
+{
+	char buf[500];
+	int res;
+
+	if (*ppos > 0)
+		return 0;
+
+	res = scnprintf(buf, sizeof(buf),
+			"version:       %s\n"
+			"build date:    %s\n",
+			DRV_VERSION, DRV_RELDATE);
+
+	return simple_read_from_buffer(data, count, ppos, buf, res);
+}
+
+static const struct file_operations usnic_debugfs_buildinfo_ops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read = usnic_debugfs_buildinfo_read
+};
+
+static ssize_t flowinfo_read(struct file *f, char __user *data,
+				size_t count, loff_t *ppos)
+{
+	struct usnic_ib_qp_grp_flow *qp_flow;
+	int n;
+	int left;
+	char *ptr;
+	char buf[512];
+
+	qp_flow = f->private_data;
+	ptr = buf;
+	left = count;
+
+	if (*ppos > 0)
+		return 0;
+
+	spin_lock(&qp_flow->qp_grp->lock);
+	n = scnprintf(ptr, left,
+			"QP Grp ID: %d Transport: %s ",
+			qp_flow->qp_grp->grp_id,
+			usnic_transport_to_str(qp_flow->trans_type));
+	UPDATE_PTR_LEFT(n, ptr, left);
+	if (qp_flow->trans_type == USNIC_TRANSPORT_ROCE_CUSTOM) {
+		n = scnprintf(ptr, left, "Port_Num:%hu\n",
+					qp_flow->usnic_roce.port_num);
+		UPDATE_PTR_LEFT(n, ptr, left);
+	} else if (qp_flow->trans_type == USNIC_TRANSPORT_IPV4_UDP) {
+		n = usnic_transport_sock_to_str(ptr, left,
+				qp_flow->udp.sock);
+		UPDATE_PTR_LEFT(n, ptr, left);
+		n = scnprintf(ptr, left, "\n");
+		UPDATE_PTR_LEFT(n, ptr, left);
+	}
+	spin_unlock(&qp_flow->qp_grp->lock);
+
+	return simple_read_from_buffer(data, count, ppos, buf, ptr - buf);
+}
+
+static const struct file_operations flowinfo_ops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read = flowinfo_read,
+};
+
+void usnic_debugfs_init(void)
+{
+	debugfs_root = debugfs_create_dir(DRV_NAME, NULL);
+	if (IS_ERR(debugfs_root)) {
+		usnic_err("Failed to create debugfs root dir, check if debugfs is enabled in kernel configuration\n");
+		goto out_clear_root;
+	}
+
+	flows_dentry = debugfs_create_dir("flows", debugfs_root);
+	if (IS_ERR_OR_NULL(flows_dentry)) {
+		usnic_err("Failed to create debugfs flow dir with err %ld\n",
+				PTR_ERR(flows_dentry));
+		goto out_free_root;
+	}
+
+	debugfs_create_file("build-info", S_IRUGO, debugfs_root,
+				NULL, &usnic_debugfs_buildinfo_ops);
+	return;
+
+out_free_root:
+	debugfs_remove_recursive(debugfs_root);
+out_clear_root:
+	debugfs_root = NULL;
+}
+
+void usnic_debugfs_exit(void)
+{
+	if (!debugfs_root)
+		return;
+
+	debugfs_remove_recursive(debugfs_root);
+	debugfs_root = NULL;
+}
+
+void usnic_debugfs_flow_add(struct usnic_ib_qp_grp_flow *qp_flow)
+{
+	if (IS_ERR_OR_NULL(flows_dentry))
+		return;
+
+	scnprintf(qp_flow->dentry_name, sizeof(qp_flow->dentry_name),
+			"%u", qp_flow->flow->flow_id);
+	qp_flow->dbgfs_dentry = debugfs_create_file(qp_flow->dentry_name,
+							S_IRUGO,
+							flows_dentry,
+							qp_flow,
+							&flowinfo_ops);
+	if (IS_ERR_OR_NULL(qp_flow->dbgfs_dentry)) {
+		usnic_err("Failed to create dbg fs entry for flow %u\n",
+				qp_flow->flow->flow_id);
+	}
+}
+
+void usnic_debugfs_flow_remove(struct usnic_ib_qp_grp_flow *qp_flow)
+{
+	if (!IS_ERR_OR_NULL(qp_flow->dbgfs_dentry))
+		debugfs_remove(qp_flow->dbgfs_dentry);
+}

drivers/infiniband/hw/usnic/usnic_debugfs.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+#ifndef USNIC_DEBUGFS_H_
+#define USNIC_DEBUGFS_H_
+
+#include "usnic_ib_qp_grp.h"
+
+void usnic_debugfs_init(void);
+
+void usnic_debugfs_exit(void);
+void usnic_debugfs_flow_add(struct usnic_ib_qp_grp_flow *qp_flow);
+void usnic_debugfs_flow_remove(struct usnic_ib_qp_grp_flow *qp_flow);
+
+#endif /*!USNIC_DEBUGFS_H_ */

drivers/infiniband/hw/usnic/usnic_fwd.c

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+
+#include "enic_api.h"
+#include "usnic_common_pkt_hdr.h"
+#include "usnic_fwd.h"
+#include "usnic_log.h"
+
+static int usnic_fwd_devcmd_locked(struct usnic_fwd_dev *ufdev, int vnic_idx,
+					enum vnic_devcmd_cmd cmd, u64 *a0,
+					u64 *a1)
+{
+	int status;
+	struct net_device *netdev = ufdev->netdev;
+
+	lockdep_assert_held(&ufdev->lock);
+
+	status = enic_api_devcmd_proxy_by_index(netdev,
+			vnic_idx,
+			cmd,
+			a0, a1,
+			1000);
+	if (status) {
+		if (status == ERR_EINVAL && cmd == CMD_DEL_FILTER) {
+			usnic_dbg("Dev %s vnic idx %u cmd %u already deleted",
+					ufdev->name, vnic_idx, cmd);
+		} else {
+			usnic_err("Dev %s vnic idx %u cmd %u failed with status %d\n",
+					ufdev->name, vnic_idx, cmd,
+					status);
+		}
+	} else {
+		usnic_dbg("Dev %s vnic idx %u cmd %u success",
+				ufdev->name, vnic_idx, cmd);
+	}
+
+	return status;
+}
+
+static int usnic_fwd_devcmd(struct usnic_fwd_dev *ufdev, int vnic_idx,
+				enum vnic_devcmd_cmd cmd, u64 *a0, u64 *a1)
+{
+	int status;
+
+	spin_lock(&ufdev->lock);
+	status = usnic_fwd_devcmd_locked(ufdev, vnic_idx, cmd, a0, a1);
+	spin_unlock(&ufdev->lock);
+
+	return status;
+}
+
+struct usnic_fwd_dev *usnic_fwd_dev_alloc(struct pci_dev *pdev)
+{
+	struct usnic_fwd_dev *ufdev;
+
+	ufdev = kzalloc(sizeof(*ufdev), GFP_KERNEL);
+	if (!ufdev)
+		return NULL;
+
+	ufdev->pdev = pdev;
+	ufdev->netdev = pci_get_drvdata(pdev);
+	spin_lock_init(&ufdev->lock);
+	strncpy(ufdev->name, netdev_name(ufdev->netdev),
+			sizeof(ufdev->name) - 1);
+
+	return ufdev;
+}
+
+void usnic_fwd_dev_free(struct usnic_fwd_dev *ufdev)
+{
+	kfree(ufdev);
+}
+
+void usnic_fwd_set_mac(struct usnic_fwd_dev *ufdev, char mac[ETH_ALEN])
+{
+	spin_lock(&ufdev->lock);
+	memcpy(&ufdev->mac, mac, sizeof(ufdev->mac));
+	spin_unlock(&ufdev->lock);
+}
+
+int usnic_fwd_add_ipaddr(struct usnic_fwd_dev *ufdev, __be32 inaddr)
+{
+	int status;
+
+	spin_lock(&ufdev->lock);
+	if (ufdev->inaddr == 0) {
+		ufdev->inaddr = inaddr;
+		status = 0;
+	} else {
+		status = -EFAULT;
+	}
+	spin_unlock(&ufdev->lock);
+
+	return status;
+}
+
+void usnic_fwd_del_ipaddr(struct usnic_fwd_dev *ufdev)
+{
+	spin_lock(&ufdev->lock);
+	ufdev->inaddr = 0;
+	spin_unlock(&ufdev->lock);
+}
+
+void usnic_fwd_carrier_up(struct usnic_fwd_dev *ufdev)
+{
+	spin_lock(&ufdev->lock);
+	ufdev->link_up = 1;
+	spin_unlock(&ufdev->lock);
+}
+
+void usnic_fwd_carrier_down(struct usnic_fwd_dev *ufdev)
+{
+	spin_lock(&ufdev->lock);
+	ufdev->link_up = 0;
+	spin_unlock(&ufdev->lock);
+}
+
+void usnic_fwd_set_mtu(struct usnic_fwd_dev *ufdev, unsigned int mtu)
+{
+	spin_lock(&ufdev->lock);
+	ufdev->mtu = mtu;
+	spin_unlock(&ufdev->lock);
+}
+
+static int usnic_fwd_dev_ready_locked(struct usnic_fwd_dev *ufdev)
+{
+	lockdep_assert_held(&ufdev->lock);
+
+	if (!ufdev->link_up)
+		return -EPERM;
+
+	return 0;
+}
+
+static int validate_filter_locked(struct usnic_fwd_dev *ufdev,
+					struct filter *filter)
+{
+
+	lockdep_assert_held(&ufdev->lock);
+
+	if (filter->type == FILTER_IPV4_5TUPLE) {
+		if (!(filter->u.ipv4.flags & FILTER_FIELD_5TUP_DST_AD))
+			return -EACCES;
+		if (!(filter->u.ipv4.flags & FILTER_FIELD_5TUP_DST_PT))
+			return -EBUSY;
+		else if (ufdev->inaddr == 0)
+			return -EINVAL;
+		else if (filter->u.ipv4.dst_port == 0)
+			return -ERANGE;
+		else if (ntohl(ufdev->inaddr) != filter->u.ipv4.dst_addr)
+			return -EFAULT;
+		else
+			return 0;
+	}
+
+	return 0;
+}
+
+static void fill_tlv(struct filter_tlv *tlv, struct filter *filter,
+		struct filter_action *action)
+{
+	tlv->type = CLSF_TLV_FILTER;
+	tlv->length = sizeof(struct filter);
+	*((struct filter *)&tlv->val) = *filter;
+
+	tlv = (struct filter_tlv *)((char *)tlv + sizeof(struct filter_tlv) +
+			sizeof(struct filter));
+	tlv->type = CLSF_TLV_ACTION;
+	tlv->length = sizeof(struct filter_action);
+	*((struct filter_action *)&tlv->val) = *action;
+}
+
+struct usnic_fwd_flow*
+usnic_fwd_alloc_flow(struct usnic_fwd_dev *ufdev, struct filter *filter,
+				struct usnic_filter_action *uaction)
+{
+	struct filter_tlv *tlv;
+	struct pci_dev *pdev;
+	struct usnic_fwd_flow *flow;
+	uint64_t a0, a1;
+	uint64_t tlv_size;
+	dma_addr_t tlv_pa;
+	int status;
+
+	pdev = ufdev->pdev;
+	tlv_size = (2*sizeof(struct filter_tlv) + sizeof(struct filter) +
+			sizeof(struct filter_action));
+
+	flow = kzalloc(sizeof(*flow), GFP_ATOMIC);
+	if (!flow)
+		return ERR_PTR(-ENOMEM);
+
+	tlv = pci_alloc_consistent(pdev, tlv_size, &tlv_pa);
+	if (!tlv) {
+		usnic_err("Failed to allocate memory\n");
+		status = -ENOMEM;
+		goto out_free_flow;
+	}
+
+	fill_tlv(tlv, filter, &uaction->action);
+
+	spin_lock(&ufdev->lock);
+	status = usnic_fwd_dev_ready_locked(ufdev);
+	if (status) {
+		usnic_err("Forwarding dev %s not ready with status %d\n",
+				ufdev->name, status);
+		goto out_free_tlv;
+	}
+
+	status = validate_filter_locked(ufdev, filter);
+	if (status) {
+		usnic_err("Failed to validate filter with status %d\n",
+				status);
+		goto out_free_tlv;
+	}
+
+	/* Issue Devcmd */
+	a0 = tlv_pa;
+	a1 = tlv_size;
+	status = usnic_fwd_devcmd_locked(ufdev, uaction->vnic_idx,
+						CMD_ADD_FILTER, &a0, &a1);
+	if (status) {
+		usnic_err("VF %s Filter add failed with status:%d",
+				ufdev->name, status);
+		status = -EFAULT;
+		goto out_free_tlv;
+	} else {
+		usnic_dbg("VF %s FILTER ID:%llu", ufdev->name, a0);
+	}
+
+	flow->flow_id = (uint32_t) a0;
+	flow->vnic_idx = uaction->vnic_idx;
+	flow->ufdev = ufdev;
+
+out_free_tlv:
+	spin_unlock(&ufdev->lock);
+	pci_free_consistent(pdev, tlv_size, tlv, tlv_pa);
+	if (!status)
+		return flow;
+out_free_flow:
+	kfree(flow);
+	return ERR_PTR(status);
+}
+
+int usnic_fwd_dealloc_flow(struct usnic_fwd_flow *flow)
+{
+	int status;
+	u64 a0, a1;
+
+	a0 = flow->flow_id;
+
+	status = usnic_fwd_devcmd(flow->ufdev, flow->vnic_idx,
+					CMD_DEL_FILTER, &a0, &a1);
+	if (status) {
+		if (status == ERR_EINVAL) {
+			usnic_dbg("Filter %u already deleted for VF Idx %u pf: %s status: %d",
+					flow->flow_id, flow->vnic_idx,
+					flow->ufdev->name, status);
+		} else {
+			usnic_err("PF %s VF Idx %u Filter: %u FILTER DELETE failed with status %d",
+					flow->ufdev->name, flow->vnic_idx,
+					flow->flow_id, status);
+		}
+		status = 0;
+		/*
+		 * Log the error and fake success to the caller because if
+		 * a flow fails to be deleted in the firmware, it is an
+		 * unrecoverable error.
+		 */
+	} else {
+		usnic_dbg("PF %s VF Idx %u Filter: %u FILTER DELETED",
+				flow->ufdev->name, flow->vnic_idx,
+				flow->flow_id);
+	}
+
+	kfree(flow);
+	return status;
+}
+
+int usnic_fwd_enable_qp(struct usnic_fwd_dev *ufdev, int vnic_idx, int qp_idx)
+{
+	int status;
+	struct net_device *pf_netdev;
+	u64 a0, a1;
+
+	pf_netdev = ufdev->netdev;
+	a0 = qp_idx;
+	a1 = CMD_QP_RQWQ;
+
+	status = usnic_fwd_devcmd(ufdev, vnic_idx, CMD_QP_ENABLE,
+						&a0, &a1);
+	if (status) {
+		usnic_err("PF %s VNIC Index %u RQ Index: %u ENABLE Failed with status %d",
+				netdev_name(pf_netdev),
+				vnic_idx,
+				qp_idx,
+				status);
+	} else {
+		usnic_dbg("PF %s VNIC Index %u RQ Index: %u ENABLED",
+				netdev_name(pf_netdev),
+				vnic_idx, qp_idx);
+	}
+
+	return status;
+}
+
+int usnic_fwd_disable_qp(struct usnic_fwd_dev *ufdev, int vnic_idx, int qp_idx)
+{
+	int status;
+	u64 a0, a1;
+	struct net_device *pf_netdev;
+
+	pf_netdev = ufdev->netdev;
+	a0 = qp_idx;
+	a1 = CMD_QP_RQWQ;
+
+	status = usnic_fwd_devcmd(ufdev, vnic_idx, CMD_QP_DISABLE,
+			&a0, &a1);
+	if (status) {
+		usnic_err("PF %s VNIC Index %u RQ Index: %u DISABLE Failed with status %d",
+				netdev_name(pf_netdev),
+				vnic_idx,
+				qp_idx,
+				status);
+	} else {
+		usnic_dbg("PF %s VNIC Index %u RQ Index: %u DISABLED",
+				netdev_name(pf_netdev),
+				vnic_idx,
+				qp_idx);
+	}
+
+	return status;
+}

drivers/infiniband/hw/usnic/usnic_fwd.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_FWD_H_
+#define USNIC_FWD_H_
+
+#include <linux/if.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/in.h>
+
+#include "usnic_abi.h"
+#include "usnic_common_pkt_hdr.h"
+#include "vnic_devcmd.h"
+
+struct usnic_fwd_dev {
+	struct pci_dev			*pdev;
+	struct net_device		*netdev;
+	spinlock_t			lock;
+	/*
+	 * The following fields can be read directly off the device.
+	 * However, they should be set by a accessor function, except name,
+	 * which cannot be changed.
+	 */
+	bool				link_up;
+	char				mac[ETH_ALEN];
+	unsigned int			mtu;
+	__be32				inaddr;
+	char				name[IFNAMSIZ+1];
+};
+
+struct usnic_fwd_flow {
+	uint32_t			flow_id;
+	struct usnic_fwd_dev		*ufdev;
+	unsigned int			vnic_idx;
+};
+
+struct usnic_filter_action {
+	int				vnic_idx;
+	struct filter_action		action;
+};
+
+struct usnic_fwd_dev *usnic_fwd_dev_alloc(struct pci_dev *pdev);
+void usnic_fwd_dev_free(struct usnic_fwd_dev *ufdev);
+
+void usnic_fwd_set_mac(struct usnic_fwd_dev *ufdev, char mac[ETH_ALEN]);
+int usnic_fwd_add_ipaddr(struct usnic_fwd_dev *ufdev, __be32 inaddr);
+void usnic_fwd_del_ipaddr(struct usnic_fwd_dev *ufdev);
+void usnic_fwd_carrier_up(struct usnic_fwd_dev *ufdev);
+void usnic_fwd_carrier_down(struct usnic_fwd_dev *ufdev);
+void usnic_fwd_set_mtu(struct usnic_fwd_dev *ufdev, unsigned int mtu);
+
+/*
+ * Allocate a flow on this forwarding device. Whoever calls this function,
+ * must monitor netdev events on ufdev's netdevice. If NETDEV_REBOOT or
+ * NETDEV_DOWN is seen, flow will no longer function and must be
+ * immediately freed by calling usnic_dealloc_flow.
+ */
+struct usnic_fwd_flow*
+usnic_fwd_alloc_flow(struct usnic_fwd_dev *ufdev, struct filter *filter,
+				struct usnic_filter_action *action);
+int usnic_fwd_dealloc_flow(struct usnic_fwd_flow *flow);
+int usnic_fwd_enable_qp(struct usnic_fwd_dev *ufdev, int vnic_idx, int qp_idx);
+int usnic_fwd_disable_qp(struct usnic_fwd_dev *ufdev, int vnic_idx, int qp_idx);
+
+static inline void usnic_fwd_init_usnic_filter(struct filter *filter,
+						uint32_t usnic_id)
+{
+	filter->type = FILTER_USNIC_ID;
+	filter->u.usnic.ethtype = USNIC_ROCE_ETHERTYPE;
+	filter->u.usnic.flags = FILTER_FIELD_USNIC_ETHTYPE |
+				FILTER_FIELD_USNIC_ID |
+				FILTER_FIELD_USNIC_PROTO;
+	filter->u.usnic.proto_version = (USNIC_ROCE_GRH_VER <<
+					 USNIC_ROCE_GRH_VER_SHIFT) |
+					 USNIC_PROTO_VER;
+	filter->u.usnic.usnic_id = usnic_id;
+}
+
+static inline void usnic_fwd_init_udp_filter(struct filter *filter,
+						uint32_t daddr, uint16_t dport)
+{
+	filter->type = FILTER_IPV4_5TUPLE;
+	filter->u.ipv4.flags = FILTER_FIELD_5TUP_PROTO;
+	filter->u.ipv4.protocol = PROTO_UDP;
+
+	if (daddr) {
+		filter->u.ipv4.flags |= FILTER_FIELD_5TUP_DST_AD;
+		filter->u.ipv4.dst_addr = daddr;
+	}
+
+	if (dport) {
+		filter->u.ipv4.flags |= FILTER_FIELD_5TUP_DST_PT;
+		filter->u.ipv4.dst_port = dport;
+	}
+}
+
+#endif /* !USNIC_FWD_H_ */

drivers/infiniband/hw/usnic/usnic_ib.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_IB_H_
+#define USNIC_IB_H_
+
+#include <linux/iommu.h>
+#include <linux/netdevice.h>
+
+#include <rdma/ib_verbs.h>
+
+
+#include "usnic.h"
+#include "usnic_abi.h"
+#include "usnic_vnic.h"
+
+#define USNIC_IB_PORT_CNT		1
+#define USNIC_IB_NUM_COMP_VECTORS	1
+
+extern unsigned int usnic_ib_share_vf;
+
+struct usnic_ib_ucontext {
+	struct ib_ucontext		ibucontext;
+	/* Protected by usnic_ib_dev->usdev_lock */
+	struct list_head		qp_grp_list;
+	struct list_head		link;
+};
+
+struct usnic_ib_pd {
+	struct ib_pd			ibpd;
+	struct usnic_uiom_pd		*umem_pd;
+};
+
+struct usnic_ib_mr {
+	struct ib_mr			ibmr;
+	struct usnic_uiom_reg		*umem;
+};
+
+struct usnic_ib_dev {
+	struct ib_device		ib_dev;
+	struct pci_dev			*pdev;
+	struct net_device		*netdev;
+	struct usnic_fwd_dev		*ufdev;
+	struct list_head		ib_dev_link;
+	struct list_head		vf_dev_list;
+	struct list_head		ctx_list;
+	struct mutex			usdev_lock;
+
+	/* provisioning information */
+	struct kref			vf_cnt;
+	unsigned int			vf_res_cnt[USNIC_VNIC_RES_TYPE_MAX];
+
+	/* sysfs vars for QPN reporting */
+	struct kobject *qpn_kobj;
+};
+
+struct usnic_ib_vf {
+	struct usnic_ib_dev		*pf;
+	spinlock_t			lock;
+	struct usnic_vnic		*vnic;
+	unsigned int			qp_grp_ref_cnt;
+	struct usnic_ib_pd		*pd;
+	struct list_head		link;
+};
+
+static inline
+struct usnic_ib_dev *to_usdev(struct ib_device *ibdev)
+{
+	return container_of(ibdev, struct usnic_ib_dev, ib_dev);
+}
+
+static inline
+struct usnic_ib_ucontext *to_ucontext(struct ib_ucontext *ibucontext)
+{
+	return container_of(ibucontext, struct usnic_ib_ucontext, ibucontext);
+}
+
+static inline
+struct usnic_ib_pd *to_upd(struct ib_pd *ibpd)
+{
+	return container_of(ibpd, struct usnic_ib_pd, ibpd);
+}
+
+static inline
+struct usnic_ib_ucontext *to_uucontext(struct ib_ucontext *ibucontext)
+{
+	return container_of(ibucontext, struct usnic_ib_ucontext, ibucontext);
+}
+
+static inline
+struct usnic_ib_mr *to_umr(struct ib_mr *ibmr)
+{
+	return container_of(ibmr, struct usnic_ib_mr, ibmr);
+}
+void usnic_ib_log_vf(struct usnic_ib_vf *vf);
+
+#define UPDATE_PTR_LEFT(N, P, L)			\
+do {							\
+	L -= (N);					\
+	P += (N);					\
+} while (0)
+
+#endif /* USNIC_IB_H_ */

drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+#include <linux/bug.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+
+#include "usnic_log.h"
+#include "usnic_vnic.h"
+#include "usnic_fwd.h"
+#include "usnic_uiom.h"
+#include "usnic_debugfs.h"
+#include "usnic_ib_qp_grp.h"
+#include "usnic_ib_sysfs.h"
+#include "usnic_transport.h"
+
+#define DFLT_RQ_IDX	0
+
+const char *usnic_ib_qp_grp_state_to_string(enum ib_qp_state state)
+{
+	switch (state) {
+	case IB_QPS_RESET:
+		return "Rst";
+	case IB_QPS_INIT:
+		return "Init";
+	case IB_QPS_RTR:
+		return "RTR";
+	case IB_QPS_RTS:
+		return "RTS";
+	case IB_QPS_SQD:
+		return "SQD";
+	case IB_QPS_SQE:
+		return "SQE";
+	case IB_QPS_ERR:
+		return "ERR";
+	default:
+		return "UNKOWN STATE";
+
+	}
+}
+
+int usnic_ib_qp_grp_dump_hdr(char *buf, int buf_sz)
+{
+	return scnprintf(buf, buf_sz, "|QPN\t|State\t|PID\t|VF Idx\t|Fil ID");
+}
+
+int usnic_ib_qp_grp_dump_rows(void *obj, char *buf, int buf_sz)
+{
+	struct usnic_ib_qp_grp *qp_grp = obj;
+	struct usnic_ib_qp_grp_flow *default_flow;
+	if (obj) {
+		default_flow = list_first_entry(&qp_grp->flows_lst,
+					struct usnic_ib_qp_grp_flow, link);
+		return scnprintf(buf, buf_sz, "|%d\t|%s\t|%d\t|%hu\t|%d",
+					qp_grp->ibqp.qp_num,
+					usnic_ib_qp_grp_state_to_string(
+							qp_grp->state),
+					qp_grp->owner_pid,
+					usnic_vnic_get_index(qp_grp->vf->vnic),
+					default_flow->flow->flow_id);
+	} else {
+		return scnprintf(buf, buf_sz, "|N/A\t|N/A\t|N/A\t|N/A\t|N/A");
+	}
+}
+
+static struct usnic_vnic_res_chunk *
+get_qp_res_chunk(struct usnic_ib_qp_grp *qp_grp)
+{
+	lockdep_assert_held(&qp_grp->lock);
+	/*
+	 * The QP res chunk, used to derive qp indices,
+	 * are just indices of the RQs
+	 */
+	return usnic_ib_qp_grp_get_chunk(qp_grp, USNIC_VNIC_RES_TYPE_RQ);
+}
+
+static int enable_qp_grp(struct usnic_ib_qp_grp *qp_grp)
+{
+
+	int status;
+	int i, vnic_idx;
+	struct usnic_vnic_res_chunk *res_chunk;
+	struct usnic_vnic_res *res;
+
+	lockdep_assert_held(&qp_grp->lock);
+
+	vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic);
+
+	res_chunk = get_qp_res_chunk(qp_grp);
+	if (IS_ERR_OR_NULL(res_chunk)) {
+		usnic_err("Unable to get qp res with err %ld\n",
+				PTR_ERR(res_chunk));
+		return res_chunk ? PTR_ERR(res_chunk) : -ENOMEM;
+	}
+
+	for (i = 0; i < res_chunk->cnt; i++) {
+		res = res_chunk->res[i];
+		status = usnic_fwd_enable_qp(qp_grp->ufdev, vnic_idx,
+						res->vnic_idx);
+		if (status) {
+			usnic_err("Failed to enable qp %d of %s:%d\n with err %d\n",
+					res->vnic_idx, qp_grp->ufdev->name,
+					vnic_idx, status);
+			goto out_err;
+		}
+	}
+
+	return 0;
+
+out_err:
+	for (i--; i >= 0; i--) {
+		res = res_chunk->res[i];
+		usnic_fwd_disable_qp(qp_grp->ufdev, vnic_idx,
+					res->vnic_idx);
+	}
+
+	return status;
+}
+
+static int disable_qp_grp(struct usnic_ib_qp_grp *qp_grp)
+{
+	int i, vnic_idx;
+	struct usnic_vnic_res_chunk *res_chunk;
+	struct usnic_vnic_res *res;
+	int status = 0;
+
+	lockdep_assert_held(&qp_grp->lock);
+	vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic);
+
+	res_chunk = get_qp_res_chunk(qp_grp);
+	if (IS_ERR_OR_NULL(res_chunk)) {
+		usnic_err("Unable to get qp res with err %ld\n",
+			PTR_ERR(res_chunk));
+		return res_chunk ? PTR_ERR(res_chunk) : -ENOMEM;
+	}
+
+	for (i = 0; i < res_chunk->cnt; i++) {
+		res = res_chunk->res[i];
+		status = usnic_fwd_disable_qp(qp_grp->ufdev, vnic_idx,
+						res->vnic_idx);
+		if (status) {
+			usnic_err("Failed to disable rq %d of %s:%d\n with err %d\n",
+					res->vnic_idx,
+					qp_grp->ufdev->name,
+					vnic_idx, status);
+		}
+	}
+
+	return status;
+
+}
+
+static int init_filter_action(struct usnic_ib_qp_grp *qp_grp,
+				struct usnic_filter_action *uaction)
+{
+	struct usnic_vnic_res_chunk *res_chunk;
+
+	res_chunk = usnic_ib_qp_grp_get_chunk(qp_grp, USNIC_VNIC_RES_TYPE_RQ);
+	if (IS_ERR_OR_NULL(res_chunk)) {
+		usnic_err("Unable to get %s with err %ld\n",
+			usnic_vnic_res_type_to_str(USNIC_VNIC_RES_TYPE_RQ),
+			PTR_ERR(res_chunk));
+		return res_chunk ? PTR_ERR(res_chunk) : -ENOMEM;
+	}
+
+	uaction->vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic);
+	uaction->action.type = FILTER_ACTION_RQ_STEERING;
+	uaction->action.u.rq_idx = res_chunk->res[DFLT_RQ_IDX]->vnic_idx;
+
+	return 0;
+}
+
+static struct usnic_ib_qp_grp_flow*
+create_roce_custom_flow(struct usnic_ib_qp_grp *qp_grp,
+			struct usnic_transport_spec *trans_spec)
+{
+	uint16_t port_num;
+	int err;
+	struct filter filter;
+	struct usnic_filter_action uaction;
+	struct usnic_ib_qp_grp_flow *qp_flow;
+	struct usnic_fwd_flow *flow;
+	enum usnic_transport_type trans_type;
+
+	trans_type = trans_spec->trans_type;
+	port_num = trans_spec->usnic_roce.port_num;
+
+	/* Reserve Port */
+	port_num = usnic_transport_rsrv_port(trans_type, port_num);
+	if (port_num == 0)
+		return ERR_PTR(-EINVAL);
+
+	/* Create Flow */
+	usnic_fwd_init_usnic_filter(&filter, port_num);
+	err = init_filter_action(qp_grp, &uaction);
+	if (err)
+		goto out_unreserve_port;
+
+	flow = usnic_fwd_alloc_flow(qp_grp->ufdev, &filter, &uaction);
+	if (IS_ERR_OR_NULL(flow)) {
+		usnic_err("Unable to alloc flow failed with err %ld\n",
+				PTR_ERR(flow));
+		err = flow ? PTR_ERR(flow) : -EFAULT;
+		goto out_unreserve_port;
+	}
+
+	/* Create Flow Handle */
+	qp_flow = kzalloc(sizeof(*qp_flow), GFP_ATOMIC);
+	if (IS_ERR_OR_NULL(qp_flow)) {
+		err = qp_flow ? PTR_ERR(qp_flow) : -ENOMEM;
+		goto out_dealloc_flow;
+	}
+	qp_flow->flow = flow;
+	qp_flow->trans_type = trans_type;
+	qp_flow->usnic_roce.port_num = port_num;
+	qp_flow->qp_grp = qp_grp;
+	return qp_flow;
+
+out_dealloc_flow:
+	usnic_fwd_dealloc_flow(flow);
+out_unreserve_port:
+	usnic_transport_unrsrv_port(trans_type, port_num);
+	return ERR_PTR(err);
+}
+
+static void release_roce_custom_flow(struct usnic_ib_qp_grp_flow *qp_flow)
+{
+	usnic_fwd_dealloc_flow(qp_flow->flow);
+	usnic_transport_unrsrv_port(qp_flow->trans_type,
+					qp_flow->usnic_roce.port_num);
+	kfree(qp_flow);
+}
+
+static struct usnic_ib_qp_grp_flow*
+create_udp_flow(struct usnic_ib_qp_grp *qp_grp,
+		struct usnic_transport_spec *trans_spec)
+{
+	struct socket *sock;
+	int sock_fd;
+	int err;
+	struct filter filter;
+	struct usnic_filter_action uaction;
+	struct usnic_ib_qp_grp_flow *qp_flow;
+	struct usnic_fwd_flow *flow;
+	enum usnic_transport_type trans_type;
+	uint32_t addr;
+	uint16_t port_num;
+	int proto;
+
+	trans_type = trans_spec->trans_type;
+	sock_fd = trans_spec->udp.sock_fd;
+
+	/* Get and check socket */
+	sock = usnic_transport_get_socket(sock_fd);
+	if (IS_ERR_OR_NULL(sock))
+		return ERR_CAST(sock);
+
+	err = usnic_transport_sock_get_addr(sock, &proto, &addr, &port_num);
+	if (err)
+		goto out_put_sock;
+
+	if (proto != IPPROTO_UDP) {
+		usnic_err("Protocol for fd %d is not UDP", sock_fd);
+		err = -EPERM;
+		goto out_put_sock;
+	}
+
+	/* Create flow */
+	usnic_fwd_init_udp_filter(&filter, addr, port_num);
+	err = init_filter_action(qp_grp, &uaction);
+	if (err)
+		goto out_put_sock;
+
+	flow = usnic_fwd_alloc_flow(qp_grp->ufdev, &filter, &uaction);
+	if (IS_ERR_OR_NULL(flow)) {
+		usnic_err("Unable to alloc flow failed with err %ld\n",
+				PTR_ERR(flow));
+		err = flow ? PTR_ERR(flow) : -EFAULT;
+		goto out_put_sock;
+	}
+
+	/* Create qp_flow */
+	qp_flow = kzalloc(sizeof(*qp_flow), GFP_ATOMIC);
+	if (IS_ERR_OR_NULL(qp_flow)) {
+		err = qp_flow ? PTR_ERR(qp_flow) : -ENOMEM;
+		goto out_dealloc_flow;
+	}
+	qp_flow->flow = flow;
+	qp_flow->trans_type = trans_type;
+	qp_flow->udp.sock = sock;
+	qp_flow->qp_grp = qp_grp;
+	return qp_flow;
+
+out_dealloc_flow:
+	usnic_fwd_dealloc_flow(flow);
+out_put_sock:
+	usnic_transport_put_socket(sock);
+	return ERR_PTR(err);
+}
+
+static void release_udp_flow(struct usnic_ib_qp_grp_flow *qp_flow)
+{
+	usnic_fwd_dealloc_flow(qp_flow->flow);
+	usnic_transport_put_socket(qp_flow->udp.sock);
+	kfree(qp_flow);
+}
+
+static struct usnic_ib_qp_grp_flow*
+create_and_add_flow(struct usnic_ib_qp_grp *qp_grp,
+			struct usnic_transport_spec *trans_spec)
+{
+	struct usnic_ib_qp_grp_flow *qp_flow;
+	enum usnic_transport_type trans_type;
+
+	trans_type = trans_spec->trans_type;
+	switch (trans_type) {
+	case USNIC_TRANSPORT_ROCE_CUSTOM:
+		qp_flow = create_roce_custom_flow(qp_grp, trans_spec);
+		break;
+	case USNIC_TRANSPORT_IPV4_UDP:
+		qp_flow = create_udp_flow(qp_grp, trans_spec);
+		break;
+	default:
+		usnic_err("Unsupported transport %u\n",
+				trans_spec->trans_type);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (!IS_ERR_OR_NULL(qp_flow)) {
+		list_add_tail(&qp_flow->link, &qp_grp->flows_lst);
+		usnic_debugfs_flow_add(qp_flow);
+	}
+
+
+	return qp_flow;
+}
+
+static void release_and_remove_flow(struct usnic_ib_qp_grp_flow *qp_flow)
+{
+	usnic_debugfs_flow_remove(qp_flow);
+	list_del(&qp_flow->link);
+
+	switch (qp_flow->trans_type) {
+	case USNIC_TRANSPORT_ROCE_CUSTOM:
+		release_roce_custom_flow(qp_flow);
+		break;
+	case USNIC_TRANSPORT_IPV4_UDP:
+		release_udp_flow(qp_flow);
+		break;
+	default:
+		WARN(1, "Unsupported transport %u\n",
+				qp_flow->trans_type);
+		break;
+	}
+}
+
+static void release_and_remove_all_flows(struct usnic_ib_qp_grp *qp_grp)
+{
+	struct usnic_ib_qp_grp_flow *qp_flow, *tmp;
+	list_for_each_entry_safe(qp_flow, tmp, &qp_grp->flows_lst, link)
+		release_and_remove_flow(qp_flow);
+}
+
+int usnic_ib_qp_grp_modify(struct usnic_ib_qp_grp *qp_grp,
+				enum ib_qp_state new_state,
+				void *data)
+{
+	int status = 0;
+	int vnic_idx;
+	struct ib_event ib_event;
+	enum ib_qp_state old_state;
+	struct usnic_transport_spec *trans_spec;
+	struct usnic_ib_qp_grp_flow *qp_flow;
+
+	old_state = qp_grp->state;
+	vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic);
+	trans_spec = (struct usnic_transport_spec *) data;
+
+	spin_lock(&qp_grp->lock);
+	switch (new_state) {
+	case IB_QPS_RESET:
+		switch (old_state) {
+		case IB_QPS_RESET:
+			/* NO-OP */
+			break;
+		case IB_QPS_INIT:
+			release_and_remove_all_flows(qp_grp);
+			status = 0;
+			break;
+		case IB_QPS_RTR:
+		case IB_QPS_RTS:
+		case IB_QPS_ERR:
+			status = disable_qp_grp(qp_grp);
+			release_and_remove_all_flows(qp_grp);
+			break;
+		default:
+			status = -EINVAL;
+		}
+		break;
+	case IB_QPS_INIT:
+		switch (old_state) {
+		case IB_QPS_RESET:
+			if (trans_spec) {
+				qp_flow = create_and_add_flow(qp_grp,
+								trans_spec);
+				if (IS_ERR_OR_NULL(qp_flow)) {
+					status = qp_flow ? PTR_ERR(qp_flow) : -EFAULT;
+					break;
+				}
+			} else {
+				/*
+				 * Optional to specify filters.
+				 */
+				status = 0;
+			}
+			break;
+		case IB_QPS_INIT:
+			if (trans_spec) {
+				qp_flow = create_and_add_flow(qp_grp,
+								trans_spec);
+				if (IS_ERR_OR_NULL(qp_flow)) {
+					status = qp_flow ? PTR_ERR(qp_flow) : -EFAULT;
+					break;
+				}
+			} else {
+				/*
+				 * Doesn't make sense to go into INIT state
+				 * from INIT state w/o adding filters.
+				 */
+				status = -EINVAL;
+			}
+			break;
+		case IB_QPS_RTR:
+			status = disable_qp_grp(qp_grp);
+			break;
+		case IB_QPS_RTS:
+			status = disable_qp_grp(qp_grp);
+			break;
+		default:
+			status = -EINVAL;
+		}
+		break;
+	case IB_QPS_RTR:
+		switch (old_state) {
+		case IB_QPS_INIT:
+			status = enable_qp_grp(qp_grp);
+			break;
+		default:
+			status = -EINVAL;
+		}
+		break;
+	case IB_QPS_RTS:
+		switch (old_state) {
+		case IB_QPS_RTR:
+			/* NO-OP FOR NOW */
+			break;
+		default:
+			status = -EINVAL;
+		}
+		break;
+	case IB_QPS_ERR:
+		ib_event.device = &qp_grp->vf->pf->ib_dev;
+		ib_event.element.qp = &qp_grp->ibqp;
+		ib_event.event = IB_EVENT_QP_FATAL;
+
+		switch (old_state) {
+		case IB_QPS_RESET:
+			qp_grp->ibqp.event_handler(&ib_event,
+					qp_grp->ibqp.qp_context);
+			break;
+		case IB_QPS_INIT:
+			release_and_remove_all_flows(qp_grp);
+			qp_grp->ibqp.event_handler(&ib_event,
+					qp_grp->ibqp.qp_context);
+			break;
+		case IB_QPS_RTR:
+		case IB_QPS_RTS:
+			status = disable_qp_grp(qp_grp);
+			release_and_remove_all_flows(qp_grp);
+			qp_grp->ibqp.event_handler(&ib_event,
+					qp_grp->ibqp.qp_context);
+			break;
+		default:
+			status = -EINVAL;
+		}
+		break;
+	default:
+		status = -EINVAL;
+	}
+	spin_unlock(&qp_grp->lock);
+
+	if (!status) {
+		qp_grp->state = new_state;
+		usnic_info("Transistioned %u from %s to %s",
+		qp_grp->grp_id,
+		usnic_ib_qp_grp_state_to_string(old_state),
+		usnic_ib_qp_grp_state_to_string(new_state));
+	} else {
+		usnic_err("Failed to transistion %u from %s to %s",
+		qp_grp->grp_id,
+		usnic_ib_qp_grp_state_to_string(old_state),
+		usnic_ib_qp_grp_state_to_string(new_state));
+	}
+
+	return status;
+}
+
+static struct usnic_vnic_res_chunk**
+alloc_res_chunk_list(struct usnic_vnic *vnic,
+			struct usnic_vnic_res_spec *res_spec, void *owner_obj)
+{
+	enum usnic_vnic_res_type res_type;
+	struct usnic_vnic_res_chunk **res_chunk_list;
+	int err, i, res_cnt, res_lst_sz;
+
+	for (res_lst_sz = 0;
+		res_spec->resources[res_lst_sz].type != USNIC_VNIC_RES_TYPE_EOL;
+		res_lst_sz++) {
+		/* Do Nothing */
+	}
+
+	res_chunk_list = kzalloc(sizeof(*res_chunk_list)*(res_lst_sz+1),
+					GFP_ATOMIC);
+	if (!res_chunk_list)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; res_spec->resources[i].type != USNIC_VNIC_RES_TYPE_EOL;
+		i++) {
+		res_type = res_spec->resources[i].type;
+		res_cnt = res_spec->resources[i].cnt;
+
+		res_chunk_list[i] = usnic_vnic_get_resources(vnic, res_type,
+					res_cnt, owner_obj);
+		if (IS_ERR_OR_NULL(res_chunk_list[i])) {
+			err = res_chunk_list[i] ?
+					PTR_ERR(res_chunk_list[i]) : -ENOMEM;
+			usnic_err("Failed to get %s from %s with err %d\n",
+				usnic_vnic_res_type_to_str(res_type),
+				usnic_vnic_pci_name(vnic),
+				err);
+			goto out_free_res;
+		}
+	}
+
+	return res_chunk_list;
+
+out_free_res:
+	for (i--; i > 0; i--)
+		usnic_vnic_put_resources(res_chunk_list[i]);
+	kfree(res_chunk_list);
+	return ERR_PTR(err);
+}
+
+static void free_qp_grp_res(struct usnic_vnic_res_chunk **res_chunk_list)
+{
+	int i;
+	for (i = 0; res_chunk_list[i]; i++)
+		usnic_vnic_put_resources(res_chunk_list[i]);
+	kfree(res_chunk_list);
+}
+
+static int qp_grp_and_vf_bind(struct usnic_ib_vf *vf,
+				struct usnic_ib_pd *pd,
+				struct usnic_ib_qp_grp *qp_grp)
+{
+	int err;
+	struct pci_dev *pdev;
+
+	lockdep_assert_held(&vf->lock);
+
+	pdev = usnic_vnic_get_pdev(vf->vnic);
+	if (vf->qp_grp_ref_cnt == 0) {
+		err = usnic_uiom_attach_dev_to_pd(pd->umem_pd, &pdev->dev);
+		if (err) {
+			usnic_err("Failed to attach %s to domain\n",
+					pci_name(pdev));
+			return err;
+		}
+		vf->pd = pd;
+	}
+	vf->qp_grp_ref_cnt++;
+
+	WARN_ON(vf->pd != pd);
+	qp_grp->vf = vf;
+
+	return 0;
+}
+
+static void qp_grp_and_vf_unbind(struct usnic_ib_qp_grp *qp_grp)
+{
+	struct pci_dev *pdev;
+	struct usnic_ib_pd *pd;
+
+	lockdep_assert_held(&qp_grp->vf->lock);
+
+	pd = qp_grp->vf->pd;
+	pdev = usnic_vnic_get_pdev(qp_grp->vf->vnic);
+	if (--qp_grp->vf->qp_grp_ref_cnt == 0) {
+		qp_grp->vf->pd = NULL;
+		usnic_uiom_detach_dev_from_pd(pd->umem_pd, &pdev->dev);
+	}
+	qp_grp->vf = NULL;
+}
+
+static void log_spec(struct usnic_vnic_res_spec *res_spec)
+{
+	char buf[512];
+	usnic_vnic_spec_dump(buf, sizeof(buf), res_spec);
+	usnic_dbg("%s\n", buf);
+}
+
+static int qp_grp_id_from_flow(struct usnic_ib_qp_grp_flow *qp_flow,
+				uint32_t *id)
+{
+	enum usnic_transport_type trans_type = qp_flow->trans_type;
+	int err;
+
+	switch (trans_type) {
+	case USNIC_TRANSPORT_ROCE_CUSTOM:
+		*id = qp_flow->usnic_roce.port_num;
+		break;
+	case USNIC_TRANSPORT_IPV4_UDP:
+		err = usnic_transport_sock_get_addr(qp_flow->udp.sock,
+							NULL, NULL,
+							(uint16_t *) id);
+		if (err)
+			return err;
+		break;
+	default:
+		usnic_err("Unsupported transport %u\n", trans_type);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+struct usnic_ib_qp_grp *
+usnic_ib_qp_grp_create(struct usnic_fwd_dev *ufdev, struct usnic_ib_vf *vf,
+			struct usnic_ib_pd *pd,
+			struct usnic_vnic_res_spec *res_spec,
+			struct usnic_transport_spec *transport_spec)
+{
+	struct usnic_ib_qp_grp *qp_grp;
+	int err;
+	enum usnic_transport_type transport = transport_spec->trans_type;
+	struct usnic_ib_qp_grp_flow *qp_flow;
+
+	lockdep_assert_held(&vf->lock);
+
+	err = usnic_vnic_res_spec_satisfied(&min_transport_spec[transport],
+						res_spec);
+	if (err) {
+		usnic_err("Spec does not meet miniumum req for transport %d\n",
+				transport);
+		log_spec(res_spec);
+		return ERR_PTR(err);
+	}
+
+	qp_grp = kzalloc(sizeof(*qp_grp), GFP_ATOMIC);
+	if (!qp_grp) {
+		usnic_err("Unable to alloc qp_grp - Out of memory\n");
+		return NULL;
+	}
+
+	qp_grp->res_chunk_list = alloc_res_chunk_list(vf->vnic, res_spec,
+							qp_grp);
+	if (IS_ERR_OR_NULL(qp_grp->res_chunk_list)) {
+		err = qp_grp->res_chunk_list ?
+				PTR_ERR(qp_grp->res_chunk_list) : -ENOMEM;
+		usnic_err("Unable to alloc res for %d with err %d\n",
+				qp_grp->grp_id, err);
+		goto out_free_qp_grp;
+	}
+
+	err = qp_grp_and_vf_bind(vf, pd, qp_grp);
+	if (err)
+		goto out_free_res;
+
+	INIT_LIST_HEAD(&qp_grp->flows_lst);
+	spin_lock_init(&qp_grp->lock);
+	qp_grp->ufdev = ufdev;
+	qp_grp->state = IB_QPS_RESET;
+	qp_grp->owner_pid = current->pid;
+
+	qp_flow = create_and_add_flow(qp_grp, transport_spec);
+	if (IS_ERR_OR_NULL(qp_flow)) {
+		usnic_err("Unable to create and add flow with err %ld\n",
+				PTR_ERR(qp_flow));
+		err = qp_flow ? PTR_ERR(qp_flow) : -EFAULT;
+		goto out_qp_grp_vf_unbind;
+	}
+
+	err = qp_grp_id_from_flow(qp_flow, &qp_grp->grp_id);
+	if (err)
+		goto out_release_flow;
+	qp_grp->ibqp.qp_num = qp_grp->grp_id;
+
+	usnic_ib_sysfs_qpn_add(qp_grp);
+
+	return qp_grp;
+
+out_release_flow:
+	release_and_remove_flow(qp_flow);
+out_qp_grp_vf_unbind:
+	qp_grp_and_vf_unbind(qp_grp);
+out_free_res:
+	free_qp_grp_res(qp_grp->res_chunk_list);
+out_free_qp_grp:
+	kfree(qp_grp);
+
+	return ERR_PTR(err);
+}
+
+void usnic_ib_qp_grp_destroy(struct usnic_ib_qp_grp *qp_grp)
+{
+
+	WARN_ON(qp_grp->state != IB_QPS_RESET);
+	lockdep_assert_held(&qp_grp->vf->lock);
+
+	release_and_remove_all_flows(qp_grp);
+	usnic_ib_sysfs_qpn_remove(qp_grp);
+	qp_grp_and_vf_unbind(qp_grp);
+	free_qp_grp_res(qp_grp->res_chunk_list);
+	kfree(qp_grp);
+}
+
+struct usnic_vnic_res_chunk*
+usnic_ib_qp_grp_get_chunk(struct usnic_ib_qp_grp *qp_grp,
+				enum usnic_vnic_res_type res_type)
+{
+	int i;
+
+	for (i = 0; qp_grp->res_chunk_list[i]; i++) {
+		if (qp_grp->res_chunk_list[i]->type == res_type)
+			return qp_grp->res_chunk_list[i];
+	}
+
+	return ERR_PTR(-EINVAL);
+}

drivers/infiniband/hw/usnic/usnic_ib_qp_grp.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_IB_QP_GRP_H_
+#define USNIC_IB_QP_GRP_H_
+
+#include <linux/debugfs.h>
+#include <rdma/ib_verbs.h>
+
+#include "usnic_ib.h"
+#include "usnic_abi.h"
+#include "usnic_fwd.h"
+#include "usnic_vnic.h"
+
+/*
+ * The qp group struct represents all the hw resources needed to present a ib_qp
+ */
+struct usnic_ib_qp_grp {
+	struct ib_qp				ibqp;
+	enum ib_qp_state			state;
+	int					grp_id;
+
+	struct usnic_fwd_dev			*ufdev;
+	struct usnic_ib_ucontext		*ctx;
+	struct list_head			flows_lst;
+
+	struct usnic_vnic_res_chunk		**res_chunk_list;
+
+	pid_t					owner_pid;
+	struct usnic_ib_vf			*vf;
+	struct list_head			link;
+
+	spinlock_t				lock;
+
+	struct kobject				kobj;
+};
+
+struct usnic_ib_qp_grp_flow {
+	struct usnic_fwd_flow		*flow;
+	enum usnic_transport_type	trans_type;
+	union {
+		struct {
+			uint16_t	port_num;
+		} usnic_roce;
+		struct {
+			struct socket	*sock;
+		} udp;
+	};
+	struct usnic_ib_qp_grp		*qp_grp;
+	struct list_head		link;
+
+	/* Debug FS */
+	struct dentry			*dbgfs_dentry;
+	char				dentry_name[32];
+};
+
+static const struct
+usnic_vnic_res_spec min_transport_spec[USNIC_TRANSPORT_MAX] = {
+	{ /*USNIC_TRANSPORT_UNKNOWN*/
+		.resources = {
+			{.type = USNIC_VNIC_RES_TYPE_EOL,	.cnt = 0,},
+		},
+	},
+	{ /*USNIC_TRANSPORT_ROCE_CUSTOM*/
+		.resources = {
+			{.type = USNIC_VNIC_RES_TYPE_WQ,	.cnt = 1,},
+			{.type = USNIC_VNIC_RES_TYPE_RQ,	.cnt = 1,},
+			{.type = USNIC_VNIC_RES_TYPE_CQ,	.cnt = 1,},
+			{.type = USNIC_VNIC_RES_TYPE_EOL,	.cnt = 0,},
+		},
+	},
+	{ /*USNIC_TRANSPORT_IPV4_UDP*/
+		.resources = {
+			{.type = USNIC_VNIC_RES_TYPE_WQ,	.cnt = 1,},
+			{.type = USNIC_VNIC_RES_TYPE_RQ,	.cnt = 1,},
+			{.type = USNIC_VNIC_RES_TYPE_CQ,	.cnt = 1,},
+			{.type = USNIC_VNIC_RES_TYPE_EOL,	.cnt = 0,},
+		},
+	},
+};
+
+const char *usnic_ib_qp_grp_state_to_string(enum ib_qp_state state);
+int usnic_ib_qp_grp_dump_hdr(char *buf, int buf_sz);
+int usnic_ib_qp_grp_dump_rows(void *obj, char *buf, int buf_sz);
+struct usnic_ib_qp_grp *
+usnic_ib_qp_grp_create(struct usnic_fwd_dev *ufdev, struct usnic_ib_vf *vf,
+			struct usnic_ib_pd *pd,
+			struct usnic_vnic_res_spec *res_spec,
+			struct usnic_transport_spec *trans_spec);
+void usnic_ib_qp_grp_destroy(struct usnic_ib_qp_grp *qp_grp);
+int usnic_ib_qp_grp_modify(struct usnic_ib_qp_grp *qp_grp,
+				enum ib_qp_state new_state,
+				void *data);
+struct usnic_vnic_res_chunk
+*usnic_ib_qp_grp_get_chunk(struct usnic_ib_qp_grp *qp_grp,
+				enum usnic_vnic_res_type type);
+static inline
+struct usnic_ib_qp_grp *to_uqp_grp(struct ib_qp *ibqp)
+{
+	return container_of(ibqp, struct usnic_ib_qp_grp, ibqp);
+}
+#endif /* USNIC_IB_QP_GRP_H_ */

drivers/infiniband/hw/usnic/usnic_ib_sysfs.c

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+
+#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_addr.h>
+
+#include "usnic_common_util.h"
+#include "usnic_ib.h"
+#include "usnic_ib_qp_grp.h"
+#include "usnic_vnic.h"
+#include "usnic_ib_verbs.h"
+#include "usnic_log.h"
+
+static ssize_t usnic_ib_show_fw_ver(struct device *device,
+					struct device_attribute *attr,
+					char *buf)
+{
+	struct usnic_ib_dev *us_ibdev =
+		container_of(device, struct usnic_ib_dev, ib_dev.dev);
+	struct ethtool_drvinfo info;
+
+	mutex_lock(&us_ibdev->usdev_lock);
+	us_ibdev->netdev->ethtool_ops->get_drvinfo(us_ibdev->netdev, &info);
+	mutex_unlock(&us_ibdev->usdev_lock);
+
+	return scnprintf(buf, PAGE_SIZE, "%s\n", info.fw_version);
+}
+
+static ssize_t usnic_ib_show_board(struct device *device,
+					struct device_attribute *attr,
+					char *buf)
+{
+	struct usnic_ib_dev *us_ibdev =
+		container_of(device, struct usnic_ib_dev, ib_dev.dev);
+	unsigned short subsystem_device_id;
+
+	mutex_lock(&us_ibdev->usdev_lock);
+	subsystem_device_id = us_ibdev->pdev->subsystem_device;
+	mutex_unlock(&us_ibdev->usdev_lock);
+
+	return scnprintf(buf, PAGE_SIZE, "%hu\n", subsystem_device_id);
+}
+
+/*
+ * Report the configuration for this PF
+ */
+static ssize_t
+usnic_ib_show_config(struct device *device, struct device_attribute *attr,
+			char *buf)
+{
+	struct usnic_ib_dev *us_ibdev;
+	char *ptr;
+	unsigned left;
+	unsigned n;
+	enum usnic_vnic_res_type res_type;
+
+	us_ibdev = container_of(device, struct usnic_ib_dev, ib_dev.dev);
+
+	/* Buffer space limit is 1 page */
+	ptr = buf;
+	left = PAGE_SIZE;
+
+	mutex_lock(&us_ibdev->usdev_lock);
+	if (atomic_read(&us_ibdev->vf_cnt.refcount) > 0) {
+		char *busname;
+
+		/*
+		 * bus name seems to come with annoying prefix.
+		 * Remove it if it is predictable
+		 */
+		busname = us_ibdev->pdev->bus->name;
+		if (strncmp(busname, "PCI Bus ", 8) == 0)
+			busname += 8;
+
+		n = scnprintf(ptr, left,
+			"%s: %s:%d.%d, %s, %pM, %u VFs\n Per VF:",
+			us_ibdev->ib_dev.name,
+			busname,
+			PCI_SLOT(us_ibdev->pdev->devfn),
+			PCI_FUNC(us_ibdev->pdev->devfn),
+			netdev_name(us_ibdev->netdev),
+			us_ibdev->ufdev->mac,
+			atomic_read(&us_ibdev->vf_cnt.refcount));
+		UPDATE_PTR_LEFT(n, ptr, left);
+
+		for (res_type = USNIC_VNIC_RES_TYPE_EOL;
+				res_type < USNIC_VNIC_RES_TYPE_MAX;
+				res_type++) {
+			if (us_ibdev->vf_res_cnt[res_type] == 0)
+				continue;
+			n = scnprintf(ptr, left, " %d %s%s",
+				us_ibdev->vf_res_cnt[res_type],
+				usnic_vnic_res_type_to_str(res_type),
+				(res_type < (USNIC_VNIC_RES_TYPE_MAX - 1)) ?
+				 "," : "");
+			UPDATE_PTR_LEFT(n, ptr, left);
+		}
+		n = scnprintf(ptr, left, "\n");
+		UPDATE_PTR_LEFT(n, ptr, left);
+	} else {
+		n = scnprintf(ptr, left, "%s: no VFs\n",
+				us_ibdev->ib_dev.name);
+		UPDATE_PTR_LEFT(n, ptr, left);
+	}
+	mutex_unlock(&us_ibdev->usdev_lock);
+
+	return ptr - buf;
+}
+
+static ssize_t
+usnic_ib_show_iface(struct device *device, struct device_attribute *attr,
+			char *buf)
+{
+	struct usnic_ib_dev *us_ibdev;
+
+	us_ibdev = container_of(device, struct usnic_ib_dev, ib_dev.dev);
+
+	return scnprintf(buf, PAGE_SIZE, "%s\n",
+			netdev_name(us_ibdev->netdev));
+}
+
+static ssize_t
+usnic_ib_show_max_vf(struct device *device, struct device_attribute *attr,
+			char *buf)
+{
+	struct usnic_ib_dev *us_ibdev;
+
+	us_ibdev = container_of(device, struct usnic_ib_dev, ib_dev.dev);
+
+	return scnprintf(buf, PAGE_SIZE, "%u\n",
+			atomic_read(&us_ibdev->vf_cnt.refcount));
+}
+
+static ssize_t
+usnic_ib_show_qp_per_vf(struct device *device, struct device_attribute *attr,
+			char *buf)
+{
+	struct usnic_ib_dev *us_ibdev;
+	int qp_per_vf;
+
+	us_ibdev = container_of(device, struct usnic_ib_dev, ib_dev.dev);
+	qp_per_vf = max(us_ibdev->vf_res_cnt[USNIC_VNIC_RES_TYPE_WQ],
+			us_ibdev->vf_res_cnt[USNIC_VNIC_RES_TYPE_RQ]);
+
+	return scnprintf(buf, PAGE_SIZE,
+				"%d\n", qp_per_vf);
+}
+
+static ssize_t
+usnic_ib_show_cq_per_vf(struct device *device, struct device_attribute *attr,
+			char *buf)
+{
+	struct usnic_ib_dev *us_ibdev;
+
+	us_ibdev = container_of(device, struct usnic_ib_dev, ib_dev.dev);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n",
+			us_ibdev->vf_res_cnt[USNIC_VNIC_RES_TYPE_CQ]);
+}
+
+static DEVICE_ATTR(fw_ver, S_IRUGO, usnic_ib_show_fw_ver, NULL);
+static DEVICE_ATTR(board_id, S_IRUGO, usnic_ib_show_board, NULL);
+static DEVICE_ATTR(config, S_IRUGO, usnic_ib_show_config, NULL);
+static DEVICE_ATTR(iface, S_IRUGO, usnic_ib_show_iface, NULL);
+static DEVICE_ATTR(max_vf, S_IRUGO, usnic_ib_show_max_vf, NULL);
+static DEVICE_ATTR(qp_per_vf, S_IRUGO, usnic_ib_show_qp_per_vf, NULL);
+static DEVICE_ATTR(cq_per_vf, S_IRUGO, usnic_ib_show_cq_per_vf, NULL);
+
+static struct device_attribute *usnic_class_attributes[] = {
+	&dev_attr_fw_ver,
+	&dev_attr_board_id,
+	&dev_attr_config,
+	&dev_attr_iface,
+	&dev_attr_max_vf,
+	&dev_attr_qp_per_vf,
+	&dev_attr_cq_per_vf,
+};
+
+struct qpn_attribute {
+	struct attribute attr;
+	ssize_t (*show)(struct usnic_ib_qp_grp *, char *buf);
+};
+
+/*
+ * Definitions for supporting QPN entries in sysfs
+ */
+static ssize_t
+usnic_ib_qpn_attr_show(struct kobject *kobj, struct attribute *attr, char *buf)
+{
+	struct usnic_ib_qp_grp *qp_grp;
+	struct qpn_attribute *qpn_attr;
+
+	qp_grp = container_of(kobj, struct usnic_ib_qp_grp, kobj);
+	qpn_attr = container_of(attr, struct qpn_attribute, attr);
+
+	return qpn_attr->show(qp_grp, buf);
+}
+
+static const struct sysfs_ops usnic_ib_qpn_sysfs_ops = {
+	.show = usnic_ib_qpn_attr_show
+};
+
+#define QPN_ATTR_RO(NAME) \
+struct qpn_attribute qpn_attr_##NAME = __ATTR_RO(NAME)
+
+static ssize_t context_show(struct usnic_ib_qp_grp *qp_grp, char *buf)
+{
+	return scnprintf(buf, PAGE_SIZE, "0x%p\n", qp_grp->ctx);
+}
+
+static ssize_t summary_show(struct usnic_ib_qp_grp *qp_grp, char *buf)
+{
+	int i, j, n;
+	int left;
+	char *ptr;
+	struct usnic_vnic_res_chunk *res_chunk;
+	struct usnic_vnic_res *vnic_res;
+
+	left = PAGE_SIZE;
+	ptr = buf;
+
+	n = scnprintf(ptr, left,
+			"QPN: %d State: (%s) PID: %u VF Idx: %hu ",
+			qp_grp->ibqp.qp_num,
+			usnic_ib_qp_grp_state_to_string(qp_grp->state),
+			qp_grp->owner_pid,
+			usnic_vnic_get_index(qp_grp->vf->vnic));
+	UPDATE_PTR_LEFT(n, ptr, left);
+
+	for (i = 0; qp_grp->res_chunk_list[i]; i++) {
+		res_chunk = qp_grp->res_chunk_list[i];
+		for (j = 0; j < res_chunk->cnt; j++) {
+			vnic_res = res_chunk->res[j];
+			n = scnprintf(ptr, left, "%s[%d] ",
+				usnic_vnic_res_type_to_str(vnic_res->type),
+				vnic_res->vnic_idx);
+			UPDATE_PTR_LEFT(n, ptr, left);
+		}
+	}
+
+	n = scnprintf(ptr, left, "\n");
+	UPDATE_PTR_LEFT(n, ptr, left);
+
+	return ptr - buf;
+}
+
+static QPN_ATTR_RO(context);
+static QPN_ATTR_RO(summary);
+
+static struct attribute *usnic_ib_qpn_default_attrs[] = {
+	&qpn_attr_context.attr,
+	&qpn_attr_summary.attr,
+	NULL
+};
+
+static struct kobj_type usnic_ib_qpn_type = {
+	.sysfs_ops = &usnic_ib_qpn_sysfs_ops,
+	.default_attrs = usnic_ib_qpn_default_attrs
+};
+
+int usnic_ib_sysfs_register_usdev(struct usnic_ib_dev *us_ibdev)
+{
+	int i;
+	int err;
+	for (i = 0; i < ARRAY_SIZE(usnic_class_attributes); ++i) {
+		err = device_create_file(&us_ibdev->ib_dev.dev,
+						usnic_class_attributes[i]);
+		if (err) {
+			usnic_err("Failed to create device file %d for %s eith err %d",
+				i, us_ibdev->ib_dev.name, err);
+			return -EINVAL;
+		}
+	}
+
+	/* create kernel object for looking at individual QPs */
+	kobject_get(&us_ibdev->ib_dev.dev.kobj);
+	us_ibdev->qpn_kobj = kobject_create_and_add("qpn",
+			&us_ibdev->ib_dev.dev.kobj);
+	if (us_ibdev->qpn_kobj == NULL) {
+		kobject_put(&us_ibdev->ib_dev.dev.kobj);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void usnic_ib_sysfs_unregister_usdev(struct usnic_ib_dev *us_ibdev)
+{
+	int i;
+	for (i = 0; i < ARRAY_SIZE(usnic_class_attributes); ++i) {
+		device_remove_file(&us_ibdev->ib_dev.dev,
+					usnic_class_attributes[i]);
+	}
+
+	kobject_put(us_ibdev->qpn_kobj);
+}
+
+void usnic_ib_sysfs_qpn_add(struct usnic_ib_qp_grp *qp_grp)
+{
+	struct usnic_ib_dev *us_ibdev;
+	int err;
+
+	us_ibdev = qp_grp->vf->pf;
+
+	err = kobject_init_and_add(&qp_grp->kobj, &usnic_ib_qpn_type,
+			kobject_get(us_ibdev->qpn_kobj),
+			"%d", qp_grp->grp_id);
+	if (err) {
+		kobject_put(us_ibdev->qpn_kobj);
+		return;
+	}
+}
+
+void usnic_ib_sysfs_qpn_remove(struct usnic_ib_qp_grp *qp_grp)
+{
+	struct usnic_ib_dev *us_ibdev;
+
+	us_ibdev = qp_grp->vf->pf;
+
+	kobject_put(&qp_grp->kobj);
+	kobject_put(us_ibdev->qpn_kobj);
+}

drivers/infiniband/hw/usnic/usnic_ib_sysfs.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_IB_SYSFS_H_
+#define USNIC_IB_SYSFS_H_
+
+#include "usnic_ib.h"
+
+int usnic_ib_sysfs_register_usdev(struct usnic_ib_dev *us_ibdev);
+void usnic_ib_sysfs_unregister_usdev(struct usnic_ib_dev *us_ibdev);
+void usnic_ib_sysfs_qpn_add(struct usnic_ib_qp_grp *qp_grp);
+void usnic_ib_sysfs_qpn_remove(struct usnic_ib_qp_grp *qp_grp);
+
+#endif /* !USNIC_IB_SYSFS_H_ */

drivers/infiniband/hw/usnic/usnic_ib_verbs.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_IB_VERBS_H_
+#define USNIC_IB_VERBS_H_
+
+#include "usnic_ib.h"
+
+enum rdma_link_layer usnic_ib_port_link_layer(struct ib_device *device,
+						u8 port_num);
+int usnic_ib_query_device(struct ib_device *ibdev,
+				struct ib_device_attr *props);
+int usnic_ib_query_port(struct ib_device *ibdev, u8 port,
+				struct ib_port_attr *props);
+int usnic_ib_query_qp(struct ib_qp *qp, struct ib_qp_attr *qp_attr,
+				int qp_attr_mask,
+				struct ib_qp_init_attr *qp_init_attr);
+int usnic_ib_query_gid(struct ib_device *ibdev, u8 port, int index,
+				union ib_gid *gid);
+int usnic_ib_query_pkey(struct ib_device *ibdev, u8 port, u16 index,
+				u16 *pkey);
+struct ib_pd *usnic_ib_alloc_pd(struct ib_device *ibdev,
+				struct ib_ucontext *context,
+				struct ib_udata *udata);
+int usnic_ib_dealloc_pd(struct ib_pd *pd);
+struct ib_qp *usnic_ib_create_qp(struct ib_pd *pd,
+					struct ib_qp_init_attr *init_attr,
+					struct ib_udata *udata);
+int usnic_ib_destroy_qp(struct ib_qp *qp);
+int usnic_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
+				int attr_mask, struct ib_udata *udata);
+struct ib_cq *usnic_ib_create_cq(struct ib_device *ibdev, int entries,
+					int vector, struct ib_ucontext *context,
+					struct ib_udata *udata);
+int usnic_ib_destroy_cq(struct ib_cq *cq);
+struct ib_mr *usnic_ib_reg_mr(struct ib_pd *pd, u64 start, u64 length,
+				u64 virt_addr, int access_flags,
+				struct ib_udata *udata);
+int usnic_ib_dereg_mr(struct ib_mr *ibmr);
+struct ib_ucontext *usnic_ib_alloc_ucontext(struct ib_device *ibdev,
+						struct ib_udata *udata);
+int usnic_ib_dealloc_ucontext(struct ib_ucontext *ibcontext);
+int usnic_ib_mmap(struct ib_ucontext *context,
+			struct vm_area_struct *vma);
+struct ib_ah *usnic_ib_create_ah(struct ib_pd *pd,
+					struct ib_ah_attr *ah_attr);
+int usnic_ib_destroy_ah(struct ib_ah *ah);
+int usnic_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
+			struct ib_send_wr **bad_wr);
+int usnic_ib_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
+			struct ib_recv_wr **bad_wr);
+int usnic_ib_poll_cq(struct ib_cq *ibcq, int num_entries,
+			struct ib_wc *wc);
+int usnic_ib_req_notify_cq(struct ib_cq *cq,
+				enum ib_cq_notify_flags flags);
+struct ib_mr *usnic_ib_get_dma_mr(struct ib_pd *pd, int acc);
+#endif /* !USNIC_IB_VERBS_H */

drivers/infiniband/hw/usnic/usnic_log.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_LOG_H_
+#define USNIC_LOG_H_
+
+#include "usnic.h"
+
+extern unsigned int usnic_log_lvl;
+
+#define USNIC_LOG_LVL_NONE		(0)
+#define USNIC_LOG_LVL_ERR		(1)
+#define USNIC_LOG_LVL_INFO		(2)
+#define USNIC_LOG_LVL_DBG		(3)
+
+#define usnic_printk(lvl, args...) \
+	do { \
+		printk(lvl "%s:%s:%d: ", DRV_NAME, __func__, \
+				__LINE__); \
+		printk(args); \
+	} while (0)
+
+#define usnic_dbg(args...) \
+	do { \
+		if (unlikely(usnic_log_lvl >= USNIC_LOG_LVL_DBG)) { \
+			usnic_printk(KERN_INFO, args); \
+	} \
+} while (0)
+
+#define usnic_info(args...) \
+do { \
+	if (usnic_log_lvl >= USNIC_LOG_LVL_INFO) { \
+			usnic_printk(KERN_INFO, args); \
+	} \
+} while (0)
+
+#define usnic_err(args...) \
+	do { \
+		if (usnic_log_lvl >= USNIC_LOG_LVL_ERR) { \
+			usnic_printk(KERN_ERR, args); \
+		} \
+	} while (0)
+#endif /* !USNIC_LOG_H_ */

drivers/infiniband/hw/usnic/usnic_transport.c

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+#include <linux/bitmap.h>
+#include <linux/file.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <net/inet_sock.h>
+
+#include "usnic_transport.h"
+#include "usnic_log.h"
+
+/* ROCE */
+static unsigned long *roce_bitmap;
+static u16 roce_next_port = 1;
+#define ROCE_BITMAP_SZ ((1 << (8 /*CHAR_BIT*/ * sizeof(u16)))/8 /*CHAR BIT*/)
+static DEFINE_SPINLOCK(roce_bitmap_lock);
+
+const char *usnic_transport_to_str(enum usnic_transport_type type)
+{
+	switch (type) {
+	case USNIC_TRANSPORT_UNKNOWN:
+		return "Unknown";
+	case USNIC_TRANSPORT_ROCE_CUSTOM:
+		return "roce custom";
+	case USNIC_TRANSPORT_IPV4_UDP:
+		return "IPv4 UDP";
+	case USNIC_TRANSPORT_MAX:
+		return "Max?";
+	default:
+		return "Not known";
+	}
+}
+
+int usnic_transport_sock_to_str(char *buf, int buf_sz,
+					struct socket *sock)
+{
+	int err;
+	uint32_t addr;
+	uint16_t port;
+	int proto;
+
+	memset(buf, 0, buf_sz);
+	err = usnic_transport_sock_get_addr(sock, &proto, &addr, &port);
+	if (err)
+		return 0;
+
+	return scnprintf(buf, buf_sz, "Proto:%u Addr:%pI4h Port:%hu",
+			proto, &addr, port);
+}
+
+/*
+ * reserve a port number.  if "0" specified, we will try to pick one
+ * starting at roce_next_port.  roce_next_port will take on the values
+ * 1..4096
+ */
+u16 usnic_transport_rsrv_port(enum usnic_transport_type type, u16 port_num)
+{
+	if (type == USNIC_TRANSPORT_ROCE_CUSTOM) {
+		spin_lock(&roce_bitmap_lock);
+		if (!port_num) {
+			port_num = bitmap_find_next_zero_area(roce_bitmap,
+						ROCE_BITMAP_SZ,
+						roce_next_port /* start */,
+						1 /* nr */,
+						0 /* align */);
+			roce_next_port = (port_num & 4095) + 1;
+		} else if (test_bit(port_num, roce_bitmap)) {
+			usnic_err("Failed to allocate port for %s\n",
+					usnic_transport_to_str(type));
+			spin_unlock(&roce_bitmap_lock);
+			goto out_fail;
+		}
+		bitmap_set(roce_bitmap, port_num, 1);
+		spin_unlock(&roce_bitmap_lock);
+	} else {
+		usnic_err("Failed to allocate port - transport %s unsupported\n",
+				usnic_transport_to_str(type));
+		goto out_fail;
+	}
+
+	usnic_dbg("Allocating port %hu for %s\n", port_num,
+			usnic_transport_to_str(type));
+	return port_num;
+
+out_fail:
+	return 0;
+}
+
+void usnic_transport_unrsrv_port(enum usnic_transport_type type, u16 port_num)
+{
+	if (type == USNIC_TRANSPORT_ROCE_CUSTOM) {
+		spin_lock(&roce_bitmap_lock);
+		if (!port_num) {
+			usnic_err("Unreserved unvalid port num 0 for %s\n",
+					usnic_transport_to_str(type));
+			goto out_roce_custom;
+		}
+
+		if (!test_bit(port_num, roce_bitmap)) {
+			usnic_err("Unreserving invalid %hu for %s\n",
+					port_num,
+					usnic_transport_to_str(type));
+			goto out_roce_custom;
+		}
+		bitmap_clear(roce_bitmap, port_num, 1);
+		usnic_dbg("Freeing port %hu for %s\n", port_num,
+				usnic_transport_to_str(type));
+out_roce_custom:
+		spin_unlock(&roce_bitmap_lock);
+	} else {
+		usnic_err("Freeing invalid port %hu for %d\n", port_num, type);
+	}
+}
+
+struct socket *usnic_transport_get_socket(int sock_fd)
+{
+	struct socket *sock;
+	int err;
+	char buf[25];
+
+	/* sockfd_lookup will internally do a fget */
+	sock = sockfd_lookup(sock_fd, &err);
+	if (!sock) {
+		usnic_err("Unable to lookup socket for fd %d with err %d\n",
+				sock_fd, err);
+		return ERR_PTR(-ENOENT);
+	}
+
+	usnic_transport_sock_to_str(buf, sizeof(buf), sock);
+	usnic_dbg("Get sock %s\n", buf);
+
+	return sock;
+}
+
+void usnic_transport_put_socket(struct socket *sock)
+{
+	char buf[100];
+
+	usnic_transport_sock_to_str(buf, sizeof(buf), sock);
+	usnic_dbg("Put sock %s\n", buf);
+	sockfd_put(sock);
+}
+
+int usnic_transport_sock_get_addr(struct socket *sock, int *proto,
+					uint32_t *addr, uint16_t *port)
+{
+	int len;
+	int err;
+	struct sockaddr_in sock_addr;
+
+	err = sock->ops->getname(sock,
+				(struct sockaddr *)&sock_addr,
+				&len, 0);
+	if (err)
+		return err;
+
+	if (sock_addr.sin_family != AF_INET)
+		return -EINVAL;
+
+	if (proto)
+		*proto = sock->sk->sk_protocol;
+	if (port)
+		*port = ntohs(((struct sockaddr_in *)&sock_addr)->sin_port);
+	if (addr)
+		*addr = ntohl(((struct sockaddr_in *)
+					&sock_addr)->sin_addr.s_addr);
+
+	return 0;
+}
+
+int usnic_transport_init(void)
+{
+	roce_bitmap = kzalloc(ROCE_BITMAP_SZ, GFP_KERNEL);
+	if (!roce_bitmap) {
+		usnic_err("Failed to allocate bit map");
+		return -ENOMEM;
+	}
+
+	/* Do not ever allocate bit 0, hence set it here */
+	bitmap_set(roce_bitmap, 0, 1);
+	return 0;
+}
+
+void usnic_transport_fini(void)
+{
+	kfree(roce_bitmap);
+}

drivers/infiniband/hw/usnic/usnic_transport.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_TRANSPORT_H_
+#define USNIC_TRANSPORT_H_
+
+#include "usnic_abi.h"
+
+const char *usnic_transport_to_str(enum usnic_transport_type trans_type);
+/*
+ * Returns number of bytes written, excluding null terminator. If
+ * nothing was written, the function returns 0.
+ */
+int usnic_transport_sock_to_str(char *buf, int buf_sz,
+					struct socket *sock);
+/*
+ * Reserve a port. If "port_num" is set, then the function will try
+ * to reserve that particular port.
+ */
+u16 usnic_transport_rsrv_port(enum usnic_transport_type type, u16 port_num);
+void usnic_transport_unrsrv_port(enum usnic_transport_type type, u16 port_num);
+/*
+ * Do a fget on the socket refered to by sock_fd and returns the socket.
+ * Socket will not be destroyed before usnic_transport_put_socket has
+ * been called.
+ */
+struct socket *usnic_transport_get_socket(int sock_fd);
+void usnic_transport_put_socket(struct socket *sock);
+/*
+ * Call usnic_transport_get_socket before calling *_sock_get_addr
+ */
+int usnic_transport_sock_get_addr(struct socket *sock, int *proto,
+					uint32_t *addr, uint16_t *port);
+int usnic_transport_init(void);
+void usnic_transport_fini(void);
+#endif /* !USNIC_TRANSPORT_H */

drivers/infiniband/hw/usnic/usnic_uiom.h

+/*
+ * Copyright (c) 2013, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef USNIC_UIOM_H_
+#define USNIC_UIOM_H_
+
+#include <linux/list.h>
+#include <linux/scatterlist.h>
+
+#include "usnic_uiom_interval_tree.h"
+
+#define USNIC_UIOM_READ			(1)
+#define USNIC_UIOM_WRITE		(2)
+
+#define USNIC_UIOM_MAX_PD_CNT		(1000)
+#define USNIC_UIOM_MAX_MR_CNT		(1000000)
+#define USNIC_UIOM_MAX_MR_SIZE		(~0UL)
+#define USNIC_UIOM_PAGE_SIZE		(PAGE_SIZE)
+
+struct usnic_uiom_dev {
+	struct device			*dev;
+	struct list_head		link;
+};
+
+struct usnic_uiom_pd {
+	struct iommu_domain		*domain;
+	spinlock_t			lock;
+	struct rb_root			rb_root;
+	struct list_head		devs;
+	int				dev_cnt;
+};
+
+struct usnic_uiom_reg {
+	struct usnic_uiom_pd		*pd;
+	unsigned long			va;
+	size_t				length;
+	int				offset;
+	int				page_size;
+	int				writable;
+	struct list_head		chunk_list;
+	struct work_struct		work;
+	struct mm_struct		*mm;
+	unsigned long			diff;
+};
+
+struct usnic_uiom_chunk {
+	struct list_head		list;
+	int				nents;
+	struct scatterlist		page_list[0];
+};
+
+struct usnic_uiom_pd *usnic_uiom_alloc_pd(void);
+void usnic_uiom_dealloc_pd(struct usnic_uiom_pd *pd);
+int usnic_uiom_attach_dev_to_pd(struct usnic_uiom_pd *pd, struct device *dev);
+void usnic_uiom_detach_dev_from_pd(struct usnic_uiom_pd *pd,
+					struct device *dev);
+struct device **usnic_uiom_get_dev_list(struct usnic_uiom_pd *pd);
+void usnic_uiom_free_dev_list(struct device **devs);
+struct usnic_uiom_reg *usnic_uiom_reg_get(struct usnic_uiom_pd *pd,
+						unsigned long addr, size_t size,
+						int access, int dmasync);
+void usnic_uiom_reg_release(struct usnic_uiom_reg *uiomr, int closing);
+int usnic_uiom_init(char *drv_name);
+void usnic_uiom_fini(void);
+#endif /* USNIC_UIOM_H_ */

drivers/infiniband/hw/usnic/usnic_uiom_interval_tree.c

+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/list_sort.h>
+
+#include <linux/interval_tree_generic.h>
+#include "usnic_uiom_interval_tree.h"
+
+#define START(node) ((node)->start)
+#define LAST(node) ((node)->last)
+
+#define MAKE_NODE(node, start, end, ref_cnt, flags, err, err_out)	\
+		do {							\
+			node = usnic_uiom_interval_node_alloc(start,	\
+					end, ref_cnt, flags);		\
+				if (!node) {				\
+					err = -ENOMEM;			\
+					goto err_out;			\
+				}					\
+		} while (0)
+
+#define MARK_FOR_ADD(node, list) (list_add_tail(&node->link, list))
+
+#define MAKE_NODE_AND_APPEND(node, start, end, ref_cnt, flags, err,	\
+				err_out, list)				\
+				do {					\
+					MAKE_NODE(node, start, end,	\
+						ref_cnt, flags, err,	\
+						err_out);		\
+					MARK_FOR_ADD(node, list);	\
+				} while (0)
+
+#define FLAGS_EQUAL(flags1, flags2, mask)				\
+			(((flags1) & (mask)) == ((flags2) & (mask)))
+
+static struct usnic_uiom_interval_node*
+usnic_uiom_interval_node_alloc(long int start, long int last, int ref_cnt,
+				int flags)
+{
+	struct usnic_uiom_interval_node *interval = kzalloc(sizeof(*interval),
+								GFP_ATOMIC);
+	if (!interval)
+		return NULL;
+
+	interval->start = start;
+	interval->last = last;
+	interval->flags = flags;
+	interval->ref_cnt = ref_cnt;
+
+	return interval;
+}
+
+static int interval_cmp(void *priv, struct list_head *a, struct list_head *b)
+{
+	struct usnic_uiom_interval_node *node_a, *node_b;
+
+	node_a = list_entry(a, struct usnic_uiom_interval_node, link);
+	node_b = list_entry(b, struct usnic_uiom_interval_node, link);
+
+	/* long to int */
+	if (node_a->start < node_b->start)
+		return -1;
+	else if (node_a->start > node_b->start)
+		return 1;
+
+	return 0;
+}
+
+static void
+find_intervals_intersection_sorted(struct rb_root *root, unsigned long start,
+					unsigned long last,
+					struct list_head *list)
+{
+	struct usnic_uiom_interval_node *node;
+
+	INIT_LIST_HEAD(list);
+
+	for (node = usnic_uiom_interval_tree_iter_first(root, start, last);
+		node;
+		node = usnic_uiom_interval_tree_iter_next(node, start, last))
+		list_add_tail(&node->link, list);
+
+	list_sort(NULL, list, interval_cmp);
+}
+
+int usnic_uiom_get_intervals_diff(unsigned long start, unsigned long last,
+					int flags, int flag_mask,
+					struct rb_root *root,
+					struct list_head *diff_set)
+{
+	struct usnic_uiom_interval_node *interval, *tmp;
+	int err = 0;
+	long int pivot = start;
+	LIST_HEAD(intersection_set);
+
+	INIT_LIST_HEAD(diff_set);
+
+	find_intervals_intersection_sorted(root, start, last,
+						&intersection_set);
+
+	list_for_each_entry(interval, &intersection_set, link) {
+		if (pivot < interval->start) {
+			MAKE_NODE_AND_APPEND(tmp, pivot, interval->start - 1,
+						1, flags, err, err_out,
+						diff_set);
+			pivot = interval->start;
+		}
+
+		/*
+		 * Invariant: Set [start, pivot] is either in diff_set or root,
+		 * but not in both.
+		 */
+
+		if (pivot > interval->last) {
+			continue;
+		} else if (pivot <= interval->last &&
+				FLAGS_EQUAL(interval->flags, flags,
+				flag_mask)) {
+			pivot = interval->last + 1;
+		}
+	}
+
+	if (pivot <= last)
+		MAKE_NODE_AND_APPEND(tmp, pivot, last, 1, flags, err, err_out,
+					diff_set);
+
+	return 0;
+
+err_out:
+	list_for_each_entry_safe(interval, tmp, diff_set, link) {
+		list_del(&interval->link);
+		kfree(interval);
+	}
+
+	return err;
+}
+
+void usnic_uiom_put_interval_set(struct list_head *intervals)
+{
+	struct usnic_uiom_interval_node *interval, *tmp;
+	list_for_each_entry_safe(interval, tmp, intervals, link)
+		kfree(interval);
+}
+
+int usnic_uiom_insert_interval(struct rb_root *root, unsigned long start,
+				unsigned long last, int flags)
+{
+	struct usnic_uiom_interval_node *interval, *tmp;
+	unsigned long istart, ilast;
+	int iref_cnt, iflags;
+	unsigned long lpivot = start;
+	int err = 0;
+	LIST_HEAD(to_add);
+	LIST_HEAD(intersection_set);
+
+	find_intervals_intersection_sorted(root, start, last,
+						&intersection_set);
+
+	list_for_each_entry(interval, &intersection_set, link) {
+		/*
+		 * Invariant - lpivot is the left edge of next interval to be
+		 * inserted
+		 */
+		istart = interval->start;
+		ilast = interval->last;
+		iref_cnt = interval->ref_cnt;
+		iflags = interval->flags;
+
+		if (istart < lpivot) {
+			MAKE_NODE_AND_APPEND(tmp, istart, lpivot - 1, iref_cnt,
+						iflags, err, err_out, &to_add);
+		} else if (istart > lpivot) {
+			MAKE_NODE_AND_APPEND(tmp, lpivot, istart - 1, 1, flags,
+						err, err_out, &to_add);
+			lpivot = istart;
+		} else {
+			lpivot = istart;
+		}
+
+		if (ilast > last) {
+			MAKE_NODE_AND_APPEND(tmp, lpivot, last, iref_cnt + 1,
+						iflags | flags, err, err_out,
+						&to_add);
+			MAKE_NODE_AND_APPEND(tmp, last + 1, ilast, iref_cnt,
+						iflags, err, err_out, &to_add);
+		} else {
+			MAKE_NODE_AND_APPEND(tmp, lpivot, ilast, iref_cnt + 1,
+						iflags | flags, err, err_out,
+						&to_add);
+		}
+
+		lpivot = ilast + 1;
+	}
+
+	if (lpivot <= last)
+		MAKE_NODE_AND_APPEND(tmp, lpivot, last, 1, flags, err, err_out,
+					&to_add);
+
+	list_for_each_entry_safe(interval, tmp, &intersection_set, link) {
+		usnic_uiom_interval_tree_remove(interval, root);
+		kfree(interval);
+	}
+
+	list_for_each_entry(interval, &to_add, link)
+		usnic_uiom_interval_tree_insert(interval, root);
+
+	return 0;
+
+err_out:
+	list_for_each_entry_safe(interval, tmp, &to_add, link)
+		kfree(interval);
+
+	return err;
+}
+
+void usnic_uiom_remove_interval(struct rb_root *root, unsigned long start,
+				unsigned long last, struct list_head *removed)
+{
+	struct usnic_uiom_interval_node *interval;
+
+	for (interval = usnic_uiom_interval_tree_iter_first(root, start, last);
+			interval;
+			interval = usnic_uiom_interval_tree_iter_next(interval,
+									start,
+									last)) {
+		if (--interval->ref_cnt == 0)
+			list_add_tail(&interval->link, removed);
+	}
+
+	list_for_each_entry(interval, removed, link)
+		usnic_uiom_interval_tree_remove(interval, root);
+}
+
+INTERVAL_TREE_DEFINE(struct usnic_uiom_interval_node, rb,
+			unsigned long, __subtree_last,
+			START, LAST, , usnic_uiom_interval_tree)

drivers/infiniband/ulp/ipoib/ipoib_main.c

@@ -104,6 +104,8 @@ int ipoib_open(struct net_device *dev)
 
 	ipoib_dbg(priv, "bringing up interface\n");
 
+	netif_carrier_off(dev);
+
 	set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
 
 	if (ipoib_pkey_dev_delay_open(dev))
@@ -1366,8 +1368,6 @@ void ipoib_setup(struct net_device *dev)
 
 	memcpy(dev->broadcast, ipv4_bcast_addr, INFINIBAND_ALEN);
 
-	netif_carrier_off(dev);
-
 	priv->dev = dev;
 
 	spin_lock_init(&priv->lock);

drivers/infiniband/ulp/ipoib/ipoib_verbs.c

@@ -192,6 +192,9 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca)
 	if (priv->hca_caps & IB_DEVICE_BLOCK_MULTICAST_LOOPBACK)
 		init_attr.create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK;
 
+	if (priv->hca_caps & IB_DEVICE_MANAGED_FLOW_STEERING)
+		init_attr.create_flags |= IB_QP_CREATE_NETIF_QP;
+
 	if (dev->features & NETIF_F_SG)
 		init_attr.cap.max_send_sge = MAX_SKB_FRAGS + 1;