From 752a50cab600c6d46c5a1921c6a6d2fb116c8a4b Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 19 Jun 2012 11:21:33 +0300
Subject: mlx4_core: Pass an invalid PCI id number to VFs

Currently, VFs have 0 in their dev->caps.function field.  This is a
valid pci id (usually of the PF).  Instead, pass an invalid PCI id to
the VF via QUERY_FW, so that if the value gets accessed in the VF
driver, we'll catch the problem.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 9c83bb8151ea..4281ce09add8 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -881,11 +881,12 @@ int mlx4_QUERY_FW(struct mlx4_dev *dev)
 		((fw_ver & 0xffff0000ull) >> 16) |
 		((fw_ver & 0x0000ffffull) << 16);
 
+	MLX4_GET(lg, outbox, QUERY_FW_PPF_ID);
+	dev->caps.function = lg;
+
 	if (mlx4_is_slave(dev))
 		goto out;
 
-	MLX4_GET(lg, outbox, QUERY_FW_PPF_ID);
-	dev->caps.function = lg;
 
 	MLX4_GET(cmd_if_rev, outbox, QUERY_FW_CMD_IF_REV_OFFSET);
 	if (cmd_if_rev < MLX4_COMMAND_INTERFACE_MIN_REV ||
@@ -966,9 +967,12 @@ int mlx4_QUERY_FW_wrapper(struct mlx4_dev *dev, int slave,
 	if (err)
 		return err;
 
-	/* for slaves, zero out everything except FW version */
+	/* for slaves, set pci PPF ID to invalid and zero out everything
+	 * else except FW version */
 	outbuf[0] = outbuf[1] = 0;
 	memset(&outbuf[8], 0, QUERY_FW_OUT_SIZE - 8);
+	outbuf[QUERY_FW_PPF_ID] = MLX4_INVALID_SLAVE_ID;
+
 	return 0;
 }
 
-- 
cgit v1.2.3


From 00f5ce99dc6ee46c3113393cc8fa12173f9bbcd7 Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 19 Jun 2012 11:21:40 +0300
Subject: mlx4: Use port management change event instead of smp_snoop

The port management change event can replace smp_snoop.  If the
capability bit for this event is set in dev-caps, the event is used
(by the driver setting the PORT_MNG_CHG_EVENT bit in the async event
mask in the MAP_EQ fw command).  In this case, when the driver passes
incoming SMP PORT_INFO SET mads to the FW, the FW generates port
management change events to signal any changes to the driver.

If the FW generates these events, smp_snoop shouldn't be invoked in
ib_process_mad(), or duplicate events will occur (once from the
FW-generated event, and once from smp_snoop).

In the case where the FW does not generate port management change
events smp_snoop needs to be invoked to create these events.  The flow
in smp_snoop has been modified to make use of the same procedures as
in the fw-generated-event event case to generate the port management
events (LID change, Client-rereg, Pkey change, and/or GID change).

Port management change event handling required changing the
mlx4_ib_event and mlx4_dispatch_event prototypes; the "param" argument
(last argument) had to be changed to unsigned long in order to
accomodate passing the EQE pointer.

We also needed to move the definition of struct mlx4_eqe from
net/mlx4.h to file device.h -- to make it available to the IB driver,
to handle port management change events.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/infiniband/hw/mlx4/mad.c             | 118 ++++++++++++++++++++-------
 drivers/infiniband/hw/mlx4/main.c            |  29 +++++--
 drivers/infiniband/hw/mlx4/mlx4_ib.h         |   9 ++
 drivers/net/ethernet/mellanox/mlx4/en_main.c |   5 +-
 drivers/net/ethernet/mellanox/mlx4/eq.c      |  22 ++++-
 drivers/net/ethernet/mellanox/mlx4/fw.c      |   1 +
 drivers/net/ethernet/mellanox/mlx4/intf.c    |   5 +-
 drivers/net/ethernet/mellanox/mlx4/mlx4.h    |  63 +-------------
 include/linux/mlx4/device.h                  |  99 +++++++++++++++++++++-
 include/linux/mlx4/driver.h                  |   3 +-
 10 files changed, 249 insertions(+), 105 deletions(-)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 84786a9fb64f..58c45fb5bd31 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -147,47 +147,49 @@ static void update_sm_ah(struct mlx4_ib_dev *dev, u8 port_num, u16 lid, u8 sl)
 }
 
 /*
- * Snoop SM MADs for port info and P_Key table sets, so we can
- * synthesize LID change and P_Key change events.
+ * Snoop SM MADs for port info, GUID info, and  P_Key table sets, so we can
+ * synthesize LID change, Client-Rereg, GID change, and P_Key change events.
  */
 static void smp_snoop(struct ib_device *ibdev, u8 port_num, struct ib_mad *mad,
-				u16 prev_lid)
+		      u16 prev_lid)
 {
-	struct ib_event event;
+	struct ib_port_info *pinfo;
+	u16 lid;
 
+	struct mlx4_ib_dev *dev = to_mdev(ibdev);
 	if ((mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED ||
 	     mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) &&
-	    mad->mad_hdr.method == IB_MGMT_METHOD_SET) {
-		if (mad->mad_hdr.attr_id == IB_SMP_ATTR_PORT_INFO) {
-			struct ib_port_info *pinfo =
-				(struct ib_port_info *) ((struct ib_smp *) mad)->data;
-			u16 lid = be16_to_cpu(pinfo->lid);
+	    mad->mad_hdr.method == IB_MGMT_METHOD_SET)
+		switch (mad->mad_hdr.attr_id) {
+		case IB_SMP_ATTR_PORT_INFO:
+			pinfo = (struct ib_port_info *) ((struct ib_smp *) mad)->data;
+			lid = be16_to_cpu(pinfo->lid);
 
-			update_sm_ah(to_mdev(ibdev), port_num,
+			update_sm_ah(dev, port_num,
 				     be16_to_cpu(pinfo->sm_lid),
 				     pinfo->neighbormtu_mastersmsl & 0xf);
 
-			event.device	       = ibdev;
-			event.element.port_num = port_num;
+			if (pinfo->clientrereg_resv_subnetto & 0x80)
+				mlx4_ib_dispatch_event(dev, port_num,
+						       IB_EVENT_CLIENT_REREGISTER);
 
-			if (pinfo->clientrereg_resv_subnetto & 0x80) {
-				event.event    = IB_EVENT_CLIENT_REREGISTER;
-				ib_dispatch_event(&event);
-			}
+			if (prev_lid != lid)
+				mlx4_ib_dispatch_event(dev, port_num,
+						       IB_EVENT_LID_CHANGE);
+			break;
 
-			if (prev_lid != lid) {
-				event.event    = IB_EVENT_LID_CHANGE;
-				ib_dispatch_event(&event);
-			}
-		}
+		case IB_SMP_ATTR_PKEY_TABLE:
+			mlx4_ib_dispatch_event(dev, port_num,
+					       IB_EVENT_PKEY_CHANGE);
+			break;
 
-		if (mad->mad_hdr.attr_id == IB_SMP_ATTR_PKEY_TABLE) {
-			event.device	       = ibdev;
-			event.event	       = IB_EVENT_PKEY_CHANGE;
-			event.element.port_num = port_num;
-			ib_dispatch_event(&event);
+		case IB_SMP_ATTR_GUID_INFO:
+			mlx4_ib_dispatch_event(dev, port_num,
+					       IB_EVENT_GID_CHANGE);
+			break;
+		default:
+			break;
 		}
-	}
 }
 
 static void node_desc_override(struct ib_device *dev,
@@ -305,7 +307,8 @@ static int ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 		return IB_MAD_RESULT_FAILURE;
 
 	if (!out_mad->mad_hdr.status) {
-		smp_snoop(ibdev, port_num, in_mad, prev_lid);
+		if (!(to_mdev(ibdev)->dev->caps.flags & MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV))
+			smp_snoop(ibdev, port_num, in_mad, prev_lid);
 		node_desc_override(ibdev, out_mad);
 	}
 
@@ -446,3 +449,62 @@ void mlx4_ib_mad_cleanup(struct mlx4_ib_dev *dev)
 			ib_destroy_ah(dev->sm_ah[p]);
 	}
 }
+
+void handle_port_mgmt_change_event(struct work_struct *work)
+{
+	struct ib_event_work *ew = container_of(work, struct ib_event_work, work);
+	struct mlx4_ib_dev *dev = ew->ib_dev;
+	struct mlx4_eqe *eqe = &(ew->ib_eqe);
+	u8 port = eqe->event.port_mgmt_change.port;
+	u32 changed_attr;
+
+	switch (eqe->subtype) {
+	case MLX4_DEV_PMC_SUBTYPE_PORT_INFO:
+		changed_attr = be32_to_cpu(eqe->event.port_mgmt_change.params.port_info.changed_attr);
+
+		/* Update the SM ah - This should be done before handling
+		   the other changed attributes so that MADs can be sent to the SM */
+		if (changed_attr & MSTR_SM_CHANGE_MASK) {
+			u16 lid = be16_to_cpu(eqe->event.port_mgmt_change.params.port_info.mstr_sm_lid);
+			u8 sl = eqe->event.port_mgmt_change.params.port_info.mstr_sm_sl & 0xf;
+			update_sm_ah(dev, port, lid, sl);
+		}
+
+		/* Check if it is a lid change event */
+		if (changed_attr & MLX4_EQ_PORT_INFO_LID_CHANGE_MASK)
+			mlx4_ib_dispatch_event(dev, port, IB_EVENT_LID_CHANGE);
+
+		/* Generate GUID changed event */
+		if (changed_attr & MLX4_EQ_PORT_INFO_GID_PFX_CHANGE_MASK)
+			mlx4_ib_dispatch_event(dev, port, IB_EVENT_GID_CHANGE);
+
+		if (changed_attr & MLX4_EQ_PORT_INFO_CLIENT_REREG_MASK)
+			mlx4_ib_dispatch_event(dev, port,
+					       IB_EVENT_CLIENT_REREGISTER);
+		break;
+
+	case MLX4_DEV_PMC_SUBTYPE_PKEY_TABLE:
+		mlx4_ib_dispatch_event(dev, port, IB_EVENT_PKEY_CHANGE);
+		break;
+	case MLX4_DEV_PMC_SUBTYPE_GUID_INFO:
+		mlx4_ib_dispatch_event(dev, port, IB_EVENT_GID_CHANGE);
+		break;
+	default:
+		pr_warn("Unsupported subtype 0x%x for "
+			"Port Management Change event\n", eqe->subtype);
+	}
+
+	kfree(ew);
+}
+
+void mlx4_ib_dispatch_event(struct mlx4_ib_dev *dev, u8 port_num,
+			    enum ib_event_type type)
+{
+	struct ib_event event;
+
+	event.device		= &dev->ib_dev;
+	event.element.port_num	= port_num;
+	event.event		= type;
+
+	ib_dispatch_event(&event);
+}
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 5266b49c46ee..4f230c26622d 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -898,7 +898,6 @@ static void update_gids_task(struct work_struct *work)
 	union ib_gid *gids;
 	int err;
 	struct mlx4_dev	*dev = gw->dev->dev;
-	struct ib_event event;
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox)) {
@@ -916,10 +915,7 @@ static void update_gids_task(struct work_struct *work)
 		pr_warn("set port command failed\n");
 	else {
 		memcpy(gw->dev->iboe.gid_table[gw->port - 1], gw->gids, sizeof gw->gids);
-		event.device = &gw->dev->ib_dev;
-		event.element.port_num = gw->port;
-		event.event    = IB_EVENT_GID_CHANGE;
-		ib_dispatch_event(&event);
+		mlx4_ib_dispatch_event(gw->dev, gw->port, IB_EVENT_GID_CHANGE);
 	}
 
 	mlx4_free_cmd_mailbox(dev, mailbox);
@@ -1383,10 +1379,18 @@ static void mlx4_ib_remove(struct mlx4_dev *dev, void *ibdev_ptr)
 }
 
 static void mlx4_ib_event(struct mlx4_dev *dev, void *ibdev_ptr,
-			  enum mlx4_dev_event event, int port)
+			  enum mlx4_dev_event event, unsigned long param)
 {
 	struct ib_event ibev;
 	struct mlx4_ib_dev *ibdev = to_mdev((struct ib_device *) ibdev_ptr);
+	struct mlx4_eqe *eqe = NULL;
+	struct ib_event_work *ew;
+	int port = 0;
+
+	if (event == MLX4_DEV_EVENT_PORT_MGMT_CHANGE)
+		eqe = (struct mlx4_eqe *)param;
+	else
+		port = (u8)param;
 
 	if (port > ibdev->num_ports)
 		return;
@@ -1405,6 +1409,19 @@ static void mlx4_ib_event(struct mlx4_dev *dev, void *ibdev_ptr,
 		ibev.event = IB_EVENT_DEVICE_FATAL;
 		break;
 
+	case MLX4_DEV_EVENT_PORT_MGMT_CHANGE:
+		ew = kmalloc(sizeof *ew, GFP_ATOMIC);
+		if (!ew) {
+			pr_err("failed to allocate memory for events work\n");
+			break;
+		}
+
+		INIT_WORK(&ew->work, handle_port_mgmt_change_event);
+		memcpy(&ew->ib_eqe, eqe, sizeof *eqe);
+		ew->ib_dev = ibdev;
+		handle_port_mgmt_change_event(&ew->work);
+		return;
+
 	default:
 		return;
 	}
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index 5f298afaa81f..23bfbf9ee0e0 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -224,6 +224,12 @@ struct mlx4_ib_dev {
 	int			eq_added;
 };
 
+struct ib_event_work {
+	struct work_struct	work;
+	struct mlx4_ib_dev	*ib_dev;
+	struct mlx4_eqe		ib_eqe;
+};
+
 static inline struct mlx4_ib_dev *to_mdev(struct ib_device *ibdev)
 {
 	return container_of(ibdev, struct mlx4_ib_dev, ib_dev);
@@ -381,4 +387,7 @@ static inline int mlx4_ib_ah_grh_present(struct mlx4_ib_ah *ah)
 int mlx4_ib_add_mc(struct mlx4_ib_dev *mdev, struct mlx4_ib_qp *mqp,
 		   union ib_gid *gid);
 
+void mlx4_ib_dispatch_event(struct mlx4_ib_dev *dev, u8 port_num,
+			    enum ib_event_type type);
+
 #endif /* MLX4_IB_H */
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c
index 69ba57270481..a52922ed85c1 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c
@@ -131,7 +131,7 @@ static void *mlx4_en_get_netdev(struct mlx4_dev *dev, void *ctx, u8 port)
 }
 
 static void mlx4_en_event(struct mlx4_dev *dev, void *endev_ptr,
-			  enum mlx4_dev_event event, int port)
+			  enum mlx4_dev_event event, unsigned long port)
 {
 	struct mlx4_en_dev *mdev = (struct mlx4_en_dev *) endev_ptr;
 	struct mlx4_en_priv *priv;
@@ -156,7 +156,8 @@ static void mlx4_en_event(struct mlx4_dev *dev, void *endev_ptr,
 		if (port < 1 || port > dev->caps.num_ports ||
 		    !mdev->pndev[port])
 			return;
-		mlx4_warn(mdev, "Unhandled event %d for port %d\n", event, port);
+		mlx4_warn(mdev, "Unhandled event %d for port %d\n", event,
+			  (int) port);
 	}
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index bce98d9c0039..9b15d0219950 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -82,6 +82,15 @@ enum {
 			       (1ull << MLX4_EVENT_TYPE_FLR_EVENT)	    | \
 			       (1ull << MLX4_EVENT_TYPE_FATAL_WARNING))
 
+static u64 get_async_ev_mask(struct mlx4_dev *dev)
+{
+	u64 async_ev_mask = MLX4_ASYNC_EVENT_MASK;
+	if (dev->caps.flags & MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV)
+		async_ev_mask |= (1ull << MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT);
+
+	return async_ev_mask;
+}
+
 static void eq_set_ci(struct mlx4_eq *eq, int req_not)
 {
 	__raw_writel((__force u32) cpu_to_be32((eq->cons_index & 0xffffff) |
@@ -473,6 +482,11 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 
 			break;
 
+		case MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT:
+			mlx4_dispatch_event(dev, MLX4_DEV_EVENT_PORT_MGMT_CHANGE,
+					    (unsigned long) eqe);
+			break;
+
 		case MLX4_EVENT_TYPE_EEC_CATAS_ERROR:
 		case MLX4_EVENT_TYPE_ECC_DETECT:
 		default:
@@ -956,7 +970,7 @@ int mlx4_init_eq_table(struct mlx4_dev *dev)
 		priv->eq_table.have_irq = 1;
 	}
 
-	err = mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+	err = mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0,
 			  priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
 	if (err)
 		mlx4_warn(dev, "MAP_EQ for async EQ %d failed (%d)\n",
@@ -996,7 +1010,7 @@ void mlx4_cleanup_eq_table(struct mlx4_dev *dev)
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	int i;
 
-	mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 1,
+	mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 1,
 		    priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
 
 	mlx4_free_irqs(dev);
@@ -1040,7 +1054,7 @@ int mlx4_test_interrupts(struct mlx4_dev *dev)
 		mlx4_cmd_use_polling(dev);
 
 		/* Map the new eq to handle all asyncronous events */
-		err = mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+		err = mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0,
 				  priv->eq_table.eq[i].eqn);
 		if (err) {
 			mlx4_warn(dev, "Failed mapping eq for interrupt test\n");
@@ -1054,7 +1068,7 @@ int mlx4_test_interrupts(struct mlx4_dev *dev)
 	}
 
 	/* Return to default */
-	mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+	mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0,
 		    priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 4281ce09add8..ee9d6b0b4d20 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -109,6 +109,7 @@ static void dump_dev_cap_flags(struct mlx4_dev *dev, u64 flags)
 		[41] = "Unicast VEP steering support",
 		[42] = "Multicast VEP steering support",
 		[48] = "Counters support",
+		[59] = "Port management change event support",
 	};
 	int i;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c
index b4e9f6f5cc04..116895ac8b35 100644
--- a/drivers/net/ethernet/mellanox/mlx4/intf.c
+++ b/drivers/net/ethernet/mellanox/mlx4/intf.c
@@ -115,7 +115,8 @@ void mlx4_unregister_interface(struct mlx4_interface *intf)
 }
 EXPORT_SYMBOL_GPL(mlx4_unregister_interface);
 
-void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int port)
+void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type,
+			 unsigned long param)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_device_context *dev_ctx;
@@ -125,7 +126,7 @@ void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int por
 
 	list_for_each_entry(dev_ctx, &priv->ctx_list, list)
 		if (dev_ctx->intf->event)
-			dev_ctx->intf->event(dev, dev_ctx->context, type, port);
+			dev_ctx->intf->event(dev, dev_ctx->context, type, param);
 
 	spin_unlock_irqrestore(&priv->ctx_lock, flags);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index e5d20220762c..4d11d12b9db4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -338,66 +338,6 @@ struct mlx4_srq_context {
 	__be64			db_rec_addr;
 };
 
-struct mlx4_eqe {
-	u8			reserved1;
-	u8			type;
-	u8			reserved2;
-	u8			subtype;
-	union {
-		u32		raw[6];
-		struct {
-			__be32	cqn;
-		} __packed comp;
-		struct {
-			u16	reserved1;
-			__be16	token;
-			u32	reserved2;
-			u8	reserved3[3];
-			u8	status;
-			__be64	out_param;
-		} __packed cmd;
-		struct {
-			__be32	qpn;
-		} __packed qp;
-		struct {
-			__be32	srqn;
-		} __packed srq;
-		struct {
-			__be32	cqn;
-			u32	reserved1;
-			u8	reserved2[3];
-			u8	syndrome;
-		} __packed cq_err;
-		struct {
-			u32	reserved1[2];
-			__be32	port;
-		} __packed port_change;
-		struct {
-			#define COMM_CHANNEL_BIT_ARRAY_SIZE	4
-			u32 reserved;
-			u32 bit_vec[COMM_CHANNEL_BIT_ARRAY_SIZE];
-		} __packed comm_channel_arm;
-		struct {
-			u8	port;
-			u8	reserved[3];
-			__be64	mac;
-		} __packed mac_update;
-		struct {
-			u8	port;
-		} __packed sw_event;
-		struct {
-			__be32	slave_id;
-		} __packed flr_event;
-		struct {
-			__be16  current_temperature;
-			__be16  warning_threshold;
-		} __packed warming;
-	}			event;
-	u8			slave_id;
-	u8			reserved3[2];
-	u8			owner;
-} __packed;
-
 struct mlx4_eq {
 	struct mlx4_dev	       *dev;
 	void __iomem	       *doorbell;
@@ -887,7 +827,8 @@ void mlx4_catas_init(void);
 int mlx4_restart_one(struct pci_dev *pdev);
 int mlx4_register_device(struct mlx4_dev *dev);
 void mlx4_unregister_device(struct mlx4_dev *dev);
-void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int port);
+void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type,
+			 unsigned long param);
 
 struct mlx4_dev_cap;
 struct mlx4_init_hca_param;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 8eadf0f14cc5..560b2201519f 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -96,7 +96,8 @@ enum {
 	MLX4_DEV_CAP_FLAG_VEP_UC_STEER	= 1LL << 41,
 	MLX4_DEV_CAP_FLAG_VEP_MC_STEER	= 1LL << 42,
 	MLX4_DEV_CAP_FLAG_COUNTERS	= 1LL << 48,
-	MLX4_DEV_CAP_FLAG_SENSE_SUPPORT	= 1LL << 55
+	MLX4_DEV_CAP_FLAG_SENSE_SUPPORT	= 1LL << 55,
+	MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV = 1LL << 59,
 };
 
 enum {
@@ -138,6 +139,7 @@ enum mlx4_event {
 	MLX4_EVENT_TYPE_COMM_CHANNEL	   = 0x18,
 	MLX4_EVENT_TYPE_FATAL_WARNING	   = 0x1b,
 	MLX4_EVENT_TYPE_FLR_EVENT	   = 0x1c,
+	MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT = 0x1d,
 	MLX4_EVENT_TYPE_NONE		   = 0xff,
 };
 
@@ -235,6 +237,24 @@ enum {
 	MLX4_MAX_FAST_REG_PAGES = 511,
 };
 
+enum {
+	MLX4_DEV_PMC_SUBTYPE_GUID_INFO	 = 0x14,
+	MLX4_DEV_PMC_SUBTYPE_PORT_INFO	 = 0x15,
+	MLX4_DEV_PMC_SUBTYPE_PKEY_TABLE	 = 0x16,
+};
+
+/* Port mgmt change event handling */
+enum {
+	MLX4_EQ_PORT_INFO_MSTR_SM_LID_CHANGE_MASK	= 1 << 0,
+	MLX4_EQ_PORT_INFO_GID_PFX_CHANGE_MASK		= 1 << 1,
+	MLX4_EQ_PORT_INFO_LID_CHANGE_MASK		= 1 << 2,
+	MLX4_EQ_PORT_INFO_CLIENT_REREG_MASK		= 1 << 3,
+	MLX4_EQ_PORT_INFO_MSTR_SM_SL_CHANGE_MASK	= 1 << 4,
+};
+
+#define MSTR_SM_CHANGE_MASK (MLX4_EQ_PORT_INFO_MSTR_SM_SL_CHANGE_MASK | \
+			     MLX4_EQ_PORT_INFO_MSTR_SM_LID_CHANGE_MASK)
+
 static inline u64 mlx4_fw_ver(u64 major, u64 minor, u64 subminor)
 {
 	return (major << 32) | (minor << 16) | subminor;
@@ -511,6 +531,81 @@ struct mlx4_dev {
 	int			num_vfs;
 };
 
+struct mlx4_eqe {
+	u8			reserved1;
+	u8			type;
+	u8			reserved2;
+	u8			subtype;
+	union {
+		u32		raw[6];
+		struct {
+			__be32	cqn;
+		} __packed comp;
+		struct {
+			u16	reserved1;
+			__be16	token;
+			u32	reserved2;
+			u8	reserved3[3];
+			u8	status;
+			__be64	out_param;
+		} __packed cmd;
+		struct {
+			__be32	qpn;
+		} __packed qp;
+		struct {
+			__be32	srqn;
+		} __packed srq;
+		struct {
+			__be32	cqn;
+			u32	reserved1;
+			u8	reserved2[3];
+			u8	syndrome;
+		} __packed cq_err;
+		struct {
+			u32	reserved1[2];
+			__be32	port;
+		} __packed port_change;
+		struct {
+			#define COMM_CHANNEL_BIT_ARRAY_SIZE	4
+			u32 reserved;
+			u32 bit_vec[COMM_CHANNEL_BIT_ARRAY_SIZE];
+		} __packed comm_channel_arm;
+		struct {
+			u8	port;
+			u8	reserved[3];
+			__be64	mac;
+		} __packed mac_update;
+		struct {
+			__be32	slave_id;
+		} __packed flr_event;
+		struct {
+			__be16  current_temperature;
+			__be16  warning_threshold;
+		} __packed warming;
+		struct {
+			u8 reserved[3];
+			u8 port;
+			union {
+				struct {
+					__be16 mstr_sm_lid;
+					__be16 port_lid;
+					__be32 changed_attr;
+					u8 reserved[3];
+					u8 mstr_sm_sl;
+					__be64 gid_prefix;
+				} __packed port_info;
+				struct {
+					__be32 block_ptr;
+					__be32 tbl_entries_mask;
+				} __packed tbl_change_info;
+			} params;
+		} __packed port_mgmt_change;
+	}			event;
+	u8			slave_id;
+	u8			reserved3[2];
+	u8			owner;
+} __packed;
+
 struct mlx4_init_port_param {
 	int			set_guid0;
 	int			set_node_guid;
@@ -536,6 +631,8 @@ struct mlx4_init_port_param {
 
 #define MLX4_INVALID_SLAVE_ID	0xFF
 
+void handle_port_mgmt_change_event(struct work_struct *work);
+
 static inline int mlx4_is_master(struct mlx4_dev *dev)
 {
 	return dev->flags & MLX4_FLAG_MASTER;
diff --git a/include/linux/mlx4/driver.h b/include/linux/mlx4/driver.h
index 5f1298b1b5ef..0f509229fb3d 100644
--- a/include/linux/mlx4/driver.h
+++ b/include/linux/mlx4/driver.h
@@ -42,13 +42,14 @@ enum mlx4_dev_event {
 	MLX4_DEV_EVENT_PORT_UP,
 	MLX4_DEV_EVENT_PORT_DOWN,
 	MLX4_DEV_EVENT_PORT_REINIT,
+	MLX4_DEV_EVENT_PORT_MGMT_CHANGE,
 };
 
 struct mlx4_interface {
 	void *			(*add)	 (struct mlx4_dev *dev);
 	void			(*remove)(struct mlx4_dev *dev, void *context);
 	void			(*event) (struct mlx4_dev *dev, void *context,
-					  enum mlx4_dev_event event, int port);
+					  enum mlx4_dev_event event, unsigned long param);
 	void *			(*get_dev)(struct mlx4_dev *dev, void *context, u8 port);
 	struct list_head	list;
 	enum mlx4_protocol	protocol;
-- 
cgit v1.2.3


From 2aca1172c2f5b27fbc37297574f716c1c15f4153 Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 19 Jun 2012 11:21:41 +0300
Subject: net/mlx4_core: Initialize IB port capabilities for all slaves

With IB SR-IOV, each slave has its own separate copy of the port
capabilities flags.  For example, the master can run a subnet manager
(which causes the IsSM bit to be set in the master's port
capabilities) without affecting the port capabilities seen by the
slaves (the IsSM bit will be seen as cleared in the slaves).

Also add a static inline mlx4_master_func_num() to enhance readability
of the code.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx4/main.c | 11 +++++++++++
 include/linux/mlx4/device.h               |  5 +++++
 2 files changed, 16 insertions(+)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index a0313de122de..83afb1541a74 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -1477,6 +1477,17 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 					  "with caps = 0\n", port, err);
 			dev->caps.ib_port_def_cap[port] = ib_port_default_caps;
 
+			/* initialize per-slave default ib port capabilities */
+			if (mlx4_is_master(dev)) {
+				int i;
+				for (i = 0; i < dev->num_slaves; i++) {
+					if (i == mlx4_master_func_num(dev))
+						continue;
+					priv->mfunc.master.slave_state[i].ib_cap_mask[port] =
+							ib_port_default_caps;
+				}
+			}
+
 			if (mlx4_is_mfunc(dev))
 				dev->caps.port_ib_mtu[port] = IB_MTU_2048;
 			else
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 560b2201519f..7fbdc89de495 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -633,6 +633,11 @@ struct mlx4_init_port_param {
 
 void handle_port_mgmt_change_event(struct work_struct *work);
 
+static inline int mlx4_master_func_num(struct mlx4_dev *dev)
+{
+	return dev->caps.function;
+}
+
 static inline int mlx4_is_master(struct mlx4_dev *dev)
 {
 	return dev->flags & MLX4_FLAG_MASTER;
-- 
cgit v1.2.3


From f457ce471c522cadf697b873e2cf46e458e90bef Mon Sep 17 00:00:00 2001
From: Dotan Barak <dotanb@dev.mellanox.co.il>
Date: Wed, 11 Jul 2012 15:39:30 +0000
Subject: mlx4_core: Remove double function declarations

Spotted four duplicate declarations in icm.h, remove them.

Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx4/icm.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.h b/drivers/net/ethernet/mellanox/mlx4/icm.h
index b10c07a1dc1a..19e4efc0b342 100644
--- a/drivers/net/ethernet/mellanox/mlx4/icm.h
+++ b/drivers/net/ethernet/mellanox/mlx4/icm.h
@@ -81,13 +81,7 @@ int mlx4_init_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table,
 			u64 virt, int obj_size,	int nobj, int reserved,
 			int use_lowmem, int use_coherent);
 void mlx4_cleanup_icm_table(struct mlx4_dev *dev, struct mlx4_icm_table *table);
-int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, int obj);
-void mlx4_table_put(struct mlx4_dev *dev, struct mlx4_icm_table *table, int obj);
 void *mlx4_table_find(struct mlx4_icm_table *table, int obj, dma_addr_t *dma_handle);
-int mlx4_table_get_range(struct mlx4_dev *dev, struct mlx4_icm_table *table,
-			 int start, int end);
-void mlx4_table_put_range(struct mlx4_dev *dev, struct mlx4_icm_table *table,
-			  int start, int end);
 
 static inline void mlx4_icm_first(struct mlx4_icm *icm,
 				  struct mlx4_icm_iter *iter)
-- 
cgit v1.2.3


From 240a9207aae24916dba7070aa7047c3732102cb8 Mon Sep 17 00:00:00 2001
From: Dotan Barak <dotanb@dev.mellanox.co.il>
Date: Wed, 11 Jul 2012 15:39:32 +0000
Subject: net/mlx4_core: Free ICM table in case of error

In mlx4_init_icm_table(), free the allocated table if we failed to
allocate memory to its entries.

Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il>
Reviewed-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx4/icm.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.c b/drivers/net/ethernet/mellanox/mlx4/icm.c
index a9ade1c3cad5..88b7b3e75ab1 100644
--- a/drivers/net/ethernet/mellanox/mlx4/icm.c
+++ b/drivers/net/ethernet/mellanox/mlx4/icm.c
@@ -413,6 +413,8 @@ err:
 			mlx4_free_icm(dev, table->icm[i], use_coherent);
 		}
 
+	kfree(table->icm);
+
 	return -ENOMEM;
 }
 
-- 
cgit v1.2.3


From 396f2feb05d7cc5549c611c05abfb4108cd1c6d6 Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 19 Jun 2012 11:21:42 +0300
Subject: mlx4_core: Implement mechanism for reserved Q_Keys

The SR-IOV special QP tunneling mechanism uses proxy special QPs
(instead of the real special QPs) for MADs on guests.  These proxy QPs
send their packets to a "tunnel" QP owned by the master.  The master
then forwards the MAD (after any required paravirtualization) to the
real special QP, which sends out the MAD.

For security reasons (i.e., to prevent guests from sending MADs to
tunnel QPs belonging to other guests), each proxy-tunnel QP pair is
assigned a unique, reserved, Q_Key.  These Q_Keys are available only
for proxy and tunnel QPs -- if the guest tries to use these Q_Keys
with other QPs, it will fail.

This patch introduces a mechanism for reserving a block of 64K Q_Keys
for proxy/tunneling use.

The patch introduces also two new fields into mlx4_dev: base_sqpn and
base_tunnel_sqpn.

In SR-IOV mode, the QP numbers for the "real," proxy, and tunnel sqps
are added to the reserved QPN area (so that they will not change).
There are 8 special QPs per port in the HCA, and each of them is
assigned both a proxy and a tunnel QP, for each VF and for the PF as
well in SR-IOV mode.

The QPNs for these QPs are arranged as follows:
 1. The real SQP numbers (8)
 2. The proxy SQPs (8 * (max number of VFs + max number of PFs)
 3. The tunnel SQPs (8 * (max number of VFs + max number of PFs)

To support these QPs, two new fields are added to struct mlx4_dev:

  base_sqp:  this is the QP number of the first of the real SQPs
  base_tunnel_sqp: this is the qp number of the first qp in the tunnel
                   sqp region. (On guests, this is the first tunnel
                   sqp of the 8 which are assigned to that guest).

In addition, in SR-IOV mode, sqp_start is the number of the first
proxy SQP in the proxy SQP region.  (In guests, this is the first
proxy SQP of the 8 which are assigned to that guest)

Note that in non-SR-IOV mode, there are no proxies and no tunnels.
In this case, sqp_start is set to sqp_base -- which minimizes code
changes.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx4/main.c | 17 +++++++++++++++++
 include/linux/mlx4/device.h               | 11 +++++++++++
 2 files changed, 28 insertions(+)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 83afb1541a74..81154a16d6b8 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -391,6 +391,23 @@ static int mlx4_how_many_lives_vf(struct mlx4_dev *dev)
 	return ret;
 }
 
+int mlx4_get_parav_qkey(struct mlx4_dev *dev, u32 qpn, u32 *qkey)
+{
+	u32 qk = MLX4_RESERVED_QKEY_BASE;
+	if (qpn >= dev->caps.base_tunnel_sqpn + 8 * MLX4_MFUNC_MAX ||
+	    qpn < dev->caps.sqp_start)
+		return -EINVAL;
+
+	if (qpn >= dev->caps.base_tunnel_sqpn)
+		/* tunnel qp */
+		qk += qpn - dev->caps.base_tunnel_sqpn;
+	else
+		qk += qpn - dev->caps.sqp_start;
+	*qkey = qk;
+	return 0;
+}
+EXPORT_SYMBOL(mlx4_get_parav_qkey);
+
 int mlx4_is_slave_active(struct mlx4_dev *dev, int slave)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 7fbdc89de495..c30a314e095c 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -56,6 +56,13 @@ enum {
 	MLX4_MAX_PORTS		= 2
 };
 
+/* base qkey for use in sriov tunnel-qp/proxy-qp communication.
+ * These qkeys must not be allowed for general use. This is a 64k range,
+ * and to test for violation, we use the mask (protect against future chg).
+ */
+#define MLX4_RESERVED_QKEY_BASE  (0xFFFF0000)
+#define MLX4_RESERVED_QKEY_MASK  (0xFFFF0000)
+
 enum {
 	MLX4_BOARD_ID_LEN = 64
 };
@@ -293,6 +300,8 @@ struct mlx4_caps {
 	int			max_qp_init_rdma;
 	int			max_qp_dest_rdma;
 	int			sqp_start;
+	u32			base_sqpn;
+	u32			base_tunnel_sqpn;
 	int			num_srqs;
 	int			max_srq_wqes;
 	int			max_srq_sge;
@@ -772,4 +781,6 @@ int mlx4_wol_write(struct mlx4_dev *dev, u64 config, int port);
 int mlx4_counter_alloc(struct mlx4_dev *dev, u32 *idx);
 void mlx4_counter_free(struct mlx4_dev *dev, u32 idx);
 
+int mlx4_get_parav_qkey(struct mlx4_dev *dev, u32 qpn, u32 *qkey);
+
 #endif /* MLX4_DEVICE_H */
-- 
cgit v1.2.3


From 105c320f6ac37af30252577d419e47b39edb5843 Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 19 Jun 2012 11:21:43 +0300
Subject: mlx4_core: Allow guests to have IB ports

Modify mlx4_dev_cap to allow IB support when SR-IOV is active.  Modify
mlx4_slave_cap to set the "rdma-supported" bit in its flags area, and
pass that to the guests (this is done in QUERY_FUNC_CAP and its
wrapper).

However, we don't activate IB support quite yet -- we leave the error
return at the start of mlx4_ib_add in the mlx4_ib driver.

In addition, set "protected fmr supported" bit to zero in the
QUERY_FUNC_CAP wrapper.

Finally, in the QUERY_FUNC_CAP wrapper, we needed to add code which
checks for the port type (IB or Ethernet).  Previously, this was not
an issue, since only Ethernet ports were supported.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c   | 78 ++++++++++++++++++++-----------
 drivers/net/ethernet/mellanox/mlx4/main.c | 26 ++++-------
 2 files changed, 59 insertions(+), 45 deletions(-)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index ee9d6b0b4d20..5549f6b3bb67 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -174,6 +174,7 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_FLAGS_OFFSET		0x0
 #define QUERY_FUNC_CAP_NUM_PORTS_OFFSET		0x1
 #define QUERY_FUNC_CAP_PF_BHVR_OFFSET		0x4
+#define QUERY_FUNC_CAP_FMR_OFFSET		0x8
 #define QUERY_FUNC_CAP_QP_QUOTA_OFFSET		0x10
 #define QUERY_FUNC_CAP_CQ_QUOTA_OFFSET		0x14
 #define QUERY_FUNC_CAP_SRQ_QUOTA_OFFSET		0x18
@@ -183,25 +184,44 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_MAX_EQ_OFFSET		0x2c
 #define QUERY_FUNC_CAP_RESERVED_EQ_OFFSET	0X30
 
+#define QUERY_FUNC_CAP_FMR_FLAG			0x80
+#define QUERY_FUNC_CAP_FLAG_RDMA		0x40
+#define QUERY_FUNC_CAP_FLAG_ETH			0x80
+
+/* when opcode modifier = 1 */
 #define QUERY_FUNC_CAP_PHYS_PORT_OFFSET		0x3
+#define QUERY_FUNC_CAP_RDMA_PROPS_OFFSET	0x8
 #define QUERY_FUNC_CAP_ETH_PROPS_OFFSET		0xc
 
+#define QUERY_FUNC_CAP_ETH_PROPS_FORCE_MAC	0x40
+#define QUERY_FUNC_CAP_ETH_PROPS_FORCE_VLAN	0x80
+
+#define QUERY_FUNC_CAP_RDMA_PROPS_FORCE_PHY_WQE_GID 0x80
+
 	if (vhcr->op_modifier == 1) {
 		field = vhcr->in_modifier;
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
 
-		field = 0; /* ensure fvl bit is not set */
+		field = 0;
+		/* ensure force vlan and force mac bits are not set */
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
+		/* ensure that phy_wqe_gid bit is not set */
+		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_RDMA_PROPS_OFFSET);
+
 	} else if (vhcr->op_modifier == 0) {
-		field = 1 << 7; /* enable only ethernet interface */
+		/* enable rdma and ethernet interfaces */
+		field = (QUERY_FUNC_CAP_FLAG_ETH | QUERY_FUNC_CAP_FLAG_RDMA);
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS_OFFSET);
 
 		field = dev->caps.num_ports;
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_NUM_PORTS_OFFSET);
 
-		size = 0; /* no PF behavious is set for now */
+		size = 0; /* no PF behaviour is set for now */
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_PF_BHVR_OFFSET);
 
+		field = 0; /* protected FMR support not available as yet */
+		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FMR_OFFSET);
+
 		size = dev->caps.num_qps;
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP_QUOTA_OFFSET);
 
@@ -254,11 +274,12 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, struct mlx4_func_cap *func_cap)
 	outbox = mailbox->buf;
 
 	MLX4_GET(field, outbox, QUERY_FUNC_CAP_FLAGS_OFFSET);
-	if (!(field & (1 << 7))) {
-		mlx4_err(dev, "The host doesn't support eth interface\n");
+	if (!(field & (QUERY_FUNC_CAP_FLAG_ETH | QUERY_FUNC_CAP_FLAG_RDMA))) {
+		mlx4_err(dev, "The host supports neither eth nor rdma interfaces\n");
 		err = -EPROTONOSUPPORT;
 		goto out;
 	}
+	func_cap->flags = field;
 
 	MLX4_GET(field, outbox, QUERY_FUNC_CAP_NUM_PORTS_OFFSET);
 	func_cap->num_ports = field;
@@ -297,17 +318,27 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, struct mlx4_func_cap *func_cap)
 		if (err)
 			goto out;
 
-		MLX4_GET(field, outbox, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
-		if (field & (1 << 7)) {
-			mlx4_err(dev, "VLAN is enforced on this port\n");
-			err = -EPROTONOSUPPORT;
-			goto out;
-		}
+		if (dev->caps.port_type[i] == MLX4_PORT_TYPE_ETH) {
+			MLX4_GET(field, outbox, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
+			if (field & QUERY_FUNC_CAP_ETH_PROPS_FORCE_VLAN) {
+				mlx4_err(dev, "VLAN is enforced on this port\n");
+				err = -EPROTONOSUPPORT;
+				goto out;
+			}
 
-		if (field & (1 << 6)) {
-			mlx4_err(dev, "Force mac is enabled on this port\n");
-			err = -EPROTONOSUPPORT;
-			goto out;
+			if (field & QUERY_FUNC_CAP_ETH_PROPS_FORCE_MAC) {
+				mlx4_err(dev, "Force mac is enabled on this port\n");
+				err = -EPROTONOSUPPORT;
+				goto out;
+			}
+		} else if (dev->caps.port_type[i] == MLX4_PORT_TYPE_IB) {
+			MLX4_GET(field, outbox, QUERY_FUNC_CAP_RDMA_PROPS_OFFSET);
+			if (field & QUERY_FUNC_CAP_RDMA_PROPS_FORCE_PHY_WQE_GID) {
+				mlx4_err(dev, "phy_wqe_gid is "
+					 "enforced on this ib port\n");
+				err = -EPROTONOSUPPORT;
+				goto out;
+			}
 		}
 
 		MLX4_GET(field, outbox, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
@@ -701,12 +732,7 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 	u8 port_type;
 	int err;
 
-#define MLX4_PORT_SUPPORT_IB		(1 << 0)
-#define MLX4_PORT_SUGGEST_TYPE		(1 << 3)
-#define MLX4_PORT_DEFAULT_SENSE		(1 << 4)
-#define MLX4_VF_PORT_ETH_ONLY_MASK	(0xff & ~MLX4_PORT_SUPPORT_IB & \
-					 ~MLX4_PORT_SUGGEST_TYPE & \
-					 ~MLX4_PORT_DEFAULT_SENSE)
+#define MLX4_VF_PORT_NO_LINK_SENSE_MASK	0xE0
 
 	err = mlx4_cmd_box(dev, 0, outbox->dma, vhcr->in_modifier, 0,
 			   MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B,
@@ -722,12 +748,10 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 		MLX4_GET(port_type, outbox->buf,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
 
-		/* Allow only Eth port, no link sensing allowed */
-		port_type &= MLX4_VF_PORT_ETH_ONLY_MASK;
-
-		/* check eth is enabled for this port */
-		if (!(port_type & 2))
-			mlx4_dbg(dev, "QUERY PORT: eth not supported by host");
+		/* No link sensing allowed */
+		port_type &= MLX4_VF_PORT_NO_LINK_SENSE_MASK;
+		/* set port type to currently operating port type */
+		port_type |= (dev->caps.port_type[vhcr->in_modifier] & 0x3);
 
 		MLX4_PUT(outbox->buf, port_type,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 81154a16d6b8..58544b72bacb 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -288,29 +288,19 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			/* if only ETH is supported - assign ETH */
 			if (dev->caps.supported_type[i] == MLX4_PORT_TYPE_ETH)
 				dev->caps.port_type[i] = MLX4_PORT_TYPE_ETH;
-			/* if only IB is supported,
-			 * assign IB only if SRIOV is off*/
+			/* if only IB is supported, assign IB */
 			else if (dev->caps.supported_type[i] ==
-				 MLX4_PORT_TYPE_IB) {
-				if (dev->flags & MLX4_FLAG_SRIOV)
-					dev->caps.port_type[i] =
-						MLX4_PORT_TYPE_NONE;
-				else
-					dev->caps.port_type[i] =
-						MLX4_PORT_TYPE_IB;
-			/* if IB and ETH are supported,
-			 * first of all check if SRIOV is on */
-			} else if (dev->flags & MLX4_FLAG_SRIOV)
-				dev->caps.port_type[i] = MLX4_PORT_TYPE_ETH;
+				 MLX4_PORT_TYPE_IB)
+				dev->caps.port_type[i] = MLX4_PORT_TYPE_IB;
 			else {
-				/* In non-SRIOV mode, we set the port type
-				 * according to user selection of port type,
-				 * if usere selected none, take the FW hint */
-				if (port_type_array[i-1] == MLX4_PORT_TYPE_NONE)
+				/* if IB and ETH are supported, we set the port
+				 * type according to user selection of port type;
+				 * if user selected none, take the FW hint */
+				if (port_type_array[i - 1] == MLX4_PORT_TYPE_NONE)
 					dev->caps.port_type[i] = dev->caps.suggested_type[i] ?
 						MLX4_PORT_TYPE_ETH : MLX4_PORT_TYPE_IB;
 				else
-					dev->caps.port_type[i] = port_type_array[i-1];
+					dev->caps.port_type[i] = port_type_array[i - 1];
 			}
 		}
 		/*
-- 
cgit v1.2.3


From 6634961c14d38ef64ec284c07aecb03d3dd03b4a Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 19 Jun 2012 11:21:44 +0300
Subject: mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct
 and paravirtualize them

To allow easy paravirtualization of P_Key and GID table sizes, keep
paravirtualized sizes in mlx4_dev->caps, but save the actual physical
sizes from FW in struct: mlx4_dev->phys_cap.

In addition, in SR-IOV mode, do the following:

1. Reduce reported P_Key table size by 1.
   This is done to reserve the highest P_Key index for internal use,
   for declaring an invalid P_Key in P_Key paravirtualization.
   We require a P_Key index which always contain an invalid P_Key
   value for this purpose (i.e., one which cannot be modified by
   the subnet manager).  The way to do this is to reduce the
   P_Key table size reported to the subnet manager by 1, so that
   it will not attempt to access the P_Key at index #127.

2. Paravirtualize the GID table size to 1. Thus, each guest sees
   only a single GID (at its paravirtualized index 0).

In addition, since we are paravirtualizing the GID table size to 1, we
add paravirtualization of the master GID event here (i.e., we do not
do ib_dispatch_event() for the GUID change event on the master, since
its (only) GUID never changes).

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
---
 drivers/infiniband/hw/mlx4/mad.c          | 10 ++++---
 drivers/net/ethernet/mellanox/mlx4/fw.c   | 43 +++++++++++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlx4/main.c | 32 ++++++++++++++++++++---
 drivers/net/ethernet/mellanox/mlx4/mlx4.h |  4 ++-
 drivers/net/ethernet/mellanox/mlx4/port.c | 11 ++++++--
 include/linux/mlx4/device.h               |  2 ++
 6 files changed, 92 insertions(+), 10 deletions(-)

(limited to 'drivers/net/ethernet/mellanox')

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 58c45fb5bd31..c27141fef1ab 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -184,8 +184,10 @@ static void smp_snoop(struct ib_device *ibdev, u8 port_num, struct ib_mad *mad,
 			break;
 
 		case IB_SMP_ATTR_GUID_INFO:
-			mlx4_ib_dispatch_event(dev, port_num,
-					       IB_EVENT_GID_CHANGE);
+			/* paravirtualized master's guid is guid 0 -- does not change */
+			if (!mlx4_is_master(dev->dev))
+				mlx4_ib_dispatch_event(dev, port_num,
+						       IB_EVENT_GID_CHANGE);
 			break;
 		default:
 			break;
@@ -487,7 +489,9 @@ void handle_port_mgmt_change_event(struct work_struct *work)
 		mlx4_ib_dispatch_event(dev, port, IB_EVENT_PKEY_CHANGE);
 		break;
 	case MLX4_DEV_PMC_SUBTYPE_GUID_INFO:
-		mlx4_ib_dispatch_event(dev, port, IB_EVENT_GID_CHANGE);
+		/* paravirtualized master's guid is guid 0 -- does not change */
+		if (!mlx4_is_master(dev->dev))
+			mlx4_ib_dispatch_event(dev, port, IB_EVENT_GID_CHANGE);
 		break;
 	default:
 		pr_warn("Unsupported subtype 0x%x for "
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 5549f6b3bb67..473d63b63b4e 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -730,9 +730,12 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 {
 	u64 def_mac;
 	u8 port_type;
+	u16 short_field;
 	int err;
 
 #define MLX4_VF_PORT_NO_LINK_SENSE_MASK	0xE0
+#define QUERY_PORT_CUR_MAX_PKEY_OFFSET	0x0c
+#define QUERY_PORT_CUR_MAX_GID_OFFSET	0x0e
 
 	err = mlx4_cmd_box(dev, 0, outbox->dma, vhcr->in_modifier, 0,
 			   MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B,
@@ -755,11 +758,51 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 
 		MLX4_PUT(outbox->buf, port_type,
 			 QUERY_PORT_SUPPORTED_TYPE_OFFSET);
+
+		short_field = 1; /* slave max gids */
+		MLX4_PUT(outbox->buf, short_field,
+			 QUERY_PORT_CUR_MAX_GID_OFFSET);
+
+		short_field = dev->caps.pkey_table_len[vhcr->in_modifier];
+		MLX4_PUT(outbox->buf, short_field,
+			 QUERY_PORT_CUR_MAX_PKEY_OFFSET);
 	}
 
 	return err;
 }
 
+int mlx4_get_slave_pkey_gid_tbl_len(struct mlx4_dev *dev, u8 port,
+				    int *gid_tbl_len, int *pkey_tbl_len)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	u32			*outbox;
+	u16			field;
+	int			err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	err =  mlx4_cmd_box(dev, 0, mailbox->dma, port, 0,
+			    MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B,
+			    MLX4_CMD_WRAPPED);
+	if (err)
+		goto out;
+
+	outbox = mailbox->buf;
+
+	MLX4_GET(field, outbox, QUERY_PORT_CUR_MAX_GID_OFFSET);
+	*gid_tbl_len = field;
+
+	MLX4_GET(field, outbox, QUERY_PORT_CUR_MAX_PKEY_OFFSET);
+	*pkey_tbl_len = field;
+
+out:
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+EXPORT_SYMBOL(mlx4_get_slave_pkey_gid_tbl_len);
+
 int mlx4_map_cmd(struct mlx4_dev *dev, u16 op, struct mlx4_icm *icm, u64 virt)
 {
 	struct mlx4_cmd_mailbox *mailbox;
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 58544b72bacb..5df3ac40a490 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -215,6 +215,10 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		dev->caps.vl_cap[i]	    = dev_cap->max_vl[i];
 		dev->caps.ib_mtu_cap[i]	    = dev_cap->ib_mtu[i];
+		dev->phys_caps.gid_phys_table_len[i]  = dev_cap->max_gids[i];
+		dev->phys_caps.pkey_phys_table_len[i] = dev_cap->max_pkeys[i];
+		/* set gid and pkey table operating lengths by default
+		 * to non-sriov values */
 		dev->caps.gid_table_len[i]  = dev_cap->max_gids[i];
 		dev->caps.pkey_table_len[i] = dev_cap->max_pkeys[i];
 		dev->caps.port_width_cap[i] = dev_cap->max_port_width[i];
@@ -498,8 +502,13 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 		return -ENODEV;
 	}
 
-	for (i = 1; i <= dev->caps.num_ports; ++i)
+	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		dev->caps.port_mask[i] = dev->caps.port_type[i];
+		if (mlx4_get_slave_pkey_gid_tbl_len(dev, i,
+						    &dev->caps.gid_table_len[i],
+						    &dev->caps.pkey_table_len[i]))
+			return -ENODEV;
+	}
 
 	if (dev->caps.uar_page_size * (dev->caps.num_uars -
 				       dev->caps.reserved_uars) >
@@ -536,7 +545,7 @@ int mlx4_change_port_types(struct mlx4_dev *dev,
 		for (port = 1; port <= dev->caps.num_ports; port++) {
 			mlx4_CLOSE_PORT(dev, port);
 			dev->caps.port_type[port] = port_types[port - 1];
-			err = mlx4_SET_PORT(dev, port);
+			err = mlx4_SET_PORT(dev, port, -1);
 			if (err) {
 				mlx4_err(dev, "Failed to set port %d, "
 					      "aborting\n", port);
@@ -722,7 +731,7 @@ static ssize_t set_port_ib_mtu(struct device *dev,
 	mlx4_unregister_device(mdev);
 	for (port = 1; port <= mdev->caps.num_ports; port++) {
 		mlx4_CLOSE_PORT(mdev, port);
-		err = mlx4_SET_PORT(mdev, port);
+		err = mlx4_SET_PORT(mdev, port, -1);
 		if (err) {
 			mlx4_err(mdev, "Failed to set port %d, "
 				      "aborting\n", port);
@@ -1173,6 +1182,17 @@ err:
 	return -EIO;
 }
 
+static void mlx4_parav_master_pf_caps(struct mlx4_dev *dev)
+{
+	int i;
+
+	for (i = 1; i <= dev->caps.num_ports; i++) {
+		dev->caps.gid_table_len[i] = 1;
+		dev->caps.pkey_table_len[i] =
+			dev->phys_caps.pkey_phys_table_len[i] - 1;
+	}
+}
+
 static int mlx4_init_hca(struct mlx4_dev *dev)
 {
 	struct mlx4_priv	  *priv = mlx4_priv(dev);
@@ -1212,6 +1232,9 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 			goto err_stop_fw;
 		}
 
+		if (mlx4_is_master(dev))
+			mlx4_parav_master_pf_caps(dev);
+
 		profile = default_profile;
 
 		icm_size = mlx4_make_profile(dev, &profile, &dev_cap,
@@ -1500,7 +1523,8 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 			else
 				dev->caps.port_ib_mtu[port] = IB_MTU_4096;
 
-			err = mlx4_SET_PORT(dev, port);
+			err = mlx4_SET_PORT(dev, port, mlx4_is_master(dev) ?
+					    dev->caps.pkey_table_len[port] : -1);
 			if (err) {
 				mlx4_err(dev, "Failed to set port %d, aborting\n",
 					port);
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 4d11d12b9db4..cde6e511899f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -969,7 +969,7 @@ int mlx4_change_port_types(struct mlx4_dev *dev,
 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table);
 void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table);
 
-int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port);
+int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz);
 /* resource tracker functions*/
 int mlx4_get_slave_from_resource_id(struct mlx4_dev *dev,
 				    enum mlx4_resource resource_type,
@@ -1012,6 +1012,8 @@ int mlx4_QUERY_PORT_wrapper(struct mlx4_dev *dev, int slave,
 			    struct mlx4_cmd_info *cmd);
 int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps);
 
+int mlx4_get_slave_pkey_gid_tbl_len(struct mlx4_dev *dev, u8 port,
+				    int *gid_tbl_len, int *pkey_tbl_len);
 
 int mlx4_QP_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 			   struct mlx4_vhcr *vhcr,
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index a8fb52992c64..90dc47542b8b 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -726,14 +726,15 @@ int mlx4_SET_PORT_wrapper(struct mlx4_dev *dev, int slave,
 enum {
 	MLX4_SET_PORT_VL_CAP	 = 4, /* bits 7:4 */
 	MLX4_SET_PORT_MTU_CAP	 = 12, /* bits 15:12 */
+	MLX4_CHANGE_PORT_PKEY_TBL_SZ = 20,
 	MLX4_CHANGE_PORT_VL_CAP	 = 21,
 	MLX4_CHANGE_PORT_MTU_CAP = 22,
 };
 
-int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port)
+int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz)
 {
 	struct mlx4_cmd_mailbox *mailbox;
-	int err, vl_cap;
+	int err, vl_cap, pkey_tbl_flag = 0;
 
 	if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH)
 		return 0;
@@ -746,11 +747,17 @@ int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port)
 
 	((__be32 *) mailbox->buf)[1] = dev->caps.ib_port_def_cap[port];
 
+	if (pkey_tbl_sz >= 0 && mlx4_is_master(dev)) {
+		pkey_tbl_flag = 1;
+		((__be16 *) mailbox->buf)[20] = cpu_to_be16(pkey_tbl_sz);
+	}
+
 	/* IB VL CAP enum isn't used by the firmware, just numerical values */
 	for (vl_cap = 8; vl_cap >= 1; vl_cap >>= 1) {
 		((__be32 *) mailbox->buf)[0] = cpu_to_be32(
 			(1 << MLX4_CHANGE_PORT_MTU_CAP) |
 			(1 << MLX4_CHANGE_PORT_VL_CAP)  |
+			(pkey_tbl_flag << MLX4_CHANGE_PORT_PKEY_TBL_SZ) |
 			(dev->caps.port_ib_mtu[port] << MLX4_SET_PORT_MTU_CAP) |
 			(vl_cap << MLX4_SET_PORT_VL_CAP));
 		err = mlx4_cmd(dev, mailbox->dma, port, 0, MLX4_CMD_SET_PORT,
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index c30a314e095c..441caf1a497d 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -268,6 +268,8 @@ static inline u64 mlx4_fw_ver(u64 major, u64 minor, u64 subminor)
 }
 
 struct mlx4_phys_caps {
+	u32			gid_phys_table_len[MLX4_MAX_PORTS + 1];
+	u32			pkey_phys_table_len[MLX4_MAX_PORTS + 1];
 	u32			num_phys_eqs;
 };
 
-- 
cgit v1.2.3