[PATCH net-next V1 00/11] net/mlx4: Mellanox driver update 08-12-2013

[PATCH net-next V1 00/11] net/mlx4: Mellanox driver update 08-12-2013

[Amir Vadai]

Hi Dave,

This patchset contains:
1. Support in ndo_get_phys_port_id added by Hadar.
2. Change the driver to use by default CQE/EQE size of 64 bytes done by Eyal.
   This doubles the packet-rate  of the NIC.
3. Configure the XPS queue mapping on driver load - added by Ido.
4. Fixes for some small bugs done by Jenny, Matan and Rana

Patchset was applied and tested against commit: "89e47d3 fddi: cleanup unsigned
to unsigned int/short"

Changes from V0:
- Found some issues in "Reuse memory in RX flow" patch from V0.
  Removing this patch from the patchset till analyzed and fixed.
- Fix some coding style issues in patch 6 "Configure the XPS queue mapping on
  driver load"
- Changed patch 9 "Add NAPI support for transmit side" to use NAPI_POLL_WEIGHT
  instead of MLX4_EN_TX_BUDGET   

Thanks,
Amir

Eugenia Emantayev (2):
  net/mlx4_en: Ignore irrelevant hypervisor events
  net/mlx4_en: Add NAPI support for transmit side

Eyal Perry (1):
  net/mlx4_core: Set CQE/EQE size to 64B by default

Hadar Hen Zion (5):
  net/mlx4_core: Remove zeroed out of explicit QUERY_FUNC_CAP fields
  net/mlx4_core: Rename QUERY_FUNC_CAP fields
  net/mlx4_core: Introduce nic_info new flag in QUERY_FUNC_CAP
  net/mlx4_core: Expose physical port id as PF/VF capability
  net/mlx4_en: Implement ndo_get_phys_port_id

Ido Shamay (1):
  net/mlx4_en: Configure the XPS queue mapping on driver load

Matan Barak (1):
  net/mlx4_core: Check port number for validity before accessing data

Rana Shahout (1):
  net/mlx4_en: Fix Supported/Advertised link mode reported by ethtool

 drivers/net/ethernet/mellanox/mlx4/en_cq.c      | 12 ++--
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 24 +++++++-
 drivers/net/ethernet/mellanox/mlx4/en_main.c    |  3 +
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c  | 29 +++++++++-
 drivers/net/ethernet/mellanox/mlx4/en_tx.c      | 48 +++++++++++++---
 drivers/net/ethernet/mellanox/mlx4/fw.c         | 74 ++++++++++++++++++++-----
 drivers/net/ethernet/mellanox/mlx4/fw.h         |  2 +
 drivers/net/ethernet/mellanox/mlx4/main.c       |  9 ++-
 drivers/net/ethernet/mellanox/mlx4/mcg.c        | 28 +++++++++-
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h    |  6 +-
 include/linux/mlx4/device.h                     |  2 +
 11 files changed, 202 insertions(+), 35 deletions(-)

-- 
1.8.3.4

[PATCH net-next V1 01/11] net/mlx4_core: Remove zeroed out of explicit QUERY_FUNC_CAP fields

[Amir Vadai]

From: Hadar Hen Zion <hadarh@mellanox.com>

All mailboxes are already zeroed by commit:
571b8b9 net/mlx4_core: Initialize all mailbox buffers to zero before use
Remove explicit zero set for force mac and force vlan fields in
mlx4_QUERY_FUNC_CAP_wrapper

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 1949282..91b50fe 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -221,12 +221,6 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_RDMA_PROPS_FORCE_PHY_WQE_GID 0x80
 
 	if (vhcr->op_modifier == 1) {
-		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);
-
 		field = vhcr->in_modifier; /* phys-port = logical-port */
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
 
-- 
1.8.3.4

[PATCH net-next V1 02/11] net/mlx4_core: Rename QUERY_FUNC_CAP fields

[Amir Vadai]

From: Hadar Hen Zion <hadarh@mellanox.com>

Use correct names for QUERY_FUNC_CAP fields: flags0 and flags1.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 91b50fe..58ca7de 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -207,18 +207,18 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 
 /* 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_FLAGS0_OFFSET		0x8
+#define QUERY_FUNC_CAP_FLAGS1_OFFSET		0xc
 
 #define QUERY_FUNC_CAP_QP0_TUNNEL		0x10
 #define QUERY_FUNC_CAP_QP0_PROXY		0x14
 #define QUERY_FUNC_CAP_QP1_TUNNEL		0x18
 #define QUERY_FUNC_CAP_QP1_PROXY		0x1c
 
-#define QUERY_FUNC_CAP_ETH_PROPS_FORCE_MAC	0x40
-#define QUERY_FUNC_CAP_ETH_PROPS_FORCE_VLAN	0x80
+#define QUERY_FUNC_CAP_FLAGS1_FORCE_MAC		0x40
+#define QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN	0x80
 
-#define QUERY_FUNC_CAP_RDMA_PROPS_FORCE_PHY_WQE_GID 0x80
+#define QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID 0x80
 
 	if (vhcr->op_modifier == 1) {
 		field = vhcr->in_modifier; /* phys-port = logical-port */
@@ -386,21 +386,21 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 	}
 
 	if (dev->caps.port_type[gen_or_port] == MLX4_PORT_TYPE_ETH) {
-		MLX4_GET(field, outbox, QUERY_FUNC_CAP_ETH_PROPS_OFFSET);
-		if (field & QUERY_FUNC_CAP_ETH_PROPS_FORCE_VLAN) {
+		MLX4_GET(field, outbox, QUERY_FUNC_CAP_FLAGS1_OFFSET);
+		if (field & QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN) {
 			mlx4_err(dev, "VLAN is enforced on this port\n");
 			err = -EPROTONOSUPPORT;
 			goto out;
 		}
 
-		if (field & QUERY_FUNC_CAP_ETH_PROPS_FORCE_MAC) {
+		if (field & QUERY_FUNC_CAP_FLAGS1_FORCE_MAC) {
 			mlx4_err(dev, "Force mac is enabled on this port\n");
 			err = -EPROTONOSUPPORT;
 			goto out;
 		}
 	} else if (dev->caps.port_type[gen_or_port] == 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_GET(field, outbox, QUERY_FUNC_CAP_FLAGS0_OFFSET);
+		if (field & QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID) {
 			mlx4_err(dev, "phy_wqe_gid is "
 				 "enforced on this ib port\n");
 			err = -EPROTONOSUPPORT;
-- 
1.8.3.4

[PATCH net-next V1 03/11] net/mlx4_core: Introduce nic_info new flag in QUERY_FUNC_CAP

[Amir Vadai]

From: Hadar Hen Zion <hadarh@mellanox.com>

Set nic_info field in QUERY_FUNC_CAP, which designates
supplementary NIC information is provided by the hypervisor.
When set, the following fields are valid: nic_num_rings,
nic_indirection_tbl_sz, cur_mac and phys_port_id.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c | 11 ++++++++---
 drivers/net/ethernet/mellanox/mlx4/fw.h |  1 +
 2 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 58ca7de..bfe91ae 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -217,10 +217,15 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 
 #define QUERY_FUNC_CAP_FLAGS1_FORCE_MAC		0x40
 #define QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN	0x80
+#define QUERY_FUNC_CAP_FLAGS1_NIC_INFO			0x10
 
 #define QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID 0x80
 
 	if (vhcr->op_modifier == 1) {
+		/* Set nic_info bit to mark new fields support */
+		field  = QUERY_FUNC_CAP_FLAGS1_NIC_INFO;
+		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS1_OFFSET);
+
 		field = vhcr->in_modifier; /* phys-port = logical-port */
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
 
@@ -385,15 +390,15 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 		goto out;
 	}
 
+	MLX4_GET(func_cap->flags1, outbox, QUERY_FUNC_CAP_FLAGS1_OFFSET);
 	if (dev->caps.port_type[gen_or_port] == MLX4_PORT_TYPE_ETH) {
-		MLX4_GET(field, outbox, QUERY_FUNC_CAP_FLAGS1_OFFSET);
-		if (field & QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN) {
+		if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_OFFSET) {
 			mlx4_err(dev, "VLAN is enforced on this port\n");
 			err = -EPROTONOSUPPORT;
 			goto out;
 		}
 
-		if (field & QUERY_FUNC_CAP_FLAGS1_FORCE_MAC) {
+		if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_FORCE_MAC) {
 			mlx4_err(dev, "Force mac is enabled on this port\n");
 			err = -EPROTONOSUPPORT;
 			goto out;
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h
index a0a368b..9d95298 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.h
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.h
@@ -140,6 +140,7 @@ struct mlx4_func_cap {
 	u32	qp1_proxy_qpn;
 	u8	physical_port;
 	u8	port_flags;
+	u8	flags1;
 };
 
 struct mlx4_adapter {
-- 
1.8.3.4

[PATCH net-next V1 04/11] net/mlx4_core: Expose physical port id as PF/VF capability

[Amir Vadai]

From: Hadar Hen Zion <hadarh@mellanox.com>

Add the infrastructure needed to support ndo_get_phys_port_id which
allows users to identify to which physical port a net-device is connected
to by reading a unique port id.
This will work for VFs and PFs.
The driver uses a new device capability - phys_port_id, The PF driver
reads the port phys_port_id from Firmware and stores it. The VF driver
reads the port phys_port_id from the PF using QUERY_FUNC_CAP command.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c   | 45 +++++++++++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlx4/fw.h   |  1 +
 drivers/net/ethernet/mellanox/mlx4/main.c |  5 ++++
 include/linux/mlx4/device.h               |  2 ++
 4 files changed, 53 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index bfe91ae..27a8434 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -214,6 +214,7 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_QP0_PROXY		0x14
 #define QUERY_FUNC_CAP_QP1_TUNNEL		0x18
 #define QUERY_FUNC_CAP_QP1_PROXY		0x1c
+#define QUERY_FUNC_CAP_PHYS_PORT_ID		0x28
 
 #define QUERY_FUNC_CAP_FLAGS1_FORCE_MAC		0x40
 #define QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN	0x80
@@ -242,6 +243,9 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 		size += 2;
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP1_PROXY);
 
+		MLX4_PUT(outbox->buf, dev->caps.phys_port_id[vhcr->in_modifier],
+			 QUERY_FUNC_CAP_PHYS_PORT_ID);
+
 	} else if (vhcr->op_modifier == 0) {
 		/* enable rdma and ethernet interfaces, and new quota locations */
 		field = (QUERY_FUNC_CAP_FLAG_ETH | QUERY_FUNC_CAP_FLAG_RDMA |
@@ -432,6 +436,10 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 	MLX4_GET(size, outbox, QUERY_FUNC_CAP_QP1_PROXY);
 	func_cap->qp1_proxy_qpn = size & 0xFFFFFF;
 
+	if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_NIC_INFO)
+		MLX4_GET(func_cap->phys_port_id, outbox,
+			 QUERY_FUNC_CAP_PHYS_PORT_ID);
+
 	/* All other resources are allocated by the master, but we still report
 	 * 'num' and 'reserved' capabilities as follows:
 	 * - num remains the maximum resource index
@@ -1712,6 +1720,43 @@ int mlx4_NOP(struct mlx4_dev *dev)
 	return mlx4_cmd(dev, 0, 0x1f, 0, MLX4_CMD_NOP, 100, MLX4_CMD_NATIVE);
 }
 
+int mlx4_get_phys_port_id(struct mlx4_dev *dev)
+{
+	u8 port;
+	u32 *outbox;
+	struct mlx4_cmd_mailbox *mailbox;
+	u32 in_mod;
+	u32 guid_hi, guid_lo;
+	int err, ret = 0;
+#define MOD_STAT_CFG_PORT_OFFSET 8
+#define MOD_STAT_CFG_GUID_H	 0X14
+#define MOD_STAT_CFG_GUID_L	 0X1c
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+	outbox = mailbox->buf;
+
+	for (port = 1; port <= dev->caps.num_ports; port++) {
+		in_mod = port << MOD_STAT_CFG_PORT_OFFSET;
+		err = mlx4_cmd_box(dev, 0, mailbox->dma, in_mod, 0x2,
+				   MLX4_CMD_MOD_STAT_CFG, MLX4_CMD_TIME_CLASS_A,
+				   MLX4_CMD_NATIVE);
+		if (err) {
+			mlx4_err(dev, "Fail to get port %d uplink guid\n",
+				 port);
+			ret = err;
+		} else {
+			MLX4_GET(guid_hi, outbox, MOD_STAT_CFG_GUID_H);
+			MLX4_GET(guid_lo, outbox, MOD_STAT_CFG_GUID_L);
+			dev->caps.phys_port_id[port] = (u64)guid_lo |
+						       (u64)guid_hi << 32;
+		}
+	}
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return ret;
+}
+
 #define MLX4_WOL_SETUP_MODE (5 << 28)
 int mlx4_wol_read(struct mlx4_dev *dev, u64 *config, int port)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h
index 9d95298..6811ee0 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.h
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.h
@@ -141,6 +141,7 @@ struct mlx4_func_cap {
 	u8	physical_port;
 	u8	port_flags;
 	u8	flags1;
+	u64	phys_port_id;
 };
 
 struct mlx4_adapter {
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index b37478d..6b13a15 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -606,6 +606,7 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 		dev->caps.qp1_tunnel[i - 1] = func_cap.qp1_tunnel_qpn;
 		dev->caps.qp1_proxy[i - 1] = func_cap.qp1_proxy_qpn;
 		dev->caps.port_mask[i] = dev->caps.port_type[i];
+		dev->caps.phys_port_id[i] = func_cap.phys_port_id;
 		if (mlx4_get_slave_pkey_gid_tbl_len(dev, i,
 						    &dev->caps.gid_table_len[i],
 						    &dev->caps.pkey_table_len[i]))
@@ -1484,6 +1485,10 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		choose_steering_mode(dev, &dev_cap);
 
+		err = mlx4_get_phys_port_id(dev);
+		if (err)
+			mlx4_err(dev, "Fail to get physical port id\n");
+
 		if (mlx4_is_master(dev))
 			mlx4_parav_master_pf_caps(dev);
 
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 7d3a523..294b7c5 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -454,6 +454,7 @@ struct mlx4_caps {
 	u32			userspace_caps; /* userspace must be aware of these */
 	u32			function_caps;  /* VFs must be aware of these */
 	u16			hca_core_clock;
+	u64			phys_port_id[MLX4_MAX_PORTS + 1];
 };
 
 struct mlx4_buf_list {
@@ -1113,6 +1114,7 @@ int mlx4_assign_eq(struct mlx4_dev *dev, char *name, struct cpu_rmap *rmap,
 		   int *vector);
 void mlx4_release_eq(struct mlx4_dev *dev, int vec);
 
+int mlx4_get_phys_port_id(struct mlx4_dev *dev);
 int mlx4_wol_read(struct mlx4_dev *dev, u64 *config, int port);
 int mlx4_wol_write(struct mlx4_dev *dev, u64 config, int port);
 
-- 
1.8.3.4

[PATCH net-next V1 05/11] net/mlx4_en: Implement ndo_get_phys_port_id

[Amir Vadai]

From: Hadar Hen Zion <hadarh@mellanox.com>

Use the port GUID read from the firmware to identify the physical port.
This port identifier is available via ndo_get_phys_port_id for both PF
and VF net-devices.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 709e5ec..d2e9666 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2164,6 +2164,27 @@ static int mlx4_en_set_vf_link_state(struct net_device *dev, int vf, int link_st
 
 	return mlx4_set_vf_link_state(mdev->dev, en_priv->port, vf, link_state);
 }
+
+#define PORT_ID_BYTE_LEN 8
+static int mlx4_en_get_phys_port_id(struct net_device *dev,
+				    struct netdev_phys_port_id *ppid)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	struct mlx4_dev *mdev = priv->mdev->dev;
+	int i;
+	u64 phys_port_id = mdev->caps.phys_port_id[priv->port];
+
+	if (!phys_port_id)
+		return -EOPNOTSUPP;
+
+	ppid->id_len = sizeof(phys_port_id);
+	for (i = PORT_ID_BYTE_LEN - 1; i >= 0; --i) {
+		ppid->id[i] =  phys_port_id & 0xff;
+		phys_port_id >>= 8;
+	}
+	return 0;
+}
+
 static const struct net_device_ops mlx4_netdev_ops = {
 	.ndo_open		= mlx4_en_open,
 	.ndo_stop		= mlx4_en_close,
@@ -2189,6 +2210,7 @@ static const struct net_device_ops mlx4_netdev_ops = {
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	.ndo_busy_poll		= mlx4_en_low_latency_recv,
 #endif
+	.ndo_get_phys_port_id	= mlx4_en_get_phys_port_id,
 };
 
 static const struct net_device_ops mlx4_netdev_ops_master = {
@@ -2217,6 +2239,7 @@ static const struct net_device_ops mlx4_netdev_ops_master = {
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= mlx4_en_filter_rfs,
 #endif
+	.ndo_get_phys_port_id	= mlx4_en_get_phys_port_id,
 };
 
 int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
-- 
1.8.3.4

[PATCH net-next V1 06/11] net/mlx4_en: Configure the XPS queue mapping on driver load

[Amir Vadai]

From: Ido Shamay <idos@mellanox.com>

Only TX rings of User Piority 0 are mapped.
TX rings of other UP's are using UP 0 mapping.
XPS is not in use when num_tc is set.

Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 6 ++++--
 drivers/net/ethernet/mellanox/mlx4/en_tx.c     | 9 ++++++++-
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h   | 5 ++++-
 3 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index d2e9666..6f92090 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1910,8 +1910,10 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
 				      prof->tx_ring_size, i, TX, node))
 			goto err;
 
-		if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i], priv->base_tx_qpn + i,
-					   prof->tx_ring_size, TXBB_SIZE, node))
+		if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i],
+					   priv->base_tx_qpn + i,
+					   prof->tx_ring_size, TXBB_SIZE,
+					   node, i))
 			goto err;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index f54ebd5..5e22d7d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -55,7 +55,7 @@ MODULE_PARM_DESC(inline_thold, "threshold for using inline data");
 
 int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_tx_ring **pring, int qpn, u32 size,
-			   u16 stride, int node)
+			   u16 stride, int node, int queue_index)
 {
 	struct mlx4_en_dev *mdev = priv->mdev;
 	struct mlx4_en_tx_ring *ring;
@@ -140,6 +140,10 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 		ring->bf_enabled = true;
 
 	ring->hwtstamp_tx_type = priv->hwtstamp_config.tx_type;
+	ring->queue_index = queue_index;
+
+	if (queue_index < priv->num_tx_rings_p_up && cpu_online(queue_index))
+		cpumask_set_cpu(queue_index, &ring->affinity_mask);
 
 	*pring = ring;
 	return 0;
@@ -206,6 +210,9 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
 
 	err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context,
 			       &ring->qp, &ring->qp_state);
+	if (!user_prio && cpu_online(ring->queue_index))
+		netif_set_xps_queue(priv->dev, &ring->affinity_mask,
+				    ring->queue_index);
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index f3758de..202d8e5 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -255,6 +255,8 @@ struct mlx4_en_tx_ring {
 	u16 poll_cnt;
 	struct mlx4_en_tx_info *tx_info;
 	u8 *bounce_buf;
+	u8 queue_index;
+	cpumask_t affinity_mask;
 	u32 last_nr_txbb;
 	struct mlx4_qp qp;
 	struct mlx4_qp_context context;
@@ -719,7 +721,8 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev);
 
 int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_tx_ring **pring,
-			   int qpn, u32 size, u16 stride, int node);
+			   int qpn, u32 size, u16 stride,
+			   int node, int queue_index);
 void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv,
 			     struct mlx4_en_tx_ring **pring);
 int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
-- 
1.8.3.4

[PATCH net-next V1 07/11] net/mlx4_core: Set CQE/EQE size to 64B by default

[Amir Vadai]

From: Eyal Perry <eyalpe@mellanox.com>

To achieve out of the box performance default is to use 64 byte CQE/EQE.
In tests that we conduct in our labs, we achieved a performance
improvement of twice the message rate. For older VF/libmlx4 support,
enable_64b_cqe_eqe must be set to 0 (disabled).


Signed-off-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 6b13a15..fbebfb0 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -96,10 +96,10 @@ MODULE_PARM_DESC(log_num_mgm_entry_size, "log mgm size, that defines the num"
 					 " To activate device managed"
 					 " flow steering when available, set to -1");
 
-static bool enable_64b_cqe_eqe;
+static bool enable_64b_cqe_eqe = true;
 module_param(enable_64b_cqe_eqe, bool, 0444);
 MODULE_PARM_DESC(enable_64b_cqe_eqe,
-		 "Enable 64 byte CQEs/EQEs when the FW supports this");
+		 "Enable 64 byte CQEs/EQEs when the FW supports this (default: True)");
 
 #define HCA_GLOBAL_CAP_MASK            0
 
-- 
1.8.3.4

[PATCH net-next V1 08/11] net/mlx4_en: Ignore irrelevant hypervisor events

[Amir Vadai]

From: Eugenia Emantayev <eugenia@mellanox.co.il>

MLX4_DEV_EVENT_SLAVE_INIT and MLX4_DEV_EVENT_SLAVE_SHUTDOWN
events used by Hypervisor to inform the PPF IB driver that
IB para-virtualization must be initialized/destroyed for a slave.
If this event is catched by ETH VF annoying but harmless error message
is printed into dmesg. Remove dmesg prints for these events.


Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c
index 0d087b0..725a4e1 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c
@@ -174,6 +174,9 @@ static void mlx4_en_event(struct mlx4_dev *dev, void *endev_ptr,
 		mlx4_err(mdev, "Internal error detected, restarting device\n");
 		break;
 
+	case MLX4_DEV_EVENT_SLAVE_INIT:
+	case MLX4_DEV_EVENT_SLAVE_SHUTDOWN:
+		break;
 	default:
 		if (port < 1 || port > dev->caps.num_ports ||
 		    !mdev->pndev[port])
-- 
1.8.3.4

[PATCH net-next V1 09/11] net/mlx4_en: Add NAPI support for transmit side

[Amir Vadai]

From: Eugenia Emantayev <eugenia@mellanox.com>

Add NAPI for TX side,
implement its support and provide NAPI callback.

CC: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_cq.c   | 12 ++++++---
 drivers/net/ethernet/mellanox/mlx4/en_tx.c   | 39 +++++++++++++++++++++++-----
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h |  1 +
 3 files changed, 41 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
index 3a098cc..70e9532 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
@@ -161,12 +161,16 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
 	cq->mcq.comp  = cq->is_tx ? mlx4_en_tx_irq : mlx4_en_rx_irq;
 	cq->mcq.event = mlx4_en_cq_event;
 
-	if (!cq->is_tx) {
+	if (cq->is_tx) {
+		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq,
+			       NAPI_POLL_WEIGHT);
+	} else {
 		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq, 64);
 		napi_hash_add(&cq->napi);
-		napi_enable(&cq->napi);
 	}
 
+	napi_enable(&cq->napi);
+
 	return 0;
 }
 
@@ -188,12 +192,12 @@ void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq **pcq)
 
 void mlx4_en_deactivate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq)
 {
+	napi_disable(&cq->napi);
 	if (!cq->is_tx) {
-		napi_disable(&cq->napi);
 		napi_hash_del(&cq->napi);
 		synchronize_rcu();
-		netif_napi_del(&cq->napi);
 	}
+	netif_napi_del(&cq->napi);
 
 	mlx4_cq_free(priv->mdev->dev, &cq->mcq);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 5e22d7d..e3adceb 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -324,7 +324,7 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
 			}
 		}
 	}
-	dev_kfree_skb_any(skb);
+	dev_kfree_skb(skb);
 	return tx_info->nr_txbb;
 }
 
@@ -361,7 +361,9 @@ int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring)
 	return cnt;
 }
 
-static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
+static int mlx4_en_process_tx_cq(struct net_device *dev,
+				 struct mlx4_en_cq *cq,
+				 int budget)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_cq *mcq = &cq->mcq;
@@ -379,9 +381,10 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
 	u32 bytes = 0;
 	int factor = priv->cqe_factor;
 	u64 timestamp = 0;
+	int done = 0;
 
 	if (!priv->port_up)
-		return;
+		return 0;
 
 	index = cons_index & size_mask;
 	cqe = &buf[(index << factor) + factor];
@@ -390,7 +393,7 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
 
 	/* Process all completed CQEs */
 	while (XNOR(cqe->owner_sr_opcode & MLX4_CQE_OWNER_MASK,
-			cons_index & size)) {
+			cons_index & size) && (done < budget)) {
 		/*
 		 * make sure we read the CQE after we read the
 		 * ownership bit
@@ -428,7 +431,7 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
 			txbbs_stamp = txbbs_skipped;
 			packets++;
 			bytes += ring->tx_info[ring_index].nr_bytes;
-		} while (ring_index != new_index);
+		} while ((++done < budget) && (ring_index != new_index));
 
 		++cons_index;
 		index = cons_index & size_mask;
@@ -454,6 +457,7 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
 		netif_tx_wake_queue(ring->tx_queue);
 		priv->port_stats.wake_queue++;
 	}
+	return done;
 }
 
 void mlx4_en_tx_irq(struct mlx4_cq *mcq)
@@ -461,10 +465,31 @@ void mlx4_en_tx_irq(struct mlx4_cq *mcq)
 	struct mlx4_en_cq *cq = container_of(mcq, struct mlx4_en_cq, mcq);
 	struct mlx4_en_priv *priv = netdev_priv(cq->dev);
 
-	mlx4_en_process_tx_cq(cq->dev, cq);
-	mlx4_en_arm_cq(priv, cq);
+	if (priv->port_up)
+		napi_schedule(&cq->napi);
+	else
+		mlx4_en_arm_cq(priv, cq);
 }
 
+/* TX CQ polling - called by NAPI */
+int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget)
+{
+	struct mlx4_en_cq *cq = container_of(napi, struct mlx4_en_cq, napi);
+	struct net_device *dev = cq->dev;
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	int done;
+
+	done = mlx4_en_process_tx_cq(dev, cq, budget);
+
+	/* If we used up all the quota - we're probably not done yet... */
+	if (done < budget) {
+		/* Done for now */
+		napi_complete(napi);
+		mlx4_en_arm_cq(priv, cq);
+		return done;
+	}
+	return budget;
+}
 
 static struct mlx4_en_tx_desc *mlx4_en_bounce_to_desc(struct mlx4_en_priv *priv,
 						      struct mlx4_en_tx_ring *ring,
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 202d8e5..123714c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -744,6 +744,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev,
 			  struct mlx4_en_cq *cq,
 			  int budget);
 int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget);
+int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget);
 void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride,
 		int is_tx, int rss, int qpn, int cqn, int user_prio,
 		struct mlx4_qp_context *context);
-- 
1.8.3.4

[PATCH net-next V1 10/11] net/mlx4_en: Fix Supported/Advertised link mode reported by ethtool

[Amir Vadai]

From: Rana Shahout <ranas@mellanox.com>

Correctly report the support for 1/10/40Gbps link-speed as
supported/advertised link mode in ethtool's get_settings command.


Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 0596f9f..a601869 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -367,8 +367,28 @@ static int mlx4_en_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	int trans_type;
 
 	cmd->autoneg = AUTONEG_DISABLE;
-	cmd->supported = SUPPORTED_10000baseT_Full;
-	cmd->advertising = ADVERTISED_10000baseT_Full;
+
+	cmd->supported = SUPPORTED_1000baseT_Full |
+		SUPPORTED_10000baseT_Full |
+		SUPPORTED_1000baseKX_Full |
+		SUPPORTED_10000baseKX4_Full |
+		SUPPORTED_10000baseKR_Full |
+		SUPPORTED_10000baseR_FEC |
+		SUPPORTED_40000baseKR4_Full |
+		SUPPORTED_40000baseCR4_Full |
+		SUPPORTED_40000baseSR4_Full |
+		SUPPORTED_40000baseLR4_Full;
+
+	cmd->advertising = ADVERTISED_1000baseT_Full |
+		ADVERTISED_10000baseT_Full |
+		ADVERTISED_1000baseKX_Full |
+		ADVERTISED_10000baseKX4_Full |
+		ADVERTISED_10000baseKR_Full |
+		ADVERTISED_10000baseR_FEC |
+		ADVERTISED_40000baseKR4_Full |
+		ADVERTISED_40000baseCR4_Full |
+		ADVERTISED_40000baseSR4_Full |
+		ADVERTISED_40000baseLR4_Full;
 
 	if (mlx4_en_QUERY_PORT(priv->mdev, priv->port))
 		return -ENOMEM;
-- 
1.8.3.4

[PATCH net-next V1 11/11] net/mlx4_core: Check port number for validity before accessing data

[Amir Vadai]

From: Matan Barak <matanb@mellanox.com>

Need to validate port number at mlx4_promisc_qp() before use.
Since port number is extracted from gid, as a cooked or corrupted gid
could lead to a crash.

Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/mcg.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c
index 4059405..7c83e6c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mcg.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c
@@ -125,9 +125,14 @@ static struct mlx4_promisc_qp *get_promisc_qp(struct mlx4_dev *dev, u8 port,
 					      enum mlx4_steer_type steer,
 					      u32 qpn)
 {
-	struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[port - 1];
+	struct mlx4_steer *s_steer;
 	struct mlx4_promisc_qp *pqp;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return NULL;
+
+	s_steer = &mlx4_priv(dev)->steer[port - 1];
+
 	list_for_each_entry(pqp, &s_steer->promisc_qps[steer], list) {
 		if (pqp->qpn == qpn)
 			return pqp;
@@ -154,6 +159,9 @@ static int new_steering_entry(struct mlx4_dev *dev, u8 port,
 	u32 prot;
 	int err;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return -EINVAL;
+
 	s_steer = &mlx4_priv(dev)->steer[port - 1];
 	new_entry = kzalloc(sizeof *new_entry, GFP_KERNEL);
 	if (!new_entry)
@@ -238,6 +246,9 @@ static int existing_steering_entry(struct mlx4_dev *dev, u8 port,
 	struct mlx4_promisc_qp *pqp;
 	struct mlx4_promisc_qp *dqp;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return -EINVAL;
+
 	s_steer = &mlx4_priv(dev)->steer[port - 1];
 
 	pqp = get_promisc_qp(dev, port, steer, qpn);
@@ -283,6 +294,9 @@ static bool check_duplicate_entry(struct mlx4_dev *dev, u8 port,
 	struct mlx4_steer_index *tmp_entry, *entry = NULL;
 	struct mlx4_promisc_qp *dqp, *tmp_dqp;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return NULL;
+
 	s_steer = &mlx4_priv(dev)->steer[port - 1];
 
 	/* if qp is not promisc, it cannot be duplicated */
@@ -324,6 +338,9 @@ static bool can_remove_steering_entry(struct mlx4_dev *dev, u8 port,
 	bool ret = false;
 	int i;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return NULL;
+
 	s_steer = &mlx4_priv(dev)->steer[port - 1];
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
@@ -378,6 +395,9 @@ static int add_promisc_qp(struct mlx4_dev *dev, u8 port,
 	int err;
 	struct mlx4_priv *priv = mlx4_priv(dev);
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return -EINVAL;
+
 	s_steer = &mlx4_priv(dev)->steer[port - 1];
 
 	mutex_lock(&priv->mcg_table.mutex);
@@ -484,6 +504,9 @@ static int remove_promisc_qp(struct mlx4_dev *dev, u8 port,
 	int loc, i;
 	int err;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return -EINVAL;
+
 	s_steer = &mlx4_priv(dev)->steer[port - 1];
 	mutex_lock(&priv->mcg_table.mutex);
 
@@ -910,6 +933,9 @@ int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	u8 port = gid[5];
 	u8 new_entry = 0;
 
+	if (port < 1 || port > dev->caps.num_ports)
+		return -EINVAL;
+
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox))
 		return PTR_ERR(mailbox);
-- 
1.8.3.4

Re: [PATCH net-next V1 10/11] net/mlx4_en: Fix Supported/Advertised link mode reported by ethtool

[Ben Hutchings]

On Tue, 2013-12-17 at 09:01 +0200, Amir Vadai wrote:
> From: Rana Shahout <ranas@mellanox.com>
> 
> Correctly report the support for 1/10/40Gbps link-speed as
> supported/advertised link mode in ethtool's get_settings command.
[...]
> +	cmd->supported = SUPPORTED_1000baseT_Full |
> +		SUPPORTED_10000baseT_Full |
> +		SUPPORTED_1000baseKX_Full |
> +		SUPPORTED_10000baseKX4_Full |
> +		SUPPORTED_10000baseKR_Full |
> +		SUPPORTED_10000baseR_FEC |
> +		SUPPORTED_40000baseKR4_Full |
> +		SUPPORTED_40000baseCR4_Full |
> +		SUPPORTED_40000baseSR4_Full |
> +		SUPPORTED_40000baseLR4_Full;
> +
> +	cmd->advertising = ADVERTISED_1000baseT_Full |
> +		ADVERTISED_10000baseT_Full |
> +		ADVERTISED_1000baseKX_Full |
> +		ADVERTISED_10000baseKX4_Full |
> +		ADVERTISED_10000baseKR_Full |
> +		ADVERTISED_10000baseR_FEC |
> +		ADVERTISED_40000baseKR4_Full |
> +		ADVERTISED_40000baseCR4_Full |
> +		ADVERTISED_40000baseSR4_Full |
> +		ADVERTISED_40000baseLR4_Full;
[...]

How is this correct?  You should report the actual capabilities of the
MAC and PHY are present, not the union of all possible modes of all the
supported boards.

(And the Kconfig description for MLX4_EN says '10Gbit Ethernet support',
so if this driver now supports some 40G boards that should be updated.)

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.

Re: [PATCH net-next V1 10/11] net/mlx4_en: Fix Supported/Advertised link mode reported by ethtool

[Amir Vadai]

On 17/12/2013 17:46, Ben Hutchings wrote:
>> +		ADVERTISED_40000baseLR4_Full;
> [...]
> 
> How is this correct?  You should report the actual capabilities of the
> MAC and PHY are present, not the union of all possible modes of all the
> supported boards.
I miss-understood this field.
Will fix it.

> 
> (And the Kconfig description for MLX4_EN says '10Gbit Ethernet support',
> so if this driver now supports some 40G boards that should be updated.)
Will fix it for V2

> 
> Ben


Thanks,
Amir

Re: [PATCH net-next V1 10/11] net/mlx4_en: Fix Supported/Advertised link mode reported by ethtool

[Amir Vadai]

On 17/12/2013 19:35, Amir Vadai wrote:
> On 17/12/2013 17:46, Ben Hutchings wrote:
>>> +		ADVERTISED_40000baseLR4_Full;
>> [...]
>>
>> How is this correct?  You should report the actual capabilities of the
>> MAC and PHY are present, not the union of all possible modes of all the
>> supported boards.
> I miss-understood this field.
> Will fix it.
> 
>>
>> (And the Kconfig description for MLX4_EN says '10Gbit Ethernet support',
>> so if this driver now supports some 40G boards that should be updated.)
> Will fix it for V2
> 
>>
>> Ben
> 
> 
> Thanks,
> Amir
> 

Hi Dave,

It seems that we need to write this patch from scratch, so I would like
to remove it from the patchset. So other patches we have in the pipe
won't be delayed.

Regarding the Kconfig description that Ben suggested, I will send it in
the next patchset.

Do you want me to send a V2 with the only difference is this patch
removed, or can you take this set?

Thanks,
Amir

Re: [PATCH net-next V1 10/11] net/mlx4_en: Fix Supported/Advertised link mode reported by ethtool

[David Miller]

From: Amir Vadai <amirv@mellanox.com>
Date: Thu, 19 Dec 2013 10:34:37 +0200

> Do you want me to send a V2 with the only difference is this patch
> removed, or can you take this set?

Please always send a complete, new, patch set.