[PATCH net-next V1 00/10] Mellanox Ethernet driver updates 2013-01-30

[PATCH net-next V1 00/10] Mellanox Ethernet driver updates 2013-01-30

[Amir Vadai]

Hi Dave,

This series contains updates to mlx4 driver.
Majority of the patches are small bug fixes. Another patch is to expose through
debugfs some HW resources to be used by external tools.

Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"

Changes from V0:
- Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
  to do it, and no need to delay the other patches.

Thanks,
Amir

Amir Vadai (1):
  net/mlx4_en: Fix transmit timeout when driver restarts port

Hadar Hen Zion (8):
  net/mlx4_core: Directly expose fields of HW flow steering rule
    control segment
  net/mlx4_core: Set correctly allow_loopback flag
  net/mlx4_en: Fix ip/udp steering rules multicast mac when attached
    via ethtool
  net/mlx4_en: Validate VLAN IDs provided in ethtool flow steering
    rules
  net/mlx4_en: Fix vlan mask for ethtool steering rules
  net/mlx4_en: Block insertion of ethtool steering rules while the
    interface is down
  net/mlx4_en: Fix ethtool rules leftovers after module unloaded
  net/mlx4_core: Use firmware driven flow steering hash mode

Matan Barak (1):
  net/mlx4_en: Don't reassign port mac address on firmware that
    supports it

 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c    |  147 ++++++++++++++------
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c     |   32 ++++-
 drivers/net/ethernet/mellanox/mlx4/fw.c            |   11 +-
 drivers/net/ethernet/mellanox/mlx4/fw.h            |    1 -
 drivers/net/ethernet/mellanox/mlx4/main.c          |   16 --
 drivers/net/ethernet/mellanox/mlx4/mcg.c           |    4 +-
 drivers/net/ethernet/mellanox/mlx4/mlx4.h          |   12 +-
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h       |    5 +-
 .../net/ethernet/mellanox/mlx4/resource_tracker.c  |    2 +-
 include/linux/mlx4/device.h                        |    3 +-
 10 files changed, 152 insertions(+), 81 deletions(-)

-- 
1.7.8.2

[PATCH net-next V1 01/10] net/mlx4_core: Directly expose fields of HW flow steering rule control segment

[Amir Vadai]

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

Some of the fields for struct mlx4_net_trans_rule_hw_ctrl were packed into u32
and accessed through bit field operations. Expose and access them directly as
u8.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/mcg.c           |    2 +-
 drivers/net/ethernet/mellanox/mlx4/mlx4.h          |    7 +++++--
 .../net/ethernet/mellanox/mlx4/resource_tracker.c  |    2 +-
 3 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c
index 1ee4db3..d7c0704 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mcg.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c
@@ -664,7 +664,7 @@ static void trans_rule_ctrl_to_hw(struct mlx4_net_trans_rule *ctrl,
 	dw |= ctrl->priority << 16;
 
 	hw->ctrl = cpu_to_be32(dw);
-	hw->vf_vep_port = cpu_to_be32(ctrl->port);
+	hw->port = ctrl->port;
 	hw->qpn = cpu_to_be32(ctrl->qpn);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 116c5c2..3263848 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -696,9 +696,12 @@ struct mlx4_steer {
 
 struct mlx4_net_trans_rule_hw_ctrl {
 	__be32 ctrl;
-	__be32 vf_vep_port;
+	u8 rsvd1;
+	u8 funcid;
+	u8 vep;
+	u8 port;
 	__be32 qpn;
-	__be32 reserved;
+	__be32 rsvd2;
 };
 
 struct mlx4_net_trans_rule_hw_ib {
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index 561ed2a..5997adc 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -3018,7 +3018,7 @@ static int add_eth_header(struct mlx4_dev *dev, int slave,
 	__be64 mac_msk = cpu_to_be64(MLX4_MAC_MASK << 16);
 
 	ctrl = (struct mlx4_net_trans_rule_hw_ctrl *)inbox->buf;
-	port = be32_to_cpu(ctrl->vf_vep_port) & 0xff;
+	port = ctrl->port;
 	eth_header = (struct mlx4_net_trans_rule_hw_eth *)(ctrl + 1);
 
 	/* Clear a space in the inbox for eth header */
-- 
1.7.8.2

[PATCH net-next V1 02/10] net/mlx4_core: Set correctly allow_loopback flag

[Amir Vadai]

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

The allow_loopback flag was wrongly set using arithmetic bit operation, change
the code to use logical bit operation.

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

diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c
index d7c0704..5268552 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mcg.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c
@@ -1157,7 +1157,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			.priority = MLX4_DOMAIN_NIC,
 		};
 
-		rule.allow_loopback = ~block_mcast_loopback;
+		rule.allow_loopback = !block_mcast_loopback;
 		rule.port = port;
 		rule.qpn = qp->qpn;
 		INIT_LIST_HEAD(&rule.list);
-- 
1.7.8.2

[PATCH net-next V1 03/10] net/mlx4_en: Fix ip/udp steering rules multicast mac when attached via ethtool

[Amir Vadai]

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

Destination mac is a mandatory specification for ip/udp steering rules.
When attaching multicast steering rules via ethtool the unicast mac of the
interface was added to the rule specification instead of the multicast mac.
The following commit sets the corresponding multicast mac for the rule multicast ip.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |  119 ++++++++++++++++-------
 1 files changed, 82 insertions(+), 37 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 03447da..86afb5b 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -35,6 +35,8 @@
 #include <linux/ethtool.h>
 #include <linux/netdevice.h>
 #include <linux/mlx4/driver.h>
+#include <linux/in.h>
+#include <net/ip.h>
 
 #include "mlx4_en.h"
 #include "en_port.h"
@@ -672,19 +674,71 @@ static int mlx4_en_validate_flow(struct net_device *dev,
 	return 0;
 }
 
+static int mlx4_en_ethtool_add_mac_rule(struct ethtool_rxnfc *cmd,
+					struct list_head *rule_list_h,
+					struct mlx4_spec_list *spec_l2,
+					unsigned char *mac)
+{
+	int err = 0;
+	__be64 mac_msk = cpu_to_be64(MLX4_MAC_MASK << 16);
+
+	spec_l2->id = MLX4_NET_TRANS_RULE_ID_ETH;
+	memcpy(spec_l2->eth.dst_mac_msk, &mac_msk, ETH_ALEN);
+	memcpy(spec_l2->eth.dst_mac, mac, ETH_ALEN);
+
+	if ((cmd->fs.flow_type & FLOW_EXT) && cmd->fs.m_ext.vlan_tci) {
+		spec_l2->eth.vlan_id = cmd->fs.h_ext.vlan_tci;
+		spec_l2->eth.vlan_id_msk = cpu_to_be16(0xfff);
+	}
+
+	list_add_tail(&spec_l2->list, rule_list_h);
+
+	return err;
+}
+
+static int mlx4_en_ethtool_add_mac_rule_by_ipv4(struct mlx4_en_priv *priv,
+						struct ethtool_rxnfc *cmd,
+						struct list_head *rule_list_h,
+						struct mlx4_spec_list *spec_l2,
+						__be32 ipv4_dst)
+{
+	__be64 be_mac = 0;
+	unsigned char mac[ETH_ALEN];
+
+	if (!ipv4_is_multicast(ipv4_dst)) {
+		if (cmd->fs.flow_type & FLOW_MAC_EXT) {
+			memcpy(&mac, cmd->fs.h_ext.h_dest, ETH_ALEN);
+		} else {
+			be_mac = cpu_to_be64((priv->mac & MLX4_MAC_MASK) << 16);
+			memcpy(&mac, &be_mac, ETH_ALEN);
+		}
+	} else {
+		ip_eth_mc_map(ipv4_dst, mac);
+	}
+
+	return mlx4_en_ethtool_add_mac_rule(cmd, rule_list_h, spec_l2, &mac[0]);
+}
+
 static int add_ip_rule(struct mlx4_en_priv *priv,
-			struct ethtool_rxnfc *cmd,
-			struct list_head *list_h)
+		       struct ethtool_rxnfc *cmd,
+		       struct list_head *list_h)
 {
-	struct mlx4_spec_list *spec_l3;
+	struct mlx4_spec_list *spec_l2 = NULL;
+	struct mlx4_spec_list *spec_l3 = NULL;
 	struct ethtool_usrip4_spec *l3_mask = &cmd->fs.m_u.usr_ip4_spec;
 
-	spec_l3 = kzalloc(sizeof *spec_l3, GFP_KERNEL);
-	if (!spec_l3) {
+	spec_l3 = kzalloc(sizeof(*spec_l3), GFP_KERNEL);
+	spec_l2 = kzalloc(sizeof(*spec_l2), GFP_KERNEL);
+	if (!spec_l2 || !spec_l3) {
 		en_err(priv, "Fail to alloc ethtool rule.\n");
+		kfree(spec_l2);
+		kfree(spec_l3);
 		return -ENOMEM;
 	}
 
+	mlx4_en_ethtool_add_mac_rule_by_ipv4(priv, cmd, list_h, spec_l2,
+					     cmd->fs.h_u.
+					     usr_ip4_spec.ip4dst);
 	spec_l3->id = MLX4_NET_TRANS_RULE_ID_IPV4;
 	spec_l3->ipv4.src_ip = cmd->fs.h_u.usr_ip4_spec.ip4src;
 	if (l3_mask->ip4src)
@@ -701,14 +755,17 @@ static int add_tcp_udp_rule(struct mlx4_en_priv *priv,
 			     struct ethtool_rxnfc *cmd,
 			     struct list_head *list_h, int proto)
 {
-	struct mlx4_spec_list *spec_l3;
-	struct mlx4_spec_list *spec_l4;
+	struct mlx4_spec_list *spec_l2 = NULL;
+	struct mlx4_spec_list *spec_l3 = NULL;
+	struct mlx4_spec_list *spec_l4 = NULL;
 	struct ethtool_tcpip4_spec *l4_mask = &cmd->fs.m_u.tcp_ip4_spec;
 
-	spec_l3 = kzalloc(sizeof *spec_l3, GFP_KERNEL);
-	spec_l4 = kzalloc(sizeof *spec_l4, GFP_KERNEL);
-	if (!spec_l4 || !spec_l3) {
+	spec_l2 = kzalloc(sizeof(*spec_l2), GFP_KERNEL);
+	spec_l3 = kzalloc(sizeof(*spec_l3), GFP_KERNEL);
+	spec_l4 = kzalloc(sizeof(*spec_l4), GFP_KERNEL);
+	if (!spec_l2 || !spec_l3 || !spec_l4) {
 		en_err(priv, "Fail to alloc ethtool rule.\n");
+		kfree(spec_l2);
 		kfree(spec_l3);
 		kfree(spec_l4);
 		return -ENOMEM;
@@ -717,12 +774,20 @@ static int add_tcp_udp_rule(struct mlx4_en_priv *priv,
 	spec_l3->id = MLX4_NET_TRANS_RULE_ID_IPV4;
 
 	if (proto == TCP_V4_FLOW) {
+		mlx4_en_ethtool_add_mac_rule_by_ipv4(priv, cmd, list_h,
+						     spec_l2,
+						     cmd->fs.h_u.
+						     tcp_ip4_spec.ip4dst);
 		spec_l4->id = MLX4_NET_TRANS_RULE_ID_TCP;
 		spec_l3->ipv4.src_ip = cmd->fs.h_u.tcp_ip4_spec.ip4src;
 		spec_l3->ipv4.dst_ip = cmd->fs.h_u.tcp_ip4_spec.ip4dst;
 		spec_l4->tcp_udp.src_port = cmd->fs.h_u.tcp_ip4_spec.psrc;
 		spec_l4->tcp_udp.dst_port = cmd->fs.h_u.tcp_ip4_spec.pdst;
 	} else {
+		mlx4_en_ethtool_add_mac_rule_by_ipv4(priv, cmd, list_h,
+						     spec_l2,
+						     cmd->fs.h_u.
+						     udp_ip4_spec.ip4dst);
 		spec_l4->id = MLX4_NET_TRANS_RULE_ID_UDP;
 		spec_l3->ipv4.src_ip = cmd->fs.h_u.udp_ip4_spec.ip4src;
 		spec_l3->ipv4.dst_ip = cmd->fs.h_u.udp_ip4_spec.ip4dst;
@@ -751,43 +816,23 @@ static int mlx4_en_ethtool_to_net_trans_rule(struct net_device *dev,
 					     struct list_head *rule_list_h)
 {
 	int err;
-	__be64 be_mac;
 	struct ethhdr *eth_spec;
-	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_spec_list *spec_l2;
-	__be64 mac_msk = cpu_to_be64(MLX4_MAC_MASK << 16);
+	struct mlx4_en_priv *priv = netdev_priv(dev);
 
 	err = mlx4_en_validate_flow(dev, cmd);
 	if (err)
 		return err;
 
-	spec_l2 = kzalloc(sizeof *spec_l2, GFP_KERNEL);
-	if (!spec_l2)
-		return -ENOMEM;
-
-	if (cmd->fs.flow_type & FLOW_MAC_EXT) {
-		memcpy(&be_mac, cmd->fs.h_ext.h_dest, ETH_ALEN);
-	} else {
-		u64 mac = priv->mac & MLX4_MAC_MASK;
-		be_mac = cpu_to_be64(mac << 16);
-	}
-
-	spec_l2->id = MLX4_NET_TRANS_RULE_ID_ETH;
-	memcpy(spec_l2->eth.dst_mac_msk, &mac_msk, ETH_ALEN);
-	if ((cmd->fs.flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) != ETHER_FLOW)
-		memcpy(spec_l2->eth.dst_mac, &be_mac, ETH_ALEN);
-
-	if ((cmd->fs.flow_type & FLOW_EXT) && cmd->fs.m_ext.vlan_tci) {
-		spec_l2->eth.vlan_id = cmd->fs.h_ext.vlan_tci;
-		spec_l2->eth.vlan_id_msk = cpu_to_be16(0xfff);
-	}
-
-	list_add_tail(&spec_l2->list, rule_list_h);
-
 	switch (cmd->fs.flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) {
 	case ETHER_FLOW:
+		spec_l2 = kzalloc(sizeof(*spec_l2), GFP_KERNEL);
+		if (!spec_l2)
+			return -ENOMEM;
+
 		eth_spec = &cmd->fs.h_u.ether_spec;
-		memcpy(&spec_l2->eth.dst_mac, eth_spec->h_dest, ETH_ALEN);
+		mlx4_en_ethtool_add_mac_rule(cmd, rule_list_h, spec_l2,
+					     &eth_spec->h_dest[0]);
 		spec_l2->eth.ether_type = eth_spec->h_proto;
 		if (eth_spec->h_proto)
 			spec_l2->eth.ether_type_enable = 1;
-- 
1.7.8.2

[PATCH net-next V1 04/10] net/mlx4_en: Validate VLAN IDs provided in ethtool flow steering rules

[Amir Vadai]

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

When attaching flow steering rules via Ethtool accept only valid vlans IDs e.g
in the range: [0,4095].

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 86afb5b..f33049f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -669,6 +669,10 @@ static int mlx4_en_validate_flow(struct net_device *dev,
 		    !(cmd->fs.m_ext.vlan_tci == 0 ||
 		      cmd->fs.m_ext.vlan_tci == cpu_to_be16(0xfff)))
 			return -EINVAL;
+		if (cmd->fs.m_ext.vlan_tci) {
+			if (be16_to_cpu(cmd->fs.h_ext.vlan_tci) >= VLAN_N_VID)
+				return -EINVAL;
+		}
 	}
 
 	return 0;
-- 
1.7.8.2

[PATCH net-next V1 05/10] net/mlx4_en: Fix vlan mask for ethtool steering rules

[Amir Vadai]

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

The vlan mask field should be validated and assigned according to the field
size which is 12 bits. Also replace the numeric 0xfff mask with existing kernel
macro.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |   13 +++++++++----
 1 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index f33049f..f36c219 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -666,12 +666,16 @@ static int mlx4_en_validate_flow(struct net_device *dev,
 
 	if ((cmd->fs.flow_type & FLOW_EXT)) {
 		if (cmd->fs.m_ext.vlan_etype ||
-		    !(cmd->fs.m_ext.vlan_tci == 0 ||
-		      cmd->fs.m_ext.vlan_tci == cpu_to_be16(0xfff)))
+		    !((cmd->fs.m_ext.vlan_tci & cpu_to_be16(VLAN_VID_MASK)) ==
+		      0 ||
+		      (cmd->fs.m_ext.vlan_tci & cpu_to_be16(VLAN_VID_MASK)) ==
+		      cpu_to_be16(VLAN_VID_MASK)))
 			return -EINVAL;
+
 		if (cmd->fs.m_ext.vlan_tci) {
 			if (be16_to_cpu(cmd->fs.h_ext.vlan_tci) >= VLAN_N_VID)
 				return -EINVAL;
+
 		}
 	}
 
@@ -690,9 +694,10 @@ static int mlx4_en_ethtool_add_mac_rule(struct ethtool_rxnfc *cmd,
 	memcpy(spec_l2->eth.dst_mac_msk, &mac_msk, ETH_ALEN);
 	memcpy(spec_l2->eth.dst_mac, mac, ETH_ALEN);
 
-	if ((cmd->fs.flow_type & FLOW_EXT) && cmd->fs.m_ext.vlan_tci) {
+	if ((cmd->fs.flow_type & FLOW_EXT) &&
+	    (cmd->fs.m_ext.vlan_tci & cpu_to_be16(VLAN_VID_MASK))) {
 		spec_l2->eth.vlan_id = cmd->fs.h_ext.vlan_tci;
-		spec_l2->eth.vlan_id_msk = cpu_to_be16(0xfff);
+		spec_l2->eth.vlan_id_msk = cpu_to_be16(VLAN_VID_MASK);
 	}
 
 	list_add_tail(&spec_l2->list, rule_list_h);
-- 
1.7.8.2

[PATCH net-next V1 06/10] net/mlx4_en: Block insertion of ethtool steering rules while the interface is down

[Amir Vadai]

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

Attaching steering rules while the interface is down is an invalid operation, block it.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index f36c219..6f8044d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -1006,7 +1006,8 @@ static int mlx4_en_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	if ((cmd->cmd == ETHTOOL_GRXCLSRLCNT ||
 	     cmd->cmd == ETHTOOL_GRXCLSRULE ||
 	     cmd->cmd == ETHTOOL_GRXCLSRLALL) &&
-	    mdev->dev->caps.steering_mode != MLX4_STEERING_MODE_DEVICE_MANAGED)
+	    (mdev->dev->caps.steering_mode !=
+	     MLX4_STEERING_MODE_DEVICE_MANAGED || !priv->port_up))
 		return -EINVAL;
 
 	switch (cmd->cmd) {
@@ -1042,7 +1043,8 @@ static int mlx4_en_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
 
-	if (mdev->dev->caps.steering_mode != MLX4_STEERING_MODE_DEVICE_MANAGED)
+	if (mdev->dev->caps.steering_mode !=
+	    MLX4_STEERING_MODE_DEVICE_MANAGED || !priv->port_up)
 		return -EINVAL;
 
 	switch (cmd->cmd) {
-- 
1.7.8.2

[PATCH net-next V1 07/10] net/mlx4_en: Fix ethtool rules leftovers after module unloaded

[Amir Vadai]

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

As part of the driver unload flow, all steering rules must be deleted,
make sure to remove the rules that were set through ethtool.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |    3 +++
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c  |   15 +++++++++++++++
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h    |    3 +++
 3 files changed, 21 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 6f8044d..738e95d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -915,6 +915,7 @@ static int mlx4_en_flow_replace(struct net_device *dev,
 		loc_rule->id = 0;
 		memset(&loc_rule->flow_spec, 0,
 		       sizeof(struct ethtool_rx_flow_spec));
+		list_del(&loc_rule->list);
 	}
 	err = mlx4_flow_attach(priv->mdev->dev, &rule, &reg_id);
 	if (err) {
@@ -925,6 +926,7 @@ static int mlx4_en_flow_replace(struct net_device *dev,
 	loc_rule->id = reg_id;
 	memcpy(&loc_rule->flow_spec, &cmd->fs,
 	       sizeof(struct ethtool_rx_flow_spec));
+	list_add_tail(&loc_rule->list, &priv->ethtool_list);
 
 out_free_list:
 	list_for_each_entry_safe(spec, tmp_spec, &rule.list, list) {
@@ -958,6 +960,7 @@ static int mlx4_en_flow_detach(struct net_device *dev,
 	}
 	rule->id = 0;
 	memset(&rule->flow_spec, 0, sizeof(struct ethtool_rx_flow_spec));
+	list_del(&rule->list);
 out:
 	return err;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 9c42812..333a7a0 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1039,6 +1039,9 @@ int mlx4_en_start_port(struct net_device *dev)
 
 	INIT_LIST_HEAD(&priv->mc_list);
 	INIT_LIST_HEAD(&priv->curr_list);
+	INIT_LIST_HEAD(&priv->ethtool_list);
+	memset(&priv->ethtool_rules[0], 0,
+	       sizeof(struct ethtool_flow_id) * MAX_NUM_OF_FS_RULES);
 
 	/* Calculate Rx buf size */
 	dev->mtu = min(dev->mtu, priv->max_mtu);
@@ -1202,6 +1205,7 @@ void mlx4_en_stop_port(struct net_device *dev)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
 	struct mlx4_en_mc_list *mclist, *tmp;
+	struct ethtool_flow_id *flow, *tmp_flow;
 	int i;
 	u8 mc_list[16] = {0};
 
@@ -1283,6 +1287,17 @@ void mlx4_en_stop_port(struct net_device *dev)
 	mlx4_put_eth_qp(mdev->dev, priv->port, priv->mac, priv->base_qpn);
 	mdev->mac_removed[priv->port] = 1;
 
+	/* Remove flow steering rules for the port*/
+	if (mdev->dev->caps.steering_mode ==
+	    MLX4_STEERING_MODE_DEVICE_MANAGED) {
+		ASSERT_RTNL();
+		list_for_each_entry_safe(flow, tmp_flow,
+					 &priv->ethtool_list, list) {
+			mlx4_flow_detach(mdev->dev, flow->id);
+			list_del(&flow->list);
+		}
+	}
+
 	/* Free RX Rings */
 	for (i = 0; i < priv->rx_ring_num; i++) {
 		mlx4_en_deactivate_rx_ring(priv, &priv->rx_ring[i]);
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 8d54412..4fb4a3e 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -427,6 +427,7 @@ struct mlx4_en_frag_info {
 #endif
 
 struct ethtool_flow_id {
+	struct list_head list;
 	struct ethtool_rx_flow_spec flow_spec;
 	u64 id;
 };
@@ -441,6 +442,8 @@ struct mlx4_en_priv {
 	struct mlx4_en_port_state port_state;
 	spinlock_t stats_lock;
 	struct ethtool_flow_id ethtool_rules[MAX_NUM_OF_FS_RULES];
+	/* To allow rules removal while port is going down */
+	struct list_head ethtool_list;
 
 	unsigned long last_moder_packets[MAX_RX_RINGS];
 	unsigned long last_moder_tx_packets;
-- 
1.7.8.2

[PATCH net-next V1 08/10] net/mlx4_core: Use firmware driven flow steering hash mode

[Amir Vadai]

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

The Firmware dynamically changes flow steering hash configuration from covering
L2 only to "full" L2/L3/L4 mode needed.  The dynamic change allows the driver
to set hard coded hash configuration which is changed by the firmware from L2
to L2/L3/L4 when attaching the first L3/L4 flow steering rule and back to L2
when there are no more such rules.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/fw.c   |    4 ++--
 drivers/net/ethernet/mellanox/mlx4/fw.h   |    1 -
 drivers/net/ethernet/mellanox/mlx4/main.c |   16 ----------------
 drivers/net/ethernet/mellanox/mlx4/mlx4.h |    5 -----
 4 files changed, 2 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 8b3d051..91acf71 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -1287,14 +1287,14 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 		/* Enable Ethernet flow steering
 		 * with udp unicast and tcp unicast
 		 */
-		MLX4_PUT(inbox, param->fs_hash_enable_bits,
+		MLX4_PUT(inbox, (u8) (MLX4_FS_UDP_UC_EN | MLX4_FS_TCP_UC_EN),
 			 INIT_HCA_FS_ETH_BITS_OFFSET);
 		MLX4_PUT(inbox, (u16) MLX4_FS_NUM_OF_L2_ADDR,
 			 INIT_HCA_FS_ETH_NUM_ADDRS_OFFSET);
 		/* Enable IPoIB flow steering
 		 * with udp unicast and tcp unicast
 		 */
-		MLX4_PUT(inbox, param->fs_hash_enable_bits,
+		MLX4_PUT(inbox, (u8) (MLX4_FS_UDP_UC_EN | MLX4_FS_TCP_UC_EN),
 			 INIT_HCA_FS_IB_BITS_OFFSET);
 		MLX4_PUT(inbox, (u16) MLX4_FS_NUM_OF_L2_ADDR,
 			 INIT_HCA_FS_IB_NUM_ADDRS_OFFSET);
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h
index dbf2f69..3af33ff 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.h
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.h
@@ -171,7 +171,6 @@ struct mlx4_init_hca_param {
 	u8  log_mpt_sz;
 	u8  log_uar_sz;
 	u8  uar_page_sz; /* log pg sz in 4k chunks */
-	u8  fs_hash_enable_bits;
 	u8  steering_mode; /* for QUERY_HCA */
 	u64 dev_cap_enabled;
 };
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index f1ee52d..e38c6b2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -1415,22 +1415,6 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 		if (mlx4_is_master(dev))
 			mlx4_parav_master_pf_caps(dev);
 
-		priv->fs_hash_mode = MLX4_FS_L2_HASH;
-
-		switch (priv->fs_hash_mode) {
-		case MLX4_FS_L2_HASH:
-			init_hca.fs_hash_enable_bits = 0;
-			break;
-
-		case MLX4_FS_L2_L3_L4_HASH:
-			/* Enable flow steering with
-			 * udp unicast and tcp unicast
-			 */
-			init_hca.fs_hash_enable_bits =
-				MLX4_FS_UDP_UC_EN | MLX4_FS_TCP_UC_EN;
-			break;
-		}
-
 		profile = default_profile;
 		if (dev->caps.steering_mode ==
 		    MLX4_STEERING_MODE_DEVICE_MANAGED)
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 3263848..172daaa 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -60,11 +60,6 @@
 #define MLX4_FS_MGM_LOG_ENTRY_SIZE	7
 #define MLX4_FS_NUM_MCG			(1 << 17)
 
-enum {
-	MLX4_FS_L2_HASH = 0,
-	MLX4_FS_L2_L3_L4_HASH,
-};
-
 #define MLX4_NUM_UP		8
 #define MLX4_NUM_TC		8
 #define MLX4_RATELIMIT_UNITS 3 /* 100 Mbps */
-- 
1.7.8.2

[PATCH net-next V1 09/10] net/mlx4_en: Don't reassign port mac address on firmware that supports it

[Amir Vadai]

From: Matan Barak <matanb@mellanox.com>

Mac reassignments should only be done when not supported by the firmware. To
accomplish that, checking firmware capability bit to know whether we should
reassign macs in the driver.

Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c |    3 ++-
 drivers/net/ethernet/mellanox/mlx4/fw.c        |    7 ++++++-
 include/linux/mlx4/device.h                    |    3 ++-
 3 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 333a7a0..7b513e9 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1285,7 +1285,8 @@ void mlx4_en_stop_port(struct net_device *dev)
 
 	/* Unregister Mac address for the port */
 	mlx4_put_eth_qp(mdev->dev, priv->port, priv->mac, priv->base_qpn);
-	mdev->mac_removed[priv->port] = 1;
+	if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAGS2_REASSIGN_MAC_EN))
+		mdev->mac_removed[priv->port] = 1;
 
 	/* Remove flow steering rules for the port*/
 	if (mdev->dev->caps.steering_mode ==
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index 91acf71..38b62c7 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -127,7 +127,8 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags)
 		[0] = "RSS support",
 		[1] = "RSS Toeplitz Hash Function support",
 		[2] = "RSS XOR Hash Function support",
-		[3] = "Device manage flow steering support"
+		[3] = "Device manage flow steering support",
+		[4] = "Automatic mac reassignment support"
 	};
 	int i;
 
@@ -478,6 +479,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_DEV_CAP_BMME_FLAGS_OFFSET		0x94
 #define QUERY_DEV_CAP_RSVD_LKEY_OFFSET		0x98
 #define QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET		0xa0
+#define QUERY_DEV_CAP_FW_REASSIGN_MAC		0x9d
 
 	dev_cap->flags2 = 0;
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
@@ -637,6 +639,9 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		 QUERY_DEV_CAP_BMME_FLAGS_OFFSET);
 	MLX4_GET(dev_cap->reserved_lkey, outbox,
 		 QUERY_DEV_CAP_RSVD_LKEY_OFFSET);
+	MLX4_GET(field, outbox, QUERY_DEV_CAP_FW_REASSIGN_MAC);
+	if (field & 1<<6)
+		dev_cap->flags2 |= MLX4_DEV_CAP_FLAGS2_REASSIGN_MAC_EN;
 	MLX4_GET(dev_cap->max_icm_sz, outbox,
 		 QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET);
 	if (dev_cap->flags & MLX4_DEV_CAP_FLAG_COUNTERS)
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 20ea939..1883e8e 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -150,7 +150,8 @@ enum {
 	MLX4_DEV_CAP_FLAG2_RSS			= 1LL <<  0,
 	MLX4_DEV_CAP_FLAG2_RSS_TOP		= 1LL <<  1,
 	MLX4_DEV_CAP_FLAG2_RSS_XOR		= 1LL <<  2,
-	MLX4_DEV_CAP_FLAG2_FS_EN		= 1LL <<  3
+	MLX4_DEV_CAP_FLAG2_FS_EN		= 1LL <<  3,
+	MLX4_DEV_CAP_FLAGS2_REASSIGN_MAC_EN	= 1LL <<  4
 };
 
 enum {
-- 
1.7.8.2

[PATCH net-next V1 10/10] net/mlx4_en: Fix transmit timeout when driver restarts port

[Amir Vadai]

Under heavy CPU load, changing, ring size/mtu/etc. could result in transmit
timeout, since stop-start port might take more than 10 seconds.
Calling netif_detach_device to prevent tx queue transmit timeout.

netif_detach_device() is not called under ndo_stop, because netif_carrier_off
will prevent the timeout, and device should not be marked as not present, or
else user won't be able to start it later on.

CC: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |    6 +++---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c  |   14 ++++++++++----
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h    |    2 +-
 3 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 738e95d..911d488 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -496,7 +496,7 @@ static int mlx4_en_set_ringparam(struct net_device *dev,
 	mutex_lock(&mdev->state_lock);
 	if (priv->port_up) {
 		port_up = 1;
-		mlx4_en_stop_port(dev);
+		mlx4_en_stop_port(dev, 1);
 	}
 
 	mlx4_en_free_resources(priv);
@@ -591,7 +591,7 @@ static int mlx4_en_set_rxfh_indir(struct net_device *dev,
 	mutex_lock(&mdev->state_lock);
 	if (priv->port_up) {
 		port_up = 1;
-		mlx4_en_stop_port(dev);
+		mlx4_en_stop_port(dev, 1);
 	}
 
 	priv->prof->rss_rings = rss_rings;
@@ -1096,7 +1096,7 @@ static int mlx4_en_set_channels(struct net_device *dev,
 	mutex_lock(&mdev->state_lock);
 	if (priv->port_up) {
 		port_up = 1;
-		mlx4_en_stop_port(dev);
+		mlx4_en_stop_port(dev, 1);
 	}
 
 	mlx4_en_free_resources(priv);
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 7b513e9..ac1c14f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1178,6 +1178,8 @@ int mlx4_en_start_port(struct net_device *dev)
 
 	priv->port_up = true;
 	netif_tx_start_all_queues(dev);
+	netif_device_attach(dev);
+
 	return 0;
 
 tx_err:
@@ -1200,7 +1202,7 @@ cq_err:
 }
 
 
-void mlx4_en_stop_port(struct net_device *dev)
+void mlx4_en_stop_port(struct net_device *dev, int detach)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
@@ -1216,9 +1218,13 @@ void mlx4_en_stop_port(struct net_device *dev)
 
 	/* Synchronize with tx routine */
 	netif_tx_lock_bh(dev);
+	if (detach)
+		netif_device_detach(dev);
 	netif_tx_stop_all_queues(dev);
 	netif_tx_unlock_bh(dev);
 
+	netif_tx_disable(dev);
+
 	/* Set port as not active */
 	priv->port_up = false;
 
@@ -1323,7 +1329,7 @@ static void mlx4_en_restart(struct work_struct *work)
 
 	mutex_lock(&mdev->state_lock);
 	if (priv->port_up) {
-		mlx4_en_stop_port(dev);
+		mlx4_en_stop_port(dev, 1);
 		for (i = 0; i < priv->tx_ring_num; i++)
 			netdev_tx_reset_queue(priv->tx_ring[i].tx_queue);
 		if (mlx4_en_start_port(dev))
@@ -1395,7 +1401,7 @@ static int mlx4_en_close(struct net_device *dev)
 
 	mutex_lock(&mdev->state_lock);
 
-	mlx4_en_stop_port(dev);
+	mlx4_en_stop_port(dev, 0);
 	netif_carrier_off(dev);
 
 	mutex_unlock(&mdev->state_lock);
@@ -1533,7 +1539,7 @@ static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu)
 			 * the port */
 			en_dbg(DRV, priv, "Change MTU called with card down!?\n");
 		} else {
-			mlx4_en_stop_port(dev);
+			mlx4_en_stop_port(dev, 1);
 			err = mlx4_en_start_port(dev);
 			if (err) {
 				en_err(priv, "Failed restarting port:%d\n",
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 4fb4a3e..43f0165 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -539,7 +539,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 			struct mlx4_en_port_profile *prof);
 
 int mlx4_en_start_port(struct net_device *dev);
-void mlx4_en_stop_port(struct net_device *dev);
+void mlx4_en_stop_port(struct net_device *dev, int detach);
 
 void mlx4_en_free_resources(struct mlx4_en_priv *priv);
 int mlx4_en_alloc_resources(struct mlx4_en_priv *priv);
-- 
1.7.8.2

Re: [PATCH net-next V1 00/10] Mellanox Ethernet driver updates 2013-01-30

[David Miller]

From: Amir Vadai <amirv@mellanox.com>
Date: Thu, 31 Jan 2013 11:07:01 +0200

> Hi Dave,
> 
> This series contains updates to mlx4 driver.
> Majority of the patches are small bug fixes. Another patch is to expose through
> debugfs some HW resources to be used by external tools.
> 
> Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"
> 
> Changes from V0:
> - Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
>   to do it, and no need to delay the other patches.

All applied, thanks.

Re: [PATCH net-next V1 00/10] Mellanox Ethernet driver updates 2013-01-30

[Ben Hutchings]

On Thu, 2013-01-31 at 13:51 -0500, David Miller wrote:
> From: Amir Vadai <amirv@mellanox.com>
> Date: Thu, 31 Jan 2013 11:07:01 +0200
> 
> > Hi Dave,
> > 
> > This series contains updates to mlx4 driver.
> > Majority of the patches are small bug fixes. Another patch is to expose through
> > debugfs some HW resources to be used by external tools.
> > 
> > Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"
> > 
> > Changes from V0:
> > - Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
> >   to do it, and no need to delay the other patches.
> 
> All applied, thanks.

The first time this series was posted (this is actually v2) I raised an
issue with the changes to interpretation of the VLAN tag in ethtool RX
NFC commands: <http://article.gmane.org/gmane.linux.network/257392/>.
The semantics are now definitely inconsistent between sfc and mlx4_en.
I asked Alexander what ixgbe (first implementation) does with the upper
4 bits, with the aim that all three implementations should be
consistent, but I haven't had an answer to that yet.

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 00/10] Mellanox Ethernet driver updates 2013-01-30

[David Miller]

From: Ben Hutchings <bhutchings@solarflare.com>
Date: Tue, 5 Feb 2013 20:08:19 +0000

> On Thu, 2013-01-31 at 13:51 -0500, David Miller wrote:
>> From: Amir Vadai <amirv@mellanox.com>
>> Date: Thu, 31 Jan 2013 11:07:01 +0200
>> 
>> > Hi Dave,
>> > 
>> > This series contains updates to mlx4 driver.
>> > Majority of the patches are small bug fixes. Another patch is to expose through
>> > debugfs some HW resources to be used by external tools.
>> > 
>> > Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"
>> > 
>> > Changes from V0:
>> > - Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
>> >   to do it, and no need to delay the other patches.
>> 
>> All applied, thanks.
> 
> The first time this series was posted (this is actually v2) I raised an
> issue with the changes to interpretation of the VLAN tag in ethtool RX
> NFC commands: <http://article.gmane.org/gmane.linux.network/257392/>.
> The semantics are now definitely inconsistent between sfc and mlx4_en.
> I asked Alexander what ixgbe (first implementation) does with the upper
> 4 bits, with the aim that all three implementations should be
> consistent, but I haven't had an answer to that yet.

I was hoping you guys would resolve this and post follow-on fixes
if necessary.

Re: [PATCH net-next V1 00/10] Mellanox Ethernet driver updates 2013-01-30

[Amir Vadai]

On 05/02/2013 22:08, Ben Hutchings wrote:
> On Thu, 2013-01-31 at 13:51 -0500, David Miller wrote:
>> From: Amir Vadai <amirv@mellanox.com>
>> Date: Thu, 31 Jan 2013 11:07:01 +0200
>>
>>> Hi Dave,
>>>
>>> This series contains updates to mlx4 driver.
>>> Majority of the patches are small bug fixes. Another patch is to expose through
>>> debugfs some HW resources to be used by external tools.
>>>
>>> Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"
>>>
>>> Changes from V0:
>>> - Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
>>>    to do it, and no need to delay the other patches.
>>
>> All applied, thanks.
>
> The first time this series was posted (this is actually v2) I raised an
> issue with the changes to interpretation of the VLAN tag in ethtool RX
> NFC commands: <http://article.gmane.org/gmane.linux.network/257392/>.
> The semantics are now definitely inconsistent between sfc and mlx4_en.
> I asked Alexander what ixgbe (first implementation) does with the upper
> 4 bits, with the aim that all three implementations should be
> consistent, but I haven't had an answer to that yet.
>
> Ben.
>

I'm not sure I understand the problem here.

mlx4_en implementation is exactly like what you describe and what 
ethtool documentation is describing.

We handle vlan_tci as the whole VLAN tag. Since we're only interested in 
the VID in our flow steering, we check that the mask sets only the 12 
bits of VID and use it.
 From looking at sfc driver - it is exactly same there.

Did I miss something?

Amir.

Re: [PATCH net-next V1 00/10] Mellanox Ethernet driver updates 2013-01-30

[Ben Hutchings]

On Wed, 2013-02-06 at 09:24 +0200, Amir Vadai wrote:
> On 05/02/2013 22:08, Ben Hutchings wrote:
> > On Thu, 2013-01-31 at 13:51 -0500, David Miller wrote:
> >> From: Amir Vadai <amirv@mellanox.com>
> >> Date: Thu, 31 Jan 2013 11:07:01 +0200
> >>
> >>> Hi Dave,
> >>>
> >>> This series contains updates to mlx4 driver.
> >>> Majority of the patches are small bug fixes. Another patch is to expose through
> >>> debugfs some HW resources to be used by external tools.
> >>>
> >>> Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"
> >>>
> >>> Changes from V0:
> >>> - Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
> >>>    to do it, and no need to delay the other patches.
> >>
> >> All applied, thanks.
> >
> > The first time this series was posted (this is actually v2) I raised an
> > issue with the changes to interpretation of the VLAN tag in ethtool RX
> > NFC commands: <http://article.gmane.org/gmane.linux.network/257392/>.
> > The semantics are now definitely inconsistent between sfc and mlx4_en.
> > I asked Alexander what ixgbe (first implementation) does with the upper
> > 4 bits, with the aim that all three implementations should be
> > consistent, but I haven't had an answer to that yet.
> >
> > Ben.
> >
> 
> I'm not sure I understand the problem here.
> 
> mlx4_en implementation is exactly like what you describe and what 
> ethtool documentation is describing.
> 
> We handle vlan_tci as the whole VLAN tag. Since we're only interested in 
> the VID in our flow steering, we check that the mask sets only the 12 
> bits of VID and use it.
>
>  From looking at sfc driver - it is exactly same there.
>
> Did I miss something?

sfc verifies that the mask (cmd->fs.m_ext.vlan_tci) is either 0 or 0xfff
(big-endian).  If any of the top 4 bits are set we reject the flow spec
as unsupported.  If I read correctly, mlx4_en was also doing that
initially, but 'net/mlx4_en: Fix vlan mask for ethtool steering rules'
changed it to ignore the top 4 bits of the mask.

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 00/10] Mellanox Ethernet driver updates 2013-01-30

[Amir Vadai]

On 06/02/2013 17:46, Ben Hutchings wrote:
> On Wed, 2013-02-06 at 09:24 +0200, Amir Vadai wrote:
>> On 05/02/2013 22:08, Ben Hutchings wrote:
>>> On Thu, 2013-01-31 at 13:51 -0500, David Miller wrote:
>>>> From: Amir Vadai <amirv@mellanox.com>
>>>> Date: Thu, 31 Jan 2013 11:07:01 +0200
>>>>
>>>>> Hi Dave,
>>>>>
>>>>> This series contains updates to mlx4 driver.
>>>>> Majority of the patches are small bug fixes. Another patch is to expose through
>>>>> debugfs some HW resources to be used by external tools.
>>>>>
>>>>> Patches done against net-next commit 1b13c97: "net/vxlan: Add ethtool drvinfo"
>>>>>
>>>>> Changes from V0:
>>>>> - Removed patch "net/mlx4_en: Add debugfs support". Need to find a standard way
>>>>>    to do it, and no need to delay the other patches.
>>>>
>>>> All applied, thanks.
>>>
>>> The first time this series was posted (this is actually v2) I raised an
>>> issue with the changes to interpretation of the VLAN tag in ethtool RX
>>> NFC commands: <http://article.gmane.org/gmane.linux.network/257392/>.
>>> The semantics are now definitely inconsistent between sfc and mlx4_en.
>>> I asked Alexander what ixgbe (first implementation) does with the upper
>>> 4 bits, with the aim that all three implementations should be
>>> consistent, but I haven't had an answer to that yet.
>>>
>>> Ben.
>>>
>>
>> I'm not sure I understand the problem here.
>>
>> mlx4_en implementation is exactly like what you describe and what 
>> ethtool documentation is describing.
>>
>> We handle vlan_tci as the whole VLAN tag. Since we're only interested in 
>> the VID in our flow steering, we check that the mask sets only the 12 
>> bits of VID and use it.
>>
>>  From looking at sfc driver - it is exactly same there.
>>
>> Did I miss something?
> 
> sfc verifies that the mask (cmd->fs.m_ext.vlan_tci) is either 0 or 0xfff
> (big-endian).  If any of the top 4 bits are set we reject the flow spec
> as unsupported.  If I read correctly, mlx4_en was also doing that
> initially, but 'net/mlx4_en: Fix vlan mask for ethtool steering rules'
> changed it to ignore the top 4 bits of the mask.
> 
> Ben.
> 

Ok - now I see.

This patch should be reverted - mlx4_en should also reject setting of
the top 4 bits in the mask.

I will send a fix soon.

Thanks,
Amir