[PATCH net-next 0/3] Mellanox Ethernet driver updates 2013-04-11

[PATCH net-next 0/3] Mellanox Ethernet driver updates 2013-04-11

[Amir Vadai]

Hi Dave,

This series contains updates to mlx4 driver.
First two patches enable SRIOV guests running old mlx4 drivers with B0 steering
only to run over hypervisor with new drivers that is using device managed flow
steering (DMFS).
Third patch is a fix to set corret MTU in SRIOV.

Thanks,
Amir


Eugenia Emantayev (1):
  net/mlx4_en: set correct MTU in SRIOV

Hadar Hen Zion (2):
  net/mlx4_core: Add helper function to translate B0 steering rules to
    DMFS
  net/mlx4_core: Translate guest B0 steering rules to DMFS

 drivers/net/ethernet/mellanox/mlx4/mcg.c           |   51 +++++++-----
 drivers/net/ethernet/mellanox/mlx4/mlx4.h          |    4 +
 drivers/net/ethernet/mellanox/mlx4/port.c          |    4 +-
 .../net/ethernet/mellanox/mlx4/resource_tracker.c  |   85 +++++++++++++++-----
 4 files changed, 99 insertions(+), 45 deletions(-)

-- 
1.7.8.2

[PATCH net-next 1/3] net/mlx4_core: Add helper function to translate B0 steering rules to DMFS

[Amir Vadai]

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

A pre-step for supporting guests that use B0 steering over a hypervisor
that runs in DMFS (device managed flow steering mode). Add helper function
which allows to translate L2 attachments / detachments provided in B0 mode
to DMFS rules.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/mcg.c  |   51 ++++++++++++++++------------
 drivers/net/ethernet/mellanox/mlx4/mlx4.h |    4 ++
 2 files changed, 33 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c
index 5268552..ffc78d2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mcg.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c
@@ -1125,28 +1125,11 @@ static int mlx4_QP_ATTACH(struct mlx4_dev *dev, struct mlx4_qp *qp,
 	return err;
 }
 
-int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
-			  u8 port, int block_mcast_loopback,
-			  enum mlx4_protocol prot, u64 *reg_id)
+int mlx4_trans_to_dmfs_attach(struct mlx4_dev *dev, struct mlx4_qp *qp,
+			      u8 gid[16], u8 port,
+			      int block_mcast_loopback,
+			      enum mlx4_protocol prot, u64 *reg_id)
 {
-
-	switch (dev->caps.steering_mode) {
-	case MLX4_STEERING_MODE_A0:
-		if (prot == MLX4_PROT_ETH)
-			return 0;
-
-	case MLX4_STEERING_MODE_B0:
-		if (prot == MLX4_PROT_ETH)
-			gid[7] |= (MLX4_MC_STEER << 1);
-
-		if (mlx4_is_mfunc(dev))
-			return mlx4_QP_ATTACH(dev, qp, gid, 1,
-					      block_mcast_loopback, prot);
-		return mlx4_qp_attach_common(dev, qp, gid,
-					     block_mcast_loopback, prot,
-					     MLX4_MC_STEER);
-
-	case MLX4_STEERING_MODE_DEVICE_MANAGED: {
 		struct mlx4_spec_list spec = { {NULL} };
 		__be64 mac_mask = cpu_to_be64(MLX4_MAC_MASK << 16);
 
@@ -1180,8 +1163,32 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 		list_add_tail(&spec.list, &rule.list);
 
 		return mlx4_flow_attach(dev, &rule, reg_id);
-	}
+}
 
+int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  u8 port, int block_mcast_loopback,
+			  enum mlx4_protocol prot, u64 *reg_id)
+{
+	switch (dev->caps.steering_mode) {
+	case MLX4_STEERING_MODE_A0:
+		if (prot == MLX4_PROT_ETH)
+			return 0;
+
+	case MLX4_STEERING_MODE_B0:
+		if (prot == MLX4_PROT_ETH)
+			gid[7] |= (MLX4_MC_STEER << 1);
+
+		if (mlx4_is_mfunc(dev))
+			return mlx4_QP_ATTACH(dev, qp, gid, 1,
+					      block_mcast_loopback, prot);
+		return mlx4_qp_attach_common(dev, qp, gid,
+					     block_mcast_loopback, prot,
+					     MLX4_MC_STEER);
+
+	case MLX4_STEERING_MODE_DEVICE_MANAGED:
+		return mlx4_trans_to_dmfs_attach(dev, qp, gid, port,
+						 block_mcast_loopback,
+						 prot, reg_id);
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index d738454..252f4ba 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -1190,6 +1190,10 @@ int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			  int block_mcast_loopback, enum mlx4_protocol prot,
 			  enum mlx4_steer_type steer);
+int mlx4_trans_to_dmfs_attach(struct mlx4_dev *dev, struct mlx4_qp *qp,
+			      u8 gid[16], u8 port,
+			      int block_mcast_loopback,
+			      enum mlx4_protocol prot, u64 *reg_id);
 int mlx4_SET_MCAST_FLTR_wrapper(struct mlx4_dev *dev, int slave,
 				struct mlx4_vhcr *vhcr,
 				struct mlx4_cmd_mailbox *inbox,
-- 
1.7.8.2

[PATCH net-next 2/3] net/mlx4_core: Translate guest B0 steering rules to DMFS

[Amir Vadai]

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

The different steering modes are global to the device, with DMFS
being introduced after SRIOV was merged. Hence, SRIOV guests running
legacy / older Linux kernels or non-Linux drivers may provide
B0 steering directives when the hypervisor is using DMFS and fail.

Under B0 only L2 steering rules are allowed, hence B0 is a subset of DMFS.
Use this fact to enable such legacy guests to run by modifying the SRIOV
B0 steering wrapper to translate guest B0 directives to DMFS ones when
the device uses DMFS. The translated B0 rule has to be kept in the
resource tracker as a B0 object to allow for lookup in case of detach.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
---
 .../net/ethernet/mellanox/mlx4/resource_tracker.c  |   85 +++++++++++++++-----
 1 files changed, 63 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index 1391b52..f2d6443 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -75,6 +75,7 @@ struct res_gid {
 	u8			gid[16];
 	enum mlx4_protocol	prot;
 	enum mlx4_steer_type	steer;
+	u64			reg_id;
 };
 
 enum res_qp_states {
@@ -2934,7 +2935,7 @@ static struct res_gid *find_gid(struct mlx4_dev *dev, int slave,
 
 static int add_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 		       u8 *gid, enum mlx4_protocol prot,
-		       enum mlx4_steer_type steer)
+		       enum mlx4_steer_type steer, u64 reg_id)
 {
 	struct res_gid *res;
 	int err;
@@ -2951,6 +2952,7 @@ static int add_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 		memcpy(res->gid, gid, 16);
 		res->prot = prot;
 		res->steer = steer;
+		res->reg_id = reg_id;
 		list_add_tail(&res->list, &rqp->mcg_list);
 		err = 0;
 	}
@@ -2961,7 +2963,7 @@ static int add_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 
 static int rem_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 		       u8 *gid, enum mlx4_protocol prot,
-		       enum mlx4_steer_type steer)
+		       enum mlx4_steer_type steer, u64 *reg_id)
 {
 	struct res_gid *res;
 	int err;
@@ -2971,6 +2973,7 @@ static int rem_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 	if (!res || res->prot != prot || res->steer != steer)
 		err = -EINVAL;
 	else {
+		*reg_id = res->reg_id;
 		list_del(&res->list);
 		kfree(res);
 		err = 0;
@@ -2980,6 +2983,37 @@ static int rem_mcg_res(struct mlx4_dev *dev, int slave, struct res_qp *rqp,
 	return err;
 }
 
+static int qp_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+		     int block_loopback, enum mlx4_protocol prot,
+		     enum mlx4_steer_type type, u64 *reg_id)
+{
+	switch (dev->caps.steering_mode) {
+	case MLX4_STEERING_MODE_DEVICE_MANAGED:
+		return mlx4_trans_to_dmfs_attach(dev, qp, gid, gid[5],
+						block_loopback, prot,
+						reg_id);
+	case MLX4_STEERING_MODE_B0:
+		return mlx4_qp_attach_common(dev, qp, gid,
+					    block_loopback, prot, type);
+	default:
+		return -EINVAL;
+	}
+}
+
+static int qp_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+		     enum mlx4_protocol prot, enum mlx4_steer_type type,
+		     u64 reg_id)
+{
+	switch (dev->caps.steering_mode) {
+	case MLX4_STEERING_MODE_DEVICE_MANAGED:
+		return mlx4_flow_detach(dev, reg_id);
+	case MLX4_STEERING_MODE_B0:
+		return mlx4_qp_detach_common(dev, qp, gid, prot, type);
+	default:
+		return -EINVAL;
+	}
+}
+
 int mlx4_QP_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 			       struct mlx4_vhcr *vhcr,
 			       struct mlx4_cmd_mailbox *inbox,
@@ -2992,14 +3026,12 @@ int mlx4_QP_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	int err;
 	int qpn;
 	struct res_qp *rqp;
+	u64 reg_id = 0;
 	int attach = vhcr->op_modifier;
 	int block_loopback = vhcr->in_modifier >> 31;
 	u8 steer_type_mask = 2;
 	enum mlx4_steer_type type = (gid[7] & steer_type_mask) >> 1;
 
-	if (dev->caps.steering_mode != MLX4_STEERING_MODE_B0)
-		return -EINVAL;
-
 	qpn = vhcr->in_modifier & 0xffffff;
 	err = get_res(dev, slave, qpn, RES_QP, &rqp);
 	if (err)
@@ -3007,30 +3039,32 @@ int mlx4_QP_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 
 	qp.qpn = qpn;
 	if (attach) {
-		err = add_mcg_res(dev, slave, rqp, gid, prot, type);
-		if (err)
+		err = qp_attach(dev, &qp, gid, block_loopback, prot,
+				type, &reg_id);
+		if (err) {
+			pr_err("Fail to attach rule to qp 0x%x\n", qpn);
 			goto ex_put;
-
-		err = mlx4_qp_attach_common(dev, &qp, gid,
-					    block_loopback, prot, type);
+		}
+		err = add_mcg_res(dev, slave, rqp, gid, prot, type, reg_id);
 		if (err)
-			goto ex_rem;
+			goto ex_detach;
 	} else {
-		err = rem_mcg_res(dev, slave, rqp, gid, prot, type);
+		err = rem_mcg_res(dev, slave, rqp, gid, prot, type, &reg_id);
 		if (err)
 			goto ex_put;
-		err = mlx4_qp_detach_common(dev, &qp, gid, prot, type);
-	}
 
+		err = qp_detach(dev, &qp, gid, prot, type, reg_id);
+		if (err)
+			pr_err("Fail to detach rule from qp 0x%x reg_id = 0x%llx\n",
+			       qpn, reg_id);
+	}
 	put_res(dev, slave, qpn, RES_QP);
-	return 0;
+	return err;
 
-ex_rem:
-	/* ignore error return below, already in error */
-	(void) rem_mcg_res(dev, slave, rqp, gid, prot, type);
+ex_detach:
+	qp_detach(dev, &qp, gid, prot, type, reg_id);
 ex_put:
 	put_res(dev, slave, qpn, RES_QP);
-
 	return err;
 }
 
@@ -3266,9 +3300,16 @@ static void detach_qp(struct mlx4_dev *dev, int slave, struct res_qp *rqp)
 	struct mlx4_qp qp; /* dummy for calling attach/detach */
 
 	list_for_each_entry_safe(rgid, tmp, &rqp->mcg_list, list) {
-		qp.qpn = rqp->local_qpn;
-		(void) mlx4_qp_detach_common(dev, &qp, rgid->gid, rgid->prot,
-					     rgid->steer);
+		switch (dev->caps.steering_mode) {
+		case MLX4_STEERING_MODE_DEVICE_MANAGED:
+			mlx4_flow_detach(dev, rgid->reg_id);
+			break;
+		case MLX4_STEERING_MODE_B0:
+			qp.qpn = rqp->local_qpn;
+			(void) mlx4_qp_detach_common(dev, &qp, rgid->gid,
+						     rgid->prot, rgid->steer);
+			break;
+		}
 		list_del(&rgid->list);
 		kfree(rgid);
 	}
-- 
1.7.8.2

[PATCH net-next 3/3] net/mlx4_en: set correct MTU in SRIOV

[Amir Vadai]

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

When setting MTU in SRIOV mode add ETH, VLAN and FCS header length
to the maximum MTU obtained from QUERY_DEV_CAP.

Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx4/port.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index 10c57c8..4b6aad3 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -32,6 +32,7 @@
 
 #include <linux/errno.h>
 #include <linux/if_ether.h>
+#include <linux/if_vlan.h>
 #include <linux/export.h>
 
 #include <linux/mlx4/cmd.h>
@@ -517,7 +518,8 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod,
 			/* Mtu is configured as the max MTU among all the
 			 * the functions on the port. */
 			mtu = be16_to_cpu(gen_context->mtu);
-			mtu = min_t(int, mtu, dev->caps.eth_mtu_cap[port]);
+			mtu = min_t(int, mtu, dev->caps.eth_mtu_cap[port] +
+				    ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN);
 			prev_mtu = slave_st->mtu[port];
 			slave_st->mtu[port] = mtu;
 			if (mtu > master->max_mtu[port])
-- 
1.7.8.2

Re: [PATCH net-next 0/3] Mellanox Ethernet driver updates 2013-04-11

[David Miller]

From: Amir Vadai <amirv@mellanox.com>
Date: Thu, 11 Apr 2013 14:56:39 +0300

> This series contains updates to mlx4 driver.
> First two patches enable SRIOV guests running old mlx4 drivers with B0 steering
> only to run over hypervisor with new drivers that is using device managed flow
> steering (DMFS).
> Third patch is a fix to set corret MTU in SRIOV.

All applied to net-next, thanks.