[pull request][net-next V2 00/15] Mellanox, mlx5 updates 2019-04-02

[pull request][net-next V2 00/15] Mellanox, mlx5 updates 2019-04-02

[Saeed Mahameed]

Hi Dave,

This series adds misc updates to mlx5 driver.

For more information please see tag log below.

Please pull and let me know if there is any problem.

v1->v2:
  remove inline keywords from static functions in c files.

Please note that the series starts with a merge of mlx5-next branch,
to resolve and avoid dependency with rdma tree.

Note: when merged with net there is going to be a small merge conflict
in en_tc.c:
++<<<<<<< HEAD
 +              err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_KERNEL,
 +                                          parse_attr, hdrs, &action, extack);
++=======
+               err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_FDB,
+                                           parse_attr, hdrs, extack);
++>>>>>>> dave/master

Resolution:
+               err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_FDB,
+                                           parse_attr, hdrs, &action, extack);

Thanks,
Saeed.

---

The following changes since commit b6460c72c36df973b4935492bacd90b2bbff7028:

  Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux (2019-04-05 14:10:16 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2019-04-02

for you to fetch changes up to 6d7ee2edaa54d676026b7f0d50ae472e62a84564:

  net/mlx5e: Unify logic of MTU boundaries (2019-04-05 14:10:35 -0700)

----------------------------------------------------------------
mlx5-updates-2019-04-02

This series provides misc updates to mlx5 driver

1) Aya Levin (1): Handle event of power detection in the PCIE slot

2) Eli Britstein (6):
  Some TC VLAN related updates and fixes to the previous VLAN modify action
  support patchset.
  Offload TC e-switch rules with egress/ingress VLAN devices

3) Max Gurtovoy (1): Fix double mutex initialization in esiwtch.c

4) Tariq Toukan (3): Misc small updates
  A write memory barrier is sufficient in EQ ci update
  Obsolete param field holding a constant value
  Unify logic of MTU boundaries

5) Tonghao Zhang (4): Misc updates to en_tc.c
  Make the log friendly when decapsulation offload not supported
  Remove 'parse_attr' argument in parse_tc_fdb_actions()
  Deletes unnecessary setting of esw_attr->parse_attr
  Return -EOPNOTSUPP when attempting to offload an unsupported action

----------------------------------------------------------------
Aya Levin (1):
      net/mlx5: Handle event of power detection in the PCIE slot

Eli Britstein (6):
      net/mlx5e: Use helpers to get headers criteria and value pointers
      net/mlx5e: Deny VLAN rewrite if there is no VLAN header match
      net/mlx5e: Allow VLAN rewrite of prio field with the same match
      net/mlx5e: Offload TC e-switch rules with egress VLAN device
      net/mlx5e: Offload TC e-switch rules with ingress VLAN device
      net/mlx5e: Do not rewrite fields with the same match

Max Gurtovoy (1):
      net/mlx5: E-Switch, Fix double mutex initialization

Tariq Toukan (3):
      net/mlx5: A write memory barrier is sufficient in EQ ci update
      net/mlx5e: Obsolete param field holding a constant value
      net/mlx5e: Unify logic of MTU boundaries

Tonghao Zhang (4):
      net/mlx5e: Make the log friendly when decapsulation offload not supported
      net/mlx5e: Remove 'parse_attr' argument in parse_tc_fdb_actions()
      net/mlx5e: Deletes unnecessary setting of esw_attr->parse_attr
      net/mlx5e: Return -EOPNOTSUPP when attempting to offload an unsupported action

 drivers/net/ethernet/mellanox/mlx5/core/en.h       |   2 +-
 .../net/ethernet/mellanox/mlx5/core/en/tc_tun.c    |   8 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  24 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   |  11 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 301 +++++++++++++++++----
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       |   2 +-
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |   2 -
 drivers/net/ethernet/mellanox/mlx5/core/events.c   |  75 +++++
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c  |   5 +-
 include/linux/mlx5/device.h                        |   1 +
 10 files changed, 349 insertions(+), 82 deletions(-)

[net-next V2 01/15] net/mlx5: Handle event of power detection in the PCIE slot

[Saeed Mahameed]

From: Aya Levin <ayal@mellanox.com>

Handle event of power state change in the PCIE slot. When the event
occurs, check if query power state and PCI power fields is supported. If
so, read these fields from MPEIN (management PCIE info) register and
issue a corresponding message.

Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/events.c  | 75 +++++++++++++++++++
 include/linux/mlx5/device.h                   |  1 +
 2 files changed, 76 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/events.c b/drivers/net/ethernet/mellanox/mlx5/core/events.c
index 5d5864e8df3c..a81e8d2168d8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/events.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/events.c
@@ -21,6 +21,7 @@ struct mlx5_event_nb {
 static int any_notifier(struct notifier_block *, unsigned long, void *);
 static int temp_warn(struct notifier_block *, unsigned long, void *);
 static int port_module(struct notifier_block *, unsigned long, void *);
+static int pcie_core(struct notifier_block *, unsigned long, void *);
 
 /* handler which forwards the event to events->nh, driver notifiers */
 static int forward_event(struct notifier_block *, unsigned long, void *);
@@ -30,6 +31,7 @@ static struct mlx5_nb events_nbs_ref[] = {
 	{.nb.notifier_call = any_notifier,  .event_type = MLX5_EVENT_TYPE_NOTIFY_ANY },
 	{.nb.notifier_call = temp_warn,     .event_type = MLX5_EVENT_TYPE_TEMP_WARN_EVENT },
 	{.nb.notifier_call = port_module,   .event_type = MLX5_EVENT_TYPE_PORT_MODULE_EVENT },
+	{.nb.notifier_call = pcie_core,     .event_type = MLX5_EVENT_TYPE_GENERAL_EVENT },
 
 	/* Events to be forwarded (as is) to mlx5 core interfaces (mlx5e/mlx5_ib) */
 	{.nb.notifier_call = forward_event,   .event_type = MLX5_EVENT_TYPE_PORT_CHANGE },
@@ -51,11 +53,14 @@ static struct mlx5_nb events_nbs_ref[] = {
 
 struct mlx5_events {
 	struct mlx5_core_dev *dev;
+	struct workqueue_struct *wq;
 	struct mlx5_event_nb  notifiers[ARRAY_SIZE(events_nbs_ref)];
 	/* driver notifier chain */
 	struct atomic_notifier_head nh;
 	/* port module events stats */
 	struct mlx5_pme_stats pme_stats;
+	/*pcie_core*/
+	struct work_struct pcie_core_work;
 };
 
 static const char *eqe_type_str(u8 type)
@@ -249,6 +254,69 @@ static int port_module(struct notifier_block *nb, unsigned long type, void *data
 	return NOTIFY_OK;
 }
 
+enum {
+	MLX5_PCI_POWER_COULD_NOT_BE_READ = 0x0,
+	MLX5_PCI_POWER_SUFFICIENT_REPORTED = 0x1,
+	MLX5_PCI_POWER_INSUFFICIENT_REPORTED = 0x2,
+};
+
+static void mlx5_pcie_event(struct work_struct *work)
+{
+	u32 out[MLX5_ST_SZ_DW(mpein_reg)] = {0};
+	u32 in[MLX5_ST_SZ_DW(mpein_reg)] = {0};
+	struct mlx5_events *events;
+	struct mlx5_core_dev *dev;
+	u8 power_status;
+	u16 pci_power;
+
+	events = container_of(work, struct mlx5_events, pcie_core_work);
+	dev  = events->dev;
+
+	if (!MLX5_CAP_MCAM_FEATURE(dev, pci_status_and_power))
+		return;
+
+	mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+			     MLX5_REG_MPEIN, 0, 0);
+	power_status = MLX5_GET(mpein_reg, out, pwr_status);
+	pci_power = MLX5_GET(mpein_reg, out, pci_power);
+
+	switch (power_status) {
+	case MLX5_PCI_POWER_COULD_NOT_BE_READ:
+		mlx5_core_info_rl(dev,
+				  "PCIe slot power capability was not advertised.\n");
+		break;
+	case MLX5_PCI_POWER_INSUFFICIENT_REPORTED:
+		mlx5_core_warn_rl(dev,
+				  "Detected insufficient power on the PCIe slot (%uW).\n",
+				  pci_power);
+		break;
+	case MLX5_PCI_POWER_SUFFICIENT_REPORTED:
+		mlx5_core_info_rl(dev,
+				  "PCIe slot advertised sufficient power (%uW).\n",
+				  pci_power);
+		break;
+	}
+}
+
+static int pcie_core(struct notifier_block *nb, unsigned long type, void *data)
+{
+	struct mlx5_event_nb    *event_nb = mlx5_nb_cof(nb,
+							struct mlx5_event_nb,
+							nb);
+	struct mlx5_events      *events   = event_nb->ctx;
+	struct mlx5_eqe         *eqe      = data;
+
+	switch (eqe->sub_type) {
+	case MLX5_GENERAL_SUBTYPE_PCI_POWER_CHANGE_EVENT:
+			queue_work(events->wq, &events->pcie_core_work);
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	return NOTIFY_OK;
+}
+
 void mlx5_get_pme_stats(struct mlx5_core_dev *dev, struct mlx5_pme_stats *stats)
 {
 	*stats = dev->priv.events->pme_stats;
@@ -277,11 +345,17 @@ int mlx5_events_init(struct mlx5_core_dev *dev)
 	ATOMIC_INIT_NOTIFIER_HEAD(&events->nh);
 	events->dev = dev;
 	dev->priv.events = events;
+	events->wq = create_singlethread_workqueue("mlx5_events");
+	if (!events->wq)
+		return -ENOMEM;
+	INIT_WORK(&events->pcie_core_work, mlx5_pcie_event);
+
 	return 0;
 }
 
 void mlx5_events_cleanup(struct mlx5_core_dev *dev)
 {
+	destroy_workqueue(dev->priv.events->wq);
 	kvfree(dev->priv.events);
 }
 
@@ -304,6 +378,7 @@ void mlx5_events_stop(struct mlx5_core_dev *dev)
 
 	for (i = ARRAY_SIZE(events_nbs_ref) - 1; i >= 0 ; i--)
 		mlx5_eq_notifier_unregister(dev, &events->notifiers[i].nb);
+	flush_workqueue(events->wq);
 }
 
 int mlx5_notifier_register(struct mlx5_core_dev *dev, struct notifier_block *nb)
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index f93a5598b942..db7dca75d726 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -361,6 +361,7 @@ enum {
 
 enum {
 	MLX5_GENERAL_SUBTYPE_DELAY_DROP_TIMEOUT = 0x1,
+	MLX5_GENERAL_SUBTYPE_PCI_POWER_CHANGE_EVENT = 0x5,
 };
 
 enum {
-- 
2.20.1

[net-next V2 02/15] net/mlx5: E-Switch, Fix double mutex initialization

[Saeed Mahameed]

From: Max Gurtovoy <maxg@mellanox.com>

Delete mutex_init call of a lock that's initialized in inner function.

Fixes: eca8cc389535 ("net/mlx5: E-Switch, Refactor offloads flow steering init/cleanup")
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 6c72f33f6d09..d684048551bd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1697,8 +1697,6 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int vf_nvports,
 {
 	int err;
 
-	mutex_init(&esw->fdb_table.offloads.fdb_prio_lock);
-
 	err = esw_offloads_steering_init(esw, total_nvports);
 	if (err)
 		return err;
-- 
2.20.1

[net-next V2 03/15] net/mlx5e: Make the log friendly when decapsulation offload not supported

[Saeed Mahameed]

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

If we try to offload decapsulation actions to VFs hw, we get the log [1].
It's not friendly, because the kind of net device is null, and we don't
know what '0' means.

[1] "mlx5_core 0000:05:01.2 vf_0: decapsulation offload is not supported for  net device (0)"

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
index fa2a3c444cdc..9ab3bd904295 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
@@ -96,7 +96,7 @@ static const char *mlx5e_netdev_kind(struct net_device *dev)
 	if (dev->rtnl_link_ops)
 		return dev->rtnl_link_ops->kind;
 	else
-		return "";
+		return "unknown";
 }
 
 static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv,
@@ -636,8 +636,10 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev,
 						headers_c, headers_v);
 	} else {
 		netdev_warn(priv->netdev,
-			    "decapsulation offload is not supported for %s net device (%d)\n",
-			    mlx5e_netdev_kind(filter_dev), tunnel_type);
+			    "decapsulation offload is not supported for %s (kind: \"%s\")\n",
+			    netdev_name(filter_dev),
+			    mlx5e_netdev_kind(filter_dev));
+
 		return -EOPNOTSUPP;
 	}
 	return err;
-- 
2.20.1

[net-next V2 04/15] net/mlx5e: Remove 'parse_attr' argument in parse_tc_fdb_actions()

[Saeed Mahameed]

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

This patch is a little improvement. Simplify the parse_tc_fdb_actions().

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 2fd425a7b156..c061e5bdf365 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2541,13 +2541,13 @@ static int parse_tc_vlan_action(struct mlx5e_priv *priv,
 
 static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				struct flow_action *flow_action,
-				struct mlx5e_tc_flow_parse_attr *parse_attr,
 				struct mlx5e_tc_flow *flow,
 				struct netlink_ext_ack *extack)
 {
 	struct pedit_headers_action hdrs[2] = {};
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5_esw_flow_attr *attr = flow->esw_attr;
+	struct mlx5e_tc_flow_parse_attr *parse_attr = attr->parse_attr;
 	struct mlx5e_rep_priv *rpriv = priv->ppriv;
 	const struct ip_tunnel_info *info = NULL;
 	const struct flow_action_entry *act;
@@ -2889,7 +2889,7 @@ __mlx5e_add_fdb_flow(struct mlx5e_priv *priv,
 	if (err)
 		goto err_free;
 
-	err = parse_tc_fdb_actions(priv, &rule->action, parse_attr, flow, extack);
+	err = parse_tc_fdb_actions(priv, &rule->action, flow, extack);
 	if (err)
 		goto err_free;
 
-- 
2.20.1

[net-next V2 05/15] net/mlx5e: Deletes unnecessary setting of esw_attr->parse_attr

[Saeed Mahameed]

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

This patch deletes unnecessary setting of the esw_attr->parse_attr
to parse_attr in parse_tc_fdb_actions() because it is already done
by the mlx5e_flow_esw_attr_init() function.

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index c061e5bdf365..26c5548ab705 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2632,7 +2632,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 					out_dev->ifindex;
 				parse_attr->tun_info[attr->out_count] = *info;
 				encap = false;
-				attr->parse_attr = parse_attr;
 				attr->dests[attr->out_count].flags |=
 					MLX5_ESW_DEST_ENCAP;
 				attr->out_count++;
-- 
2.20.1

[net-next V2 06/15] net/mlx5e: Return -EOPNOTSUPP when attempting to offload an unsupported action

[Saeed Mahameed]

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

* Now the encapsulation is not supported for mlx5 VFs. When we try to
offload that action, the -EINVAL is returned, but not -EOPNOTSUPP.
This patch changes the returned value and ignore to confuse user.
The command is shown as below [1].

* When max modify header action is zero, we return -EOPNOTSUPP
directly. In this way, we can ignore wrong message info (e.g.
"mlx5: parsed 0 pedit actions, can't do more"). This happens when
offloading pedit actions on mlx(cx4) VFs. The command is shown as below [2].

For example: (p2p1_0 is VF net device)
[1]
$ tc filter add dev p2p1_0 protocol ip  parent ffff: prio 1 flower skip_sw \
    src_mac e4:11:22:33:44:01    \
    action tunnel_key set        \
    src_ip 1.1.1.100        \
    dst_ip 1.1.1.200        \
    dst_port 4789 id 100        \
    action mirred egress redirect dev vxlan0

[2]
$ tc filter add dev p2p1_0 parent ffff: protocol ip prio 1 \
    flower skip_sw dst_mac 00:10:56:fb:64:e8 \
    dst_ip 1.1.1.100 src_ip 1.1.1.200 \
    action pedit ex munge eth src set 00:10:56:b4:5d:20

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 27 ++++++++++++++-----
 1 file changed, 20 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 26c5548ab705..7c1ea0a17024 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2029,6 +2029,15 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs,
 	return 0;
 }
 
+static int mlx5e_flow_namespace_max_modify_action(struct mlx5_core_dev *mdev,
+						  int namespace)
+{
+	if (namespace == MLX5_FLOW_NAMESPACE_FDB) /* FDB offloading */
+		return MLX5_CAP_ESW_FLOWTABLE_FDB(mdev, max_modify_header_actions);
+	else /* namespace is MLX5_FLOW_NAMESPACE_KERNEL - NIC offloading */
+		return MLX5_CAP_FLOWTABLE_NIC_RX(mdev, max_modify_header_actions);
+}
+
 static int alloc_mod_hdr_actions(struct mlx5e_priv *priv,
 				 struct pedit_headers_action *hdrs,
 				 int namespace,
@@ -2040,11 +2049,7 @@ static int alloc_mod_hdr_actions(struct mlx5e_priv *priv,
 		hdrs[TCA_PEDIT_KEY_EX_CMD_ADD].pedits;
 	action_size = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto);
 
-	if (namespace == MLX5_FLOW_NAMESPACE_FDB) /* FDB offloading */
-		max_actions = MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev, max_modify_header_actions);
-	else /* namespace is MLX5_FLOW_NAMESPACE_KERNEL - NIC offloading */
-		max_actions = MLX5_CAP_FLOWTABLE_NIC_RX(priv->mdev, max_modify_header_actions);
-
+	max_actions = mlx5e_flow_namespace_max_modify_action(priv->mdev, namespace);
 	/* can get up to crazingly 16 HW actions in 32 bits pedit SW key */
 	max_actions = min(max_actions, nkeys * 16);
 
@@ -2077,6 +2082,12 @@ static int parse_tc_pedit_action(struct mlx5e_priv *priv,
 		goto out_err;
 	}
 
+	if (!mlx5e_flow_namespace_max_modify_action(priv->mdev, namespace)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "The pedit offload action is not supported");
+		goto out_err;
+	}
+
 	mask = act->mangle.mask;
 	val = act->mangle.val;
 	offset = act->mangle.offset;
@@ -2362,7 +2373,8 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv,
 			}
 			break;
 		default:
-			return -EINVAL;
+			NL_SET_ERR_MSG_MOD(extack, "The offload action is not supported");
+			return -EOPNOTSUPP;
 		}
 	}
 
@@ -2710,7 +2722,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			break;
 			}
 		default:
-			return -EINVAL;
+			NL_SET_ERR_MSG_MOD(extack, "The offload action is not supported");
+			return -EOPNOTSUPP;
 		}
 	}
 
-- 
2.20.1

[net-next V2 07/15] net/mlx5e: Use helpers to get headers criteria and value pointers

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

The headers criteria and value pointers may be either of the inner
packet, if a tunnel exists, or of the outer. Simplify the code by using
helper functions to retrieve them.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 34 ++++++++++++++-----
 1 file changed, 25 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 7c1ea0a17024..81f8ac569a0e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1438,6 +1438,26 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv,
 	return 0;
 }
 
+static void *get_match_headers_criteria(u32 flags,
+					struct mlx5_flow_spec *spec)
+{
+	return (flags & MLX5_FLOW_CONTEXT_ACTION_DECAP) ?
+		MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+			     inner_headers) :
+		MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+			     outer_headers);
+}
+
+static void *get_match_headers_value(u32 flags,
+				     struct mlx5_flow_spec *spec)
+{
+	return (flags & MLX5_FLOW_CONTEXT_ACTION_DECAP) ?
+		MLX5_ADDR_OF(fte_match_param, spec->match_value,
+			     inner_headers) :
+		MLX5_ADDR_OF(fte_match_param, spec->match_value,
+			     outer_headers);
+}
+
 static int __parse_cls_flower(struct mlx5e_priv *priv,
 			      struct mlx5_flow_spec *spec,
 			      struct tc_cls_flower_offload *f,
@@ -1503,10 +1523,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 		/* In decap flow, header pointers should point to the inner
 		 * headers, outer header were already set by parse_tunnel_attr
 		 */
-		headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
-					 inner_headers);
-		headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
-					 inner_headers);
+		headers_c = get_match_headers_criteria(MLX5_FLOW_CONTEXT_ACTION_DECAP,
+						       spec);
+		headers_v = get_match_headers_value(MLX5_FLOW_CONTEXT_ACTION_DECAP,
+						    spec);
 	}
 
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
@@ -2184,11 +2204,7 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec,
 	u16 ethertype;
 	int i;
 
-	if (actions & MLX5_FLOW_CONTEXT_ACTION_DECAP)
-		headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, inner_headers);
-	else
-		headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers);
-
+	headers_v = get_match_headers_value(actions, spec);
 	ethertype = MLX5_GET(fte_match_set_lyr_2_4, headers_v, ethertype);
 
 	/* for non-IP we only re-write MACs, so we're okay */
-- 
2.20.1

[net-next V2 08/15] net/mlx5e: Deny VLAN rewrite if there is no VLAN header match

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

Rewrite of the packet in the VLAN offset may corrupt the packet if it's
not VLAN tagged. Deny the rewrite in this case.

Fixes: 37410902874c ("net/mlx5e: Support VLAN modify action")
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 81f8ac569a0e..3a0854fbcfd4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2292,8 +2292,19 @@ static int add_vlan_rewrite_action(struct mlx5e_priv *priv, int namespace,
 		.mangle.mask = ~(u32)be16_to_cpu(*(__be16 *)&mask16),
 		.mangle.val = (u32)be16_to_cpu(*(__be16 *)&val16),
 	};
+	void *headers_c, *headers_v;
 	int err;
 
+	headers_c = get_match_headers_criteria(*action, &parse_attr->spec);
+	headers_v = get_match_headers_value(*action, &parse_attr->spec);
+
+	if (!(MLX5_GET(fte_match_set_lyr_2_4, headers_c, cvlan_tag) &&
+	      MLX5_GET(fte_match_set_lyr_2_4, headers_v, cvlan_tag))) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "VLAN rewrite action must have VLAN protocol match");
+		return -EOPNOTSUPP;
+	}
+
 	if (act->vlan.prio) {
 		NL_SET_ERR_MSG_MOD(extack, "Setting VLAN prio is not supported");
 		return -EOPNOTSUPP;
-- 
2.20.1

[net-next V2 09/15] net/mlx5e: Allow VLAN rewrite of prio field with the same match

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

Changing the prio field of the VLAN is not supported. With
commit 37410902874c ("net/mlx5e: Support VLAN modify action") zero
value indicated "no-change". Allow the vid rewrite if the prio match
is the same as the prio set value.

Fixes: 37410902874c ("net/mlx5e: Support VLAN modify action")
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 3a0854fbcfd4..611b636c603e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2292,6 +2292,7 @@ static int add_vlan_rewrite_action(struct mlx5e_priv *priv, int namespace,
 		.mangle.mask = ~(u32)be16_to_cpu(*(__be16 *)&mask16),
 		.mangle.val = (u32)be16_to_cpu(*(__be16 *)&val16),
 	};
+	u8 match_prio_mask, match_prio_val;
 	void *headers_c, *headers_v;
 	int err;
 
@@ -2305,8 +2306,11 @@ static int add_vlan_rewrite_action(struct mlx5e_priv *priv, int namespace,
 		return -EOPNOTSUPP;
 	}
 
-	if (act->vlan.prio) {
-		NL_SET_ERR_MSG_MOD(extack, "Setting VLAN prio is not supported");
+	match_prio_mask = MLX5_GET(fte_match_set_lyr_2_4, headers_c, first_prio);
+	match_prio_val = MLX5_GET(fte_match_set_lyr_2_4, headers_v, first_prio);
+	if (act->vlan.prio != (match_prio_val & match_prio_mask)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Changing VLAN prio is not supported");
 		return -EOPNOTSUPP;
 	}
 
-- 
2.20.1

[net-next V2 10/15] net/mlx5e: Offload TC e-switch rules with egress VLAN device

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

Upon redirection to an uplink VLAN device, emulate vlan push actions
according to the VLAN properties of the VLAN device and redirect to
the uplink.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 611b636c603e..affa54d10a97 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2582,6 +2582,32 @@ static int parse_tc_vlan_action(struct mlx5e_priv *priv,
 	return 0;
 }
 
+static int add_vlan_push_action(struct mlx5e_priv *priv,
+				struct mlx5_esw_flow_attr *attr,
+				struct net_device **out_dev,
+				u32 *action)
+{
+	struct net_device *vlan_dev = *out_dev;
+	struct flow_action_entry vlan_act = {
+		.id = FLOW_ACTION_VLAN_PUSH,
+		.vlan.vid = vlan_dev_vlan_id(vlan_dev),
+		.vlan.proto = vlan_dev_vlan_proto(vlan_dev),
+		.vlan.prio = 0,
+	};
+	int err;
+
+	err = parse_tc_vlan_action(priv, &vlan_act, attr, action);
+	if (err)
+		return err;
+
+	*out_dev = dev_get_by_index_rcu(dev_net(vlan_dev),
+					dev_get_iflink(vlan_dev));
+	if (is_vlan_dev(*out_dev))
+		err = add_vlan_push_action(priv, attr, out_dev, action);
+
+	return err;
+}
+
 static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				struct flow_action *flow_action,
 				struct mlx5e_tc_flow *flow,
@@ -2662,6 +2688,14 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				    uplink_upper == out_dev)
 					out_dev = uplink_dev;
 
+				if (is_vlan_dev(out_dev)) {
+					err = add_vlan_push_action(priv, attr,
+								   &out_dev,
+								   &action);
+					if (err)
+						return err;
+				}
+
 				if (!mlx5e_eswitch_rep(out_dev))
 					return -EOPNOTSUPP;
 
-- 
2.20.1

[net-next V2 11/15] net/mlx5e: Offload TC e-switch rules with ingress VLAN device

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

Offload TC rule on a VLAN device by matching the VLAN properties
of the VLAN device and emulating vlan pop actions.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_rep.c  |  3 +-
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 46 +++++++++++++++++--
 2 files changed, 44 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index a66b6ed80b30..94c39655c665 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -795,7 +795,8 @@ static int mlx5e_nic_rep_netdevice_event(struct notifier_block *nb,
 	struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
 	struct net_device *netdev = netdev_notifier_info_to_dev(ptr);
 
-	if (!mlx5e_tc_tun_device_to_offload(priv, netdev))
+	if (!mlx5e_tc_tun_device_to_offload(priv, netdev) &&
+	    !is_vlan_dev(netdev))
 		return NOTIFY_OK;
 
 	switch (event) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index affa54d10a97..716fabf6fde7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1541,11 +1541,23 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 		if (match.mask->n_proto)
 			*match_level = MLX5_MATCH_L2;
 	}
-
-	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN) ||
+	    is_vlan_dev(filter_dev)) {
+		struct flow_dissector_key_vlan filter_dev_mask;
+		struct flow_dissector_key_vlan filter_dev_key;
 		struct flow_match_vlan match;
 
-		flow_rule_match_vlan(rule, &match);
+		if (is_vlan_dev(filter_dev)) {
+			match.key = &filter_dev_key;
+			match.key->vlan_id = vlan_dev_vlan_id(filter_dev);
+			match.key->vlan_tpid = vlan_dev_vlan_proto(filter_dev);
+			match.key->vlan_priority = 0;
+			match.mask = &filter_dev_mask;
+			memset(match.mask, 0xff, sizeof(*match.mask));
+			match.mask->vlan_priority = 0;
+		} else {
+			flow_rule_match_vlan(rule, &match);
+		}
 		if (match.mask->vlan_id ||
 		    match.mask->vlan_priority ||
 		    match.mask->vlan_tpid) {
@@ -2252,7 +2264,8 @@ static bool actions_match_supported(struct mlx5e_priv *priv,
 		actions = flow->nic_attr->action;
 
 	if (flow->flags & MLX5E_TC_FLOW_EGRESS &&
-	    !(actions & MLX5_FLOW_CONTEXT_ACTION_DECAP))
+	    !((actions & MLX5_FLOW_CONTEXT_ACTION_DECAP) ||
+	      (actions & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP)))
 		return false;
 
 	if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
@@ -2608,6 +2621,25 @@ static int add_vlan_push_action(struct mlx5e_priv *priv,
 	return err;
 }
 
+static int add_vlan_pop_action(struct mlx5e_priv *priv,
+			       struct mlx5_esw_flow_attr *attr,
+			       u32 *action)
+{
+	int nest_level = vlan_get_encap_level(attr->parse_attr->filter_dev);
+	struct flow_action_entry vlan_act = {
+		.id = FLOW_ACTION_VLAN_POP,
+	};
+	int err = 0;
+
+	while (nest_level--) {
+		err = parse_tc_vlan_action(priv, &vlan_act, attr, action);
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
 static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				struct flow_action *flow_action,
 				struct mlx5e_tc_flow *flow,
@@ -2695,6 +2727,12 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 					if (err)
 						return err;
 				}
+				if (is_vlan_dev(parse_attr->filter_dev)) {
+					err = add_vlan_pop_action(priv, attr,
+								  &action);
+					if (err)
+						return err;
+				}
 
 				if (!mlx5e_eswitch_rep(out_dev))
 					return -EOPNOTSUPP;
-- 
2.20.1

[net-next V2 12/15] net/mlx5e: Do not rewrite fields with the same match

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

If we have a match for the same value of a rewrite field, there is no
point for the rewrite. In order to save rewrite actions, and avoid
entirely rewrite actions (if all rewrites are the same), ignore such
rewrite fields.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 136 +++++++++++++-----
 1 file changed, 104 insertions(+), 32 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 716fabf6fde7..ff9756c3dee5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1907,39 +1907,73 @@ struct mlx5_fields {
 	u8  field;
 	u8  size;
 	u32 offset;
+	u32 match_offset;
 };
 
-#define OFFLOAD(fw_field, size, field, off) \
-		{MLX5_ACTION_IN_FIELD_OUT_ ## fw_field, size, offsetof(struct pedit_headers, field) + (off)}
+#define OFFLOAD(fw_field, size, field, off, match_field) \
+		{MLX5_ACTION_IN_FIELD_OUT_ ## fw_field, size, \
+		 offsetof(struct pedit_headers, field) + (off), \
+		 MLX5_BYTE_OFF(fte_match_set_lyr_2_4, match_field)}
+
+static bool cmp_val_mask(void *valp, void *maskp, void *matchvalp,
+			 void *matchmaskp, int size)
+{
+	bool same = false;
+
+	switch (size) {
+	case sizeof(u8):
+		same = ((*(u8 *)valp) & (*(u8 *)maskp)) ==
+		       ((*(u8 *)matchvalp) & (*(u8 *)matchmaskp));
+		break;
+	case sizeof(u16):
+		same = ((*(u16 *)valp) & (*(u16 *)maskp)) ==
+		       ((*(u16 *)matchvalp) & (*(u16 *)matchmaskp));
+		break;
+	case sizeof(u32):
+		same = ((*(u32 *)valp) & (*(u32 *)maskp)) ==
+		       ((*(u32 *)matchvalp) & (*(u32 *)matchmaskp));
+		break;
+	}
+
+	return same;
+}
 
 static struct mlx5_fields fields[] = {
-	OFFLOAD(DMAC_47_16, 4, eth.h_dest[0], 0),
-	OFFLOAD(DMAC_15_0,  2, eth.h_dest[4], 0),
-	OFFLOAD(SMAC_47_16, 4, eth.h_source[0], 0),
-	OFFLOAD(SMAC_15_0,  2, eth.h_source[4], 0),
-	OFFLOAD(ETHERTYPE,  2, eth.h_proto, 0),
-	OFFLOAD(FIRST_VID,  2, vlan.h_vlan_TCI, 0),
-
-	OFFLOAD(IP_TTL, 1, ip4.ttl,   0),
-	OFFLOAD(SIPV4,  4, ip4.saddr, 0),
-	OFFLOAD(DIPV4,  4, ip4.daddr, 0),
-
-	OFFLOAD(SIPV6_127_96, 4, ip6.saddr.s6_addr32[0], 0),
-	OFFLOAD(SIPV6_95_64,  4, ip6.saddr.s6_addr32[1], 0),
-	OFFLOAD(SIPV6_63_32,  4, ip6.saddr.s6_addr32[2], 0),
-	OFFLOAD(SIPV6_31_0,   4, ip6.saddr.s6_addr32[3], 0),
-	OFFLOAD(DIPV6_127_96, 4, ip6.daddr.s6_addr32[0], 0),
-	OFFLOAD(DIPV6_95_64,  4, ip6.daddr.s6_addr32[1], 0),
-	OFFLOAD(DIPV6_63_32,  4, ip6.daddr.s6_addr32[2], 0),
-	OFFLOAD(DIPV6_31_0,   4, ip6.daddr.s6_addr32[3], 0),
-	OFFLOAD(IPV6_HOPLIMIT, 1, ip6.hop_limit, 0),
-
-	OFFLOAD(TCP_SPORT, 2, tcp.source,  0),
-	OFFLOAD(TCP_DPORT, 2, tcp.dest,    0),
-	OFFLOAD(TCP_FLAGS, 1, tcp.ack_seq, 5),
-
-	OFFLOAD(UDP_SPORT, 2, udp.source, 0),
-	OFFLOAD(UDP_DPORT, 2, udp.dest,   0),
+	OFFLOAD(DMAC_47_16, 4, eth.h_dest[0], 0, dmac_47_16),
+	OFFLOAD(DMAC_15_0,  2, eth.h_dest[4], 0, dmac_15_0),
+	OFFLOAD(SMAC_47_16, 4, eth.h_source[0], 0, smac_47_16),
+	OFFLOAD(SMAC_15_0,  2, eth.h_source[4], 0, smac_15_0),
+	OFFLOAD(ETHERTYPE,  2, eth.h_proto, 0, ethertype),
+	OFFLOAD(FIRST_VID,  2, vlan.h_vlan_TCI, 0, first_vid),
+
+	OFFLOAD(IP_TTL, 1, ip4.ttl,   0, ttl_hoplimit),
+	OFFLOAD(SIPV4,  4, ip4.saddr, 0, src_ipv4_src_ipv6.ipv4_layout.ipv4),
+	OFFLOAD(DIPV4,  4, ip4.daddr, 0, dst_ipv4_dst_ipv6.ipv4_layout.ipv4),
+
+	OFFLOAD(SIPV6_127_96, 4, ip6.saddr.s6_addr32[0], 0,
+		src_ipv4_src_ipv6.ipv6_layout.ipv6[0]),
+	OFFLOAD(SIPV6_95_64,  4, ip6.saddr.s6_addr32[1], 0,
+		src_ipv4_src_ipv6.ipv6_layout.ipv6[4]),
+	OFFLOAD(SIPV6_63_32,  4, ip6.saddr.s6_addr32[2], 0,
+		src_ipv4_src_ipv6.ipv6_layout.ipv6[8]),
+	OFFLOAD(SIPV6_31_0,   4, ip6.saddr.s6_addr32[3], 0,
+		src_ipv4_src_ipv6.ipv6_layout.ipv6[12]),
+	OFFLOAD(DIPV6_127_96, 4, ip6.daddr.s6_addr32[0], 0,
+		dst_ipv4_dst_ipv6.ipv6_layout.ipv6[0]),
+	OFFLOAD(DIPV6_95_64,  4, ip6.daddr.s6_addr32[1], 0,
+		dst_ipv4_dst_ipv6.ipv6_layout.ipv6[4]),
+	OFFLOAD(DIPV6_63_32,  4, ip6.daddr.s6_addr32[2], 0,
+		dst_ipv4_dst_ipv6.ipv6_layout.ipv6[8]),
+	OFFLOAD(DIPV6_31_0,   4, ip6.daddr.s6_addr32[3], 0,
+		dst_ipv4_dst_ipv6.ipv6_layout.ipv6[12]),
+	OFFLOAD(IPV6_HOPLIMIT, 1, ip6.hop_limit, 0, ttl_hoplimit),
+
+	OFFLOAD(TCP_SPORT, 2, tcp.source,  0, tcp_sport),
+	OFFLOAD(TCP_DPORT, 2, tcp.dest,    0, tcp_dport),
+	OFFLOAD(TCP_FLAGS, 1, tcp.ack_seq, 5, tcp_flags),
+
+	OFFLOAD(UDP_SPORT, 2, udp.source, 0, udp_sport),
+	OFFLOAD(UDP_DPORT, 2, udp.dest,   0, udp_dport),
 };
 
 /* On input attr->max_mod_hdr_actions tells how many HW actions can be parsed at
@@ -1948,9 +1982,14 @@ static struct mlx5_fields fields[] = {
  */
 static int offload_pedit_fields(struct pedit_headers_action *hdrs,
 				struct mlx5e_tc_flow_parse_attr *parse_attr,
+				u32 *action_flags,
 				struct netlink_ext_ack *extack)
 {
 	struct pedit_headers *set_masks, *add_masks, *set_vals, *add_vals;
+	void *headers_c = get_match_headers_criteria(*action_flags,
+						     &parse_attr->spec);
+	void *headers_v = get_match_headers_value(*action_flags,
+						  &parse_attr->spec);
 	int i, action_size, nactions, max_actions, first, last, next_z;
 	void *s_masks_p, *a_masks_p, *vals_p;
 	struct mlx5_fields *f;
@@ -1974,6 +2013,8 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs,
 	nactions = parse_attr->num_mod_hdr_actions;
 
 	for (i = 0; i < ARRAY_SIZE(fields); i++) {
+		bool skip;
+
 		f = &fields[i];
 		/* avoid seeing bits set from previous iterations */
 		s_mask = 0;
@@ -2002,19 +2043,34 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs,
 			return -EOPNOTSUPP;
 		}
 
+		skip = false;
 		if (s_mask) {
+			void *match_mask = headers_c + f->match_offset;
+			void *match_val = headers_v + f->match_offset;
+
 			cmd  = MLX5_ACTION_TYPE_SET;
 			mask = s_mask;
 			vals_p = (void *)set_vals + f->offset;
+			/* don't rewrite if we have a match on the same value */
+			if (cmp_val_mask(vals_p, s_masks_p, match_val,
+					 match_mask, f->size))
+				skip = true;
 			/* clear to denote we consumed this field */
 			memset(s_masks_p, 0, f->size);
 		} else {
+			u32 zero = 0;
+
 			cmd  = MLX5_ACTION_TYPE_ADD;
 			mask = a_mask;
 			vals_p = (void *)add_vals + f->offset;
+			/* add 0 is no change */
+			if (!memcmp(vals_p, &zero, f->size))
+				skip = true;
 			/* clear to denote we consumed this field */
 			memset(a_masks_p, 0, f->size);
 		}
+		if (skip)
+			continue;
 
 		field_bsize = f->size * BITS_PER_BYTE;
 
@@ -2138,6 +2194,7 @@ static int parse_tc_pedit_action(struct mlx5e_priv *priv,
 static int alloc_tc_pedit_action(struct mlx5e_priv *priv, int namespace,
 				 struct mlx5e_tc_flow_parse_attr *parse_attr,
 				 struct pedit_headers_action *hdrs,
+				 u32 *action_flags,
 				 struct netlink_ext_ack *extack)
 {
 	struct pedit_headers *cmd_masks;
@@ -2150,7 +2207,7 @@ static int alloc_tc_pedit_action(struct mlx5e_priv *priv, int namespace,
 			goto out_err;
 	}
 
-	err = offload_pedit_fields(hdrs, parse_attr, extack);
+	err = offload_pedit_fields(hdrs, parse_attr, action_flags, extack);
 	if (err < 0)
 		goto out_dealloc_parsed_actions;
 
@@ -2425,9 +2482,14 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv,
 	if (hdrs[TCA_PEDIT_KEY_EX_CMD_SET].pedits ||
 	    hdrs[TCA_PEDIT_KEY_EX_CMD_ADD].pedits) {
 		err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_KERNEL,
-					    parse_attr, hdrs, extack);
+					    parse_attr, hdrs, &action, extack);
 		if (err)
 			return err;
+		/* in case all pedit actions are skipped, remove the MOD_HDR
+		 * flag.
+		 */
+		if (parse_attr->num_mod_hdr_actions == 0)
+			action &= ~MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
 	}
 
 	attr->action = action;
@@ -2833,9 +2895,19 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 	if (hdrs[TCA_PEDIT_KEY_EX_CMD_SET].pedits ||
 	    hdrs[TCA_PEDIT_KEY_EX_CMD_ADD].pedits) {
 		err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_KERNEL,
-					    parse_attr, hdrs, extack);
+					    parse_attr, hdrs, &action, extack);
 		if (err)
 			return err;
+		/* in case all pedit actions are skipped, remove the MOD_HDR
+		 * flag. we might have set split_count either by pedit or
+		 * pop/push. if there is no pop/push either, reset it too.
+		 */
+		if (parse_attr->num_mod_hdr_actions == 0) {
+			action &= ~MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+			if (!((action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) ||
+			      (action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH)))
+				attr->split_count = 0;
+		}
 	}
 
 	attr->action = action;
-- 
2.20.1

[net-next V2 13/15] net/mlx5: A write memory barrier is sufficient in EQ ci update

[Saeed Mahameed]

From: Tariq Toukan <tariqt@mellanox.com>

Soften the memory barrier call of mb() by a sufficient wmb() in the
consumer index update of the event queues.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 46a747f7c162..e9837aeb7088 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -707,7 +707,7 @@ void mlx5_eq_update_ci(struct mlx5_eq *eq, u32 cc, bool arm)
 
 	__raw_writel((__force u32)cpu_to_be32(val), addr);
 	/* We still want ordering, just not swabbing, so add a barrier */
-	mb();
+	wmb();
 }
 EXPORT_SYMBOL(mlx5_eq_update_ci);
 
-- 
2.20.1

[net-next V2 14/15] net/mlx5e: Obsolete param field holding a constant value

[Saeed Mahameed]

From: Tariq Toukan <tariqt@mellanox.com>

The LRO WQE size is a constant, obsolete the parameter field
that holds it.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      | 1 -
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 3 +--
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index b0b68dde3017..7ce5edeaba33 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -241,7 +241,6 @@ struct mlx5e_params {
 	struct net_dim_cq_moder rx_cq_moderation;
 	struct net_dim_cq_moder tx_cq_moderation;
 	bool lro_en;
-	u32 lro_wqe_sz;
 	u8  tx_min_inline_mode;
 	bool vlan_strip_disable;
 	bool scatter_fcs_en;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index e08a1eb04e22..35ce60b8214a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -204,7 +204,6 @@ static u16 mlx5e_get_rq_headroom(struct mlx5_core_dev *mdev,
 void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev,
 			       struct mlx5e_params *params)
 {
-	params->lro_wqe_sz = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ;
 	params->log_rq_mtu_frames = is_kdump_kernel() ?
 		MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE :
 		MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE;
@@ -2637,7 +2636,7 @@ static void mlx5e_build_tir_ctx_lro(struct mlx5e_params *params, void *tirc)
 		 MLX5_TIRC_LRO_ENABLE_MASK_IPV4_LRO |
 		 MLX5_TIRC_LRO_ENABLE_MASK_IPV6_LRO);
 	MLX5_SET(tirc, tirc, lro_max_ip_payload_size,
-		 (params->lro_wqe_sz - ROUGH_MAX_L2_L3_HDR_SZ) >> 8);
+		 (MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ - ROUGH_MAX_L2_L3_HDR_SZ) >> 8);
 	MLX5_SET(tirc, tirc, lro_timeout_period_usecs, params->lro_timeout);
 }
 
-- 
2.20.1

[net-next V2 15/15] net/mlx5e: Unify logic of MTU boundaries

[Saeed Mahameed]

From: Tariq Toukan <tariqt@mellanox.com>

Expose a new helper that wraps the logic for setting the
netdevice's MTU boundaries.
Use it for the different components (Eth, rep, IPoIB).

Set the netdevice min MTU to ETH_MIN_MTU, and the max according
to both the FW capability and the kernel definition.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h  |  1 +
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 21 ++++++++++++++-----
 .../net/ethernet/mellanox/mlx5/core/en_rep.c  |  8 +------
 .../ethernet/mellanox/mlx5/core/ipoib/ipoib.c |  5 ++---
 4 files changed, 20 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 7ce5edeaba33..bd855ab8dfe2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1086,6 +1086,7 @@ mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *prof
 int mlx5e_attach_netdev(struct mlx5e_priv *priv);
 void mlx5e_detach_netdev(struct mlx5e_priv *priv);
 void mlx5e_destroy_netdev(struct mlx5e_priv *priv);
+void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv);
 void mlx5e_build_nic_params(struct mlx5_core_dev *mdev,
 			    struct mlx5e_rss_params *rss_params,
 			    struct mlx5e_params *params,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 35ce60b8214a..ba705392b46b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -2811,6 +2811,21 @@ int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv)
 	return 0;
 }
 
+void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv)
+{
+	struct mlx5e_params *params = &priv->channels.params;
+	struct net_device *netdev   = priv->netdev;
+	struct mlx5_core_dev *mdev  = priv->mdev;
+	u16 max_mtu;
+
+	/* MTU range: 68 - hw-specific max */
+	netdev->min_mtu = ETH_MIN_MTU;
+
+	mlx5_query_port_max_mtu(mdev, &max_mtu, 1);
+	netdev->max_mtu = min_t(unsigned int, MLX5E_HW2SW_MTU(params, max_mtu),
+				ETH_MAX_MTU);
+}
+
 static void mlx5e_netdev_set_tcs(struct net_device *netdev)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
@@ -4912,7 +4927,6 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv)
 {
 	struct net_device *netdev = priv->netdev;
 	struct mlx5_core_dev *mdev = priv->mdev;
-	u16 max_mtu;
 
 	mlx5e_init_l2_addr(priv);
 
@@ -4920,10 +4934,7 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv)
 	if (!netif_running(netdev))
 		mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN);
 
-	/* MTU range: 68 - hw-specific max */
-	netdev->min_mtu = ETH_MIN_MTU;
-	mlx5_query_port_max_mtu(priv->mdev, &max_mtu, 1);
-	netdev->max_mtu = MLX5E_HW2SW_MTU(&priv->channels.params, max_mtu);
+	mlx5e_set_netdev_mtu_boundaries(priv);
 	mlx5e_set_dev_port_mtu(priv);
 
 	mlx5_lag_add(mdev, netdev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 94c39655c665..6bfdefa8b9f4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1624,13 +1624,7 @@ static void mlx5e_cleanup_rep_tx(struct mlx5e_priv *priv)
 
 static void mlx5e_vf_rep_enable(struct mlx5e_priv *priv)
 {
-	struct net_device *netdev = priv->netdev;
-	struct mlx5_core_dev *mdev = priv->mdev;
-	u16 max_mtu;
-
-	netdev->min_mtu = ETH_MIN_MTU;
-	mlx5_query_port_max_mtu(mdev, &max_mtu, 1);
-	netdev->max_mtu = MLX5E_HW2SW_MTU(&priv->channels.params, max_mtu);
+	mlx5e_set_netdev_mtu_boundaries(priv);
 }
 
 static int uplink_rep_async_event(struct notifier_block *nb, unsigned long event, void *data)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index 4eac42555c7d..9b03ae1e1e10 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -77,15 +77,14 @@ int mlx5i_init(struct mlx5_core_dev *mdev,
 	       void *ppriv)
 {
 	struct mlx5e_priv *priv  = mlx5i_epriv(netdev);
-	u16 max_mtu;
 	int err;
 
 	err = mlx5e_netdev_init(netdev, priv, mdev, profile, ppriv);
 	if (err)
 		return err;
 
-	mlx5_query_port_max_mtu(mdev, &max_mtu, 1);
-	netdev->mtu = max_mtu;
+	mlx5e_set_netdev_mtu_boundaries(priv);
+	netdev->mtu = netdev->max_mtu;
 
 	mlx5e_build_nic_params(mdev, &priv->rss_params, &priv->channels.params,
 			       mlx5e_get_netdev_max_channels(netdev),
-- 
2.20.1

Re: [pull request][net-next V2 00/15] Mellanox, mlx5 updates 2019-04-02

[David Miller]

From: Saeed Mahameed <saeedm@mellanox.com>
Date: Fri,  5 Apr 2019 14:17:04 -0700

> This series adds misc updates to mlx5 driver.
> 
> For more information please see tag log below.
> 
> Please pull and let me know if there is any problem.
> 
> v1->v2:
>   remove inline keywords from static functions in c files.

Pulled, thanks Saeed.

The parse_tc_fdb_actions function could use some reverse christmas
tree fixing, just FYI.

> Please note that the series starts with a merge of mlx5-next branch,
> to resolve and avoid dependency with rdma tree.
> 
> Note: when merged with net there is going to be a small merge conflict
> in en_tc.c:
> ++<<<<<<< HEAD
>  +              err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_KERNEL,
>  +                                          parse_attr, hdrs, &action, extack);
> ++=======
> +               err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_FDB,
> +                                           parse_attr, hdrs, extack);
> ++>>>>>>> dave/master
> 
> Resolution:
> +               err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_FDB,
> +                                           parse_attr, hdrs, &action, extack);

Thanks for this, it helped a lot.