[pull request][net-next 00/15] Mellanox, mlx5 updates 2019-04-30

[pull request][net-next 00/15] Mellanox, mlx5 updates 2019-04-30

[Saeed Mahameed]

Hi Dave,

This series provides misc updates to mlx5 driver.
There is one patch of this series that is touching outside mlx5 driver:

ethtool.h: Add SFF-8436 and SFF-8636 max EEPROM length definitions
Added max EEPROM length defines for ethtool usage:
     #define ETH_MODULE_SFF_8636_MAX_LEN     640
     #define ETH_MODULE_SFF_8436_MAX_LEN     640
    
These definitions used to determine the EEPROM data
length when reading high eeprom pages.

For more information please see tag log below.

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

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

Thanks,
Saeed.

---
The following changes since commit db5ac5de59bac721cf1e210731cc88a043e6bbdc:

  Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux (2019-04-30 13:32:52 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2019-04-30

for you to fetch changes up to 7a3cc3c6d3e54ed1f6001272c4f15a4544390abe:

  net/mlx5: E-Switch, Use atomic rep state to serialize state change (2019-04-30 13:32:59 -0700)

----------------------------------------------------------------
mlx5-updates-2019-04-30

mlx5 misc updates:

1) Bodong Wang and Parav Pandit (6):
   - Remove unused mlx5_query_nic_vport_vlans
   - vport macros refactoring
   - Fix vport access in E-Switch
   - Use atomic rep state to serialize state change

2) Eli Britstein (2):
   - prio tag mode support, added ACLs and replace TC vlan pop with
     vlan 0 rewrite when prio tag mode is enabled.

3) Erez Alfasi (2):
   - ethtool: Add SFF-8436 and SFF-8636 max EEPROM length definitions
   - mlx5e: ethtool, Add support for EEPROM high pages query

4) Masahiro Yamada (1):
   - remove meaningless CFLAGS_tracepoint.o

5) Maxim Mikityanskiy (1):
   - Put the common XDP code into a function

6) Tariq Toukan (2):
   - Turn on HW tunnel offload in all TIRs

7) Vlad Buslov (1):
   - Return error when trying to insert existing flower filter

----------------------------------------------------------------
Bodong Wang (4):
      net/mlx5: Remove unused mlx5_query_nic_vport_vlans
      net/mlx5: E-Switch, Use getter to access all vport array
      net/mlx5: E-Switch, Fix the check of legal vport
      net/mlx5: E-Switch, Use atomic rep state to serialize state change

Eli Britstein (2):
      net/mlx5e: ACLs for priority tag mode
      net/mlx5e: Replace TC VLAN pop with VLAN 0 rewrite in prio tag mode

Erez Alfasi (2):
      ethtool: Add SFF-8436 and SFF-8636 max EEPROM length definitions
      net/mlx5e: ethtool, Add support for EEPROM high pages query

Masahiro Yamada (1):
      net/mlx5e: remove meaningless CFLAGS_tracepoint.o

Maxim Mikityanskiy (1):
      net/mlx5e: Put the common XDP code into a function

Parav Pandit (2):
      net/mlx5: Reuse mlx5_esw_for_each_vf_vport macro in two files
      net/mlx5: Use available mlx5_vport struct

Tariq Toukan (2):
      net/mlx5e: Take common TIR context settings into a function
      net/mlx5e: Turn on HW tunnel offload in all TIRs

Vlad Buslov (1):
      net/mlx5e: Return error when trying to insert existing flower filter

 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 -
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c   |  63 +++---
 .../net/ethernet/mellanox/mlx5/core/en_ethtool.c   |   8 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  52 +++--
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |  37 ++++
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c  | 224 +++++++++----------
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |  57 +++++
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 240 ++++++++++++++++-----
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c  |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/port.c     |  40 +++-
 drivers/net/ethernet/mellanox/mlx5/core/vport.c    |  61 ------
 include/linux/mlx5/eswitch.h                       |   2 +-
 include/linux/mlx5/port.h                          |   1 +
 include/linux/mlx5/vport.h                         |   4 -
 include/uapi/linux/ethtool.h                       |   3 +
 17 files changed, 485 insertions(+), 312 deletions(-)

[net-next 01/15] net/mlx5e: Take common TIR context settings into a function

[Saeed Mahameed]

From: Tariq Toukan <tariqt@mellanox.com>

Many TIR context settings are common to different TIR types,
take them into a common function.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 49 ++++++++-----------
 1 file changed, 21 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 236cce59181d..d713ab2e7a2d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -2674,22 +2674,6 @@ static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv)
 	return err;
 }
 
-static void mlx5e_build_inner_indir_tir_ctx(struct mlx5e_priv *priv,
-					    enum mlx5e_traffic_types tt,
-					    u32 *tirc)
-{
-	MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn);
-
-	mlx5e_build_tir_ctx_lro(&priv->channels.params, tirc);
-
-	MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT);
-	MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqt.rqtn);
-	MLX5_SET(tirc, tirc, tunneled_offload_en, 0x1);
-
-	mlx5e_build_indir_tir_ctx_hash(&priv->rss_params,
-				       &tirc_default_config[tt], tirc, true);
-}
-
 static int mlx5e_set_mtu(struct mlx5_core_dev *mdev,
 			 struct mlx5e_params *params, u16 mtu)
 {
@@ -3110,32 +3094,41 @@ static void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv)
 		mlx5e_destroy_tis(priv->mdev, priv->tisn[tc]);
 }
 
-static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv,
-				      enum mlx5e_traffic_types tt,
-				      u32 *tirc)
+static void mlx5e_build_indir_tir_ctx_common(struct mlx5e_priv *priv,
+					     u32 rqtn, u32 *tirc)
 {
 	MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn);
+	MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT);
+	MLX5_SET(tirc, tirc, indirect_table, rqtn);
 
 	mlx5e_build_tir_ctx_lro(&priv->channels.params, tirc);
+}
 
-	MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT);
-	MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqt.rqtn);
-
+static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv,
+				      enum mlx5e_traffic_types tt,
+				      u32 *tirc)
+{
+	mlx5e_build_indir_tir_ctx_common(priv, priv->indir_rqt.rqtn, tirc);
 	mlx5e_build_indir_tir_ctx_hash(&priv->rss_params,
 				       &tirc_default_config[tt], tirc, false);
 }
 
 static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 rqtn, u32 *tirc)
 {
-	MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn);
-
-	mlx5e_build_tir_ctx_lro(&priv->channels.params, tirc);
-
-	MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT);
-	MLX5_SET(tirc, tirc, indirect_table, rqtn);
+	mlx5e_build_indir_tir_ctx_common(priv, rqtn, tirc);
 	MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_INVERTED_XOR8);
 }
 
+static void mlx5e_build_inner_indir_tir_ctx(struct mlx5e_priv *priv,
+					    enum mlx5e_traffic_types tt,
+					    u32 *tirc)
+{
+	mlx5e_build_indir_tir_ctx_common(priv, priv->indir_rqt.rqtn, tirc);
+	mlx5e_build_indir_tir_ctx_hash(&priv->rss_params,
+				       &tirc_default_config[tt], tirc, true);
+	MLX5_SET(tirc, tirc, tunneled_offload_en, 0x1);
+}
+
 int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc)
 {
 	struct mlx5e_tir *tir;
-- 
2.20.1

[net-next 02/15] net/mlx5e: Turn on HW tunnel offload in all TIRs

[Saeed Mahameed]

From: Tariq Toukan <tariqt@mellanox.com>

Hardware requires that all TIRs that steer traffic to the same RQ
should share identical tunneled_offload_en value.
For that, the tunneled_offload_en bit should be set/unset (according to
the HW capability) for all TIRs', not only the ones dedicated for
tunneled (inner) traffic.

Fixes: 1b223dd39162 ("net/mlx5e: Fix checksum handling for non-stripped vlan packets")
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 +
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c     | 5 ++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c      | 1 +
 drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 1 +
 4 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 7e0c3d4de108..3a183d690e23 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -240,6 +240,7 @@ struct mlx5e_params {
 	bool rx_cqe_compress_def;
 	struct net_dim_cq_moder rx_cq_moderation;
 	struct net_dim_cq_moder tx_cq_moderation;
+	bool tunneled_offload_en;
 	bool lro_en;
 	u8  tx_min_inline_mode;
 	bool vlan_strip_disable;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index d713ab2e7a2d..457cc39423f2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3100,6 +3100,8 @@ static void mlx5e_build_indir_tir_ctx_common(struct mlx5e_priv *priv,
 	MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn);
 	MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT);
 	MLX5_SET(tirc, tirc, indirect_table, rqtn);
+	MLX5_SET(tirc, tirc, tunneled_offload_en,
+		 priv->channels.params.tunneled_offload_en);
 
 	mlx5e_build_tir_ctx_lro(&priv->channels.params, tirc);
 }
@@ -3126,7 +3128,6 @@ static void mlx5e_build_inner_indir_tir_ctx(struct mlx5e_priv *priv,
 	mlx5e_build_indir_tir_ctx_common(priv, priv->indir_rqt.rqtn, tirc);
 	mlx5e_build_indir_tir_ctx_hash(&priv->rss_params,
 				       &tirc_default_config[tt], tirc, true);
-	MLX5_SET(tirc, tirc, tunneled_offload_en, 0x1);
 }
 
 int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc)
@@ -4572,6 +4573,8 @@ void mlx5e_build_nic_params(struct mlx5_core_dev *mdev,
 
 	/* RSS */
 	mlx5e_build_rss_params(rss_params, params->num_channels);
+	params->tunneled_offload_en =
+		mlx5e_tunnel_inner_ft_supported(mdev);
 }
 
 static void mlx5e_set_netdev_dev_addr(struct net_device *netdev)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 2ebca9bd5cf8..91e24f1cead8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1375,6 +1375,7 @@ static void mlx5e_build_rep_params(struct net_device *netdev)
 	mlx5e_set_rx_cq_mode_params(params, cq_period_mode);
 
 	params->num_tc                = 1;
+	params->tunneled_offload_en = false;
 
 	mlx5_query_min_inline(mdev, &params->tx_min_inline_mode);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index 9b03ae1e1e10..ada1b7c0e0b8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -68,6 +68,7 @@ static void mlx5i_build_nic_params(struct mlx5_core_dev *mdev,
 
 	params->lro_en = false;
 	params->hard_mtu = MLX5_IB_GRH_BYTES + MLX5_IPOIB_HARD_LEN;
+	params->tunneled_offload_en = false;
 }
 
 /* Called directly after IPoIB netdevice was created to initialize SW structs */
-- 
2.20.1

[net-next 03/15] net/mlx5e: ACLs for priority tag mode

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

Current ConnectX HW is unable to perform VLAN pop in TX path and VLAN
push on RX path. As a workaround, untagged packets are tagged with
VID 0x000 allowing pop/push actions to be exchanged with VLAN rewrite
actions.
Use the ingress ACL table, preceding the FDB, to push VLAN 0x000 ID tag
for untagged packets and the egress ACL table, succeeding the FDB, to
pop VLAN 0x000 ID tag.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c |  24 +--
 .../net/ethernet/mellanox/mlx5/core/eswitch.h |  12 ++
 .../mellanox/mlx5/core/eswitch_offloads.c     | 169 ++++++++++++++++++
 3 files changed, 193 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 8a67fd197b79..f0ef4ac51b45 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -922,8 +922,8 @@ static void esw_vport_change_handler(struct work_struct *work)
 	mutex_unlock(&esw->state_lock);
 }
 
-static int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
-				       struct mlx5_vport *vport)
+int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
+				struct mlx5_vport *vport)
 {
 	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
 	struct mlx5_flow_group *vlan_grp = NULL;
@@ -1006,8 +1006,8 @@ static int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
 	return err;
 }
 
-static void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw,
-					   struct mlx5_vport *vport)
+void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw,
+				    struct mlx5_vport *vport)
 {
 	if (!IS_ERR_OR_NULL(vport->egress.allowed_vlan))
 		mlx5_del_flow_rules(vport->egress.allowed_vlan);
@@ -1019,8 +1019,8 @@ static void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw,
 	vport->egress.drop_rule = NULL;
 }
 
-static void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw,
-					 struct mlx5_vport *vport)
+void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw,
+				  struct mlx5_vport *vport)
 {
 	if (IS_ERR_OR_NULL(vport->egress.acl))
 		return;
@@ -1036,8 +1036,8 @@ static void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw,
 	vport->egress.acl = NULL;
 }
 
-static int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
-					struct mlx5_vport *vport)
+int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
+				 struct mlx5_vport *vport)
 {
 	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
 	struct mlx5_core_dev *dev = esw->dev;
@@ -1168,8 +1168,8 @@ static int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
 	return err;
 }
 
-static void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw,
-					    struct mlx5_vport *vport)
+void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw,
+				     struct mlx5_vport *vport)
 {
 	if (!IS_ERR_OR_NULL(vport->ingress.drop_rule))
 		mlx5_del_flow_rules(vport->ingress.drop_rule);
@@ -1181,8 +1181,8 @@ static void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw,
 	vport->ingress.allow_rule = NULL;
 }
 
-static void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw,
-					  struct mlx5_vport *vport)
+void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw,
+				   struct mlx5_vport *vport)
 {
 	if (IS_ERR_OR_NULL(vport->ingress.acl))
 		return;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 9ecc8c9eee39..424d45cf9454 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -227,6 +227,18 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int vf_nvports,
 		      int total_nvports);
 void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw);
 int esw_offloads_init_reps(struct mlx5_eswitch *esw);
+void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw,
+				     struct mlx5_vport *vport);
+int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
+				 struct mlx5_vport *vport);
+void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw,
+				    struct mlx5_vport *vport);
+int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
+				struct mlx5_vport *vport);
+void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw,
+				  struct mlx5_vport *vport);
+void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw,
+				   struct mlx5_vport *vport);
 
 /* E-Switch API */
 int mlx5_eswitch_init(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 840630305ede..f371e79cbc9f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1601,6 +1601,167 @@ static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw)
 	mlx5_devcom_unregister_component(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
 }
 
+static int esw_vport_ingress_prio_tag_config(struct mlx5_eswitch *esw,
+					     struct mlx5_vport *vport)
+{
+	struct mlx5_core_dev *dev = esw->dev;
+	struct mlx5_flow_act flow_act = {0};
+	struct mlx5_flow_spec *spec;
+	int err = 0;
+
+	/* For prio tag mode, there is only 1 FTEs:
+	 * 1) Untagged packets - push prio tag VLAN, allow
+	 * Unmatched traffic is allowed by default
+	 */
+
+	if (!MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support))
+		return -EOPNOTSUPP;
+
+	esw_vport_cleanup_ingress_rules(esw, vport);
+
+	err = esw_vport_enable_ingress_acl(esw, vport);
+	if (err) {
+		mlx5_core_warn(esw->dev,
+			       "failed to enable prio tag ingress acl (%d) on vport[%d]\n",
+			       err, vport->vport);
+		return err;
+	}
+
+	esw_debug(esw->dev,
+		  "vport[%d] configure ingress rules\n", vport->vport);
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec) {
+		err = -ENOMEM;
+		goto out_no_mem;
+	}
+
+	/* Untagged packets - push prio tag VLAN, allow */
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag);
+	MLX5_SET(fte_match_param, spec->match_value, outer_headers.cvlan_tag, 0);
+	spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
+			  MLX5_FLOW_CONTEXT_ACTION_ALLOW;
+	flow_act.vlan[0].ethtype = ETH_P_8021Q;
+	flow_act.vlan[0].vid = 0;
+	flow_act.vlan[0].prio = 0;
+	vport->ingress.allow_rule =
+		mlx5_add_flow_rules(vport->ingress.acl, spec,
+				    &flow_act, NULL, 0);
+	if (IS_ERR(vport->ingress.allow_rule)) {
+		err = PTR_ERR(vport->ingress.allow_rule);
+		esw_warn(esw->dev,
+			 "vport[%d] configure ingress untagged allow rule, err(%d)\n",
+			 vport->vport, err);
+		vport->ingress.allow_rule = NULL;
+		goto out;
+	}
+
+out:
+	kvfree(spec);
+out_no_mem:
+	if (err)
+		esw_vport_cleanup_ingress_rules(esw, vport);
+	return err;
+}
+
+static int esw_vport_egress_prio_tag_config(struct mlx5_eswitch *esw,
+					    struct mlx5_vport *vport)
+{
+	struct mlx5_flow_act flow_act = {0};
+	struct mlx5_flow_spec *spec;
+	int err = 0;
+
+	/* For prio tag mode, there is only 1 FTEs:
+	 * 1) prio tag packets - pop the prio tag VLAN, allow
+	 * Unmatched traffic is allowed by default
+	 */
+
+	esw_vport_cleanup_egress_rules(esw, vport);
+
+	err = esw_vport_enable_egress_acl(esw, vport);
+	if (err) {
+		mlx5_core_warn(esw->dev,
+			       "failed to enable egress acl (%d) on vport[%d]\n",
+			       err, vport->vport);
+		return err;
+	}
+
+	esw_debug(esw->dev,
+		  "vport[%d] configure prio tag egress rules\n", vport->vport);
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec) {
+		err = -ENOMEM;
+		goto out_no_mem;
+	}
+
+	/* prio tag vlan rule - pop it so VF receives untagged packets */
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_value, outer_headers.cvlan_tag);
+	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.first_vid);
+	MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, 0);
+
+	spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_VLAN_POP |
+			  MLX5_FLOW_CONTEXT_ACTION_ALLOW;
+	vport->egress.allowed_vlan =
+		mlx5_add_flow_rules(vport->egress.acl, spec,
+				    &flow_act, NULL, 0);
+	if (IS_ERR(vport->egress.allowed_vlan)) {
+		err = PTR_ERR(vport->egress.allowed_vlan);
+		esw_warn(esw->dev,
+			 "vport[%d] configure egress pop prio tag vlan rule failed, err(%d)\n",
+			 vport->vport, err);
+		vport->egress.allowed_vlan = NULL;
+		goto out;
+	}
+
+out:
+	kvfree(spec);
+out_no_mem:
+	if (err)
+		esw_vport_cleanup_egress_rules(esw, vport);
+	return err;
+}
+
+static int esw_prio_tag_acls_config(struct mlx5_eswitch *esw, int nvports)
+{
+	int i, j;
+	int err;
+
+	mlx5_esw_for_each_vf_vport(esw, i, nvports) {
+		err = esw_vport_ingress_prio_tag_config(esw, &esw->vports[i]);
+		if (err)
+			goto err_ingress;
+		err = esw_vport_egress_prio_tag_config(esw, &esw->vports[i]);
+		if (err)
+			goto err_egress;
+	}
+
+	return 0;
+
+err_egress:
+	esw_vport_disable_ingress_acl(esw, &esw->vports[i]);
+err_ingress:
+	mlx5_esw_for_each_vf_vport_reverse(esw, j, i - 1) {
+		esw_vport_disable_egress_acl(esw, &esw->vports[j]);
+		esw_vport_disable_ingress_acl(esw, &esw->vports[j]);
+	}
+
+	return err;
+}
+
+static void esw_prio_tag_acls_cleanup(struct mlx5_eswitch *esw)
+{
+	int i;
+
+	mlx5_esw_for_each_vf_vport(esw, i, esw->nvports) {
+		esw_vport_disable_egress_acl(esw, &esw->vports[i]);
+		esw_vport_disable_ingress_acl(esw, &esw->vports[i]);
+	}
+}
+
 static int esw_offloads_steering_init(struct mlx5_eswitch *esw, int nvports)
 {
 	int err;
@@ -1608,6 +1769,12 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw, int nvports)
 	memset(&esw->fdb_table.offloads, 0, sizeof(struct offloads_fdb));
 	mutex_init(&esw->fdb_table.offloads.fdb_prio_lock);
 
+	if (MLX5_CAP_GEN(esw->dev, prio_tag_required)) {
+		err = esw_prio_tag_acls_config(esw, nvports);
+		if (err)
+			return err;
+	}
+
 	err = esw_create_offloads_fdb_tables(esw, nvports);
 	if (err)
 		return err;
@@ -1636,6 +1803,8 @@ static void esw_offloads_steering_cleanup(struct mlx5_eswitch *esw)
 	esw_destroy_vport_rx_group(esw);
 	esw_destroy_offloads_table(esw);
 	esw_destroy_offloads_fdb_tables(esw);
+	if (MLX5_CAP_GEN(esw->dev, prio_tag_required))
+		esw_prio_tag_acls_cleanup(esw);
 }
 
 static void esw_host_params_event_handler(struct work_struct *work)
-- 
2.20.1

[net-next 04/15] net/mlx5e: Replace TC VLAN pop with VLAN 0 rewrite in prio tag mode

[Saeed Mahameed]

From: Eli Britstein <elibr@mellanox.com>

Current ConnectX HW is unable to perform VLAN pop in TX path and VLAN
push on RX path. To workaround that limitation untagged packets are
tagged with VLAN ID 0x000 (priority tag) and pop/push actions are
replaced by VLAN re-write actions (which are supported by the HW).
Replace TC VLAN pop action with a VLAN priority tag header rewrite.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 36 +++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 72891cc7f32a..c79db55f8a76 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2436,6 +2436,30 @@ static int add_vlan_rewrite_action(struct mlx5e_priv *priv, int namespace,
 	return err;
 }
 
+static int
+add_vlan_prio_tag_rewrite_action(struct mlx5e_priv *priv,
+				 struct mlx5e_tc_flow_parse_attr *parse_attr,
+				 struct pedit_headers_action *hdrs,
+				 u32 *action, struct netlink_ext_ack *extack)
+{
+	const struct flow_action_entry prio_tag_act = {
+		.vlan.vid = 0,
+		.vlan.prio =
+			MLX5_GET(fte_match_set_lyr_2_4,
+				 get_match_headers_value(*action,
+							 &parse_attr->spec),
+				 first_prio) &
+			MLX5_GET(fte_match_set_lyr_2_4,
+				 get_match_headers_criteria(*action,
+							    &parse_attr->spec),
+				 first_prio),
+	};
+
+	return add_vlan_rewrite_action(priv, MLX5_FLOW_NAMESPACE_FDB,
+				       &prio_tag_act, parse_attr, hdrs, action,
+				       extack);
+}
+
 static int parse_tc_nic_actions(struct mlx5e_priv *priv,
 				struct flow_action *flow_action,
 				struct mlx5e_tc_flow_parse_attr *parse_attr,
@@ -2947,6 +2971,18 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 		}
 	}
 
+	if (MLX5_CAP_GEN(esw->dev, prio_tag_required) &&
+	    action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) {
+		/* For prio tag mode, replace vlan pop with rewrite vlan prio
+		 * tag rewrite.
+		 */
+		action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+		err = add_vlan_prio_tag_rewrite_action(priv, parse_attr, hdrs,
+						       &action, extack);
+		if (err)
+			return err;
+	}
+
 	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_FDB,
-- 
2.20.1

[net-next 05/15] net/mlx5e: Return error when trying to insert existing flower filter

[Saeed Mahameed]

From: Vlad Buslov <vladbu@mellanox.com>

With unlocked TC it is possible to have spurious deletes and inserts of
same filter. TC layer needs drivers to always return error when flow
insertion failed in order to correctly calculate "in_hw_count" for each
filter. Fix mlx5e_configure_flower() to return -EEXIST when TC tries to
insert a filter that is already provisioned to the driver.

Signed-off-by: Vlad Buslov <vladbu@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 | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index c79db55f8a76..122f457091a2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3364,6 +3364,7 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
 		netdev_warn_once(priv->netdev,
 				 "flow cookie %lx already exists, ignoring\n",
 				 f->cookie);
+		err = -EEXIST;
 		goto out;
 	}
 
-- 
2.20.1

[net-next 06/15] ethtool: Add SFF-8436 and SFF-8636 max EEPROM length definitions

[Saeed Mahameed]

From: Erez Alfasi <ereza@mellanox.com>

Added max EEPROM length defines for ethtool usage:
 #define ETH_MODULE_SFF_8636_MAX_LEN     640
 #define ETH_MODULE_SFF_8436_MAX_LEN     640

These definitions are exists in ethtool and used to
determine the EEPROM data length when reading high
pages as well.

For example, SFF-8636 EEPROM data from page 03h
needs to be stored at data[512] - data[639].

Signed-off-by: Erez Alfasi <ereza@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/uapi/linux/ethtool.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 818ad368b586..3534ce157ae9 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -1712,6 +1712,9 @@ static inline int ethtool_validate_duplex(__u8 duplex)
 #define ETH_MODULE_SFF_8436		0x4
 #define ETH_MODULE_SFF_8436_LEN		256
 
+#define ETH_MODULE_SFF_8636_MAX_LEN     640
+#define ETH_MODULE_SFF_8436_MAX_LEN     640
+
 /* Reset flags */
 /* The reset() operation must clear the flags for the components which
  * were actually reset.  On successful return, the flags indicate the
-- 
2.20.1

[net-next 07/15] net/mlx5e: ethtool, Add support for EEPROM high pages query

[Saeed Mahameed]

From: Erez Alfasi <ereza@mellanox.com>

Add the support to read additional EEPROM information from high pages.
Information for modules such as SFF-8436 and SFF-8636:
 1) Application select table
 2) User writable EEPROM
 3) Thresholds and alarms

Signed-off-by: Erez Alfasi <ereza@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/en_ethtool.c  |  8 ++--
 .../net/ethernet/mellanox/mlx5/core/port.c    | 40 ++++++++++++++++---
 include/linux/mlx5/port.h                     |  1 +
 3 files changed, 39 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 78dc8fe2a83c..7efaa58ae034 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -1561,7 +1561,7 @@ static int mlx5e_get_module_info(struct net_device *netdev,
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 	struct mlx5_core_dev *dev = priv->mdev;
 	int size_read = 0;
-	u8 data[4];
+	u8 data[4] = {0};
 
 	size_read = mlx5_query_module_eeprom(dev, 0, 2, data);
 	if (size_read < 2)
@@ -1571,17 +1571,17 @@ static int mlx5e_get_module_info(struct net_device *netdev,
 	switch (data[0]) {
 	case MLX5_MODULE_ID_QSFP:
 		modinfo->type       = ETH_MODULE_SFF_8436;
-		modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN;
 		break;
 	case MLX5_MODULE_ID_QSFP_PLUS:
 	case MLX5_MODULE_ID_QSFP28:
 		/* data[1] = revision id */
 		if (data[0] == MLX5_MODULE_ID_QSFP28 || data[1] >= 0x3) {
 			modinfo->type       = ETH_MODULE_SFF_8636;
-			modinfo->eeprom_len = ETH_MODULE_SFF_8636_LEN;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN;
 		} else {
 			modinfo->type       = ETH_MODULE_SFF_8436;
-			modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN;
+			modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN;
 		}
 		break;
 	case MLX5_MODULE_ID_SFP:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index 361468e0435d..cc262b30aed5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -293,15 +293,36 @@ static int mlx5_query_module_num(struct mlx5_core_dev *dev, int *module_num)
 	return 0;
 }
 
+static int mlx5_eeprom_page(int offset)
+{
+	if (offset < MLX5_EEPROM_PAGE_LENGTH)
+		/* Addresses between 0-255 - page 00 */
+		return 0;
+
+	/* Addresses between 256 - 639 belongs to pages 01, 02 and 03
+	 * For example, offset = 400 belongs to page 02:
+	 * 1 + ((400 - 256)/128) = 2
+	 */
+	return 1 + ((offset - MLX5_EEPROM_PAGE_LENGTH) /
+		    MLX5_EEPROM_HIGH_PAGE_LENGTH);
+}
+
+static int mlx5_eeprom_high_page_offset(int page_num)
+{
+	if (!page_num) /* Page 0 always start from low page */
+		return 0;
+
+	/* High page */
+	return page_num * MLX5_EEPROM_HIGH_PAGE_LENGTH;
+}
+
 int mlx5_query_module_eeprom(struct mlx5_core_dev *dev,
 			     u16 offset, u16 size, u8 *data)
 {
+	int module_num, page_num, status, err;
 	u32 out[MLX5_ST_SZ_DW(mcia_reg)];
 	u32 in[MLX5_ST_SZ_DW(mcia_reg)];
-	int module_num;
 	u16 i2c_addr;
-	int status;
-	int err;
 	void *ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0);
 
 	err = mlx5_query_module_num(dev, &module_num);
@@ -311,8 +332,15 @@ int mlx5_query_module_eeprom(struct mlx5_core_dev *dev,
 	memset(in, 0, sizeof(in));
 	size = min_t(int, size, MLX5_EEPROM_MAX_BYTES);
 
-	if (offset < MLX5_EEPROM_PAGE_LENGTH &&
-	    offset + size > MLX5_EEPROM_PAGE_LENGTH)
+	/* Get the page number related to the given offset */
+	page_num = mlx5_eeprom_page(offset);
+
+	/* Set the right offset according to the page number,
+	 * For page_num > 0, relative offset is always >= 128 (high page).
+	 */
+	offset -= mlx5_eeprom_high_page_offset(page_num);
+
+	if (offset + size > MLX5_EEPROM_PAGE_LENGTH)
 		/* Cross pages read, read until offset 256 in low page */
 		size -= offset + size - MLX5_EEPROM_PAGE_LENGTH;
 
@@ -321,7 +349,7 @@ int mlx5_query_module_eeprom(struct mlx5_core_dev *dev,
 	MLX5_SET(mcia_reg, in, l, 0);
 	MLX5_SET(mcia_reg, in, module, module_num);
 	MLX5_SET(mcia_reg, in, i2c_device_address, i2c_addr);
-	MLX5_SET(mcia_reg, in, page_number, 0);
+	MLX5_SET(mcia_reg, in, page_number, page_num);
 	MLX5_SET(mcia_reg, in, device_address, offset);
 	MLX5_SET(mcia_reg, in, size, size);
 
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 64e78394fc9c..de9a272c9f3d 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -60,6 +60,7 @@ enum mlx5_an_status {
 #define MLX5_I2C_ADDR_LOW		0x50
 #define MLX5_I2C_ADDR_HIGH		0x51
 #define MLX5_EEPROM_PAGE_LENGTH		256
+#define MLX5_EEPROM_HIGH_PAGE_LENGTH	128
 
 enum mlx5e_link_mode {
 	MLX5E_1000BASE_CX_SGMII	 = 0,
-- 
2.20.1

[net-next 08/15] net/mlx5e: Put the common XDP code into a function

[Saeed Mahameed]

From: Maxim Mikityanskiy <maximmi@mellanox.com>

The same code that returns XDP frames and releases pages is used both in
mlx5e_poll_xdpsq_cq and mlx5e_free_xdpsq_descs. Create a function that
cleans up an MPWQE.

Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en/xdp.c  | 63 ++++++++-----------
 1 file changed, 27 insertions(+), 36 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 399957104f9d..eb8ef78e5626 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -275,12 +275,33 @@ static bool mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_info *
 	return true;
 }
 
+static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq,
+				  struct mlx5e_xdp_wqe_info *wi,
+				  struct mlx5e_rq *rq,
+				  bool recycle)
+{
+	struct mlx5e_xdp_info_fifo *xdpi_fifo = &sq->db.xdpi_fifo;
+	u16 i;
+
+	for (i = 0; i < wi->num_pkts; i++) {
+		struct mlx5e_xdp_info xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+
+		if (rq) {
+			/* XDP_TX */
+			mlx5e_page_release(rq, &xdpi.di, recycle);
+		} else {
+			/* XDP_REDIRECT */
+			dma_unmap_single(sq->pdev, xdpi.dma_addr,
+					 xdpi.xdpf->len, DMA_TO_DEVICE);
+			xdp_return_frame(xdpi.xdpf);
+		}
+	}
+}
+
 bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq, struct mlx5e_rq *rq)
 {
-	struct mlx5e_xdp_info_fifo *xdpi_fifo;
 	struct mlx5e_xdpsq *sq;
 	struct mlx5_cqe64 *cqe;
-	bool is_redirect;
 	u16 sqcc;
 	int i;
 
@@ -293,9 +314,6 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq, struct mlx5e_rq *rq)
 	if (!cqe)
 		return false;
 
-	is_redirect = !rq;
-	xdpi_fifo = &sq->db.xdpi_fifo;
-
 	/* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
 	 * otherwise a cq overrun may occur
 	 */
@@ -317,7 +335,7 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq, struct mlx5e_rq *rq)
 
 		do {
 			struct mlx5e_xdp_wqe_info *wi;
-			u16 ci, j;
+			u16 ci;
 
 			last_wqe = (sqcc == wqe_counter);
 			ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc);
@@ -325,19 +343,7 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq, struct mlx5e_rq *rq)
 
 			sqcc += wi->num_wqebbs;
 
-			for (j = 0; j < wi->num_pkts; j++) {
-				struct mlx5e_xdp_info xdpi =
-					mlx5e_xdpi_fifo_pop(xdpi_fifo);
-
-				if (is_redirect) {
-					dma_unmap_single(sq->pdev, xdpi.dma_addr,
-							 xdpi.xdpf->len, DMA_TO_DEVICE);
-					xdp_return_frame(xdpi.xdpf);
-				} else {
-					/* Recycle RX page */
-					mlx5e_page_release(rq, &xdpi.di, true);
-				}
-			}
+			mlx5e_free_xdpsq_desc(sq, wi, rq, true);
 		} while (!last_wqe);
 	} while ((++i < MLX5E_TX_CQ_POLL_BUDGET) && (cqe = mlx5_cqwq_get_cqe(&cq->wq)));
 
@@ -354,31 +360,16 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq, struct mlx5e_rq *rq)
 
 void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq)
 {
-	struct mlx5e_xdp_info_fifo *xdpi_fifo = &sq->db.xdpi_fifo;
-	bool is_redirect = !rq;
-
 	while (sq->cc != sq->pc) {
 		struct mlx5e_xdp_wqe_info *wi;
-		u16 ci, i;
+		u16 ci;
 
 		ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->cc);
 		wi = &sq->db.wqe_info[ci];
 
 		sq->cc += wi->num_wqebbs;
 
-		for (i = 0; i < wi->num_pkts; i++) {
-			struct mlx5e_xdp_info xdpi =
-				mlx5e_xdpi_fifo_pop(xdpi_fifo);
-
-			if (is_redirect) {
-				dma_unmap_single(sq->pdev, xdpi.dma_addr,
-						 xdpi.xdpf->len, DMA_TO_DEVICE);
-				xdp_return_frame(xdpi.xdpf);
-			} else {
-				/* Recycle RX page */
-				mlx5e_page_release(rq, &xdpi.di, false);
-			}
-		}
+		mlx5e_free_xdpsq_desc(sq, wi, rq, false);
 	}
 }
 
-- 
2.20.1

[net-next 09/15] net/mlx5e: remove meaningless CFLAGS_tracepoint.o

[Saeed Mahameed]

From: Masahiro Yamada <yamada.masahiro@socionext.com>

CFLAGS_tracepoint.o specifies CFLAGS for compiling tracepoint.c but
it does not exist under drivers/net/ethernet/mellanox/mlx5/core/.

CFLAGS_tracepoint.o is unused.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 12278f024ffc..243368dc23db 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -58,5 +58,3 @@ mlx5_core-$(CONFIG_MLX5_EN_IPSEC) += en_accel/ipsec.o en_accel/ipsec_rxtx.o \
 				     en_accel/ipsec_stats.o
 
 mlx5_core-$(CONFIG_MLX5_EN_TLS) += en_accel/tls.o en_accel/tls_rxtx.o en_accel/tls_stats.o
-
-CFLAGS_tracepoint.o := -I$(src)
-- 
2.20.1

[net-next 10/15] net/mlx5: Remove unused mlx5_query_nic_vport_vlans

[Saeed Mahameed]

From: Bodong Wang <bodong@mellanox.com>

mlx5_query_nic_vport_vlans() is not used anymore. Hence remove it.
This patch doesn't change any functionality.

Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/vport.c   | 61 -------------------
 include/linux/mlx5/vport.h                    |  4 --
 2 files changed, 65 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
index ef95feca9961..95cdc8cbcba4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
@@ -371,67 +371,6 @@ int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev,
 }
 EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mac_list);
 
-int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev,
-			       u16 vport,
-			       u16 vlans[],
-			       int *size)
-{
-	u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)];
-	void *nic_vport_ctx;
-	int req_list_size;
-	int max_list_size;
-	int out_sz;
-	void *out;
-	int err;
-	int i;
-
-	req_list_size = *size;
-	max_list_size = 1 << MLX5_CAP_GEN(dev, log_max_vlan_list);
-	if (req_list_size > max_list_size) {
-		mlx5_core_warn(dev, "Requested list size (%d) > (%d) max list size\n",
-			       req_list_size, max_list_size);
-		req_list_size = max_list_size;
-	}
-
-	out_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) +
-			req_list_size * MLX5_ST_SZ_BYTES(vlan_layout);
-
-	memset(in, 0, sizeof(in));
-	out = kzalloc(out_sz, GFP_KERNEL);
-	if (!out)
-		return -ENOMEM;
-
-	MLX5_SET(query_nic_vport_context_in, in, opcode,
-		 MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT);
-	MLX5_SET(query_nic_vport_context_in, in, allowed_list_type,
-		 MLX5_NVPRT_LIST_TYPE_VLAN);
-	MLX5_SET(query_nic_vport_context_in, in, vport_number, vport);
-
-	if (vport)
-		MLX5_SET(query_nic_vport_context_in, in, other_vport, 1);
-
-	err = mlx5_cmd_exec(dev, in, sizeof(in), out, out_sz);
-	if (err)
-		goto out;
-
-	nic_vport_ctx = MLX5_ADDR_OF(query_nic_vport_context_out, out,
-				     nic_vport_context);
-	req_list_size = MLX5_GET(nic_vport_context, nic_vport_ctx,
-				 allowed_list_size);
-
-	*size = req_list_size;
-	for (i = 0; i < req_list_size; i++) {
-		void *vlan_addr = MLX5_ADDR_OF(nic_vport_context,
-					       nic_vport_ctx,
-					       current_uc_mac_address[i]);
-		vlans[i] = MLX5_GET(vlan_layout, vlan_addr, vlan);
-	}
-out:
-	kfree(out);
-	return err;
-}
-EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_vlans);
-
 int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev,
 				u16 vlans[],
 				int list_size)
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 0eef548b9946..3d1c6cdbbba7 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -118,10 +118,6 @@ int mlx5_modify_nic_vport_promisc(struct mlx5_core_dev *mdev,
 				  int promisc_uc,
 				  int promisc_mc,
 				  int promisc_all);
-int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev,
-			       u16 vport,
-			       u16 vlans[],
-			       int *size);
 int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev,
 				u16 vlans[],
 				int list_size);
-- 
2.20.1

[net-next 11/15] net/mlx5: Reuse mlx5_esw_for_each_vf_vport macro in two files

[Saeed Mahameed]

From: Parav Pandit <parav@mellanox.com>

Currently mlx5_esw_for_each_vf_vport iterates over mlx5_vport entries in
eswitch.c
Same macro in eswitch_offloads.c iterates over vport number in
eswitch_offloads.c

Instead of duplicate macro names, to avoid confusion and to reuse the
same macro in both files, move it to eswitch.h.

To iterate over vport numbers where there is no need to iterate over
mlx5_vport, but only a vport number is needed, rename those macros in
eswitch_offloads.c to mlx5_esw_for_each_vf_num_vport*.

While at it, keep all vport and vport rep iterators together.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c | 13 -----
 .../net/ethernet/mellanox/mlx5/core/eswitch.h | 42 ++++++++++++++
 .../mellanox/mlx5/core/eswitch_offloads.c     | 55 ++++++-------------
 3 files changed, 58 insertions(+), 52 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index f0ef4ac51b45..01dc89e9f91d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -72,19 +72,6 @@ static void esw_cleanup_vepa_rules(struct mlx5_eswitch *esw);
 			    MC_ADDR_CHANGE | \
 			    PROMISC_CHANGE)
 
-/* The vport getter/iterator are only valid after esw->total_vports
- * and vport->vport are initialized in mlx5_eswitch_init.
- */
-#define mlx5_esw_for_all_vports(esw, i, vport)			\
-	for ((i) = MLX5_VPORT_PF;				\
-	     (vport) = &(esw)->vports[i],			\
-	     (i) < (esw)->total_vports; (i)++)
-
-#define mlx5_esw_for_each_vf_vport(esw, i, vport, nvfs)	\
-	for ((i) = MLX5_VPORT_FIRST_VF;				\
-	     (vport) = &(esw)->vports[i],			\
-	     (i) <= (nvfs); (i)++)
-
 static struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
 						 u16 vport_num)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 424d45cf9454..fc512a5d0c4c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -446,6 +446,48 @@ static inline int mlx5_eswitch_index_to_vport_num(struct mlx5_eswitch *esw,
 /* TODO: This mlx5e_tc function shouldn't be called by eswitch */
 void mlx5e_tc_clean_fdb_peer_flows(struct mlx5_eswitch *esw);
 
+/* The vport getter/iterator are only valid after esw->total_vports
+ * and vport->vport are initialized in mlx5_eswitch_init.
+ */
+#define mlx5_esw_for_all_vports(esw, i, vport)		\
+	for ((i) = MLX5_VPORT_PF;			\
+	     (vport) = &(esw)->vports[i],		\
+	     (i) < (esw)->total_vports; (i)++)
+
+#define mlx5_esw_for_each_vf_vport(esw, i, vport, nvfs)	\
+	for ((i) = MLX5_VPORT_FIRST_VF;			\
+	     (vport) = &(esw)->vports[(i)],		\
+	     (i) <= (nvfs); (i)++)
+
+#define mlx5_esw_for_each_vf_vport_reverse(esw, i, vport, nvfs)	\
+	for ((i) = (nvfs);					\
+	     (vport) = &(esw)->vports[(i)],			\
+	     (i) >= MLX5_VPORT_FIRST_VF; (i)--)
+
+/* The rep getter/iterator are only valid after esw->total_vports
+ * and vport->vport are initialized in mlx5_eswitch_init.
+ */
+#define mlx5_esw_for_all_reps(esw, i, rep)			\
+	for ((i) = MLX5_VPORT_PF;				\
+	     (rep) = &(esw)->offloads.vport_reps[i],		\
+	     (i) < (esw)->total_vports; (i)++)
+
+#define mlx5_esw_for_each_vf_rep(esw, i, rep, nvfs)		\
+	for ((i) = MLX5_VPORT_FIRST_VF;				\
+	     (rep) = &(esw)->offloads.vport_reps[i],		\
+	     (i) <= (nvfs); (i)++)
+
+#define mlx5_esw_for_each_vf_rep_reverse(esw, i, rep, nvfs)	\
+	for ((i) = (nvfs);					\
+	     (rep) = &(esw)->offloads.vport_reps[i],		\
+	     (i) >= MLX5_VPORT_FIRST_VF; (i)--)
+
+#define mlx5_esw_for_each_vf_vport_num(esw, vport, nvfs)	\
+	for ((vport) = MLX5_VPORT_FIRST_VF; (vport) <= (nvfs); (vport)++)
+
+#define mlx5_esw_for_each_vf_vport_num_reverse(esw, vport, nvfs)	\
+	for ((vport) = (nvfs); (vport) >= MLX5_VPORT_FIRST_VF; (vport)--)
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index f371e79cbc9f..e88feaa293f6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -54,32 +54,6 @@
 
 #define UPLINK_REP_INDEX 0
 
-/* The rep getter/iterator are only valid after esw->total_vports
- * and vport->vport are initialized in mlx5_eswitch_init.
- */
-#define mlx5_esw_for_all_reps(esw, i, rep)			\
-	for ((i) = MLX5_VPORT_PF;				\
-	     (rep) = &(esw)->offloads.vport_reps[i],		\
-	     (i) < (esw)->total_vports; (i)++)
-
-#define mlx5_esw_for_each_vf_rep(esw, i, rep, nvfs)		\
-	for ((i) = MLX5_VPORT_FIRST_VF;				\
-	     (rep) = &(esw)->offloads.vport_reps[i],		\
-	     (i) <= (nvfs); (i)++)
-
-#define mlx5_esw_for_each_vf_rep_reverse(esw, i, rep, nvfs)	\
-	for ((i) = (nvfs);					\
-	     (rep) = &(esw)->offloads.vport_reps[i],		\
-	     (i) >= MLX5_VPORT_FIRST_VF; (i)--)
-
-#define mlx5_esw_for_each_vf_vport(esw, vport, nvfs)		\
-	for ((vport) = MLX5_VPORT_FIRST_VF;			\
-	     (vport) <= (nvfs); (vport)++)
-
-#define mlx5_esw_for_each_vf_vport_reverse(esw, vport, nvfs)	\
-	for ((vport) = (nvfs);					\
-	     (vport) >= MLX5_VPORT_FIRST_VF; (vport)--)
-
 static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw,
 						     u16 vport_num)
 {
@@ -659,7 +633,7 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
 		flows[mlx5_eswitch_ecpf_idx(esw)] = flow;
 	}
 
-	mlx5_esw_for_each_vf_vport(esw, i, mlx5_core_max_vfs(esw->dev)) {
+	mlx5_esw_for_each_vf_vport_num(esw, i, mlx5_core_max_vfs(esw->dev)) {
 		MLX5_SET(fte_match_set_misc, misc, source_port, i);
 		flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
 					   spec, &flow_act, &dest, 1);
@@ -677,7 +651,7 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
 
 add_vf_flow_err:
 	nvports = --i;
-	mlx5_esw_for_each_vf_vport_reverse(esw, i, nvports)
+	mlx5_esw_for_each_vf_vport_num_reverse(esw, i, nvports)
 		mlx5_del_flow_rules(flows[i]);
 
 	if (mlx5_ecpf_vport_exists(esw->dev))
@@ -700,7 +674,8 @@ static void esw_del_fdb_peer_miss_rules(struct mlx5_eswitch *esw)
 
 	flows = esw->fdb_table.offloads.peer_miss_rules;
 
-	mlx5_esw_for_each_vf_vport_reverse(esw, i, mlx5_core_max_vfs(esw->dev))
+	mlx5_esw_for_each_vf_vport_num_reverse(esw, i,
+					       mlx5_core_max_vfs(esw->dev))
 		mlx5_del_flow_rules(flows[i]);
 
 	if (mlx5_ecpf_vport_exists(esw->dev))
@@ -1727,14 +1702,15 @@ static int esw_vport_egress_prio_tag_config(struct mlx5_eswitch *esw,
 
 static int esw_prio_tag_acls_config(struct mlx5_eswitch *esw, int nvports)
 {
+	struct mlx5_vport *vport = NULL;
 	int i, j;
 	int err;
 
-	mlx5_esw_for_each_vf_vport(esw, i, nvports) {
-		err = esw_vport_ingress_prio_tag_config(esw, &esw->vports[i]);
+	mlx5_esw_for_each_vf_vport(esw, i, vport, nvports) {
+		err = esw_vport_ingress_prio_tag_config(esw, vport);
 		if (err)
 			goto err_ingress;
-		err = esw_vport_egress_prio_tag_config(esw, &esw->vports[i]);
+		err = esw_vport_egress_prio_tag_config(esw, vport);
 		if (err)
 			goto err_egress;
 	}
@@ -1742,11 +1718,11 @@ static int esw_prio_tag_acls_config(struct mlx5_eswitch *esw, int nvports)
 	return 0;
 
 err_egress:
-	esw_vport_disable_ingress_acl(esw, &esw->vports[i]);
+	esw_vport_disable_ingress_acl(esw, vport);
 err_ingress:
-	mlx5_esw_for_each_vf_vport_reverse(esw, j, i - 1) {
-		esw_vport_disable_egress_acl(esw, &esw->vports[j]);
-		esw_vport_disable_ingress_acl(esw, &esw->vports[j]);
+	mlx5_esw_for_each_vf_vport_reverse(esw, j, vport, i - 1) {
+		esw_vport_disable_egress_acl(esw, vport);
+		esw_vport_disable_ingress_acl(esw, vport);
 	}
 
 	return err;
@@ -1754,11 +1730,12 @@ static int esw_prio_tag_acls_config(struct mlx5_eswitch *esw, int nvports)
 
 static void esw_prio_tag_acls_cleanup(struct mlx5_eswitch *esw)
 {
+	struct mlx5_vport *vport;
 	int i;
 
-	mlx5_esw_for_each_vf_vport(esw, i, esw->nvports) {
-		esw_vport_disable_egress_acl(esw, &esw->vports[i]);
-		esw_vport_disable_ingress_acl(esw, &esw->vports[i]);
+	mlx5_esw_for_each_vf_vport(esw, i, vport, esw->nvports) {
+		esw_vport_disable_egress_acl(esw, vport);
+		esw_vport_disable_ingress_acl(esw, vport);
 	}
 }
 
-- 
2.20.1

[net-next 12/15] net/mlx5: Use available mlx5_vport struct

[Saeed Mahameed]

From: Parav Pandit <parav@mellanox.com>

Several functions need to access mlx5_vport and vport_num.
When these functions are called, caller already has mlx5_vport*
available.
Hence pass such mlx5_vport pointer.

This is preparation patch to add error checks to
mlx5_eswitch_get_vport() and to return error status.
By doing so, reduce places where error check of mlx5_eswitch_get_vport()
can be avoided.

While doing such change, mlx5_eswitch_query_vport_drop_stats() gets
corrected to work on vport, instead of vport_idx.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c | 109 ++++++++----------
 1 file changed, 51 insertions(+), 58 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 01dc89e9f91d..41afc7bf8bd8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -631,9 +631,8 @@ static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
 
 /* Apply vport UC/MC list to HW l2 table and FDB table */
 static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw,
-				      u16 vport_num, int list_type)
+				      struct mlx5_vport *vport, int list_type)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
 	vport_addr_action vport_addr_add;
 	vport_addr_action vport_addr_del;
@@ -666,9 +665,8 @@ static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw,
 
 /* Sync vport UC/MC list from vport context */
 static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
-				       u16 vport_num, int list_type)
+				       struct mlx5_vport *vport, int list_type)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
 	u8 (*mac_list)[ETH_ALEN];
 	struct l2addr_node *node;
@@ -697,12 +695,12 @@ static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
 	if (!vport->enabled)
 		goto out;
 
-	err = mlx5_query_nic_vport_mac_list(esw->dev, vport_num, list_type,
+	err = mlx5_query_nic_vport_mac_list(esw->dev, vport->vport, list_type,
 					    mac_list, &size);
 	if (err)
 		goto out;
 	esw_debug(esw->dev, "vport[%d] context update %s list size (%d)\n",
-		  vport_num, is_uc ? "UC" : "MC", size);
+		  vport->vport, is_uc ? "UC" : "MC", size);
 
 	for (i = 0; i < size; i++) {
 		if (is_uc && !is_valid_ether_addr(mac_list[i]))
@@ -740,10 +738,10 @@ static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
 		if (!addr) {
 			esw_warn(esw->dev,
 				 "Failed to add MAC(%pM) to vport[%d] DB\n",
-				 mac_list[i], vport_num);
+				 mac_list[i], vport->vport);
 			continue;
 		}
-		addr->vport = vport_num;
+		addr->vport = vport->vport;
 		addr->action = MLX5_ACTION_ADD;
 	}
 out:
@@ -753,9 +751,9 @@ static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
 /* Sync vport UC/MC list from vport context
  * Must be called after esw_update_vport_addr_list
  */
-static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw, u16 vport_num)
+static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw,
+					struct mlx5_vport *vport)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	struct l2addr_node *node;
 	struct vport_addr *addr;
 	struct hlist_head *hash;
@@ -778,20 +776,20 @@ static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw, u16 vport_num)
 		if (!addr) {
 			esw_warn(esw->dev,
 				 "Failed to add allmulti MAC(%pM) to vport[%d] DB\n",
-				 mac, vport_num);
+				 mac, vport->vport);
 			continue;
 		}
-		addr->vport = vport_num;
+		addr->vport = vport->vport;
 		addr->action = MLX5_ACTION_ADD;
 		addr->mc_promisc = true;
 	}
 }
 
 /* Apply vport rx mode to HW FDB table */
-static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num,
+static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw,
+				    struct mlx5_vport *vport,
 				    bool promisc, bool mc_promisc)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	struct esw_mc_addr *allmulti_addr = &esw->mc_promisc;
 
 	if (IS_ERR_OR_NULL(vport->allmulti_rule) != mc_promisc)
@@ -799,7 +797,7 @@ static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num,
 
 	if (mc_promisc) {
 		vport->allmulti_rule =
-				esw_fdb_set_vport_allmulti_rule(esw, vport_num);
+			esw_fdb_set_vport_allmulti_rule(esw, vport->vport);
 		if (!allmulti_addr->uplink_rule)
 			allmulti_addr->uplink_rule =
 				esw_fdb_set_vport_allmulti_rule(esw,
@@ -822,8 +820,8 @@ static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num,
 		return;
 
 	if (promisc) {
-		vport->promisc_rule = esw_fdb_set_vport_promisc_rule(esw,
-								     vport_num);
+		vport->promisc_rule =
+			esw_fdb_set_vport_promisc_rule(esw, vport->vport);
 	} else if (vport->promisc_rule) {
 		mlx5_del_flow_rules(vport->promisc_rule);
 		vport->promisc_rule = NULL;
@@ -831,23 +829,23 @@ static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num,
 }
 
 /* Sync vport rx mode from vport context */
-static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num)
+static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw,
+				     struct mlx5_vport *vport)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	int promisc_all = 0;
 	int promisc_uc = 0;
 	int promisc_mc = 0;
 	int err;
 
 	err = mlx5_query_nic_vport_promisc(esw->dev,
-					   vport_num,
+					   vport->vport,
 					   &promisc_uc,
 					   &promisc_mc,
 					   &promisc_all);
 	if (err)
 		return;
 	esw_debug(esw->dev, "vport[%d] context update rx mode promisc_all=%d, all_multi=%d\n",
-		  vport_num, promisc_all, promisc_mc);
+		  vport->vport, promisc_all, promisc_mc);
 
 	if (!vport->info.trusted || !vport->enabled) {
 		promisc_uc = 0;
@@ -855,7 +853,7 @@ static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num)
 		promisc_all = 0;
 	}
 
-	esw_apply_vport_rx_mode(esw, vport_num, promisc_all,
+	esw_apply_vport_rx_mode(esw, vport, promisc_all,
 				(promisc_all || promisc_mc));
 }
 
@@ -870,27 +868,21 @@ static void esw_vport_change_handle_locked(struct mlx5_vport *vport)
 		  vport->vport, mac);
 
 	if (vport->enabled_events & UC_ADDR_CHANGE) {
-		esw_update_vport_addr_list(esw, vport->vport,
-					   MLX5_NVPRT_LIST_TYPE_UC);
-		esw_apply_vport_addr_list(esw, vport->vport,
-					  MLX5_NVPRT_LIST_TYPE_UC);
+		esw_update_vport_addr_list(esw, vport, MLX5_NVPRT_LIST_TYPE_UC);
+		esw_apply_vport_addr_list(esw, vport, MLX5_NVPRT_LIST_TYPE_UC);
 	}
 
-	if (vport->enabled_events & MC_ADDR_CHANGE) {
-		esw_update_vport_addr_list(esw, vport->vport,
-					   MLX5_NVPRT_LIST_TYPE_MC);
-	}
+	if (vport->enabled_events & MC_ADDR_CHANGE)
+		esw_update_vport_addr_list(esw, vport, MLX5_NVPRT_LIST_TYPE_MC);
 
 	if (vport->enabled_events & PROMISC_CHANGE) {
-		esw_update_vport_rx_mode(esw, vport->vport);
+		esw_update_vport_rx_mode(esw, vport);
 		if (!IS_ERR_OR_NULL(vport->allmulti_rule))
-			esw_update_vport_mc_promisc(esw, vport->vport);
+			esw_update_vport_mc_promisc(esw, vport);
 	}
 
-	if (vport->enabled_events & (PROMISC_CHANGE | MC_ADDR_CHANGE)) {
-		esw_apply_vport_addr_list(esw, vport->vport,
-					  MLX5_NVPRT_LIST_TYPE_MC);
-	}
+	if (vport->enabled_events & (PROMISC_CHANGE | MC_ADDR_CHANGE))
+		esw_apply_vport_addr_list(esw, vport, MLX5_NVPRT_LIST_TYPE_MC);
 
 	esw_debug(esw->dev, "vport[%d] Context Changed: Done\n", vport->vport);
 	if (vport->enabled)
@@ -1407,10 +1399,10 @@ static void esw_destroy_tsar(struct mlx5_eswitch *esw)
 	esw->qos.enabled = false;
 }
 
-static int esw_vport_enable_qos(struct mlx5_eswitch *esw, int vport_num,
+static int esw_vport_enable_qos(struct mlx5_eswitch *esw,
+				struct mlx5_vport *vport,
 				u32 initial_max_rate, u32 initial_bw_share)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	u32 sched_ctx[MLX5_ST_SZ_DW(scheduling_context)] = {0};
 	struct mlx5_core_dev *dev = esw->dev;
 	void *vport_elem;
@@ -1427,7 +1419,7 @@ static int esw_vport_enable_qos(struct mlx5_eswitch *esw, int vport_num,
 		 SCHEDULING_CONTEXT_ELEMENT_TYPE_VPORT);
 	vport_elem = MLX5_ADDR_OF(scheduling_context, sched_ctx,
 				  element_attributes);
-	MLX5_SET(vport_element, vport_elem, vport_number, vport_num);
+	MLX5_SET(vport_element, vport_elem, vport_number, vport->vport);
 	MLX5_SET(scheduling_context, sched_ctx, parent_element_id,
 		 esw->qos.root_tsar_id);
 	MLX5_SET(scheduling_context, sched_ctx, max_average_bw,
@@ -1440,7 +1432,7 @@ static int esw_vport_enable_qos(struct mlx5_eswitch *esw, int vport_num,
 						 &vport->qos.esw_tsar_ix);
 	if (err) {
 		esw_warn(esw->dev, "E-Switch create TSAR vport element failed (vport=%d,err=%d)\n",
-			 vport_num, err);
+			 vport->vport, err);
 		return err;
 	}
 
@@ -1448,10 +1440,10 @@ static int esw_vport_enable_qos(struct mlx5_eswitch *esw, int vport_num,
 	return 0;
 }
 
-static void esw_vport_disable_qos(struct mlx5_eswitch *esw, int vport_num)
+static void esw_vport_disable_qos(struct mlx5_eswitch *esw,
+				  struct mlx5_vport *vport)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
-	int err = 0;
+	int err;
 
 	if (!vport->qos.enabled)
 		return;
@@ -1461,15 +1453,15 @@ static void esw_vport_disable_qos(struct mlx5_eswitch *esw, int vport_num)
 						  vport->qos.esw_tsar_ix);
 	if (err)
 		esw_warn(esw->dev, "E-Switch destroy TSAR vport element failed (vport=%d,err=%d)\n",
-			 vport_num, err);
+			 vport->vport, err);
 
 	vport->qos.enabled = false;
 }
 
-static int esw_vport_qos_config(struct mlx5_eswitch *esw, int vport_num,
+static int esw_vport_qos_config(struct mlx5_eswitch *esw,
+				struct mlx5_vport *vport,
 				u32 max_rate, u32 bw_share)
 {
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	u32 sched_ctx[MLX5_ST_SZ_DW(scheduling_context)] = {0};
 	struct mlx5_core_dev *dev = esw->dev;
 	void *vport_elem;
@@ -1486,7 +1478,7 @@ static int esw_vport_qos_config(struct mlx5_eswitch *esw, int vport_num,
 		 SCHEDULING_CONTEXT_ELEMENT_TYPE_VPORT);
 	vport_elem = MLX5_ADDR_OF(scheduling_context, sched_ctx,
 				  element_attributes);
-	MLX5_SET(vport_element, vport_elem, vport_number, vport_num);
+	MLX5_SET(vport_element, vport_elem, vport_number, vport->vport);
 	MLX5_SET(scheduling_context, sched_ctx, parent_element_id,
 		 esw->qos.root_tsar_id);
 	MLX5_SET(scheduling_context, sched_ctx, max_average_bw,
@@ -1502,7 +1494,7 @@ static int esw_vport_qos_config(struct mlx5_eswitch *esw, int vport_num,
 						 bitmask);
 	if (err) {
 		esw_warn(esw->dev, "E-Switch modify TSAR vport element failed (vport=%d,err=%d)\n",
-			 vport_num, err);
+			 vport->vport, err);
 		return err;
 	}
 
@@ -1605,7 +1597,7 @@ static void esw_enable_vport(struct mlx5_eswitch *esw, struct mlx5_vport *vport,
 	esw_apply_vport_conf(esw, vport);
 
 	/* Attach vport to the eswitch rate limiter */
-	if (esw_vport_enable_qos(esw, vport_num, vport->info.max_rate,
+	if (esw_vport_enable_qos(esw, vport, vport->info.max_rate,
 				 vport->qos.bw_share))
 		esw_warn(esw->dev, "Failed to attach vport %d to eswitch rate limiter", vport_num);
 
@@ -1650,7 +1642,7 @@ static void esw_disable_vport(struct mlx5_eswitch *esw,
 	 */
 	esw_vport_change_handle_locked(vport);
 	vport->enabled_events = 0;
-	esw_vport_disable_qos(esw, vport_num);
+	esw_vport_disable_qos(esw, vport);
 	if (esw->manager_vport != vport_num &&
 	    esw->mode == SRIOV_LEGACY) {
 		mlx5_modify_vport_admin_state(esw->dev,
@@ -2271,7 +2263,7 @@ static int normalize_vports_min_rate(struct mlx5_eswitch *esw, u32 divider)
 		if (bw_share == evport->qos.bw_share)
 			continue;
 
-		err = esw_vport_qos_config(esw, evport->vport, vport_max_rate,
+		err = esw_vport_qos_config(esw, evport, vport_max_rate,
 					   bw_share);
 		if (!err)
 			evport->qos.bw_share = bw_share;
@@ -2325,7 +2317,7 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 	if (max_rate == evport->info.max_rate)
 		goto unlock;
 
-	err = esw_vport_qos_config(esw, vport, max_rate, evport->qos.bw_share);
+	err = esw_vport_qos_config(esw, evport, max_rate, evport->qos.bw_share);
 	if (!err)
 		evport->info.max_rate = max_rate;
 
@@ -2335,11 +2327,10 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 }
 
 static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
-					       int vport_idx,
+					       struct mlx5_vport *vport,
 					       struct mlx5_vport_drop_stats *stats)
 {
 	struct mlx5_eswitch *esw = dev->priv.eswitch;
-	struct mlx5_vport *vport = &esw->vports[vport_idx];
 	u64 rx_discard_vport_down, tx_discard_vport_down;
 	u64 bytes = 0;
 	int err = 0;
@@ -2359,7 +2350,7 @@ static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
 	    !MLX5_CAP_GEN(dev, transmit_discard_vport_down))
 		return 0;
 
-	err = mlx5_query_vport_down_stats(dev, vport_idx, 1,
+	err = mlx5_query_vport_down_stats(dev, vport->vport, 1,
 					  &rx_discard_vport_down,
 					  &tx_discard_vport_down);
 	if (err)
@@ -2374,9 +2365,10 @@ static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
 }
 
 int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
-				 int vport,
+				 int vport_num,
 				 struct ifla_vf_stats *vf_stats)
 {
+	struct mlx5_vport *vport;
 	int outlen = MLX5_ST_SZ_BYTES(query_vport_counter_out);
 	u32 in[MLX5_ST_SZ_DW(query_vport_counter_in)] = {0};
 	struct mlx5_vport_drop_stats stats = {0};
@@ -2385,9 +2377,10 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport))
+	if (!LEGAL_VPORT(esw, vport_num))
 		return -EINVAL;
 
+	vport = mlx5_eswitch_get_vport(esw, vport_num);
 	out = kvzalloc(outlen, GFP_KERNEL);
 	if (!out)
 		return -ENOMEM;
@@ -2395,7 +2388,7 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
 	MLX5_SET(query_vport_counter_in, in, opcode,
 		 MLX5_CMD_OP_QUERY_VPORT_COUNTER);
 	MLX5_SET(query_vport_counter_in, in, op_mod, 0);
-	MLX5_SET(query_vport_counter_in, in, vport_number, vport);
+	MLX5_SET(query_vport_counter_in, in, vport_number, vport->vport);
 	MLX5_SET(query_vport_counter_in, in, other_vport, 1);
 
 	memset(out, 0, outlen);
-- 
2.20.1

[net-next 13/15] net/mlx5: E-Switch, Use getter to access all vport array

[Saeed Mahameed]

From: Bodong Wang <bodong@mellanox.com>

Some functions issue vport commands and access vport array using
vport_index/vport_num interchangeably which is OK for VFs vports.
However, this creates potential bug if those vports are not VFs
(E.g, uplink, sf) where their vport_index don't equal to vport_num.

Prepare code to access mlx5_vport structure using a getter function.

Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Vu Pham <vuhuong@mellanox.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c  | 18 +++++++++---------
 .../net/ethernet/mellanox/mlx5/core/eswitch.h  |  2 ++
 2 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 41afc7bf8bd8..0107ce8bf659 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -72,8 +72,8 @@ static void esw_cleanup_vepa_rules(struct mlx5_eswitch *esw);
 			    MC_ADDR_CHANGE | \
 			    PROMISC_CHANGE)
 
-static struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
-						 u16 vport_num)
+struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
+					  u16 vport_num)
 {
 	u16 idx = mlx5_eswitch_vport_num_to_index(esw, vport_num);
 
@@ -1916,7 +1916,7 @@ int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	if (evport->info.spoofchk && !is_valid_ether_addr(mac))
 		mlx5_core_warn(esw->dev,
@@ -1960,7 +1960,7 @@ int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	err = mlx5_modify_vport_admin_state(esw->dev,
 					    MLX5_VPORT_STATE_OP_MOD_ESW_VPORT,
@@ -1989,7 +1989,7 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
 	if (!LEGAL_VPORT(esw, vport))
 		return -EINVAL;
 
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	memset(ivi, 0, sizeof(*ivi));
 	ivi->vf = vport - 1;
@@ -2020,7 +2020,7 @@ int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	err = modify_esw_vport_cvlan(esw->dev, vport, vlan, qos, set_flags);
 	if (err)
@@ -2064,7 +2064,7 @@ int mlx5_eswitch_set_vport_spoofchk(struct mlx5_eswitch *esw,
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 	pschk = evport->info.spoofchk;
 	evport->info.spoofchk = spoofchk;
 	if (pschk && !is_valid_ether_addr(evport->info.mac))
@@ -2213,7 +2213,7 @@ int mlx5_eswitch_set_vport_trust(struct mlx5_eswitch *esw,
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 	evport->info.trusted = setting;
 	if (evport->enabled)
 		esw_vport_change_handle_locked(evport);
@@ -2299,7 +2299,7 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 		return -EOPNOTSUPP;
 
 	mutex_lock(&esw->state_lock);
-	evport = &esw->vports[vport];
+	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	if (min_rate == evport->info.min_rate)
 		goto set_max_rate;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index fc512a5d0c4c..2e6b37d4fc7f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -488,6 +488,8 @@ void mlx5e_tc_clean_fdb_peer_flows(struct mlx5_eswitch *esw);
 #define mlx5_esw_for_each_vf_vport_num_reverse(esw, vport, nvfs)	\
 	for ((vport) = (nvfs); (vport) >= MLX5_VPORT_FIRST_VF; (vport)--)
 
+struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
+					  u16 vport_num);
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
-- 
2.20.1

[net-next 14/15] net/mlx5: E-Switch, Fix the check of legal vport

[Saeed Mahameed]

From: Bodong Wang <bodong@mellanox.com>

The check of legal vport is to ensure the vport number falls between
0 and total number of vports. Along with the introduction of uplink
rep, enabled vports are not consecutive any more.
Therefore, rely on the eswitch vport getter function to check if it's
a valid vport.

As the getter function relies on eswitch, add the check of vport
group manager and validation the presence of eswitch structure.
Remove the redundant check in the function calls.

Since the vport array will be allocated once eswitch is initialized
and will be kept alive if eswitch presents, no need to protect it with
the state lock.

Fixes: 5ae5162066d8 ("net/mlx5: E-Switch, Assign a different position for uplink rep and vport")
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c | 86 +++++++++----------
 .../net/ethernet/mellanox/mlx5/core/eswitch.h |  5 +-
 2 files changed, 46 insertions(+), 45 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 0107ce8bf659..9ea0ccfe5ef5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -72,12 +72,22 @@ static void esw_cleanup_vepa_rules(struct mlx5_eswitch *esw);
 			    MC_ADDR_CHANGE | \
 			    PROMISC_CHANGE)
 
-struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
-					  u16 vport_num)
+struct mlx5_vport *__must_check
+mlx5_eswitch_get_vport(struct mlx5_eswitch *esw, u16 vport_num)
 {
-	u16 idx = mlx5_eswitch_vport_num_to_index(esw, vport_num);
+	u16 idx;
+
+	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager))
+		return ERR_PTR(-EPERM);
+
+	idx = mlx5_eswitch_vport_num_to_index(esw, vport_num);
+
+	if (idx > esw->total_vports - 1) {
+		esw_debug(esw->dev, "vport out of range: num(0x%x), idx(0x%x)\n",
+			  vport_num, idx);
+		return ERR_PTR(-EINVAL);
+	}
 
-	WARN_ON(idx > esw->total_vports - 1);
 	return &esw->vports[idx];
 }
 
@@ -1667,6 +1677,9 @@ static int eswitch_vport_event(struct notifier_block *nb,
 
 	vport_num = be16_to_cpu(eqe->data.vport_change.vport_num);
 	vport = mlx5_eswitch_get_vport(esw, vport_num);
+	if (IS_ERR(vport))
+		return NOTIFY_OK;
+
 	if (vport->enabled)
 		queue_work(esw->work_queue, &vport->vport_change_handler);
 
@@ -1901,22 +1914,19 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
 }
 
 /* Vport Administration */
-#define LEGAL_VPORT(esw, vport) (vport >= 0 && vport < esw->total_vports)
-
 int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
 			       int vport, u8 mac[ETH_ALEN])
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 	u64 node_guid;
 	int err = 0;
 
-	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager))
-		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport) || is_multicast_ether_addr(mac))
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
+	if (is_multicast_ether_addr(mac))
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	if (evport->info.spoofchk && !is_valid_ether_addr(mac))
 		mlx5_core_warn(esw->dev,
@@ -1951,16 +1961,15 @@ int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
 int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
 				 int vport, int link_state)
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 	int err = 0;
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport))
-		return -EINVAL;
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
 
 	mutex_lock(&esw->state_lock);
-	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	err = mlx5_modify_vport_admin_state(esw->dev,
 					    MLX5_VPORT_STATE_OP_MOD_ESW_VPORT,
@@ -1982,14 +1991,10 @@ int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
 int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
 				  int vport, struct ifla_vf_info *ivi)
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 
-	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager))
-		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport))
-		return -EINVAL;
-
-	evport = mlx5_eswitch_get_vport(esw, vport);
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
 
 	memset(ivi, 0, sizeof(*ivi));
 	ivi->vf = vport - 1;
@@ -2011,16 +2016,17 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
 int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 				  int vport, u16 vlan, u8 qos, u8 set_flags)
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 	int err = 0;
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport) || (vlan > 4095) || (qos > 7))
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
+	if (vlan > 4095 || qos > 7)
 		return -EINVAL;
 
 	mutex_lock(&esw->state_lock);
-	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	err = modify_esw_vport_cvlan(esw->dev, vport, vlan, qos, set_flags);
 	if (err)
@@ -2054,17 +2060,16 @@ int mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 int mlx5_eswitch_set_vport_spoofchk(struct mlx5_eswitch *esw,
 				    int vport, bool spoofchk)
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 	bool pschk;
 	int err = 0;
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport))
-		return -EINVAL;
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
 
 	mutex_lock(&esw->state_lock);
-	evport = mlx5_eswitch_get_vport(esw, vport);
 	pschk = evport->info.spoofchk;
 	evport->info.spoofchk = spoofchk;
 	if (pschk && !is_valid_ether_addr(evport->info.mac))
@@ -2205,15 +2210,14 @@ int mlx5_eswitch_get_vepa(struct mlx5_eswitch *esw, u8 *setting)
 int mlx5_eswitch_set_vport_trust(struct mlx5_eswitch *esw,
 				 int vport, bool setting)
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport))
-		return -EINVAL;
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
 
 	mutex_lock(&esw->state_lock);
-	evport = mlx5_eswitch_get_vport(esw, vport);
 	evport->info.trusted = setting;
 	if (evport->enabled)
 		esw_vport_change_handle_locked(evport);
@@ -2277,7 +2281,7 @@ static int normalize_vports_min_rate(struct mlx5_eswitch *esw, u32 divider)
 int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 				u32 max_rate, u32 min_rate)
 {
-	struct mlx5_vport *evport;
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 	u32 fw_max_bw_share;
 	u32 previous_min_rate;
 	u32 divider;
@@ -2287,8 +2291,8 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport))
-		return -EINVAL;
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
 
 	fw_max_bw_share = MLX5_CAP_QOS(esw->dev, max_tsar_bw_share);
 	min_rate_supported = MLX5_CAP_QOS(esw->dev, esw_bw_share) &&
@@ -2299,7 +2303,6 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 		return -EOPNOTSUPP;
 
 	mutex_lock(&esw->state_lock);
-	evport = mlx5_eswitch_get_vport(esw, vport);
 
 	if (min_rate == evport->info.min_rate)
 		goto set_max_rate;
@@ -2368,19 +2371,16 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
 				 int vport_num,
 				 struct ifla_vf_stats *vf_stats)
 {
-	struct mlx5_vport *vport;
+	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
 	int outlen = MLX5_ST_SZ_BYTES(query_vport_counter_out);
 	u32 in[MLX5_ST_SZ_DW(query_vport_counter_in)] = {0};
 	struct mlx5_vport_drop_stats stats = {0};
 	int err = 0;
 	u32 *out;
 
-	if (!ESW_ALLOWED(esw))
-		return -EPERM;
-	if (!LEGAL_VPORT(esw, vport_num))
-		return -EINVAL;
+	if (IS_ERR(vport))
+		return PTR_ERR(vport);
 
-	vport = mlx5_eswitch_get_vport(esw, vport_num);
 	out = kvzalloc(outlen, GFP_KERNEL);
 	if (!out)
 		return -ENOMEM;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 2e6b37d4fc7f..ed3fad689ec9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -488,8 +488,9 @@ void mlx5e_tc_clean_fdb_peer_flows(struct mlx5_eswitch *esw);
 #define mlx5_esw_for_each_vf_vport_num_reverse(esw, vport, nvfs)	\
 	for ((vport) = (nvfs); (vport) >= MLX5_VPORT_FIRST_VF; (vport)--)
 
-struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
-					  u16 vport_num);
+struct mlx5_vport *__must_check
+mlx5_eswitch_get_vport(struct mlx5_eswitch *esw, u16 vport_num);
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
-- 
2.20.1

[net-next 15/15] net/mlx5: E-Switch, Use atomic rep state to serialize state change

[Saeed Mahameed]

From: Bodong Wang <bodong@mellanox.com>

When the state of rep was introduced, it was also designed to prevent
duplicate unloading of the same rep. Considering the following two
flows when an eswitch manager is at switchdev mode with n VF reps loaded.

+--------------------------------------+--------------------------------+
| cpu-0                                | cpu-1                          |
| --------                             | --------                       |
| mlx5_ib_remove                       | mlx5_eswitch_disable_sriov     |
|  mlx5_ib_unregister_vport_reps       |  esw_offloads_cleanup          |
|   mlx5_eswitch_unregister_vport_reps |   esw_offloads_unload_all_reps |
|    __unload_reps_all_vport           |    __unload_reps_all_vport     |
+--------------------------------------+--------------------------------+

These two flows will try to unload the same rep. Per original design,
once one flow unloads the rep, the state moves to REGISTERED. The 2nd
flow will no longer needs to do the unload and bails out. However, as
read and write of the state is not atomic, when 1st flow is doing the
unload, the state is still LOADED, 2nd flow is able to do the same
unload action. Kernel crash will happen.

To solve this, driver should do atomic test-and-set for the state. So
that only one flow can change the rep state from LOADED to REGISTERED,
and proceed to do the actual unloading.

Since the state is changing to atomic type, all other read/write should
be atomic action as well.

Fixes: f121e0ea9586 (net/mlx5: E-Switch, Add state to eswitch vport representors)
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Vu Pham <vuhuong@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../mellanox/mlx5/core/eswitch_offloads.c     | 36 +++++++++----------
 include/linux/mlx5/eswitch.h                  |  2 +-
 2 files changed, 18 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index e88feaa293f6..e09ae27485ee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -333,7 +333,7 @@ static int esw_set_global_vlan_pop(struct mlx5_eswitch *esw, u8 val)
 	esw_debug(esw->dev, "%s applying global %s policy\n", __func__, val ? "pop" : "none");
 	for (vf_vport = 1; vf_vport < esw->enabled_vports; vf_vport++) {
 		rep = &esw->offloads.vport_reps[vf_vport];
-		if (rep->rep_if[REP_ETH].state != REP_LOADED)
+		if (atomic_read(&rep->rep_if[REP_ETH].state) != REP_LOADED)
 			continue;
 
 		err = __mlx5_eswitch_set_vport_vlan(esw, rep->vport, 0, 0, val);
@@ -1277,7 +1277,8 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
 		ether_addr_copy(rep->hw_id, hw_id);
 
 		for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++)
-			rep->rep_if[rep_type].state = REP_UNREGISTERED;
+			atomic_set(&rep->rep_if[rep_type].state,
+				   REP_UNREGISTERED);
 	}
 
 	return 0;
@@ -1286,11 +1287,9 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
 static void __esw_offloads_unload_rep(struct mlx5_eswitch *esw,
 				      struct mlx5_eswitch_rep *rep, u8 rep_type)
 {
-	if (rep->rep_if[rep_type].state != REP_LOADED)
-		return;
-
-	rep->rep_if[rep_type].unload(rep);
-	rep->rep_if[rep_type].state = REP_REGISTERED;
+	if (atomic_cmpxchg(&rep->rep_if[rep_type].state,
+			   REP_LOADED, REP_REGISTERED) == REP_LOADED)
+		rep->rep_if[rep_type].unload(rep);
 }
 
 static void __unload_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type)
@@ -1351,16 +1350,15 @@ static int __esw_offloads_load_rep(struct mlx5_eswitch *esw,
 {
 	int err = 0;
 
-	if (rep->rep_if[rep_type].state != REP_REGISTERED)
-		return 0;
-
-	err = rep->rep_if[rep_type].load(esw->dev, rep);
-	if (err)
-		return err;
-
-	rep->rep_if[rep_type].state = REP_LOADED;
+	if (atomic_cmpxchg(&rep->rep_if[rep_type].state,
+			   REP_REGISTERED, REP_LOADED) == REP_REGISTERED) {
+		err = rep->rep_if[rep_type].load(esw->dev, rep);
+		if (err)
+			atomic_set(&rep->rep_if[rep_type].state,
+				   REP_REGISTERED);
+	}
 
-	return 0;
+	return err;
 }
 
 static int __load_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type)
@@ -2217,7 +2215,7 @@ void mlx5_eswitch_register_vport_reps(struct mlx5_eswitch *esw,
 		rep_if->get_proto_dev = __rep_if->get_proto_dev;
 		rep_if->priv = __rep_if->priv;
 
-		rep_if->state = REP_REGISTERED;
+		atomic_set(&rep_if->state, REP_REGISTERED);
 	}
 }
 EXPORT_SYMBOL(mlx5_eswitch_register_vport_reps);
@@ -2232,7 +2230,7 @@ void mlx5_eswitch_unregister_vport_reps(struct mlx5_eswitch *esw, u8 rep_type)
 		__unload_reps_all_vport(esw, max_vf, rep_type);
 
 	mlx5_esw_for_all_reps(esw, i, rep)
-		rep->rep_if[rep_type].state = REP_UNREGISTERED;
+		atomic_set(&rep->rep_if[rep_type].state, REP_UNREGISTERED);
 }
 EXPORT_SYMBOL(mlx5_eswitch_unregister_vport_reps);
 
@@ -2252,7 +2250,7 @@ void *mlx5_eswitch_get_proto_dev(struct mlx5_eswitch *esw,
 
 	rep = mlx5_eswitch_get_rep(esw, vport);
 
-	if (rep->rep_if[rep_type].state == REP_LOADED &&
+	if (atomic_read(&rep->rep_if[rep_type].state) == REP_LOADED &&
 	    rep->rep_if[rep_type].get_proto_dev)
 		return rep->rep_if[rep_type].get_proto_dev(rep);
 	return NULL;
diff --git a/include/linux/mlx5/eswitch.h b/include/linux/mlx5/eswitch.h
index 96d8435421de..0ca77dd1429c 100644
--- a/include/linux/mlx5/eswitch.h
+++ b/include/linux/mlx5/eswitch.h
@@ -35,7 +35,7 @@ struct mlx5_eswitch_rep_if {
 	void		       (*unload)(struct mlx5_eswitch_rep *rep);
 	void		       *(*get_proto_dev)(struct mlx5_eswitch_rep *rep);
 	void			*priv;
-	u8			state;
+	atomic_t		state;
 };
 
 struct mlx5_eswitch_rep {
-- 
2.20.1

Re: [pull request][net-next 00/15] Mellanox, mlx5 updates 2019-04-30

[Saeed Mahameed]

On Tue, 2019-04-30 at 20:39 +0000, Saeed Mahameed wrote:
> Hi Dave,
> 
> This series provides misc updates to mlx5 driver.
> There is one patch of this series that is touching outside mlx5
> driver:
> 
> ethtool.h: Add SFF-8436 and SFF-8636 max EEPROM length definitions
> Added max EEPROM length defines for ethtool usage:
>      #define ETH_MODULE_SFF_8636_MAX_LEN     640
>      #define ETH_MODULE_SFF_8436_MAX_LEN     640
>     
> These definitions used to determine the EEPROM data
> length when reading high eeprom pages.
> 
> For more information please see tag log below.
> 
> Please pull and let me know if there is any problem.
> 
> Please note that the series starts with a merge of mlx5-next branch,
> to resolve and avoid dependency with rdma tree.
> 
> Thanks,
> Saeed.
> 

Hi Dave, i will be sending V2 to include latest fix from mlx5-next
branch.

otherwise mlx5 will be broken.

Thanks,
Saeed.

Re: [pull request][net-next 00/15] Mellanox, mlx5 updates 2019-04-30

[David Miller]

From: Saeed Mahameed <saeedm@mellanox.com>
Date: Wed, 1 May 2019 21:43:41 +0000

> Hi Dave, i will be sending V2 to include latest fix from mlx5-next
> branch.

Ok.