[pull request][net-next v0 00/14] mlx5 updates 2021-12-02

[pull request][net-next v0 00/14] mlx5 updates 2021-12-02

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

Hi Dave, Hi Jakub,

This series provides misc updates.
For more information please see tag log below.

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

Thanks,
Saeed.


The following changes since commit fc993be36f9ea7fc286d84d8471a1a20e871aad4:

  Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net (2021-12-02 11:44:56 -0800)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2021-12-02

for you to fetch changes up to b247f32aecad09e6cf7edff7739e6f2c9dc5fca9:

  net/mlx5: Dynamically resize flow counters query buffer (2021-12-02 16:53:16 -0800)

----------------------------------------------------------------
mlx5-updates-2021-12-02

Misc updates to mlx5 driver

1) Various code cleanups
2) Error path handling fixes of latest features
3) Print more information on pci error handling
4) Dynamically resize flow counters query buffer

----------------------------------------------------------------
Arnd Bergmann (2):
      mlx5: fix psample_sample_packet link error
      mlx5: fix mlx5i_grp_sw_update_stats() stack usage

Avihai Horon (1):
      net/mlx5: Dynamically resize flow counters query buffer

Ben Ben-Ishay (1):
      net/mlx5e: SHAMPO, clean MLX5E_MAX_KLM_PER_WQE macro

Christophe JAILLET (1):
      net/mlx5: Fix some error handling paths in 'mlx5e_tc_add_fdb_flow()'

Dan Carpenter (1):
      net/mlx5: SF, silence an uninitialized variable warning

Roi Dayan (5):
      net/mlx5e: TC, Remove redundant action stack var
      net/mlx5e: Remove redundant actions arg from validate_goto_chain()
      net/mlx5e: Remove redundant actions arg from vlan push/pop funcs
      net/mlx5e: TC, Move common flow_action checks into function
      net/mlx5e: TC, Set flow attr ip_version earlier

Saeed Mahameed (1):
      net/mlx5: Print more info on pci error handlers

Tariq Toukan (1):
      net/mlx5e: Hide function mlx5e_num_channels_changed

Wei Yongjun (1):
      net/mlx5: Fix error return code in esw_qos_create()

 drivers/net/ethernet/mellanox/mlx5/core/Kconfig    |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |   3 +-
 .../net/ethernet/mellanox/mlx5/core/en/params.c    |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 175 ++++++++++-----------
 drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c  |   1 +
 .../net/ethernet/mellanox/mlx5/core/fs_counters.c  |  74 +++++++--
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c  |  12 +-
 drivers/net/ethernet/mellanox/mlx5/core/main.c     |  51 ++++--
 .../net/ethernet/mellanox/mlx5/core/sf/hw_table.c  |   2 +-
 include/linux/mlx5/driver.h                        |   4 +
 12 files changed, 202 insertions(+), 127 deletions(-)

[net-next v0 01/14] mlx5: fix psample_sample_packet link error

[Saeed Mahameed]

From: Arnd Bergmann <arnd@arndb.de>

When PSAMPLE is a loadable module, built-in drivers cannot use it:

aarch64-linux-ld: drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.o: in function `mlx5e_tc_sample_skb':
sample.c:(.text+0xd68): undefined reference to `psample_sample_packet'

Add the same dependency here that is used for MLXSW

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
index 92056452a9e3..4ba1a78c6515 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -115,6 +115,7 @@ config MLX5_TC_CT
 config MLX5_TC_SAMPLE
 	bool "MLX5 TC sample offload support"
 	depends on MLX5_CLS_ACT
+	depends on PSAMPLE=y || PSAMPLE=n || MLX5_CORE=m
 	default y
 	help
 	  Say Y here if you want to support offloading sample rules via tc
-- 
2.31.1

[net-next v0 02/14] mlx5: fix mlx5i_grp_sw_update_stats() stack usage

[Saeed Mahameed]

From: Arnd Bergmann <arnd@arndb.de>

The mlx5e_sw_stats structure has grown to the point of triggering
a warning when put on the stack of a function:

mlx5/core/ipoib/ipoib.c: In function 'mlx5i_grp_sw_update_stats':
mlx5/core/ipoib/ipoib.c:136:1: error: the frame size of 1028 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]

In this case, only five of the structure members are actually set,
so it's sufficient to have those as separate local variables.
As en_rep.c uses 'struct rtnl_link_stats64' for this, just use
the same one here for consistency.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c    | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index ea1efdecc88c..051b20ec7bdb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -110,7 +110,7 @@ void mlx5i_cleanup(struct mlx5e_priv *priv)
 
 static void mlx5i_grp_sw_update_stats(struct mlx5e_priv *priv)
 {
-	struct mlx5e_sw_stats s = { 0 };
+	struct rtnl_link_stats64 s = {};
 	int i, j;
 
 	for (i = 0; i < priv->stats_nch; i++) {
@@ -128,11 +128,17 @@ static void mlx5i_grp_sw_update_stats(struct mlx5e_priv *priv)
 
 			s.tx_packets           += sq_stats->packets;
 			s.tx_bytes             += sq_stats->bytes;
-			s.tx_queue_dropped     += sq_stats->dropped;
+			s.tx_dropped           += sq_stats->dropped;
 		}
 	}
 
-	memcpy(&priv->stats.sw, &s, sizeof(s));
+	memset(&priv->stats.sw, 0, sizeof(s));
+
+	priv->stats.sw.rx_packets = s.rx_packets;
+	priv->stats.sw.rx_bytes = s.rx_bytes;
+	priv->stats.sw.tx_packets = s.tx_packets;
+	priv->stats.sw.tx_bytes = s.tx_bytes;
+	priv->stats.sw.tx_queue_dropped = s.tx_dropped;
 }
 
 void mlx5i_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
-- 
2.31.1

[net-next v0 03/14] net/mlx5: Fix error return code in esw_qos_create()

[Saeed Mahameed]

From: Wei Yongjun <weiyongjun1@huawei.com>

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Fixes: 85c5f7c9200e ("net/mlx5: E-switch, Create QoS on demand")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
index eead33defa80..11bbcd5f5b8b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
@@ -590,6 +590,7 @@ static int esw_qos_create(struct mlx5_eswitch *esw, struct netlink_ext_ack *exta
 		if (IS_ERR(esw->qos.group0)) {
 			esw_warn(dev, "E-Switch create rate group 0 failed (%ld)\n",
 				 PTR_ERR(esw->qos.group0));
+			err = PTR_ERR(esw->qos.group0);
 			goto err_group0;
 		}
 	}
-- 
2.31.1

[net-next v0 04/14] net/mlx5: Fix some error handling paths in 'mlx5e_tc_add_fdb_flow()'

[Saeed Mahameed]

From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>

All the error handling paths of 'mlx5e_tc_add_fdb_flow()' end to 'err_out'
where 'flow_flag_set(flow, FAILED);' is called.

All but the new error handling paths added by the commits given in the
Fixes tag below.

Fix these error handling paths and branch to 'err_out'.

Fixes: 166f431ec6be ("net/mlx5e: Add indirect tc offload of ovs internal port")
Fixes: b16eb3c81fe2 ("net/mlx5: Support internal port as decap route device")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index cb66c08783c2..ca74ed616382 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1441,7 +1441,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 							MLX5_FLOW_NAMESPACE_FDB, VPORT_TO_REG,
 							metadata);
 			if (err)
-				return err;
+				goto err_out;
 		}
 	}
 
@@ -1457,13 +1457,15 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 		if (attr->chain) {
 			NL_SET_ERR_MSG_MOD(extack,
 					   "Internal port rule is only supported on chain 0");
-			return -EOPNOTSUPP;
+			err = -EOPNOTSUPP;
+			goto err_out;
 		}
 
 		if (attr->dest_chain) {
 			NL_SET_ERR_MSG_MOD(extack,
 					   "Internal port rule offload doesn't support goto action");
-			return -EOPNOTSUPP;
+			err = -EOPNOTSUPP;
+			goto err_out;
 		}
 
 		int_port = mlx5e_tc_int_port_get(mlx5e_get_int_port_priv(priv),
@@ -1471,8 +1473,10 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 						 flow_flag_test(flow, EGRESS) ?
 						 MLX5E_TC_INT_PORT_EGRESS :
 						 MLX5E_TC_INT_PORT_INGRESS);
-		if (IS_ERR(int_port))
-			return PTR_ERR(int_port);
+		if (IS_ERR(int_port)) {
+			err = PTR_ERR(int_port);
+			goto err_out;
+		}
 
 		esw_attr->int_port = int_port;
 	}
-- 
2.31.1

[net-next v0 05/14] net/mlx5: SF, silence an uninitialized variable warning

[Saeed Mahameed]

From: Dan Carpenter <dan.carpenter@oracle.com>

This code sometimes calls mlx5_sf_hw_table_hwc_init() when "ext_base_id"
is uninitialized.  It's not used on that path, but it generates a static
checker warning to pass uninitialized variables to another function.
It may also generate runtime UBSan  warnings depending on if the
mlx5_sf_hw_table_hwc_init() function is inlined or not.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c
index 252d6017387d..17aa348989cb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c
@@ -247,7 +247,7 @@ int mlx5_sf_hw_table_init(struct mlx5_core_dev *dev)
 {
 	struct mlx5_sf_hw_table *table;
 	u16 max_ext_fn = 0;
-	u16 ext_base_id;
+	u16 ext_base_id = 0;
 	u16 max_fn = 0;
 	u16 base_id;
 	int err;
-- 
2.31.1

[net-next v0 06/14] net/mlx5: Print more info on pci error handlers

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

In case mlx5_pci_err_detected was called with state equals to
pci_channel_io_perm_failure, the driver will never come back up.

It is nice to know why the driver went to zombie land, so print some
useful information on pci err handlers.

Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/main.c    | 51 ++++++++++++++-----
 1 file changed, 37 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 7df9c7f8d9c8..d97c9e86d7b3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1604,12 +1604,28 @@ static void remove_one(struct pci_dev *pdev)
 	mlx5_devlink_free(devlink);
 }
 
+#define mlx5_pci_trace(dev, fmt, ...) ({ \
+	struct mlx5_core_dev *__dev = (dev); \
+	mlx5_core_info(__dev, "%s Device state = %d health sensors: %d pci_status: %d. " fmt, \
+		       __func__, __dev->state, mlx5_health_check_fatal_sensors(__dev), \
+		       __dev->pci_status, ##__VA_ARGS__); \
+})
+
+static const char *result2str(enum pci_ers_result result)
+{
+	return  result == PCI_ERS_RESULT_NEED_RESET ? "need reset" :
+		result == PCI_ERS_RESULT_DISCONNECT ? "disconnect" :
+		result == PCI_ERS_RESULT_RECOVERED  ? "recovered" :
+		"unknown";
+}
+
 static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev,
 					      pci_channel_state_t state)
 {
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
+	enum pci_ers_result res;
 
-	mlx5_core_info(dev, "%s was called\n", __func__);
+	mlx5_pci_trace(dev, "Enter, pci channel state = %d\n", state);
 
 	mlx5_enter_error_state(dev, false);
 	mlx5_error_sw_reset(dev);
@@ -1617,8 +1633,11 @@ static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev,
 	mlx5_drain_health_wq(dev);
 	mlx5_pci_disable_device(dev);
 
-	return state == pci_channel_io_perm_failure ?
+	res = state == pci_channel_io_perm_failure ?
 		PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_NEED_RESET;
+
+	mlx5_pci_trace(dev, "Exit, result = %d, %s\n",  res, result2str(res));
+	return res;
 }
 
 /* wait for the device to show vital signs by waiting
@@ -1652,28 +1671,34 @@ static int wait_vital(struct pci_dev *pdev)
 
 static pci_ers_result_t mlx5_pci_slot_reset(struct pci_dev *pdev)
 {
+	enum pci_ers_result res = PCI_ERS_RESULT_DISCONNECT;
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
 	int err;
 
-	mlx5_core_info(dev, "%s was called\n", __func__);
+	mlx5_pci_trace(dev, "Enter\n");
 
 	err = mlx5_pci_enable_device(dev);
 	if (err) {
 		mlx5_core_err(dev, "%s: mlx5_pci_enable_device failed with error code: %d\n",
 			      __func__, err);
-		return PCI_ERS_RESULT_DISCONNECT;
+		goto out;
 	}
 
 	pci_set_master(pdev);
 	pci_restore_state(pdev);
 	pci_save_state(pdev);
 
-	if (wait_vital(pdev)) {
-		mlx5_core_err(dev, "%s: wait_vital timed out\n", __func__);
-		return PCI_ERS_RESULT_DISCONNECT;
+	err = wait_vital(pdev);
+	if (err) {
+		mlx5_core_err(dev, "%s: wait vital failed with error code: %d\n",
+			      __func__, err);
+		goto out;
 	}
 
-	return PCI_ERS_RESULT_RECOVERED;
+	res = PCI_ERS_RESULT_RECOVERED;
+out:
+	mlx5_pci_trace(dev, "Exit, err = %d, result = %d, %s\n", err, res, result2str(res));
+	return res;
 }
 
 static void mlx5_pci_resume(struct pci_dev *pdev)
@@ -1681,14 +1706,12 @@ static void mlx5_pci_resume(struct pci_dev *pdev)
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
 	int err;
 
-	mlx5_core_info(dev, "%s was called\n", __func__);
+	mlx5_pci_trace(dev, "Enter, loading driver..\n");
 
 	err = mlx5_load_one(dev);
-	if (err)
-		mlx5_core_err(dev, "%s: mlx5_load_one failed with error code: %d\n",
-			      __func__, err);
-	else
-		mlx5_core_info(dev, "%s: device recovered\n", __func__);
+
+	mlx5_pci_trace(dev, "Done, err = %d, device %s\n", err,
+		       !err ? "recovered" : "Failed");
 }
 
 static const struct pci_error_handlers mlx5_err_handler = {
-- 
2.31.1

[net-next v0 07/14] net/mlx5e: SHAMPO, clean MLX5E_MAX_KLM_PER_WQE macro

[Saeed Mahameed]

From: Ben Ben-Ishay <benishay@nvidia.com>

This commit reduces unused variable from MLX5E_MAX_KLM_PER_WQE macro that
introduced by commit d7b896acbdcb ("net/mlx5e: Add support to klm_umr_wqe").

Signed-off-by: Ben Ben-Ishay <benishay@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h        | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/params.c | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c     | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 48b12ee44b8d..1834efa64c1d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -173,7 +173,7 @@ struct page_pool;
 #define MLX5E_KLM_ENTRIES_PER_WQE(wqe_size)\
 	ALIGN_DOWN(MLX5E_KLM_MAX_ENTRIES_PER_WQE(wqe_size), MLX5_UMR_KLM_ALIGNMENT)
 
-#define MLX5E_MAX_KLM_PER_WQE(mdev) \
+#define MLX5E_MAX_KLM_PER_WQE \
 	MLX5E_KLM_ENTRIES_PER_WQE(MLX5E_TX_MPW_MAX_NUM_DS << MLX5_MKEY_BSF_OCTO_SIZE)
 
 #define MLX5E_MSG_LEVEL			NETIF_MSG_LINK
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index f8c29022dbb2..66180ffb4606 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -717,7 +717,7 @@ static u32 mlx5e_shampo_icosq_sz(struct mlx5_core_dev *mdev,
 	int wq_size = BIT(MLX5_GET(wq, wqc, log_wq_sz));
 	u32 wqebbs;
 
-	max_klm_per_umr = MLX5E_MAX_KLM_PER_WQE(mdev);
+	max_klm_per_umr = MLX5E_MAX_KLM_PER_WQE;
 	max_hd_per_wqe = mlx5e_shampo_hd_per_wqe(mdev, params, rq_param);
 	max_num_of_umr_per_wqe = max_hd_per_wqe / max_klm_per_umr;
 	rest = max_hd_per_wqe % max_klm_per_umr;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index b84ed9a7855c..7e05d7592bce 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -619,7 +619,7 @@ static int mlx5e_alloc_rx_hd_mpwqe(struct mlx5e_rq *rq)
 	struct mlx5e_icosq *sq = rq->icosq;
 	int i, err, max_klm_entries, len;
 
-	max_klm_entries = MLX5E_MAX_KLM_PER_WQE(rq->mdev);
+	max_klm_entries = MLX5E_MAX_KLM_PER_WQE;
 	klm_entries = bitmap_find_window(shampo->bitmap,
 					 shampo->hd_per_wqe,
 					 shampo->hd_per_wq, shampo->pi);
-- 
2.31.1

[net-next v0 08/14] net/mlx5e: Hide function mlx5e_num_channels_changed

[Saeed Mahameed]

From: Tariq Toukan <tariqt@nvidia.com>

No calls for mlx5e_num_channels_changed() out of en_main.c,
turn it static and remove from header.
Keep the wrapper function mlx5e_num_channels_changed_ctx exposed.

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

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 1834efa64c1d..e77c4159713f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1057,7 +1057,6 @@ int mlx5e_safe_switch_params(struct mlx5e_priv *priv,
 			     mlx5e_fp_preactivate preactivate,
 			     void *context, bool reset);
 int mlx5e_update_tx_netdev_queues(struct mlx5e_priv *priv);
-int mlx5e_num_channels_changed(struct mlx5e_priv *priv);
 int mlx5e_num_channels_changed_ctx(struct mlx5e_priv *priv, void *context);
 void mlx5e_activate_priv_channels(struct mlx5e_priv *priv);
 void mlx5e_deactivate_priv_channels(struct mlx5e_priv *priv);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 65571593ec5c..496977e7406e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -2598,7 +2598,7 @@ static void mlx5e_set_default_xps_cpumasks(struct mlx5e_priv *priv,
 	}
 }
 
-int mlx5e_num_channels_changed(struct mlx5e_priv *priv)
+static int mlx5e_num_channels_changed(struct mlx5e_priv *priv)
 {
 	u16 count = priv->channels.params.num_channels;
 	int err;
-- 
2.31.1

[net-next v0 09/14] net/mlx5e: TC, Remove redundant action stack var

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Remove the action stack var from parse tc fdb actions
and prase tc nic actions, use the flow attr action var directly.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 89 +++++++++----------
 1 file changed, 42 insertions(+), 47 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index ca74ed616382..d869907fdb70 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3457,7 +3457,6 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 	struct pedit_headers_action hdrs[2] = {};
 	const struct flow_action_entry *act;
 	struct mlx5_nic_flow_attr *nic_attr;
-	u32 action = 0;
 	int err, i;
 
 	if (!flow_action_has_entries(flow_action)) {
@@ -3478,12 +3477,12 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 	flow_action_for_each(i, act, flow_action) {
 		switch (act->id) {
 		case FLOW_ACTION_ACCEPT:
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			break;
 		case FLOW_ACTION_DROP:
-			action |= MLX5_FLOW_CONTEXT_ACTION_DROP |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_DROP |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			break;
 		case FLOW_ACTION_MANGLE:
 		case FLOW_ACTION_ADD:
@@ -3492,19 +3491,19 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 			if (err)
 				return err;
 
-			action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
 			break;
 		case FLOW_ACTION_VLAN_MANGLE:
 			err = add_vlan_rewrite_action(priv,
 						      MLX5_FLOW_NAMESPACE_KERNEL,
 						      act, parse_attr, hdrs,
-						      &action, extack);
+						      &attr->action, extack);
 			if (err)
 				return err;
 
 			break;
 		case FLOW_ACTION_CSUM:
-			if (csum_offload_supported(priv, action,
+			if (csum_offload_supported(priv, attr->action,
 						   act->csum_flags,
 						   extack))
 				break;
@@ -3517,8 +3516,8 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 			    same_hw_devs(priv, netdev_priv(peer_dev))) {
 				parse_attr->mirred_ifindex[0] = peer_dev->ifindex;
 				flow_flag_set(flow, HAIRPIN);
-				action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-					  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+				attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+						MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			} else {
 				NL_SET_ERR_MSG_MOD(extack,
 						   "device is not on same HW, can't offload");
@@ -3538,17 +3537,17 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 			}
 
 			nic_attr->flow_tag = mark;
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 			}
 			break;
 		case FLOW_ACTION_GOTO:
-			err = validate_goto_chain(priv, flow, act, action,
+			err = validate_goto_chain(priv, flow, act, attr->action,
 						  extack);
 			if (err)
 				return err;
 
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			attr->dest_chain = act->chain_index;
 			break;
 		case FLOW_ACTION_CT:
@@ -3567,8 +3566,6 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 		}
 	}
 
-	attr->action = action;
-
 	if (attr->dest_chain && parse_attr->mirred_ifindex[0]) {
 		NL_SET_ERR_MSG(extack, "Mirroring goto chain rules isn't supported");
 		return -EOPNOTSUPP;
@@ -3886,7 +3883,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 	const struct flow_action_entry *act;
 	struct mlx5_esw_flow_attr *esw_attr;
 	bool encap = false, decap = false;
-	u32 action = attr->action;
 	int err, i, if_count = 0;
 	bool ptype_host = false;
 	bool mpls_push = false;
@@ -3908,8 +3904,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 	flow_action_for_each(i, act, flow_action) {
 		switch (act->id) {
 		case FLOW_ACTION_ACCEPT:
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			attr->flags |= MLX5_ESW_ATTR_FLAG_ACCEPT;
 			break;
 		case FLOW_ACTION_PTYPE:
@@ -3922,8 +3918,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			ptype_host = true;
 			break;
 		case FLOW_ACTION_DROP:
-			action |= MLX5_FLOW_CONTEXT_ACTION_DROP |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_DROP |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			break;
 		case FLOW_ACTION_TRAP:
 			if (!flow_offload_has_one_action(flow_action)) {
@@ -3931,8 +3927,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 						   "action trap is supported as a sole action only");
 				return -EOPNOTSUPP;
 			}
-			action |= (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				   MLX5_FLOW_CONTEXT_ACTION_COUNT);
+			attr->action |= (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					 MLX5_FLOW_CONTEXT_ACTION_COUNT);
 			attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH;
 			break;
 		case FLOW_ACTION_MPLS_PUSH:
@@ -3963,7 +3959,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			}
 
 			parse_attr->eth.h_proto = act->mpls_pop.proto;
-			action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
 			flow_flag_set(flow, L3_TO_L2_DECAP);
 			break;
 		case FLOW_ACTION_MANGLE:
@@ -3974,12 +3970,12 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				return err;
 
 			if (!flow_flag_test(flow, L3_TO_L2_DECAP)) {
-				action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+				attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
 				esw_attr->split_count = esw_attr->out_count;
 			}
 			break;
 		case FLOW_ACTION_CSUM:
-			if (csum_offload_supported(priv, action,
+			if (csum_offload_supported(priv, attr->action,
 						   act->csum_flags, extack))
 				break;
 
@@ -4015,12 +4011,12 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				return -EOPNOTSUPP;
 			}
 
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 
 			err = mlx5e_set_fwd_to_int_port_actions(priv, attr, out_dev->ifindex,
 								MLX5E_TC_INT_PORT_INGRESS,
-								&action, esw_attr->out_count);
+								&attr->action, esw_attr->out_count);
 			if (err)
 				return err;
 
@@ -4065,8 +4061,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				return -EOPNOTSUPP;
 			}
 
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			if (encap) {
 				parse_attr->mirred_ifindex[esw_attr->out_count] =
 					out_dev->ifindex;
@@ -4102,14 +4098,14 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				if (is_vlan_dev(out_dev)) {
 					err = add_vlan_push_action(priv, attr,
 								   &out_dev,
-								   &action, extack);
+								   &attr->action, extack);
 					if (err)
 						return err;
 				}
 
 				if (is_vlan_dev(parse_attr->filter_dev)) {
 					err = add_vlan_pop_action(priv, attr,
-								  &action, extack);
+								  &attr->action, extack);
 					if (err)
 						return err;
 				}
@@ -4142,7 +4138,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 				err = mlx5e_set_fwd_to_int_port_actions(priv, attr,
 									out_dev->ifindex,
 									MLX5E_TC_INT_PORT_EGRESS,
-									&action,
+									&attr->action,
 									esw_attr->out_count);
 				if (err)
 					return err;
@@ -4180,15 +4176,16 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 		case FLOW_ACTION_VLAN_PUSH:
 		case FLOW_ACTION_VLAN_POP:
 			if (act->id == FLOW_ACTION_VLAN_PUSH &&
-			    (action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP)) {
+			    (attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP)) {
 				/* Replace vlan pop+push with vlan modify */
-				action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+				attr->action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
 				err = add_vlan_rewrite_action(priv,
 							      MLX5_FLOW_NAMESPACE_FDB,
 							      act, parse_attr, hdrs,
-							      &action, extack);
+							      &attr->action, extack);
 			} else {
-				err = parse_tc_vlan_action(priv, act, esw_attr, &action, extack);
+				err = parse_tc_vlan_action(priv, act, esw_attr, &attr->action,
+							   extack);
 			}
 			if (err)
 				return err;
@@ -4199,7 +4196,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			err = add_vlan_rewrite_action(priv,
 						      MLX5_FLOW_NAMESPACE_FDB,
 						      act, parse_attr, hdrs,
-						      &action, extack);
+						      &attr->action, extack);
 			if (err)
 				return err;
 
@@ -4209,13 +4206,13 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			decap = true;
 			break;
 		case FLOW_ACTION_GOTO:
-			err = validate_goto_chain(priv, flow, act, action,
+			err = validate_goto_chain(priv, flow, act, attr->action,
 						  extack);
 			if (err)
 				return err;
 
-			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+					MLX5_FLOW_CONTEXT_ACTION_COUNT;
 			attr->dest_chain = act->chain_index;
 			break;
 		case FLOW_ACTION_CT:
@@ -4262,19 +4259,17 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 	attr->ip_version = mlx5e_tc_get_ip_version(&parse_attr->spec, true);
 
 	if (MLX5_CAP_GEN(esw->dev, prio_tag_required) &&
-	    action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) {
+	    attr->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;
+		attr->action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
 		err = add_vlan_prio_tag_rewrite_action(priv, parse_attr, hdrs,
-						       &action, extack);
+						       &attr->action, extack);
 		if (err)
 			return err;
 	}
 
-	attr->action = action;
-
 	err = actions_prepare_mod_hdr_actions(priv, flow, attr, hdrs, extack);
 	if (err)
 		return err;
-- 
2.31.1

[net-next v0 10/14] net/mlx5e: Remove redundant actions arg from validate_goto_chain()

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Passing actions is redundant and can be retrieved from flow.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 14 +++++---------
 1 file changed, 5 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 d869907fdb70..90fca3555563 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3359,11 +3359,9 @@ add_vlan_prio_tag_rewrite_action(struct mlx5e_priv *priv,
 static int validate_goto_chain(struct mlx5e_priv *priv,
 			       struct mlx5e_tc_flow *flow,
 			       const struct flow_action_entry *act,
-			       u32 actions,
 			       struct netlink_ext_ack *extack)
 {
 	bool is_esw = mlx5e_is_eswitch_flow(flow);
-	struct mlx5_flow_attr *attr = flow->attr;
 	bool ft_flow = mlx5e_is_ft_flow(flow);
 	u32 dest_chain = act->chain_index;
 	struct mlx5_fs_chains *chains;
@@ -3384,7 +3382,7 @@ static int validate_goto_chain(struct mlx5e_priv *priv,
 	}
 
 	if (!mlx5_chains_backwards_supported(chains) &&
-	    dest_chain <= attr->chain) {
+	    dest_chain <= flow->attr->chain) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "Goto lower numbered chain isn't supported");
 		return -EOPNOTSUPP;
@@ -3396,8 +3394,8 @@ static int validate_goto_chain(struct mlx5e_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
-	if (actions & (MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT |
-		       MLX5_FLOW_CONTEXT_ACTION_DECAP) &&
+	if (flow->attr->action & (MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT |
+				  MLX5_FLOW_CONTEXT_ACTION_DECAP) &&
 	    !reformat_and_fwd) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "Goto chain is not allowed if action has reformat or decap");
@@ -3541,8 +3539,7 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 			}
 			break;
 		case FLOW_ACTION_GOTO:
-			err = validate_goto_chain(priv, flow, act, attr->action,
-						  extack);
+			err = validate_goto_chain(priv, flow, act, extack);
 			if (err)
 				return err;
 
@@ -4206,8 +4203,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			decap = true;
 			break;
 		case FLOW_ACTION_GOTO:
-			err = validate_goto_chain(priv, flow, act, attr->action,
-						  extack);
+			err = validate_goto_chain(priv, flow, act, extack);
 			if (err)
 				return err;
 
-- 
2.31.1

[net-next v0 11/14] net/mlx5e: Remove redundant actions arg from vlan push/pop funcs

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Passing actions is redundant and can be retrieved from flow attr.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 90fca3555563..3c5e9efb9873 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3682,7 +3682,6 @@ static struct net_device *get_fdb_out_dev(struct net_device *uplink_dev,
 static int add_vlan_push_action(struct mlx5e_priv *priv,
 				struct mlx5_flow_attr *attr,
 				struct net_device **out_dev,
-				u32 *action,
 				struct netlink_ext_ack *extack)
 {
 	struct net_device *vlan_dev = *out_dev;
@@ -3694,7 +3693,7 @@ static int add_vlan_push_action(struct mlx5e_priv *priv,
 	};
 	int err;
 
-	err = parse_tc_vlan_action(priv, &vlan_act, attr->esw_attr, action, extack);
+	err = parse_tc_vlan_action(priv, &vlan_act, attr->esw_attr, &attr->action, extack);
 	if (err)
 		return err;
 
@@ -3705,14 +3704,13 @@ static int add_vlan_push_action(struct mlx5e_priv *priv,
 		return -ENODEV;
 
 	if (is_vlan_dev(*out_dev))
-		err = add_vlan_push_action(priv, attr, out_dev, action, extack);
+		err = add_vlan_push_action(priv, attr, out_dev, extack);
 
 	return err;
 }
 
 static int add_vlan_pop_action(struct mlx5e_priv *priv,
 			       struct mlx5_flow_attr *attr,
-			       u32 *action,
 			       struct netlink_ext_ack *extack)
 {
 	struct flow_action_entry vlan_act = {
@@ -3723,7 +3721,8 @@ static int add_vlan_pop_action(struct mlx5e_priv *priv,
 	nest_level = attr->parse_attr->filter_dev->lower_level -
 						priv->netdev->lower_level;
 	while (nest_level--) {
-		err = parse_tc_vlan_action(priv, &vlan_act, attr->esw_attr, action, extack);
+		err = parse_tc_vlan_action(priv, &vlan_act, attr->esw_attr,
+					   &attr->action, extack);
 		if (err)
 			return err;
 	}
@@ -4093,16 +4092,13 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 					return -ENODEV;
 
 				if (is_vlan_dev(out_dev)) {
-					err = add_vlan_push_action(priv, attr,
-								   &out_dev,
-								   &attr->action, extack);
+					err = add_vlan_push_action(priv, attr, &out_dev, extack);
 					if (err)
 						return err;
 				}
 
 				if (is_vlan_dev(parse_attr->filter_dev)) {
-					err = add_vlan_pop_action(priv, attr,
-								  &attr->action, extack);
+					err = add_vlan_pop_action(priv, attr, extack);
 					if (err)
 						return err;
 				}
-- 
2.31.1

[net-next v0 12/14] net/mlx5e: TC, Move common flow_action checks into function

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Remove duplicate checks on flow_action by using common function.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 44 ++++++++++---------
 1 file changed, 24 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 3c5e9efb9873..c7f1c93709cd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3444,6 +3444,24 @@ actions_prepare_mod_hdr_actions(struct mlx5e_priv *priv,
 	return 0;
 }
 
+static int
+flow_action_supported(struct flow_action *flow_action,
+		      struct netlink_ext_ack *extack)
+{
+	if (!flow_action_has_entries(flow_action)) {
+		NL_SET_ERR_MSG_MOD(extack, "Flow action doesn't have any entries");
+		return -EINVAL;
+	}
+
+	if (!flow_action_hw_stats_check(flow_action, extack,
+					FLOW_ACTION_HW_STATS_DELAYED_BIT)) {
+		NL_SET_ERR_MSG_MOD(extack, "Flow action HW stats type is not supported");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
 static int
 parse_tc_nic_actions(struct mlx5e_priv *priv,
 		     struct flow_action *flow_action,
@@ -3457,16 +3475,9 @@ parse_tc_nic_actions(struct mlx5e_priv *priv,
 	struct mlx5_nic_flow_attr *nic_attr;
 	int err, i;
 
-	if (!flow_action_has_entries(flow_action)) {
-		NL_SET_ERR_MSG_MOD(extack, "Flow action doesn't have any entries");
-		return -EINVAL;
-	}
-
-	if (!flow_action_hw_stats_check(flow_action, extack,
-					FLOW_ACTION_HW_STATS_DELAYED_BIT)) {
-		NL_SET_ERR_MSG_MOD(extack, "Flow action HW stats type is not supported");
-		return -EOPNOTSUPP;
-	}
+	err = flow_action_supported(flow_action, extack);
+	if (err)
+		return err;
 
 	nic_attr = attr->nic_attr;
 	nic_attr->flow_tag = MLX5_FS_DEFAULT_FLOW_TAG;
@@ -3883,16 +3894,9 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 	bool ptype_host = false;
 	bool mpls_push = false;
 
-	if (!flow_action_has_entries(flow_action)) {
-		NL_SET_ERR_MSG_MOD(extack, "Flow action doesn't have any entries");
-		return -EINVAL;
-	}
-
-	if (!flow_action_hw_stats_check(flow_action, extack,
-					FLOW_ACTION_HW_STATS_DELAYED_BIT)) {
-		NL_SET_ERR_MSG_MOD(extack, "Flow action HW stats type is not supported");
-		return -EOPNOTSUPP;
-	}
+	err = flow_action_supported(flow_action, extack);
+	if (err)
+		return err;
 
 	esw_attr = attr->esw_attr;
 	parse_attr = attr->parse_attr;
-- 
2.31.1

[net-next v0 13/14] net/mlx5e: TC, Set flow attr ip_version earlier

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Setting flow attr ip_version is not related to parsing tc flow actions.
It needs to be set after parsing flower matches which changes the spec.
So move it outside parse_tc_fdb_actions() and set it in
__mlx5e_add_fdb_flow().

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index c7f1c93709cd..3e3419190c55 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -4251,9 +4251,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
-	/* always set IP version for indirect table handling */
-	attr->ip_version = mlx5e_tc_get_ip_version(&parse_attr->spec, true);
-
 	if (MLX5_CAP_GEN(esw->dev, prio_tag_required) &&
 	    attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) {
 		/* For prio tag mode, replace vlan pop with rewrite vlan prio
@@ -4488,6 +4485,9 @@ __mlx5e_add_fdb_flow(struct mlx5e_priv *priv,
 	if (err)
 		goto err_free;
 
+	/* always set IP version for indirect table handling */
+	flow->attr->ip_version = mlx5e_tc_get_ip_version(&parse_attr->spec, true);
+
 	err = parse_tc_fdb_actions(priv, &rule->action, flow, extack);
 	if (err)
 		goto err_free;
-- 
2.31.1

[net-next v0 14/14] net/mlx5: Dynamically resize flow counters query buffer

[Saeed Mahameed]

From: Avihai Horon <avihaih@nvidia.com>

The flow counters bulk query buffer is allocated once during
mlx5_fc_init_stats(). For PFs and VFs this buffer usually takes a little
more than 512KB of memory, which is aligned to the next power of 2, to
1MB. For SFs, this buffer is reduced and takes around 128 Bytes.

The buffer size determines the maximum number of flow counters that
can be queried at a time. Thus, having a bigger buffer can improve
performance for users that need to query many flow counters.

There are cases that don't use many flow counters and don't need a big
buffer (e.g. SFs, VFs). Since this size is critical with large scale,
in these cases the buffer size should be reduced.

In order to reduce memory consumption while maintaining query
performance, change the query buffer's allocation scheme to the
following:
- First allocate the buffer with small initial size.
- If the number of counters surpasses the initial size, resize the
  buffer to the maximum size.

The buffer only grows and isn't shrank, because users with many flow
counters don't care about the buffer size and we don't want to add
resize overhead if the current number of counters drops.

This solution is preferable to the current one, which is less accurate
and only addresses SFs.

Signed-off-by: Avihai Horon <avihaih@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/fs_counters.c | 74 +++++++++++++++----
 include/linux/mlx5/driver.h                   |  4 +
 2 files changed, 64 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c
index 7e0e04cf26f8..b406e0367af6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c
@@ -38,9 +38,10 @@
 #include "fs_cmd.h"
 
 #define MLX5_FC_STATS_PERIOD msecs_to_jiffies(1000)
+#define MLX5_FC_BULK_QUERY_ALLOC_PERIOD msecs_to_jiffies(180 * 1000)
 /* Max number of counters to query in bulk read is 32K */
 #define MLX5_SW_MAX_COUNTERS_BULK BIT(15)
-#define MLX5_SF_NUM_COUNTERS_BULK 8
+#define MLX5_INIT_COUNTERS_BULK 8
 #define MLX5_FC_POOL_MAX_THRESHOLD BIT(18)
 #define MLX5_FC_POOL_USED_BUFF_RATIO 10
 
@@ -145,13 +146,15 @@ static void mlx5_fc_stats_remove(struct mlx5_core_dev *dev,
 	spin_unlock(&fc_stats->counters_idr_lock);
 }
 
-static int get_max_bulk_query_len(struct mlx5_core_dev *dev)
+static int get_init_bulk_query_len(struct mlx5_core_dev *dev)
 {
-	int num_counters_bulk = mlx5_core_is_sf(dev) ?
-					MLX5_SF_NUM_COUNTERS_BULK :
-					MLX5_SW_MAX_COUNTERS_BULK;
+	return min_t(int, MLX5_INIT_COUNTERS_BULK,
+		     (1 << MLX5_CAP_GEN(dev, log_max_flow_counter_bulk)));
+}
 
-	return min_t(int, num_counters_bulk,
+static int get_max_bulk_query_len(struct mlx5_core_dev *dev)
+{
+	return min_t(int, MLX5_SW_MAX_COUNTERS_BULK,
 		     (1 << MLX5_CAP_GEN(dev, log_max_flow_counter_bulk)));
 }
 
@@ -177,7 +180,7 @@ static void mlx5_fc_stats_query_counter_range(struct mlx5_core_dev *dev,
 {
 	struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats;
 	bool query_more_counters = (first->id <= last_id);
-	int max_bulk_len = get_max_bulk_query_len(dev);
+	int cur_bulk_len = fc_stats->bulk_query_len;
 	u32 *data = fc_stats->bulk_query_out;
 	struct mlx5_fc *counter = first;
 	u32 bulk_base_id;
@@ -189,7 +192,7 @@ static void mlx5_fc_stats_query_counter_range(struct mlx5_core_dev *dev,
 		bulk_base_id = counter->id & ~0x3;
 
 		/* number of counters to query inc. the last counter */
-		bulk_len = min_t(int, max_bulk_len,
+		bulk_len = min_t(int, cur_bulk_len,
 				 ALIGN(last_id - bulk_base_id + 1, 4));
 
 		err = mlx5_cmd_fc_bulk_query(dev, bulk_base_id, bulk_len,
@@ -230,6 +233,41 @@ static void mlx5_fc_release(struct mlx5_core_dev *dev, struct mlx5_fc *counter)
 		mlx5_fc_free(dev, counter);
 }
 
+static void mlx5_fc_stats_bulk_query_size_increase(struct mlx5_core_dev *dev)
+{
+	struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats;
+	int max_bulk_len = get_max_bulk_query_len(dev);
+	unsigned long now = jiffies;
+	u32 *bulk_query_out_tmp;
+	int max_out_len;
+
+	if (fc_stats->bulk_query_alloc_failed &&
+	    time_before(now, fc_stats->next_bulk_query_alloc))
+		return;
+
+	max_out_len = mlx5_cmd_fc_get_bulk_query_out_len(max_bulk_len);
+	bulk_query_out_tmp = kzalloc(max_out_len, GFP_KERNEL);
+	if (!bulk_query_out_tmp) {
+		mlx5_core_warn_once(dev,
+				    "Can't increase flow counters bulk query buffer size, insufficient memory, bulk_size(%d)\n",
+				    max_bulk_len);
+		fc_stats->bulk_query_alloc_failed = true;
+		fc_stats->next_bulk_query_alloc =
+			now + MLX5_FC_BULK_QUERY_ALLOC_PERIOD;
+		return;
+	}
+
+	kfree(fc_stats->bulk_query_out);
+	fc_stats->bulk_query_out = bulk_query_out_tmp;
+	fc_stats->bulk_query_len = max_bulk_len;
+	if (fc_stats->bulk_query_alloc_failed) {
+		mlx5_core_info(dev,
+			       "Flow counters bulk query buffer size increased, bulk_size(%d)\n",
+			       max_bulk_len);
+		fc_stats->bulk_query_alloc_failed = false;
+	}
+}
+
 static void mlx5_fc_stats_work(struct work_struct *work)
 {
 	struct mlx5_core_dev *dev = container_of(work, struct mlx5_core_dev,
@@ -247,15 +285,22 @@ static void mlx5_fc_stats_work(struct work_struct *work)
 		queue_delayed_work(fc_stats->wq, &fc_stats->work,
 				   fc_stats->sampling_interval);
 
-	llist_for_each_entry(counter, addlist, addlist)
+	llist_for_each_entry(counter, addlist, addlist) {
 		mlx5_fc_stats_insert(dev, counter);
+		fc_stats->num_counters++;
+	}
 
 	llist_for_each_entry_safe(counter, tmp, dellist, dellist) {
 		mlx5_fc_stats_remove(dev, counter);
 
 		mlx5_fc_release(dev, counter);
+		fc_stats->num_counters--;
 	}
 
+	if (fc_stats->bulk_query_len < get_max_bulk_query_len(dev) &&
+	    fc_stats->num_counters > get_init_bulk_query_len(dev))
+		mlx5_fc_stats_bulk_query_size_increase(dev);
+
 	if (time_before(now, fc_stats->next_query) ||
 	    list_empty(&fc_stats->counters))
 		return;
@@ -378,8 +423,8 @@ EXPORT_SYMBOL(mlx5_fc_destroy);
 int mlx5_init_fc_stats(struct mlx5_core_dev *dev)
 {
 	struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats;
-	int max_bulk_len;
-	int max_out_len;
+	int init_bulk_len;
+	int init_out_len;
 
 	spin_lock_init(&fc_stats->counters_idr_lock);
 	idr_init(&fc_stats->counters_idr);
@@ -387,11 +432,12 @@ int mlx5_init_fc_stats(struct mlx5_core_dev *dev)
 	init_llist_head(&fc_stats->addlist);
 	init_llist_head(&fc_stats->dellist);
 
-	max_bulk_len = get_max_bulk_query_len(dev);
-	max_out_len = mlx5_cmd_fc_get_bulk_query_out_len(max_bulk_len);
-	fc_stats->bulk_query_out = kzalloc(max_out_len, GFP_KERNEL);
+	init_bulk_len = get_init_bulk_query_len(dev);
+	init_out_len = mlx5_cmd_fc_get_bulk_query_out_len(init_bulk_len);
+	fc_stats->bulk_query_out = kzalloc(init_out_len, GFP_KERNEL);
 	if (!fc_stats->bulk_query_out)
 		return -ENOMEM;
+	fc_stats->bulk_query_len = init_bulk_len;
 
 	fc_stats->wq = create_singlethread_workqueue("mlx5_fc");
 	if (!fc_stats->wq)
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index a623ec635947..78655d8d13a7 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -478,6 +478,10 @@ struct mlx5_fc_stats {
 	unsigned long next_query;
 	unsigned long sampling_interval; /* jiffies */
 	u32 *bulk_query_out;
+	int bulk_query_len;
+	size_t num_counters;
+	bool bulk_query_alloc_failed;
+	unsigned long next_bulk_query_alloc;
 	struct mlx5_fc_pool fc_pool;
 };
 
-- 
2.31.1

Re: [net-next v0 01/14] mlx5: fix psample_sample_packet link error

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (master)
by Saeed Mahameed <saeedm@nvidia.com>:

On Thu,  2 Dec 2021 16:56:09 -0800 you wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> When PSAMPLE is a loadable module, built-in drivers cannot use it:
> 
> aarch64-linux-ld: drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.o: in function `mlx5e_tc_sample_skb':
> sample.c:(.text+0xd68): undefined reference to `psample_sample_packet'
> 
> [...]

Here is the summary with links:
  - [net-next,v0,01/14] mlx5: fix psample_sample_packet link error
    https://git.kernel.org/netdev/net-next/c/7a7dd5114f53
  - [net-next,v0,02/14] mlx5: fix mlx5i_grp_sw_update_stats() stack usage
    https://git.kernel.org/netdev/net-next/c/d2b8c7ba3c79
  - [net-next,v0,03/14] net/mlx5: Fix error return code in esw_qos_create()
    https://git.kernel.org/netdev/net-next/c/baf5c001300e
  - [net-next,v0,04/14] net/mlx5: Fix some error handling paths in 'mlx5e_tc_add_fdb_flow()'
    https://git.kernel.org/netdev/net-next/c/31108d142f36
  - [net-next,v0,05/14] net/mlx5: SF, silence an uninitialized variable warning
    https://git.kernel.org/netdev/net-next/c/c64d01b3ceba
  - [net-next,v0,06/14] net/mlx5: Print more info on pci error handlers
    https://git.kernel.org/netdev/net-next/c/fad1783a6d66
  - [net-next,v0,07/14] net/mlx5e: SHAMPO, clean MLX5E_MAX_KLM_PER_WQE macro
    https://git.kernel.org/netdev/net-next/c/3ef1f8e795ba
  - [net-next,v0,08/14] net/mlx5e: Hide function mlx5e_num_channels_changed
    https://git.kernel.org/netdev/net-next/c/e9542221c4f5
  - [net-next,v0,09/14] net/mlx5e: TC, Remove redundant action stack var
    https://git.kernel.org/netdev/net-next/c/9745dbe03669
  - [net-next,v0,10/14] net/mlx5e: Remove redundant actions arg from validate_goto_chain()
    https://git.kernel.org/netdev/net-next/c/3cc78411f3f4
  - [net-next,v0,11/14] net/mlx5e: Remove redundant actions arg from vlan push/pop funcs
    https://git.kernel.org/netdev/net-next/c/70a140ea6f79
  - [net-next,v0,12/14] net/mlx5e: TC, Move common flow_action checks into function
    https://git.kernel.org/netdev/net-next/c/df990477242f
  - [net-next,v0,13/14] net/mlx5e: TC, Set flow attr ip_version earlier
    https://git.kernel.org/netdev/net-next/c/d4bb053139e7
  - [net-next,v0,14/14] net/mlx5: Dynamically resize flow counters query buffer
    https://git.kernel.org/netdev/net-next/c/b247f32aecad

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html