[pull request][net-next 00/15] mlx5 updates 2023-02-07

[pull request][net-next 00/15] mlx5 updates 2023-02-07

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

This series adds minor misc updates to mlx5.
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 383d9f87a06dd923c4fd0fdcb65b58258851f545:

  Merge branch 'net-core-use-a-dedicated-kmem_cache-for-skb-head-allocs' (2023-02-07 11:00:03 -0800)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2023-02-07

for you to fetch changes up to f7133135235dbd11e7cb5fe62fe5d05ce5e82eeb:

  net/mlx5: fw_tracer, Add support for unrecognized string (2023-02-07 16:29:56 -0800)

----------------------------------------------------------------
mlx5-updates-2023-02-07

Minor updates to mlx5 driver:

1) Minor and trivial code Cleanups

2) Minor fixes for net-next

3) From Shay: dynamic FW trace strings update.

----------------------------------------------------------------
Arnd Bergmann (1):
      mlx5: reduce stack usage in mlx5_setup_tc

Gal Pressman (2):
      net/mlx5e: Remove incorrect debugfs_create_dir NULL check in hairpin
      net/mlx5e: Remove incorrect debugfs_create_dir NULL check in TLS

Leon Romanovsky (1):
      net/mlx5e: Don't listen to remove flows event

Maher Sanalla (1):
      net/mlx5: Fix memory leak in error flow of port set buffer

Maor Dickman (1):
      net/mlx5: fs_core, Remove redundant variable err

Moshe Shemesh (1):
      net/mlx5: fw reset: Skip device ID check if PCI link up failed

Roi Dayan (3):
      net/mlx5e: Remove redundant code for handling vlan actions
      net/mlx5: fs, Remove redundant vport_number assignment
      net/mlx5: fs, Remove redundant assignment of size

Shay Drory (5):
      net/mlx5: Remove redundant health work lock
      net/mlx5: fw_tracer: Fix debug print
      net/mlx5: fw_tracer, allow 0 size string DBs
      net/mlx5: fw_tracer, Add support for strings DB update event
      net/mlx5: fw_tracer, Add support for unrecognized string

 .../ethernet/mellanox/mlx5/core/diag/fw_tracer.c   |  79 +++++++-
 .../ethernet/mellanox/mlx5/core/diag/fw_tracer.h   |   9 +
 .../ethernet/mellanox/mlx5/core/en/port_buffer.c   |   4 +-
 .../ethernet/mellanox/mlx5/core/en/tc/act/vlan.c   |  35 +---
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    |   4 +-
 .../ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c |   2 -
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  17 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |  11 --
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |   5 -
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 208 +--------------------
 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c   |   7 -
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  |   3 +-
 drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/health.c   |  28 +--
 include/linux/mlx5/device.h                        |   1 +
 include/linux/mlx5/driver.h                        |   2 -
 16 files changed, 117 insertions(+), 299 deletions(-)

[net-next 01/15] mlx5: reduce stack usage in mlx5_setup_tc

[Saeed Mahameed]

From: Arnd Bergmann <arnd@arndb.de>

Clang warns about excessive stack usage on 32-bit targets:

drivers/net/ethernet/mellanox/mlx5/core/en_main.c:3597:12: error: stack frame size (1184) exceeds limit (1024) in 'mlx5e_setup_tc' [-Werror,-Wframe-larger-than]
static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type,

It turns out that both the mlx5e_setup_tc_mqprio_dcb() function and
the mlx5e_safe_switch_params() function it calls have a copy of
'struct mlx5e_params' on the stack, and this structure is fairly
large.

Use dynamic allocation for the inner one.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_main.c   | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 3973d86905e8..6b9267c19e00 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3002,32 +3002,37 @@ int mlx5e_safe_switch_params(struct mlx5e_priv *priv,
 			     mlx5e_fp_preactivate preactivate,
 			     void *context, bool reset)
 {
-	struct mlx5e_channels new_chs = {};
+	struct mlx5e_channels *new_chs;
 	int err;
 
 	reset &= test_bit(MLX5E_STATE_OPENED, &priv->state);
 	if (!reset)
 		return mlx5e_switch_priv_params(priv, params, preactivate, context);
 
-	new_chs.params = *params;
+	new_chs = kzalloc(sizeof(*new_chs), GFP_KERNEL);
+	if (!new_chs)
+		return -ENOMEM;
+	new_chs->params = *params;
 
-	mlx5e_selq_prepare_params(&priv->selq, &new_chs.params);
+	mlx5e_selq_prepare_params(&priv->selq, &new_chs->params);
 
-	err = mlx5e_open_channels(priv, &new_chs);
+	err = mlx5e_open_channels(priv, new_chs);
 	if (err)
 		goto err_cancel_selq;
 
-	err = mlx5e_switch_priv_channels(priv, &new_chs, preactivate, context);
+	err = mlx5e_switch_priv_channels(priv, new_chs, preactivate, context);
 	if (err)
 		goto err_close;
 
+	kfree(new_chs);
 	return 0;
 
 err_close:
-	mlx5e_close_channels(&new_chs);
+	mlx5e_close_channels(new_chs);
 
 err_cancel_selq:
 	mlx5e_selq_cancel(&priv->selq);
+	kfree(new_chs);
 	return err;
 }
 
-- 
2.39.1

Re: [net-next 01/15] mlx5: reduce stack usage in mlx5_setup_tc

[patchwork-bot+netdevbpf]

Hello:

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

On Tue,  7 Feb 2023 16:36:58 -0800 you wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> Clang warns about excessive stack usage on 32-bit targets:
> 
> drivers/net/ethernet/mellanox/mlx5/core/en_main.c:3597:12: error: stack frame size (1184) exceeds limit (1024) in 'mlx5e_setup_tc' [-Werror,-Wframe-larger-than]
> static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type,
> 
> [...]

Here is the summary with links:
  - [net-next,01/15] mlx5: reduce stack usage in mlx5_setup_tc
    https://git.kernel.org/netdev/net-next/c/7802886274cc
  - [net-next,02/15] net/mlx5: Remove redundant health work lock
    https://git.kernel.org/netdev/net-next/c/67257cba905d
  - [net-next,03/15] net/mlx5: fw reset: Skip device ID check if PCI link up failed
    https://git.kernel.org/netdev/net-next/c/114b295470e7
  - [net-next,04/15] net/mlx5e: Don't listen to remove flows event
    https://git.kernel.org/netdev/net-next/c/a2a73ea14b1a
  - [net-next,05/15] net/mlx5e: Remove redundant code for handling vlan actions
    https://git.kernel.org/netdev/net-next/c/633ad4b29c21
  - [net-next,06/15] net/mlx5: fs, Remove redundant vport_number assignment
    https://git.kernel.org/netdev/net-next/c/2e762e65998e
  - [net-next,07/15] net/mlx5e: Remove incorrect debugfs_create_dir NULL check in hairpin
    https://git.kernel.org/netdev/net-next/c/afae6254c5ea
  - [net-next,08/15] net/mlx5e: Remove incorrect debugfs_create_dir NULL check in TLS
    https://git.kernel.org/netdev/net-next/c/1e985a8d887c
  - [net-next,09/15] net/mlx5: Fix memory leak in error flow of port set buffer
    https://git.kernel.org/netdev/net-next/c/e3e01c1c1598
  - [net-next,10/15] net/mlx5: fs_core, Remove redundant variable err
    https://git.kernel.org/netdev/net-next/c/08929f32da0f
  - [net-next,11/15] net/mlx5: fs, Remove redundant assignment of size
    https://git.kernel.org/netdev/net-next/c/beeebdc52caf
  - [net-next,12/15] net/mlx5: fw_tracer: Fix debug print
    https://git.kernel.org/netdev/net-next/c/988c23522739
  - [net-next,13/15] net/mlx5: fw_tracer, allow 0 size string DBs
    https://git.kernel.org/netdev/net-next/c/b0118ced6b2e
  - [net-next,14/15] net/mlx5: fw_tracer, Add support for strings DB update event
    https://git.kernel.org/netdev/net-next/c/7dfcd110a458
  - [net-next,15/15] net/mlx5: fw_tracer, Add support for unrecognized string
    https://git.kernel.org/netdev/net-next/c/f7133135235d

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

[net-next 02/15] net/mlx5: Remove redundant health work lock

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

Commit 90e7cb78b815 ("net/mlx5: fix missing mutex_unlock in
mlx5_fw_fatal_reporter_err_work()") introduced another checking of
MLX5_DROP_HEALTH_NEW_WORK. At this point, the first check of
MLX5_DROP_HEALTH_NEW_WORK is redundant and so is the lock that
protects it.

Remove the lock and rename MLX5_DROP_HEALTH_NEW_WORK to reflect these
changes.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/health.c  | 28 +++++--------------
 include/linux/mlx5/driver.h                   |  2 --
 2 files changed, 7 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index 879555ba847d..1e8bee906c31 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -62,7 +62,7 @@ enum {
 };
 
 enum {
-	MLX5_DROP_NEW_HEALTH_WORK,
+	MLX5_DROP_HEALTH_WORK,
 };
 
 enum  {
@@ -675,7 +675,7 @@ static void mlx5_fw_fatal_reporter_err_work(struct work_struct *work)
 	devlink = priv_to_devlink(dev);
 
 	mutex_lock(&dev->intf_state_mutex);
-	if (test_bit(MLX5_DROP_NEW_HEALTH_WORK, &health->flags)) {
+	if (test_bit(MLX5_DROP_HEALTH_WORK, &health->flags)) {
 		mlx5_core_err(dev, "health works are not permitted at this stage\n");
 		mutex_unlock(&dev->intf_state_mutex);
 		return;
@@ -771,14 +771,8 @@ static unsigned long get_next_poll_jiffies(struct mlx5_core_dev *dev)
 void mlx5_trigger_health_work(struct mlx5_core_dev *dev)
 {
 	struct mlx5_core_health *health = &dev->priv.health;
-	unsigned long flags;
 
-	spin_lock_irqsave(&health->wq_lock, flags);
-	if (!test_bit(MLX5_DROP_NEW_HEALTH_WORK, &health->flags))
-		queue_work(health->wq, &health->fatal_report_work);
-	else
-		mlx5_core_err(dev, "new health works are not permitted at this stage\n");
-	spin_unlock_irqrestore(&health->wq_lock, flags);
+	queue_work(health->wq, &health->fatal_report_work);
 }
 
 #define MLX5_MSEC_PER_HOUR (MSEC_PER_SEC * 60 * 60)
@@ -858,7 +852,7 @@ void mlx5_start_health_poll(struct mlx5_core_dev *dev)
 
 	timer_setup(&health->timer, poll_health, 0);
 	health->fatal_error = MLX5_SENSOR_NO_ERR;
-	clear_bit(MLX5_DROP_NEW_HEALTH_WORK, &health->flags);
+	clear_bit(MLX5_DROP_HEALTH_WORK, &health->flags);
 	health->health = &dev->iseg->health;
 	health->health_counter = &dev->iseg->health_counter;
 
@@ -869,13 +863,9 @@ void mlx5_start_health_poll(struct mlx5_core_dev *dev)
 void mlx5_stop_health_poll(struct mlx5_core_dev *dev, bool disable_health)
 {
 	struct mlx5_core_health *health = &dev->priv.health;
-	unsigned long flags;
 
-	if (disable_health) {
-		spin_lock_irqsave(&health->wq_lock, flags);
-		set_bit(MLX5_DROP_NEW_HEALTH_WORK, &health->flags);
-		spin_unlock_irqrestore(&health->wq_lock, flags);
-	}
+	if (disable_health)
+		set_bit(MLX5_DROP_HEALTH_WORK, &health->flags);
 
 	del_timer_sync(&health->timer);
 }
@@ -891,11 +881,8 @@ void mlx5_start_health_fw_log_up(struct mlx5_core_dev *dev)
 void mlx5_drain_health_wq(struct mlx5_core_dev *dev)
 {
 	struct mlx5_core_health *health = &dev->priv.health;
-	unsigned long flags;
 
-	spin_lock_irqsave(&health->wq_lock, flags);
-	set_bit(MLX5_DROP_NEW_HEALTH_WORK, &health->flags);
-	spin_unlock_irqrestore(&health->wq_lock, flags);
+	set_bit(MLX5_DROP_HEALTH_WORK, &health->flags);
 	cancel_delayed_work_sync(&health->update_fw_log_ts_work);
 	cancel_work_sync(&health->report_work);
 	cancel_work_sync(&health->fatal_report_work);
@@ -928,7 +915,6 @@ int mlx5_health_init(struct mlx5_core_dev *dev)
 	kfree(name);
 	if (!health->wq)
 		goto out_err;
-	spin_lock_init(&health->wq_lock);
 	INIT_WORK(&health->fatal_report_work, mlx5_fw_fatal_reporter_err_work);
 	INIT_WORK(&health->report_work, mlx5_fw_reporter_err_work);
 	INIT_DELAYED_WORK(&health->update_fw_log_ts_work, mlx5_health_log_ts_update);
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index cd529e051b4d..91e8160ed087 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -430,8 +430,6 @@ struct mlx5_core_health {
 	u8				synd;
 	u32				fatal_error;
 	u32				crdump_size;
-	/* wq spinlock to synchronize draining */
-	spinlock_t			wq_lock;
 	struct workqueue_struct	       *wq;
 	unsigned long			flags;
 	struct work_struct		fatal_report_work;
-- 
2.39.1

[net-next 03/15] net/mlx5: fw reset: Skip device ID check if PCI link up failed

[Saeed Mahameed]

From: Moshe Shemesh <moshe@nvidia.com>

In case where after reset the PCI link is not ready within timeout, skip
reading device ID as if there is no PCI link up we can't have FW
response to pci config cycles either.

This also fixes err value not used and overwritten in such flow.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index 1da4da564e6d..63290da84010 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -371,6 +371,7 @@ static int mlx5_pci_link_toggle(struct mlx5_core_dev *dev)
 		mlx5_core_err(dev, "PCI link not ready (0x%04x) after %llu ms\n",
 			      reg16, mlx5_tout_ms(dev, PCI_TOGGLE));
 		err = -ETIMEDOUT;
+		goto restore;
 	}
 
 	do {
-- 
2.39.1

[net-next 04/15] net/mlx5e: Don't listen to remove flows event

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

remove_flow_enable event isn't really needed as it will be
triggered once user and/or XFRM core explicitly asked to delete
state. In such situation, we won't be interested in any event
at all.

So don't trigger and listen to remove_flow_enable event.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c  | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 57ac0f663fcd..7fb3835befbd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -92,7 +92,6 @@ static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn,
 		MLX5_SET(ipsec_aso, aso_ctx, remove_flow_pkt_cnt,
 			 lower_32_bits(attrs->hard_packet_limit));
 		MLX5_SET(ipsec_aso, aso_ctx, hard_lft_arm, 1);
-		MLX5_SET(ipsec_aso, aso_ctx, remove_flow_enable, 1);
 	}
 
 	if (attrs->soft_packet_limit != XFRM_INF) {
@@ -329,8 +328,7 @@ static void mlx5e_ipsec_handle_event(struct work_struct *_work)
 
 	if (attrs->soft_packet_limit != XFRM_INF)
 		if (!MLX5_GET(ipsec_aso, aso->ctx, soft_lft_arm) ||
-		    !MLX5_GET(ipsec_aso, aso->ctx, hard_lft_arm) ||
-		    !MLX5_GET(ipsec_aso, aso->ctx, remove_flow_enable))
+		    !MLX5_GET(ipsec_aso, aso->ctx, hard_lft_arm))
 			xfrm_state_check_expire(sa_entry->x);
 
 unlock:
-- 
2.39.1

[net-next 05/15] net/mlx5e: Remove redundant code for handling vlan actions

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Remove unused code which was used only with deprecated HW
which didn't support vlan actions.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/en/tc/act/vlan.c       |  35 +--
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   |   9 -
 .../net/ethernet/mellanox/mlx5/core/eswitch.h |   5 -
 .../mellanox/mlx5/core/eswitch_offloads.c     | 208 +-----------------
 4 files changed, 14 insertions(+), 243 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c
index b86ac604d0c2..2e0d88b513aa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c
@@ -44,19 +44,17 @@ parse_tc_vlan_action(struct mlx5e_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
+	if (!mlx5_eswitch_vlan_actions_supported(priv->mdev, vlan_idx)) {
+		NL_SET_ERR_MSG_MOD(extack, "firmware vlan actions is not supported");
+		return -EOPNOTSUPP;
+	}
+
 	switch (act->id) {
 	case FLOW_ACTION_VLAN_POP:
-		if (vlan_idx) {
-			if (!mlx5_eswitch_vlan_actions_supported(priv->mdev,
-								 MLX5_FS_VLAN_DEPTH)) {
-				NL_SET_ERR_MSG_MOD(extack, "vlan pop action is not supported");
-				return -EOPNOTSUPP;
-			}
-
+		if (vlan_idx)
 			*action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP_2;
-		} else {
+		else
 			*action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
-		}
 		break;
 	case FLOW_ACTION_VLAN_PUSH:
 		attr->vlan_vid[vlan_idx] = act->vlan.vid;
@@ -65,25 +63,10 @@ parse_tc_vlan_action(struct mlx5e_priv *priv,
 		if (!attr->vlan_proto[vlan_idx])
 			attr->vlan_proto[vlan_idx] = htons(ETH_P_8021Q);
 
-		if (vlan_idx) {
-			if (!mlx5_eswitch_vlan_actions_supported(priv->mdev,
-								 MLX5_FS_VLAN_DEPTH)) {
-				NL_SET_ERR_MSG_MOD(extack,
-						   "vlan push action is not supported for vlan depth > 1");
-				return -EOPNOTSUPP;
-			}
-
+		if (vlan_idx)
 			*action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2;
-		} else {
-			if (!mlx5_eswitch_vlan_actions_supported(priv->mdev, 1) &&
-			    (act->vlan.proto != htons(ETH_P_8021Q) ||
-			     act->vlan.prio)) {
-				NL_SET_ERR_MSG_MOD(extack, "vlan push action is not supported");
-				return -EOPNOTSUPP;
-			}
-
+		else
 			*action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
-		}
 		break;
 	case FLOW_ACTION_VLAN_POP_ETH:
 		parse_state->eth_pop = true;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 4e6f5caf8ab6..bf4cff8b1d42 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1884,7 +1884,6 @@ post_process_attr(struct mlx5e_tc_flow *flow,
 		  struct mlx5_flow_attr *attr,
 		  struct netlink_ext_ack *extack)
 {
-	struct mlx5_eswitch *esw = flow->priv->mdev->priv.eswitch;
 	bool vf_tun;
 	int err = 0;
 
@@ -1896,12 +1895,6 @@ post_process_attr(struct mlx5e_tc_flow *flow,
 	if (err)
 		goto err_out;
 
-	if (mlx5e_is_eswitch_flow(flow)) {
-		err = mlx5_eswitch_add_vlan_action(esw, attr);
-		if (err)
-			goto err_out;
-	}
-
 	if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) {
 		err = mlx5e_tc_attach_mod_hdr(flow->priv, flow, attr);
 		if (err)
@@ -2105,8 +2098,6 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv,
 	if (mlx5_flow_has_geneve_opt(flow))
 		mlx5_geneve_tlv_option_del(priv->mdev->geneve);
 
-	mlx5_eswitch_del_vlan_action(esw, attr);
-
 	if (flow->decap_route)
 		mlx5e_detach_decap_route(priv, flow);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 5b5a215a7dc5..fd03f076551b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -222,7 +222,6 @@ struct mlx5_eswitch_fdb {
 			struct mlx5_flow_handle **send_to_vport_meta_rules;
 			struct mlx5_flow_handle *miss_rule_uni;
 			struct mlx5_flow_handle *miss_rule_multi;
-			int vlan_push_pop_refcount;
 
 			struct mlx5_fs_chains *esw_chains_priv;
 			struct {
@@ -520,10 +519,6 @@ int mlx5_devlink_port_fn_migratable_set(struct devlink_port *port, bool enable,
 					struct netlink_ext_ack *extack);
 void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type);
 
-int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
-				 struct mlx5_flow_attr *attr);
-int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
-				 struct mlx5_flow_attr *attr);
 int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 				  u16 vport, u16 vlan, u8 qos, u8 set_flags);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 3a82e385544d..8fb09143e9e8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -579,16 +579,16 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
 	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
 		return ERR_PTR(-EOPNOTSUPP);
 
+	if (!mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
+		return ERR_PTR(-EOPNOTSUPP);
+
 	dest = kcalloc(MLX5_MAX_FLOW_FWD_VPORTS + 1, sizeof(*dest), GFP_KERNEL);
 	if (!dest)
 		return ERR_PTR(-ENOMEM);
 
 	flow_act.action = attr->action;
-	/* if per flow vlan pop/push is emulated, don't set that into the firmware */
-	if (!mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
-		flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
-				     MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
-	else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
+
+	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
 		flow_act.vlan[0].ethtype = ntohs(esw_attr->vlan_proto[0]);
 		flow_act.vlan[0].vid = esw_attr->vlan_vid[0];
 		flow_act.vlan[0].prio = esw_attr->vlan_prio[0];
@@ -829,204 +829,6 @@ mlx5_eswitch_del_fwd_rule(struct mlx5_eswitch *esw,
 	__mlx5_eswitch_del_rule(esw, rule, attr, true);
 }
 
-static int esw_set_global_vlan_pop(struct mlx5_eswitch *esw, u8 val)
-{
-	struct mlx5_eswitch_rep *rep;
-	unsigned long i;
-	int err = 0;
-
-	esw_debug(esw->dev, "%s applying global %s policy\n", __func__, val ? "pop" : "none");
-	mlx5_esw_for_each_host_func_vport(esw, i, rep, esw->esw_funcs.num_vfs) {
-		if (atomic_read(&rep->rep_data[REP_ETH].state) != REP_LOADED)
-			continue;
-
-		err = __mlx5_eswitch_set_vport_vlan(esw, rep->vport, 0, 0, val);
-		if (err)
-			goto out;
-	}
-
-out:
-	return err;
-}
-
-static struct mlx5_eswitch_rep *
-esw_vlan_action_get_vport(struct mlx5_esw_flow_attr *attr, bool push, bool pop)
-{
-	struct mlx5_eswitch_rep *in_rep, *out_rep, *vport = NULL;
-
-	in_rep  = attr->in_rep;
-	out_rep = attr->dests[0].rep;
-
-	if (push)
-		vport = in_rep;
-	else if (pop)
-		vport = out_rep;
-	else
-		vport = in_rep;
-
-	return vport;
-}
-
-static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr,
-				     bool push, bool pop, bool fwd)
-{
-	struct mlx5_eswitch_rep *in_rep, *out_rep;
-
-	if ((push || pop) && !fwd)
-		goto out_notsupp;
-
-	in_rep  = attr->in_rep;
-	out_rep = attr->dests[0].rep;
-
-	if (push && in_rep->vport == MLX5_VPORT_UPLINK)
-		goto out_notsupp;
-
-	if (pop && out_rep->vport == MLX5_VPORT_UPLINK)
-		goto out_notsupp;
-
-	/* vport has vlan push configured, can't offload VF --> wire rules w.o it */
-	if (!push && !pop && fwd)
-		if (in_rep->vlan && out_rep->vport == MLX5_VPORT_UPLINK)
-			goto out_notsupp;
-
-	/* protects against (1) setting rules with different vlans to push and
-	 * (2) setting rules w.o vlans (attr->vlan = 0) && w. vlans to push (!= 0)
-	 */
-	if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid[0]))
-		goto out_notsupp;
-
-	return 0;
-
-out_notsupp:
-	return -EOPNOTSUPP;
-}
-
-int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
-				 struct mlx5_flow_attr *attr)
-{
-	struct offloads_fdb *offloads = &esw->fdb_table.offloads;
-	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
-	struct mlx5_eswitch_rep *vport = NULL;
-	bool push, pop, fwd;
-	int err = 0;
-
-	/* nop if we're on the vlan push/pop non emulation mode */
-	if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
-		return 0;
-
-	push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH);
-	pop  = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
-	fwd  = !!((attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) &&
-		   !attr->dest_chain);
-
-	mutex_lock(&esw->state_lock);
-
-	err = esw_add_vlan_action_check(esw_attr, push, pop, fwd);
-	if (err)
-		goto unlock;
-
-	attr->flags &= ~MLX5_ATTR_FLAG_VLAN_HANDLED;
-
-	vport = esw_vlan_action_get_vport(esw_attr, push, pop);
-
-	if (!push && !pop && fwd) {
-		/* tracks VF --> wire rules without vlan push action */
-		if (esw_attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) {
-			vport->vlan_refcount++;
-			attr->flags |= MLX5_ATTR_FLAG_VLAN_HANDLED;
-		}
-
-		goto unlock;
-	}
-
-	if (!push && !pop)
-		goto unlock;
-
-	if (!(offloads->vlan_push_pop_refcount)) {
-		/* it's the 1st vlan rule, apply global vlan pop policy */
-		err = esw_set_global_vlan_pop(esw, SET_VLAN_STRIP);
-		if (err)
-			goto out;
-	}
-	offloads->vlan_push_pop_refcount++;
-
-	if (push) {
-		if (vport->vlan_refcount)
-			goto skip_set_push;
-
-		err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, esw_attr->vlan_vid[0],
-						    0, SET_VLAN_INSERT | SET_VLAN_STRIP);
-		if (err)
-			goto out;
-		vport->vlan = esw_attr->vlan_vid[0];
-skip_set_push:
-		vport->vlan_refcount++;
-	}
-out:
-	if (!err)
-		attr->flags |= MLX5_ATTR_FLAG_VLAN_HANDLED;
-unlock:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
-int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
-				 struct mlx5_flow_attr *attr)
-{
-	struct offloads_fdb *offloads = &esw->fdb_table.offloads;
-	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
-	struct mlx5_eswitch_rep *vport = NULL;
-	bool push, pop, fwd;
-	int err = 0;
-
-	/* nop if we're on the vlan push/pop non emulation mode */
-	if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
-		return 0;
-
-	if (!(attr->flags & MLX5_ATTR_FLAG_VLAN_HANDLED))
-		return 0;
-
-	push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH);
-	pop  = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
-	fwd  = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST);
-
-	mutex_lock(&esw->state_lock);
-
-	vport = esw_vlan_action_get_vport(esw_attr, push, pop);
-
-	if (!push && !pop && fwd) {
-		/* tracks VF --> wire rules without vlan push action */
-		if (esw_attr->dests[0].rep->vport == MLX5_VPORT_UPLINK)
-			vport->vlan_refcount--;
-
-		goto out;
-	}
-
-	if (push) {
-		vport->vlan_refcount--;
-		if (vport->vlan_refcount)
-			goto skip_unset_push;
-
-		vport->vlan = 0;
-		err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport,
-						    0, 0, SET_VLAN_STRIP);
-		if (err)
-			goto out;
-	}
-
-skip_unset_push:
-	offloads->vlan_push_pop_refcount--;
-	if (offloads->vlan_push_pop_refcount)
-		goto out;
-
-	/* no more vlan rules, stop global vlan pop policy */
-	err = esw_set_global_vlan_pop(esw, 0);
-
-out:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
 struct mlx5_flow_handle *
 mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *on_esw,
 				    struct mlx5_eswitch *from_esw,
-- 
2.39.1

[net-next 06/15] net/mlx5: fs, Remove redundant vport_number assignment

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

vport_number and other_vport being reassigned outside the if clause anyway.
remove the redundant assignment.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index 32d4c967469c..e6874298ba92 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -412,11 +412,6 @@ static int mlx5_cmd_create_flow_group(struct mlx5_flow_root_namespace *ns,
 		 MLX5_CMD_OP_CREATE_FLOW_GROUP);
 	MLX5_SET(create_flow_group_in, in, table_type, ft->type);
 	MLX5_SET(create_flow_group_in, in, table_id, ft->id);
-	if (ft->vport) {
-		MLX5_SET(create_flow_group_in, in, vport_number, ft->vport);
-		MLX5_SET(create_flow_group_in, in, other_vport, 1);
-	}
-
 	MLX5_SET(create_flow_group_in, in, vport_number, ft->vport);
 	MLX5_SET(create_flow_group_in, in, other_vport,
 		 !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT));
-- 
2.39.1

[net-next 07/15] net/mlx5e: Remove incorrect debugfs_create_dir NULL check in hairpin

[Saeed Mahameed]

From: Gal Pressman <gal@nvidia.com>

Remove the NULL check on debugfs_create_dir() return value as the
function returns an ERR pointer on failure, not NULL.
The check is not replaced with a IS_ERR_OR_NULL() as
debugfs_create_file(), and debugfs functions in general don't need error
checking.

Fixes: 0e414518d6d8 ("net/mlx5e: Add hairpin debugfs files")
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index bf4cff8b1d42..e2ec80ebde58 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1124,8 +1124,6 @@ static void mlx5e_tc_debugfs_init(struct mlx5e_tc_table *tc,
 		return;
 
 	tc->dfs_root = debugfs_create_dir("tc", dfs_root);
-	if (!tc->dfs_root)
-		return;
 
 	debugfs_create_file("hairpin_num_queues", 0644, tc->dfs_root,
 			    &tc->hairpin_params, &fops_hairpin_queues);
-- 
2.39.1

[net-next 08/15] net/mlx5e: Remove incorrect debugfs_create_dir NULL check in TLS

[Saeed Mahameed]

From: Gal Pressman <gal@nvidia.com>

Remove the NULL check on debugfs_create_dir() return value as the
function returns an ERR pointer on failure, not NULL.
The check is not replaced with a IS_ERR_OR_NULL() as
debugfs_create_file(), and debugfs functions in general don't need error
checking.

Fixes: 0fedee1ae9ef ("net/mlx5e: kTLS, Add debugfs")
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c
index e80b43b7aac9..60b3e08a1028 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c
@@ -899,8 +899,6 @@ static void mlx5e_tls_tx_debugfs_init(struct mlx5e_tls *tls,
 		return;
 
 	tls->debugfs.dfs_tx = debugfs_create_dir("tx", dfs_root);
-	if (!tls->debugfs.dfs_tx)
-		return;
 
 	debugfs_create_size_t("pool_size", 0400, tls->debugfs.dfs_tx,
 			      &tls->tx_pool->size);
-- 
2.39.1

[net-next 09/15] net/mlx5: Fix memory leak in error flow of port set buffer

[Saeed Mahameed]

From: Maher Sanalla <msanalla@nvidia.com>

In the cited commit, shared buffer updates were added whenever
port buffer gets updated.

However, in case the shared buffer update fails, exiting early from
port_set_buffer() is performed without freeing previously-allocated memory.

Fix it by jumping to out label where memory is freed before returning
with error.

Fixes: a440030d8946 ("net/mlx5e: Update shared buffer along with device buffer changes")
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c
index 57f4b1b50421..7ac1ad9c46de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c
@@ -314,11 +314,11 @@ static int port_set_buffer(struct mlx5e_priv *priv,
 	err = port_update_shared_buffer(priv->mdev, current_headroom_size,
 					new_headroom_size);
 	if (err)
-		return err;
+		goto out;
 
 	err = port_update_pool_cfg(priv->mdev, port_buffer);
 	if (err)
-		return err;
+		goto out;
 
 	err = mlx5e_port_set_pbmc(mdev, in);
 out:
-- 
2.39.1

[net-next 10/15] net/mlx5: fs_core, Remove redundant variable err

[Saeed Mahameed]

From: Maor Dickman <maord@nvidia.com>

Local variable "err" is not used so it is safe to remove.

Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index dd43a940499b..fad479df12e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -1776,7 +1776,6 @@ static int build_match_list(struct match_list *match_head,
 {
 	struct rhlist_head *tmp, *list;
 	struct mlx5_flow_group *g;
-	int err = 0;
 
 	rcu_read_lock();
 	INIT_LIST_HEAD(&match_head->list);
@@ -1802,7 +1801,7 @@ static int build_match_list(struct match_list *match_head,
 		list_add_tail(&curr_match->list, &match_head->list);
 	}
 	rcu_read_unlock();
-	return err;
+	return 0;
 }
 
 static u64 matched_fgs_get_version(struct list_head *match_head)
-- 
2.39.1

[net-next 11/15] net/mlx5: fs, Remove redundant assignment of size

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

size is being reassigned in the line after.
remove the redundant assignment.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index e6874298ba92..374c17445e54 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -272,8 +272,6 @@ static int mlx5_cmd_create_flow_table(struct mlx5_flow_root_namespace *ns,
 	unsigned int size;
 	int err;
 
-	if (ft_attr->max_fte != POOL_NEXT_SIZE)
-		size = roundup_pow_of_two(ft_attr->max_fte);
 	size = mlx5_ft_pool_get_avail_sz(dev, ft->type, ft_attr->max_fte);
 	if (!size)
 		return -ENOSPC;
-- 
2.39.1

[net-next 12/15] net/mlx5: fw_tracer: Fix debug print

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

The debug message specify tdsn, but takes as an argument the
tmsn. The correct argument is tmsn, hence, fix the print.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 21831386b26e..4ce75cef46c8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -602,7 +602,7 @@ static int mlx5_tracer_handle_string_trace(struct mlx5_fw_tracer *tracer,
 	} else {
 		cur_string = mlx5_tracer_message_get(tracer, tracer_event);
 		if (!cur_string) {
-			pr_debug("%s Got string event for unknown string tdsm: %d\n",
+			pr_debug("%s Got string event for unknown string tmsn: %d\n",
 				 __func__, tracer_event->string_event.tmsn);
 			return -1;
 		}
-- 
2.39.1

[net-next 13/15] net/mlx5: fw_tracer, allow 0 size string DBs

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

Device can expose string DB of size 0 which means this string DB is
currently not in use. Therefore, allow for 0 size string DBs.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 4ce75cef46c8..de98357dfd14 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -233,6 +233,8 @@ static int mlx5_fw_tracer_allocate_strings_db(struct mlx5_fw_tracer *tracer)
 	int i;
 
 	for (i = 0; i < num_string_db; i++) {
+		if (!string_db_size_out[i])
+			continue;
 		tracer->str_db.buffer[i] = kzalloc(string_db_size_out[i], GFP_KERNEL);
 		if (!tracer->str_db.buffer[i])
 			goto free_strings_db;
@@ -278,6 +280,8 @@ static void mlx5_tracer_read_strings_db(struct work_struct *work)
 	}
 
 	for (i = 0; i < num_string_db; i++) {
+		if (!tracer->str_db.size_out[i])
+			continue;
 		offset = 0;
 		MLX5_SET(mtrc_stdb, in, string_db_index, i);
 		num_of_reads = tracer->str_db.size_out[i] /
@@ -384,6 +388,8 @@ static struct tracer_string_format *mlx5_tracer_get_string(struct mlx5_fw_tracer
 	str_ptr = tracer_event->string_event.string_param;
 
 	for (i = 0; i < tracer->str_db.num_string_db; i++) {
+		if (!tracer->str_db.size_out[i])
+			continue;
 		if (str_ptr > tracer->str_db.base_address_out[i] &&
 		    str_ptr < tracer->str_db.base_address_out[i] +
 		    tracer->str_db.size_out[i]) {
-- 
2.39.1

[net-next 14/15] net/mlx5: fw_tracer, Add support for strings DB update event

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

In case a new string DB is added to the FW, the FW publishes an event
notifying the strings DB have updated.

Add support in driver for handling this event.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/diag/fw_tracer.c       | 46 ++++++++++++++++---
 .../mellanox/mlx5/core/diag/fw_tracer.h       |  8 ++++
 include/linux/mlx5/device.h                   |  1 +
 3 files changed, 49 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index de98357dfd14..46f1e866bf64 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -936,6 +936,14 @@ int mlx5_fw_tracer_get_saved_traces_objects(struct mlx5_fw_tracer *tracer,
 	return err;
 }
 
+static void mlx5_fw_tracer_update_db(struct work_struct *work)
+{
+	struct mlx5_fw_tracer *tracer =
+			container_of(work, struct mlx5_fw_tracer, update_db_work);
+
+	mlx5_fw_tracer_reload(tracer);
+}
+
 /* Create software resources (Buffers, etc ..) */
 struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
 {
@@ -963,6 +971,8 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
 	INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
 	INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
 	INIT_WORK(&tracer->handle_traces_work, mlx5_fw_tracer_handle_traces);
+	INIT_WORK(&tracer->update_db_work, mlx5_fw_tracer_update_db);
+	mutex_init(&tracer->state_lock);
 
 
 	err = mlx5_query_mtrc_caps(tracer);
@@ -1009,11 +1019,15 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
 	if (IS_ERR_OR_NULL(tracer))
 		return 0;
 
-	dev = tracer->dev;
-
 	if (!tracer->str_db.loaded)
 		queue_work(tracer->work_queue, &tracer->read_fw_strings_work);
 
+	mutex_lock(&tracer->state_lock);
+	if (test_and_set_bit(MLX5_TRACER_STATE_UP, &tracer->state))
+		goto unlock;
+
+	dev = tracer->dev;
+
 	err = mlx5_core_alloc_pd(dev, &tracer->buff.pdn);
 	if (err) {
 		mlx5_core_warn(dev, "FWTracer: Failed to allocate PD %d\n", err);
@@ -1034,6 +1048,8 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
 		mlx5_core_warn(dev, "FWTracer: Failed to start tracer %d\n", err);
 		goto err_notifier_unregister;
 	}
+unlock:
+	mutex_unlock(&tracer->state_lock);
 	return 0;
 
 err_notifier_unregister:
@@ -1043,6 +1059,7 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
 	mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
 err_cancel_work:
 	cancel_work_sync(&tracer->read_fw_strings_work);
+	mutex_unlock(&tracer->state_lock);
 	return err;
 }
 
@@ -1052,17 +1069,27 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
 	if (IS_ERR_OR_NULL(tracer))
 		return;
 
+	mutex_lock(&tracer->state_lock);
+	if (!test_and_clear_bit(MLX5_TRACER_STATE_UP, &tracer->state))
+		goto unlock;
+
 	mlx5_core_dbg(tracer->dev, "FWTracer: Cleanup, is owner ? (%d)\n",
 		      tracer->owner);
 	mlx5_eq_notifier_unregister(tracer->dev, &tracer->nb);
 	cancel_work_sync(&tracer->ownership_change_work);
 	cancel_work_sync(&tracer->handle_traces_work);
+	/* It is valid to get here from update_db_work. Hence, don't wait for
+	 * update_db_work to finished.
+	 */
+	cancel_work(&tracer->update_db_work);
 
 	if (tracer->owner)
 		mlx5_fw_tracer_ownership_release(tracer);
 
 	mlx5_core_destroy_mkey(tracer->dev, tracer->buff.mkey);
 	mlx5_core_dealloc_pd(tracer->dev, tracer->buff.pdn);
+unlock:
+	mutex_unlock(&tracer->state_lock);
 }
 
 /* Free software resources (Buffers, etc ..) */
@@ -1079,6 +1106,7 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
 	mlx5_fw_tracer_clean_saved_traces_array(tracer);
 	mlx5_fw_tracer_free_strings_db(tracer);
 	mlx5_fw_tracer_destroy_log_buf(tracer);
+	mutex_destroy(&tracer->state_lock);
 	destroy_workqueue(tracer->work_queue);
 	kvfree(tracer);
 }
@@ -1088,6 +1116,8 @@ static int mlx5_fw_tracer_recreate_strings_db(struct mlx5_fw_tracer *tracer)
 	struct mlx5_core_dev *dev;
 	int err;
 
+	if (test_and_set_bit(MLX5_TRACER_RECREATE_DB, &tracer->state))
+		return 0;
 	cancel_work_sync(&tracer->read_fw_strings_work);
 	mlx5_fw_tracer_clean_ready_list(tracer);
 	mlx5_fw_tracer_clean_print_hash(tracer);
@@ -1098,17 +1128,18 @@ static int mlx5_fw_tracer_recreate_strings_db(struct mlx5_fw_tracer *tracer)
 	err = mlx5_query_mtrc_caps(tracer);
 	if (err) {
 		mlx5_core_dbg(dev, "FWTracer: Failed to query capabilities %d\n", err);
-		return err;
+		goto out;
 	}
 
 	err = mlx5_fw_tracer_allocate_strings_db(tracer);
 	if (err) {
 		mlx5_core_warn(dev, "FWTracer: Allocate strings DB failed %d\n", err);
-		return err;
+		goto out;
 	}
 	mlx5_fw_tracer_init_saved_traces_array(tracer);
-
-	return 0;
+out:
+	clear_bit(MLX5_TRACER_RECREATE_DB, &tracer->state);
+	return err;
 }
 
 int mlx5_fw_tracer_reload(struct mlx5_fw_tracer *tracer)
@@ -1148,6 +1179,9 @@ static int fw_tracer_event(struct notifier_block *nb, unsigned long action, void
 	case MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE:
 		queue_work(tracer->work_queue, &tracer->handle_traces_work);
 		break;
+	case MLX5_TRACER_SUBTYPE_STRINGS_DB_UPDATE:
+		queue_work(tracer->work_queue, &tracer->update_db_work);
+		break;
 	default:
 		mlx5_core_dbg(dev, "FWTracer: Event with unrecognized subtype: sub_type %d\n",
 			      eqe->sub_type);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 4762b55b0b0e..873c1eb38e49 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -63,6 +63,11 @@ struct mlx5_fw_trace_data {
 	char msg[TRACE_STR_MSG];
 };
 
+enum mlx5_fw_tracer_state {
+	MLX5_TRACER_STATE_UP = BIT(0),
+	MLX5_TRACER_RECREATE_DB = BIT(1),
+};
+
 struct mlx5_fw_tracer {
 	struct mlx5_core_dev *dev;
 	struct mlx5_nb        nb;
@@ -104,6 +109,9 @@ struct mlx5_fw_tracer {
 	struct work_struct handle_traces_work;
 	struct hlist_head hash[MESSAGE_HASH_SIZE];
 	struct list_head ready_strings_list;
+	struct work_struct update_db_work;
+	struct mutex state_lock; /* Synchronize update work with reload flows */
+	unsigned long state;
 };
 
 struct tracer_string_format {
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index bc531bd9804f..70b3eacf2bbb 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -367,6 +367,7 @@ enum mlx5_driver_event {
 enum {
 	MLX5_TRACER_SUBTYPE_OWNERSHIP_CHANGE = 0x0,
 	MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE = 0x1,
+	MLX5_TRACER_SUBTYPE_STRINGS_DB_UPDATE = 0x2,
 };
 
 enum {
-- 
2.39.1

[net-next 15/15] net/mlx5: fw_tracer, Add support for unrecognized string

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

In case FW is publishing a string which isn't found in the driver's
string DBs, keep the string as raw data.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/diag/fw_tracer.c       | 25 +++++++++++++++++--
 .../mellanox/mlx5/core/diag/fw_tracer.h       |  1 +
 2 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 46f1e866bf64..f7864d51d4c4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -465,6 +465,7 @@ static void poll_trace(struct mlx5_fw_tracer *tracer,
 
 	tracer_event->event_id = MLX5_GET(tracer_event, trace, event_id);
 	tracer_event->lost_event = MLX5_GET(tracer_event, trace, lost);
+	tracer_event->out = trace;
 
 	switch (tracer_event->event_id) {
 	case TRACER_EVENT_TYPE_TIMESTAMP:
@@ -587,6 +588,26 @@ void mlx5_tracer_print_trace(struct tracer_string_format *str_frmt,
 	mlx5_tracer_clean_message(str_frmt);
 }
 
+static int mlx5_tracer_handle_raw_string(struct mlx5_fw_tracer *tracer,
+					 struct tracer_event *tracer_event)
+{
+	struct tracer_string_format *cur_string;
+
+	cur_string = mlx5_tracer_message_insert(tracer, tracer_event);
+	if (!cur_string)
+		return -1;
+
+	cur_string->event_id = tracer_event->event_id;
+	cur_string->timestamp = tracer_event->string_event.timestamp;
+	cur_string->lost = tracer_event->lost_event;
+	cur_string->string = "0x%08x%08x";
+	cur_string->num_of_params = 2;
+	cur_string->params[0] = upper_32_bits(*tracer_event->out);
+	cur_string->params[1] = lower_32_bits(*tracer_event->out);
+	list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+	return 0;
+}
+
 static int mlx5_tracer_handle_string_trace(struct mlx5_fw_tracer *tracer,
 					   struct tracer_event *tracer_event)
 {
@@ -595,7 +616,7 @@ static int mlx5_tracer_handle_string_trace(struct mlx5_fw_tracer *tracer,
 	if (tracer_event->string_event.tdsn == 0) {
 		cur_string = mlx5_tracer_get_string(tracer, tracer_event);
 		if (!cur_string)
-			return -1;
+			return mlx5_tracer_handle_raw_string(tracer, tracer_event);
 
 		cur_string->num_of_params = mlx5_tracer_get_num_of_params(cur_string->string);
 		cur_string->last_param_num = 0;
@@ -610,7 +631,7 @@ static int mlx5_tracer_handle_string_trace(struct mlx5_fw_tracer *tracer,
 		if (!cur_string) {
 			pr_debug("%s Got string event for unknown string tmsn: %d\n",
 				 __func__, tracer_event->string_event.tmsn);
-			return -1;
+			return mlx5_tracer_handle_raw_string(tracer, tracer_event);
 		}
 		cur_string->last_param_num += 1;
 		if (cur_string->last_param_num > TRACER_MAX_PARAMS) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 873c1eb38e49..5c548bb74f07 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -166,6 +166,7 @@ struct tracer_event {
 		struct tracer_string_event string_event;
 		struct tracer_timestamp_event timestamp_event;
 	};
+	u64 *out;
 };
 
 struct mlx5_ifc_tracer_event_bits {
-- 
2.39.1