[pull request][net-next 00/15] mlx5 updates 2021-03-11

[pull request][net-next 00/15] mlx5 updates 2021-03-11

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

Hi Dave, Jakub,

This series provides some cleanups to mlx5 driver
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 34bb975126419e86bc3b95e200dc41de6c6ca69c:

  net: fddi: skfp: Mundane typo fixes throughout the file smt.h (2021-03-10 15:42:22 -0800)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2021-03-11

for you to fetch changes up to 9f4d9283388d9069dcf4d68ae9a212d7f0e6a985:

  net/mlx5e: Alloc flow spec using kvzalloc instead of kzalloc (2021-03-11 14:35:15 -0800)

----------------------------------------------------------------
mlx5-updates-2021-03-11

Cleanups for mlx5 driver

1) Fix build warnings form Arnd and Vlad
2) Leon improves locking for driver load/unload flows
3) From Roi, Lockdep false dependency warning
4) Other trivial cleanups

----------------------------------------------------------------
Arnd Bergmann (1):
      net/mlx5e: fix mlx5e_tc_tun_update_header_ipv6 dummy definition

Eli Cohen (1):
      net/mlx5: Avoid unnecessary operation

Leon Romanovsky (5):
      net/mlx5: Remove impossible checks of interface state
      net/mlx5: Separate probe vs. reload flows
      net/mlx5: Remove second FW tracer check
      net/mlx5: Don't rely on interface state bit
      net/mlx5: Check returned value from health recover sequence

Roi Dayan (3):
      net/mlx5e: CT, Avoid false lock dependency warning
      net/mlx5: SF, Fix return type
      net/mlx5e: Alloc flow spec using kvzalloc instead of kzalloc

Saeed Mahameed (3):
      net/mlx5: Don't skip vport check
      net/mlx5e: mlx5_tc_ct_init does not fail
      net/mlx5e: rep: Improve reg_cX conditions

Vlad Buslov (2):
      net/mlx5e: Add missing include
      net/mlx5: Fix indir stable stubs

 drivers/net/ethernet/mellanox/mlx5/core/devlink.c  |   6 +-
 .../ethernet/mellanox/mlx5/core/diag/fw_tracer.c   |   5 +-
 .../net/ethernet/mellanox/mlx5/core/en/rep/tc.c    |   8 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c |  24 +++-
 .../net/ethernet/mellanox/mlx5/core/en/tc_tun.h    |  10 +-
 .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c  |   1 +
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         |   4 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |   8 --
 .../ethernet/mellanox/mlx5/core/esw/devlink_port.c |   2 +-
 .../ethernet/mellanox/mlx5/core/esw/indir_table.h  |   6 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c  |   6 +
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |  10 +-
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  |   6 +-
 drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c |  13 +-
 drivers/net/ethernet/mellanox/mlx5/core/health.c   |   6 +-
 drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c  |   5 -
 drivers/net/ethernet/mellanox/mlx5/core/main.c     | 143 ++++++++++++++-------
 .../net/ethernet/mellanox/mlx5/core/mlx5_core.h    |   8 +-
 .../net/ethernet/mellanox/mlx5/core/sf/dev/dev.h   |   2 +-
 .../ethernet/mellanox/mlx5/core/sf/dev/driver.c    |  14 +-
 20 files changed, 169 insertions(+), 118 deletions(-)

[net-next 01/15] net/mlx5: Don't skip vport check

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

Users of mlx5_eswitch_get_vport() are required to check return value
prior to passing mlx5_vport further. Fix all the places to do not skip
that check.

Reviewed-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c          | 6 ++++++
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 6 ++++++
 3 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c
index cb1e181f4c6a..7bfc84238b3d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c
@@ -120,7 +120,7 @@ struct devlink_port *mlx5_esw_offloads_devlink_port(struct mlx5_eswitch *esw, u1
 	struct mlx5_vport *vport;
 
 	vport = mlx5_eswitch_get_vport(esw, vport_num);
-	return vport->dl_port;
+	return IS_ERR(vport) ? ERR_CAST(vport) : vport->dl_port;
 }
 
 int mlx5_esw_devlink_sf_port_register(struct mlx5_eswitch *esw, struct devlink_port *dl_port,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index aba17835465b..fe1e06d95a12 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1141,6 +1141,8 @@ int mlx5_esw_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num,
 	struct mlx5_vport *vport;
 
 	vport = mlx5_eswitch_get_vport(esw, vport_num);
+	if (IS_ERR(vport))
+		return PTR_ERR(vport);
 
 	if (!vport->qos.enabled)
 		return -EOPNOTSUPP;
@@ -1279,6 +1281,8 @@ int mlx5_esw_vport_enable(struct mlx5_eswitch *esw, u16 vport_num,
 	int ret;
 
 	vport = mlx5_eswitch_get_vport(esw, vport_num);
+	if (IS_ERR(vport))
+		return PTR_ERR(vport);
 
 	mutex_lock(&esw->state_lock);
 	WARN_ON(vport->enabled);
@@ -1326,6 +1330,8 @@ void mlx5_esw_vport_disable(struct mlx5_eswitch *esw, u16 vport_num)
 	struct mlx5_vport *vport;
 
 	vport = mlx5_eswitch_get_vport(esw, vport_num);
+	if (IS_ERR(vport))
+		return;
 
 	mutex_lock(&esw->state_lock);
 	if (!vport->enabled)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 94cb0217b4f3..703753ac2e02 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -2554,6 +2554,9 @@ static int esw_create_uplink_offloads_acl_tables(struct mlx5_eswitch *esw)
 	struct mlx5_vport *vport;
 
 	vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK);
+	if (IS_ERR(vport))
+		return PTR_ERR(vport);
+
 	return esw_vport_create_offloads_acl_tables(esw, vport);
 }
 
@@ -2562,6 +2565,9 @@ static void esw_destroy_uplink_offloads_acl_tables(struct mlx5_eswitch *esw)
 	struct mlx5_vport *vport;
 
 	vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK);
+	if (IS_ERR(vport))
+		return;
+
 	esw_vport_destroy_offloads_acl_tables(esw, vport);
 }
 
-- 
2.29.2

Re: [net-next 01/15] net/mlx5: Don't skip vport check

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (refs/heads/master):

On Thu, 11 Mar 2021 14:37:09 -0800 you wrote:
> From: Saeed Mahameed <saeedm@nvidia.com>
> 
> Users of mlx5_eswitch_get_vport() are required to check return value
> prior to passing mlx5_vport further. Fix all the places to do not skip
> that check.
> 
> Reviewed-by: Eli Cohen <elic@nvidia.com>
> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
> 
> [...]

Here is the summary with links:
  - [net-next,01/15] net/mlx5: Don't skip vport check
    https://git.kernel.org/netdev/net-next/c/7bef147a6ab6
  - [net-next,02/15] net/mlx5: Remove impossible checks of interface state
    https://git.kernel.org/netdev/net-next/c/d89edb360705
  - [net-next,03/15] net/mlx5: Separate probe vs. reload flows
    https://git.kernel.org/netdev/net-next/c/6dea2f7eff96
  - [net-next,04/15] net/mlx5: Remove second FW tracer check
    https://git.kernel.org/netdev/net-next/c/7e615b997802
  - [net-next,05/15] net/mlx5: Don't rely on interface state bit
    https://git.kernel.org/netdev/net-next/c/7ad67a20f28f
  - [net-next,06/15] net/mlx5: Check returned value from health recover sequence
    https://git.kernel.org/netdev/net-next/c/fe06992b04a9
  - [net-next,07/15] net/mlx5e: CT, Avoid false lock dependency warning
    https://git.kernel.org/netdev/net-next/c/76e68d950a17
  - [net-next,08/15] net/mlx5e: fix mlx5e_tc_tun_update_header_ipv6 dummy definition
    https://git.kernel.org/netdev/net-next/c/87f77a679797
  - [net-next,09/15] net/mlx5e: Add missing include
    https://git.kernel.org/netdev/net-next/c/5632817b144f
  - [net-next,10/15] net/mlx5: Fix indir stable stubs
    https://git.kernel.org/netdev/net-next/c/fbeab6be054c
  - [net-next,11/15] net/mlx5e: mlx5_tc_ct_init does not fail
    https://git.kernel.org/netdev/net-next/c/51ada5a52379
  - [net-next,12/15] net/mlx5: SF, Fix return type
    https://git.kernel.org/netdev/net-next/c/3094552bcd72
  - [net-next,13/15] net/mlx5e: rep: Improve reg_cX conditions
    https://git.kernel.org/netdev/net-next/c/03e219c4cf84
  - [net-next,14/15] net/mlx5: Avoid unnecessary operation
    https://git.kernel.org/netdev/net-next/c/61e9508f1e5e
  - [net-next,15/15] net/mlx5e: Alloc flow spec using kvzalloc instead of kzalloc
    https://git.kernel.org/netdev/net-next/c/9f4d9283388d

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 impossible checks of interface state

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

The interface state is constant at this stage and checked
before calling to the register/unregister reserved GIDs.

There is no need to double check it.

Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c
index a68738c8f4bc..97324d9d4f2a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c
@@ -55,10 +55,6 @@ void mlx5_cleanup_reserved_gids(struct mlx5_core_dev *dev)
 
 int mlx5_core_reserve_gids(struct mlx5_core_dev *dev, unsigned int count)
 {
-	if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) {
-		mlx5_core_err(dev, "Cannot reserve GIDs when interfaces are up\n");
-		return -EPERM;
-	}
 	if (dev->roce.reserved_gids.start < count) {
 		mlx5_core_warn(dev, "GID table exhausted attempting to reserve %d more GIDs\n",
 			       count);
@@ -79,7 +75,6 @@ int mlx5_core_reserve_gids(struct mlx5_core_dev *dev, unsigned int count)
 
 void mlx5_core_unreserve_gids(struct mlx5_core_dev *dev, unsigned int count)
 {
-	WARN(test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state), "Unreserving GIDs when interfaces are up");
 	WARN(count > dev->roce.reserved_gids.count, "Unreserving %u GIDs when only %u reserved",
 	     count, dev->roce.reserved_gids.count);
 
-- 
2.29.2

[net-next 03/15] net/mlx5: Separate probe vs. reload flows

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

The mix between probe/unprobe and reload flows causes to have an extra
mutex lock intf_state_mutex that generates LOCKDEP warning between it
and devlink_mutex. As a preparation for the future removal, separate
those flows.

Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/devlink.c |   6 +-
 .../ethernet/mellanox/mlx5/core/fw_reset.c    |   6 +-
 .../net/ethernet/mellanox/mlx5/core/main.c    | 136 ++++++++++++------
 .../ethernet/mellanox/mlx5/core/mlx5_core.h   |   6 +-
 .../mellanox/mlx5/core/sf/dev/driver.c        |  14 +-
 5 files changed, 107 insertions(+), 61 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index d7d8a68ef23d..6729720e1ab7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -148,7 +148,7 @@ static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change,
 
 	switch (action) {
 	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
-		mlx5_unload_one(dev, false);
+		mlx5_unload_one(dev);
 		return 0;
 	case DEVLINK_RELOAD_ACTION_FW_ACTIVATE:
 		if (limit == DEVLINK_RELOAD_LIMIT_NO_RESET)
@@ -170,13 +170,13 @@ static int mlx5_devlink_reload_up(struct devlink *devlink, enum devlink_reload_a
 	*actions_performed = BIT(action);
 	switch (action) {
 	case DEVLINK_RELOAD_ACTION_DRIVER_REINIT:
-		return mlx5_load_one(dev, false);
+		return mlx5_load_one(dev);
 	case DEVLINK_RELOAD_ACTION_FW_ACTIVATE:
 		if (limit == DEVLINK_RELOAD_LIMIT_NO_RESET)
 			break;
 		/* On fw_activate action, also driver is reloaded and reinit performed */
 		*actions_performed |= BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT);
-		return mlx5_load_one(dev, false);
+		return mlx5_load_one(dev);
 	default:
 		/* Unsupported action should not get to this function */
 		WARN_ON(1);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index f9042e147c7f..255bd0059da1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -104,7 +104,7 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev)
 	if (test_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags)) {
 		complete(&fw_reset->done);
 	} else {
-		mlx5_load_one(dev, false);
+		mlx5_load_one(dev);
 		devlink_remote_reload_actions_performed(priv_to_devlink(dev), 0,
 							BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
 							BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE));
@@ -119,7 +119,7 @@ static void mlx5_sync_reset_reload_work(struct work_struct *work)
 	int err;
 
 	mlx5_enter_error_state(dev, true);
-	mlx5_unload_one(dev, false);
+	mlx5_unload_one(dev);
 	err = mlx5_health_wait_pci_up(dev);
 	if (err)
 		mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n");
@@ -342,7 +342,7 @@ static void mlx5_sync_reset_now_event(struct work_struct *work)
 	}
 
 	mlx5_enter_error_state(dev, true);
-	mlx5_unload_one(dev, false);
+	mlx5_unload_one(dev);
 done:
 	fw_reset->ret = err;
 	mlx5_fw_reset_complete_reload(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index c568896cfb23..363bc3e917c2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1235,7 +1235,7 @@ static void mlx5_unload(struct mlx5_core_dev *dev)
 	mlx5_put_uars_page(dev, dev->priv.uar);
 }
 
-int mlx5_load_one(struct mlx5_core_dev *dev, bool boot)
+int mlx5_init_one(struct mlx5_core_dev *dev)
 {
 	int err = 0;
 
@@ -1247,16 +1247,14 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot)
 	/* remove any previous indication of internal error */
 	dev->state = MLX5_DEVICE_STATE_UP;
 
-	err = mlx5_function_setup(dev, boot);
+	err = mlx5_function_setup(dev, true);
 	if (err)
 		goto err_function;
 
-	if (boot) {
-		err = mlx5_init_once(dev);
-		if (err) {
-			mlx5_core_err(dev, "sw objs init failed\n");
-			goto function_teardown;
-		}
+	err = mlx5_init_once(dev);
+	if (err) {
+		mlx5_core_err(dev, "sw objs init failed\n");
+		goto function_teardown;
 	}
 
 	err = mlx5_load(dev);
@@ -1265,16 +1263,11 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot)
 
 	set_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 
-	if (boot) {
-		err = mlx5_devlink_register(priv_to_devlink(dev), dev->device);
-		if (err)
-			goto err_devlink_reg;
-
-		err = mlx5_register_device(dev);
-	} else {
-		err = mlx5_attach_device(dev);
-	}
+	err = mlx5_devlink_register(priv_to_devlink(dev), dev->device);
+	if (err)
+		goto err_devlink_reg;
 
+	err = mlx5_register_device(dev);
 	if (err)
 		goto err_register;
 
@@ -1282,16 +1275,14 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot)
 	return 0;
 
 err_register:
-	if (boot)
-		mlx5_devlink_unregister(priv_to_devlink(dev));
+	mlx5_devlink_unregister(priv_to_devlink(dev));
 err_devlink_reg:
 	clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 	mlx5_unload(dev);
 err_load:
-	if (boot)
-		mlx5_cleanup_once(dev);
+	mlx5_cleanup_once(dev);
 function_teardown:
-	mlx5_function_teardown(dev, boot);
+	mlx5_function_teardown(dev, true);
 err_function:
 	dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR;
 out:
@@ -1299,33 +1290,84 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot)
 	return err;
 }
 
-void mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup)
+void mlx5_uninit_one(struct mlx5_core_dev *dev)
 {
 	mutex_lock(&dev->intf_state_mutex);
 
-	if (cleanup) {
-		mlx5_unregister_device(dev);
-		mlx5_devlink_unregister(priv_to_devlink(dev));
-	} else {
-		mlx5_detach_device(dev);
-	}
+	mlx5_unregister_device(dev);
+	mlx5_devlink_unregister(priv_to_devlink(dev));
 
 	if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) {
 		mlx5_core_warn(dev, "%s: interface is down, NOP\n",
 			       __func__);
-		if (cleanup)
-			mlx5_cleanup_once(dev);
+		mlx5_cleanup_once(dev);
 		goto out;
 	}
 
 	clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
+	mlx5_unload(dev);
+	mlx5_cleanup_once(dev);
+	mlx5_function_teardown(dev, true);
+out:
+	mutex_unlock(&dev->intf_state_mutex);
+}
+
+int mlx5_load_one(struct mlx5_core_dev *dev)
+{
+	int err = 0;
+
+	mutex_lock(&dev->intf_state_mutex);
+	if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) {
+		mlx5_core_warn(dev, "interface is up, NOP\n");
+		goto out;
+	}
+	/* remove any previous indication of internal error */
+	dev->state = MLX5_DEVICE_STATE_UP;
+
+	err = mlx5_function_setup(dev, false);
+	if (err)
+		goto err_function;
+
+	err = mlx5_load(dev);
+	if (err)
+		goto err_load;
+
+	set_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 
+	err = mlx5_attach_device(dev);
+	if (err)
+		goto err_attach;
+
+	mutex_unlock(&dev->intf_state_mutex);
+	return 0;
+
+err_attach:
+	clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 	mlx5_unload(dev);
+err_load:
+	mlx5_function_teardown(dev, false);
+err_function:
+	dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR;
+out:
+	mutex_unlock(&dev->intf_state_mutex);
+	return err;
+}
 
-	if (cleanup)
-		mlx5_cleanup_once(dev);
+void mlx5_unload_one(struct mlx5_core_dev *dev)
+{
+	mutex_lock(&dev->intf_state_mutex);
+
+	mlx5_detach_device(dev);
+
+	if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) {
+		mlx5_core_warn(dev, "%s: interface is down, NOP\n",
+			       __func__);
+		goto out;
+	}
 
-	mlx5_function_teardown(dev, cleanup);
+	clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
+	mlx5_unload(dev);
+	mlx5_function_teardown(dev, false);
 out:
 	mutex_unlock(&dev->intf_state_mutex);
 }
@@ -1397,7 +1439,7 @@ void mlx5_mdev_uninit(struct mlx5_core_dev *dev)
 	mutex_destroy(&dev->intf_state_mutex);
 }
 
-static int init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+static int probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct mlx5_core_dev *dev;
 	struct devlink *devlink;
@@ -1433,11 +1475,11 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto pci_init_err;
 	}
 
-	err = mlx5_load_one(dev, true);
+	err = mlx5_init_one(dev);
 	if (err) {
-		mlx5_core_err(dev, "mlx5_load_one failed with error code %d\n",
+		mlx5_core_err(dev, "mlx5_init_one failed with error code %d\n",
 			      err);
-		goto err_load_one;
+		goto err_init_one;
 	}
 
 	err = mlx5_crdump_enable(dev);
@@ -1449,7 +1491,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		devlink_reload_enable(devlink);
 	return 0;
 
-err_load_one:
+err_init_one:
 	mlx5_pci_close(dev);
 pci_init_err:
 	mlx5_mdev_uninit(dev);
@@ -1469,7 +1511,7 @@ static void remove_one(struct pci_dev *pdev)
 	devlink_reload_disable(devlink);
 	mlx5_crdump_disable(dev);
 	mlx5_drain_health_wq(dev);
-	mlx5_unload_one(dev, true);
+	mlx5_uninit_one(dev);
 	mlx5_pci_close(dev);
 	mlx5_mdev_uninit(dev);
 	mlx5_adev_idx_free(dev->priv.adev_idx);
@@ -1485,7 +1527,7 @@ static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev,
 
 	mlx5_enter_error_state(dev, false);
 	mlx5_error_sw_reset(dev);
-	mlx5_unload_one(dev, false);
+	mlx5_unload_one(dev);
 	mlx5_drain_health_wq(dev);
 	mlx5_pci_disable_device(dev);
 
@@ -1555,7 +1597,7 @@ static void mlx5_pci_resume(struct pci_dev *pdev)
 
 	mlx5_core_info(dev, "%s was called\n", __func__);
 
-	err = mlx5_load_one(dev, false);
+	err = mlx5_load_one(dev);
 	if (err)
 		mlx5_core_err(dev, "%s: mlx5_load_one failed with error code: %d\n",
 			      __func__, err);
@@ -1627,7 +1669,7 @@ static void shutdown(struct pci_dev *pdev)
 	mlx5_core_info(dev, "Shutdown was called\n");
 	err = mlx5_try_fast_unload(dev);
 	if (err)
-		mlx5_unload_one(dev, false);
+		mlx5_unload_one(dev);
 	mlx5_pci_disable_device(dev);
 }
 
@@ -1635,7 +1677,7 @@ static int mlx5_suspend(struct pci_dev *pdev, pm_message_t state)
 {
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
 
-	mlx5_unload_one(dev, false);
+	mlx5_unload_one(dev);
 
 	return 0;
 }
@@ -1644,7 +1686,7 @@ static int mlx5_resume(struct pci_dev *pdev)
 {
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
 
-	return mlx5_load_one(dev, false);
+	return mlx5_load_one(dev);
 }
 
 static const struct pci_device_id mlx5_core_pci_table[] = {
@@ -1676,7 +1718,7 @@ MODULE_DEVICE_TABLE(pci, mlx5_core_pci_table);
 void mlx5_disable_device(struct mlx5_core_dev *dev)
 {
 	mlx5_error_sw_reset(dev);
-	mlx5_unload_one(dev, false);
+	mlx5_unload_one(dev);
 }
 
 void mlx5_recover_device(struct mlx5_core_dev *dev)
@@ -1689,7 +1731,7 @@ void mlx5_recover_device(struct mlx5_core_dev *dev)
 static struct pci_driver mlx5_core_driver = {
 	.name           = KBUILD_MODNAME,
 	.id_table       = mlx5_core_pci_table,
-	.probe          = init_one,
+	.probe          = probe_one,
 	.remove         = remove_one,
 	.suspend        = mlx5_suspend,
 	.resume         = mlx5_resume,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index efe403c7e354..02993a51b114 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -267,8 +267,10 @@ static inline bool mlx5_core_is_sf(const struct mlx5_core_dev *dev)
 
 int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx);
 void mlx5_mdev_uninit(struct mlx5_core_dev *dev);
-void mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup);
-int mlx5_load_one(struct mlx5_core_dev *dev, bool boot);
+int mlx5_init_one(struct mlx5_core_dev *dev);
+void mlx5_uninit_one(struct mlx5_core_dev *dev);
+void mlx5_unload_one(struct mlx5_core_dev *dev);
+int mlx5_load_one(struct mlx5_core_dev *dev);
 
 int mlx5_vport_get_other_func_cap(struct mlx5_core_dev *dev, u16 function_id, void *out);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
index c4bf555c25ea..42c8ee03fe3e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
@@ -41,14 +41,15 @@ static int mlx5_sf_dev_probe(struct auxiliary_device *adev, const struct auxilia
 		goto remap_err;
 	}
 
-	err = mlx5_load_one(mdev, true);
+	err = mlx5_init_one(mdev);
 	if (err) {
-		mlx5_core_warn(mdev, "mlx5_load_one err=%d\n", err);
-		goto load_one_err;
+		mlx5_core_warn(mdev, "mlx5_init_one err=%d\n", err);
+		goto init_one_err;
 	}
+	devlink_reload_enable(devlink);
 	return 0;
 
-load_one_err:
+init_one_err:
 	iounmap(mdev->iseg);
 remap_err:
 	mlx5_mdev_uninit(mdev);
@@ -63,7 +64,8 @@ static void mlx5_sf_dev_remove(struct auxiliary_device *adev)
 	struct devlink *devlink;
 
 	devlink = priv_to_devlink(sf_dev->mdev);
-	mlx5_unload_one(sf_dev->mdev, true);
+	devlink_reload_disable(devlink);
+	mlx5_uninit_one(sf_dev->mdev);
 	iounmap(sf_dev->mdev->iseg);
 	mlx5_mdev_uninit(sf_dev->mdev);
 	mlx5_devlink_free(devlink);
@@ -73,7 +75,7 @@ static void mlx5_sf_dev_shutdown(struct auxiliary_device *adev)
 {
 	struct mlx5_sf_dev *sf_dev = container_of(adev, struct mlx5_sf_dev, adev);
 
-	mlx5_unload_one(sf_dev->mdev, false);
+	mlx5_unload_one(sf_dev->mdev);
 }
 
 static const struct auxiliary_device_id mlx5_sf_dev_id_table[] = {
-- 
2.29.2

[net-next 04/15] net/mlx5: Remove second FW tracer check

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

The FW tracer check is called twice, so delete one of them.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c       | 7 +------
 2 files changed, 2 insertions(+), 7 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 2eb022ad7fd0..49e106719392 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -1100,7 +1100,7 @@ int mlx5_fw_tracer_reload(struct mlx5_fw_tracer *tracer)
 	int err;
 
 	if (IS_ERR_OR_NULL(tracer))
-		return -EINVAL;
+		return 0;
 
 	dev = tracer->dev;
 	mlx5_fw_tracer_cleanup(tracer);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index 255bd0059da1..d5d57630015f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -199,16 +199,11 @@ static void mlx5_fw_live_patch_event(struct work_struct *work)
 	struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset,
 						      fw_live_patch_work);
 	struct mlx5_core_dev *dev = fw_reset->dev;
-	struct mlx5_fw_tracer *tracer;
 
 	mlx5_core_info(dev, "Live patch updated firmware version: %d.%d.%d\n", fw_rev_maj(dev),
 		       fw_rev_min(dev), fw_rev_sub(dev));
 
-	tracer = dev->tracer;
-	if (IS_ERR_OR_NULL(tracer))
-		return;
-
-	if (mlx5_fw_tracer_reload(tracer))
+	if (mlx5_fw_tracer_reload(dev->tracer))
 		mlx5_core_err(dev, "Failed to reload FW tracer\n");
 }
 
-- 
2.29.2

[net-next 05/15] net/mlx5: Don't rely on interface state bit

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

The check of MLX5_INTERFACE_STATE_UP is completely useless, because
the FW tracer cleanup is called on every change of the interface
and it ensures that notifier is disabled together with canceling
all the pending works.

Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 3 +--
 1 file changed, 1 insertion(+), 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 49e106719392..01a1d02dcf15 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -1126,8 +1126,7 @@ static int fw_tracer_event(struct notifier_block *nb, unsigned long action, void
 
 	switch (eqe->sub_type) {
 	case MLX5_TRACER_SUBTYPE_OWNERSHIP_CHANGE:
-		if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state))
-			queue_work(tracer->work_queue, &tracer->ownership_change_work);
+		queue_work(tracer->work_queue, &tracer->ownership_change_work);
 		break;
 	case MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE:
 		if (likely(tracer->str_db.loaded))
-- 
2.29.2

[net-next 06/15] net/mlx5: Check returned value from health recover sequence

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

MLX5_INTERFACE_STATE_UP is far from being reliable check for success to
recover, because it can be changed any time and health logic doesn't
have any locks to protect from it.

The locks are not needed here because health recover is good to have,
but not must to success, so rely on the returned value from the
mlx5_recover_device() as a marker for success/failure.

Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/health.c    | 6 +++---
 drivers/net/ethernet/mellanox/mlx5/core/main.c      | 7 +++++--
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h | 2 +-
 3 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index 0c32c485eb58..a0a851640804 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -335,12 +335,12 @@ static int mlx5_health_try_recover(struct mlx5_core_dev *dev)
 		return -EIO;
 	}
 	mlx5_core_err(dev, "starting health recovery flow\n");
-	mlx5_recover_device(dev);
-	if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state) ||
-	    mlx5_health_check_fatal_sensors(dev)) {
+	if (mlx5_recover_device(dev) || mlx5_health_check_fatal_sensors(dev)) {
 		mlx5_core_err(dev, "health recovery failed\n");
 		return -EIO;
 	}
+
+	mlx5_core_info(dev, "health revovery succeded\n");
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 363bc3e917c2..e3a417d17707 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1721,11 +1721,14 @@ void mlx5_disable_device(struct mlx5_core_dev *dev)
 	mlx5_unload_one(dev);
 }
 
-void mlx5_recover_device(struct mlx5_core_dev *dev)
+int mlx5_recover_device(struct mlx5_core_dev *dev)
 {
+	int ret = -EIO;
+
 	mlx5_pci_disable_device(dev);
 	if (mlx5_pci_slot_reset(dev->pdev) == PCI_ERS_RESULT_RECOVERED)
-		mlx5_pci_resume(dev->pdev);
+		ret = mlx5_load_one(dev);
+	return ret;
 }
 
 static struct pci_driver mlx5_core_driver = {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 02993a51b114..37c8ec7d2217 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -134,7 +134,7 @@ void mlx5_error_sw_reset(struct mlx5_core_dev *dev);
 u32 mlx5_health_check_fatal_sensors(struct mlx5_core_dev *dev);
 int mlx5_health_wait_pci_up(struct mlx5_core_dev *dev);
 void mlx5_disable_device(struct mlx5_core_dev *dev);
-void mlx5_recover_device(struct mlx5_core_dev *dev);
+int mlx5_recover_device(struct mlx5_core_dev *dev);
 int mlx5_sriov_init(struct mlx5_core_dev *dev);
 void mlx5_sriov_cleanup(struct mlx5_core_dev *dev);
 int mlx5_sriov_attach(struct mlx5_core_dev *dev);
-- 
2.29.2

[net-next 07/15] net/mlx5e: CT, Avoid false lock dependency warning

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

To avoid false lock dependency warning set the ct_entries_ht lock
class different than the lock class of the ht being used when deleting
last flow from a group and then deleting a group, we get into del_sw_flow_group()
which call rhashtable_destroy on fg->ftes_hash which will take ht->mutex but
it's different than the ht->mutex here.

======================================================
WARNING: possible circular locking dependency detected
5.10.0-rc2+ #8 Tainted: G           O
------------------------------------------------------
revalidator23/24009 is trying to acquire lock:
ffff888128d83828 (&node->lock){++++}-{3:3}, at: mlx5_del_flow_rules+0x83/0x7a0 [mlx5_core]

but task is already holding lock:
ffff8881081ef518 (&ht->mutex){+.+.}-{3:3}, at: rhashtable_free_and_destroy+0x37/0x720

which lock already depends on the new lock.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Paul Blakey <paulb@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index f3f6eb081948..f81da2dc6af1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -1539,6 +1539,14 @@ mlx5_tc_ct_free_pre_ct_tables(struct mlx5_ct_ft *ft)
 	mlx5_tc_ct_free_pre_ct(ft, &ft->pre_ct);
 }
 
+/* To avoid false lock dependency warning set the ct_entries_ht lock
+ * class different than the lock class of the ht being used when deleting
+ * last flow from a group and then deleting a group, we get into del_sw_flow_group()
+ * which call rhashtable_destroy on fg->ftes_hash which will take ht->mutex but
+ * it's different than the ht->mutex here.
+ */
+static struct lock_class_key ct_entries_ht_lock_key;
+
 static struct mlx5_ct_ft *
 mlx5_tc_ct_add_ft_cb(struct mlx5_tc_ct_priv *ct_priv, u16 zone,
 		     struct nf_flowtable *nf_ft)
@@ -1573,6 +1581,8 @@ mlx5_tc_ct_add_ft_cb(struct mlx5_tc_ct_priv *ct_priv, u16 zone,
 	if (err)
 		goto err_init;
 
+	lockdep_set_class(&ft->ct_entries_ht.mutex, &ct_entries_ht_lock_key);
+
 	err = rhashtable_insert_fast(&ct_priv->zone_ht, &ft->node,
 				     zone_params);
 	if (err)
-- 
2.29.2

[net-next 08/15] net/mlx5e: fix mlx5e_tc_tun_update_header_ipv6 dummy definition

[Saeed Mahameed]

From: Arnd Bergmann <arnd@arndb.de>

The alternative implementation of this function in a header file
is declared as a global symbol, and gets added to every .c file
that includes it, which leads to a link error:

arm-linux-gnueabi-ld: drivers/net/ethernet/mellanox/mlx5/core/en_rx.o: in function `mlx5e_tc_tun_update_header_ipv6':
en_rx.c:(.text+0x0): multiple definition of `mlx5e_tc_tun_update_header_ipv6'; drivers/net/ethernet/mellanox/mlx5/core/en_main.o:en_main.c:(.text+0x0): first defined here

Mark it 'static inline' like the other functions here.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
index 67de2bf36861..89d5ca91566e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
@@ -76,10 +76,12 @@ int mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
 static inline int
 mlx5e_tc_tun_create_header_ipv6(struct mlx5e_priv *priv,
 				struct net_device *mirred_dev,
-				struct mlx5e_encap_entry *e) { return -EOPNOTSUPP; }
-int mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
-				    struct net_device *mirred_dev,
-				    struct mlx5e_encap_entry *e)
+				struct mlx5e_encap_entry *e)
+{ return -EOPNOTSUPP; }
+static inline int
+mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
+				struct net_device *mirred_dev,
+				struct mlx5e_encap_entry *e)
 { return -EOPNOTSUPP; }
 #endif
 int mlx5e_tc_tun_route_lookup(struct mlx5e_priv *priv,
-- 
2.29.2

[net-next 09/15] net/mlx5e: Add missing include

[Saeed Mahameed]

From: Vlad Buslov <vladbu@nvidia.com>

When CONFIG_IPV6 is disabled the header nexthop.h is not included by
fib_notifier.h which causes tc_tun_encap.c to fail to compile:

   In file included from drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:5:
   In file included from drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h:7:
   In file included from drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h:7:
   In file included from drivers/net/ethernet/mellanox/mlx5/core/en_tc.h:40:
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h:78:5: warning: no previous prototype for function 'mlx5e_tc_tun_update_header_ipv6' [-Wmissing-prototypes]
   int mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
       ^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h:78:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
   int mlx5e_tc_tun_update_header_ipv6(struct mlx5e_priv *priv,
   ^
   static
>> drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1510:12: error: implicit declaration of function 'fib_info_nh' [-Werror,-Wimplicit-function-declaration]
           fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
                     ^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1510:12: note: did you mean 'fib_info_put'?
   include/net/ip_fib.h:528:20: note: 'fib_info_put' declared here
   static inline void fib_info_put(struct fib_info *fi)
                      ^
>> drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1510:42: error: member reference type 'int' is not a pointer
           fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
                     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~  ^
   include/net/ip_fib.h:113:21: note: expanded from macro 'fib_nh_dev'
   #define fib_nh_dev              nh_common.nhc_dev
                                   ^
>> drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1552:13: error: incomplete definition of type 'struct fib6_entry_notifier_info'
           fen_info = container_of(info, struct fib6_entry_notifier_info, info);
                      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   include/linux/kernel.h:694:51: note: expanded from macro 'container_of'
           BUILD_BUG_ON_MSG(!__same_type(*(ptr), ((type *)0)->member) &&   \
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~
   include/linux/compiler_types.h:256:74: note: expanded from macro '__same_type'
   #define __same_type(a, b) __builtin_types_compatible_p(typeof(a), typeof(b))
                                                                            ^
   include/linux/build_bug.h:39:58: note: expanded from macro 'BUILD_BUG_ON_MSG'
   #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
                                       ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~
   include/linux/compiler_types.h:320:22: note: expanded from macro 'compiletime_assert'
           _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
           ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   include/linux/compiler_types.h:308:23: note: expanded from macro '_compiletime_assert'
           __compiletime_assert(condition, msg, prefix, suffix)
           ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   include/linux/compiler_types.h:300:9: note: expanded from macro '__compiletime_assert'
                   if (!(condition))                                       \
                         ^~~~~~~~~
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1546:9: note: forward declaration of 'struct fib6_entry_notifier_info'
           struct fib6_entry_notifier_info *fen_info;
                  ^
>> drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1552:13: error: offsetof of incomplete type 'struct fib6_entry_notifier_info'
           fen_info = container_of(info, struct fib6_entry_notifier_info, info);
                      ^                  ~~~~~~
   include/linux/kernel.h:697:21: note: expanded from macro 'container_of'
           ((type *)(__mptr - offsetof(type, member))); })
                              ^        ~~~~
   include/linux/stddef.h:17:32: note: expanded from macro 'offsetof'
   #define offsetof(TYPE, MEMBER)  __compiler_offsetof(TYPE, MEMBER)
                                   ^                   ~~~~
   include/linux/compiler_types.h:140:35: note: expanded from macro '__compiler_offsetof'
   #define __compiler_offsetof(a, b)       __builtin_offsetof(a, b)
                                           ^                  ~
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1546:9: note: forward declaration of 'struct fib6_entry_notifier_info'
           struct fib6_entry_notifier_info *fen_info;
                  ^
>> drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1552:11: error: assigning to 'struct fib6_entry_notifier_info *' from incompatible type 'void'
           fen_info = container_of(info, struct fib6_entry_notifier_info, info);
                    ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1553:12: error: implicit declaration of function 'fib6_info_nh_dev' [-Werror,-Wimplicit-function-declaration]
           fib_dev = fib6_info_nh_dev(fen_info->rt);
                     ^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1553:37: error: incomplete definition of type 'struct fib6_entry_notifier_info'
           fib_dev = fib6_info_nh_dev(fen_info->rt);
                                      ~~~~~~~~^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1546:9: note: forward declaration of 'struct fib6_entry_notifier_info'
           struct fib6_entry_notifier_info *fen_info;
                  ^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1555:14: error: incomplete definition of type 'struct fib6_entry_notifier_info'
               fen_info->rt->fib6_dst.plen != 128)
               ~~~~~~~~^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1546:9: note: forward declaration of 'struct fib6_entry_notifier_info'
           struct fib6_entry_notifier_info *fen_info;
                  ^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1562:39: error: incomplete definition of type 'struct fib6_entry_notifier_info'
           memcpy(&key.endpoint_ip.v6, &fen_info->rt->fib6_dst.addr,
                                        ~~~~~~~~^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1546:9: note: forward declaration of 'struct fib6_entry_notifier_info'
           struct fib6_entry_notifier_info *fen_info;
                  ^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1563:24: error: incomplete definition of type 'struct fib6_entry_notifier_info'
                  sizeof(fen_info->rt->fib6_dst.addr));
                         ~~~~~~~~^
   drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c:1546:9: note: forward declaration of 'struct fib6_entry_notifier_info'
           struct fib6_entry_notifier_info *fen_info;
                  ^
   1 warning and 10 errors generated.

Manually include net/nexthop.h in tc_tun_encap.c.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
index 6a116335bb21..32d06fe94acc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
@@ -2,6 +2,7 @@
 /* Copyright (c) 2021 Mellanox Technologies. */
 
 #include <net/fib_notifier.h>
+#include <net/nexthop.h>
 #include "tc_tun_encap.h"
 #include "en_tc.h"
 #include "tc_tun.h"
-- 
2.29.2

[net-next 10/15] net/mlx5: Fix indir stable stubs

[Saeed Mahameed]

From: Vlad Buslov <vladbu@nvidia.com>

Some of the stubs for CONFIG_MLX5_CLS_ACT==disabled are missing "static
inline" in their definition which causes the following compilation
warnings:

   In file included from drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c:41:
>> drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h:34:1: warning: no previous prototype for function 'mlx5_esw_indir_table_init' [-Wmissing-prototypes]
   mlx5_esw_indir_table_init(void)
   ^
   drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h:33:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
   struct mlx5_esw_indir_table *
   ^
   static
>> drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h:40:1: warning: no previous prototype for function 'mlx5_esw_indir_table_destroy' [-Wmissing-prototypes]
   mlx5_esw_indir_table_destroy(struct mlx5_esw_indir_table *indir)
   ^
   drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h:39:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
   void
   ^
   static
>> drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h:61:1: warning: no previous prototype for function 'mlx5_esw_indir_table_needed' [-Wmissing-prototypes]
   mlx5_esw_indir_table_needed(struct mlx5_eswitch *esw,
   ^
   drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h:60:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
   bool
   ^
   static
   3 warnings generated.

Add "static inline" prefix to signatures of stubs that were missing it.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h
index cb9eafd1b4ee..21d56b49d14b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/indir_table.h
@@ -30,13 +30,13 @@ mlx5_esw_indir_table_decap_vport(struct mlx5_flow_attr *attr);
 
 #else
 /* indir API stubs */
-struct mlx5_esw_indir_table *
+static inline struct mlx5_esw_indir_table *
 mlx5_esw_indir_table_init(void)
 {
 	return NULL;
 }
 
-void
+static inline void
 mlx5_esw_indir_table_destroy(struct mlx5_esw_indir_table *indir)
 {
 }
@@ -57,7 +57,7 @@ mlx5_esw_indir_table_put(struct mlx5_eswitch *esw,
 {
 }
 
-bool
+static inline bool
 mlx5_esw_indir_table_needed(struct mlx5_eswitch *esw,
 			    struct mlx5_flow_attr *attr,
 			    u16 vport_num,
-- 
2.29.2

[net-next 11/15] net/mlx5e: mlx5_tc_ct_init does not fail

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

mlx5_tc_ct_init() either returns a valid pointer or a NULL, either way
the caller can continue, remove IS_ERR check from callers as it has no
effect.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 0da69b98f38f..dc126389291d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -4646,10 +4646,6 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 
 	tc->ct = mlx5_tc_ct_init(priv, tc->chains, &priv->fs.tc.mod_hdr,
 				 MLX5_FLOW_NAMESPACE_KERNEL);
-	if (IS_ERR(tc->ct)) {
-		err = PTR_ERR(tc->ct);
-		goto err_ct;
-	}
 
 	tc->netdevice_nb.notifier_call = mlx5e_tc_netdev_event;
 	err = register_netdevice_notifier_dev_net(priv->netdev,
@@ -4665,7 +4661,6 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 
 err_reg:
 	mlx5_tc_ct_clean(tc->ct);
-err_ct:
 	mlx5_chains_destroy(tc->chains);
 err_chains:
 	rhashtable_destroy(&tc->ht);
@@ -4724,8 +4719,6 @@ int mlx5e_tc_esw_init(struct rhashtable *tc_ht)
 					       esw_chains(esw),
 					       &esw->offloads.mod_hdr,
 					       MLX5_FLOW_NAMESPACE_FDB);
-	if (IS_ERR(uplink_priv->ct_priv))
-		goto err_ct;
 
 	mapping = mapping_create(sizeof(struct tunnel_match_key),
 				 TUNNEL_INFO_BITS_MASK, true);
@@ -4765,7 +4758,6 @@ int mlx5e_tc_esw_init(struct rhashtable *tc_ht)
 	mapping_destroy(uplink_priv->tunnel_mapping);
 err_tun_mapping:
 	mlx5_tc_ct_clean(uplink_priv->ct_priv);
-err_ct:
 	netdev_warn(priv->netdev,
 		    "Failed to initialize tc (eswitch), err: %d", err);
 	return err;
-- 
2.29.2

[net-next 12/15] net/mlx5: SF, Fix return type

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Fix the following coccicheck warnings:

drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h:50:8-9: WARNING:
return of 0/1 in function 'mlx5_sf_dev_allocated' with return type bool

Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h
index 4de02902aef1..149fd9e698cf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h
@@ -47,7 +47,7 @@ static inline void mlx5_sf_driver_unregister(void)
 
 static inline bool mlx5_sf_dev_allocated(const struct mlx5_core_dev *dev)
 {
-	return 0;
+	return false;
 }
 
 #endif
-- 
2.29.2

[net-next 13/15] net/mlx5e: rep: Improve reg_cX conditions

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

There is no point of calculating reg_c1 or overriding reg_c0 if we are
going to abort the function.

Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
index 065126370acd..fcae3c0a4e9f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
@@ -621,11 +621,7 @@ bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe,
 	int err;
 
 	reg_c0 = (be32_to_cpu(cqe->sop_drop_qpn) & MLX5E_TC_FLOW_ID_MASK);
-	if (reg_c0 == MLX5_FS_DEFAULT_FLOW_TAG)
-		reg_c0 = 0;
-	reg_c1 = be32_to_cpu(cqe->ft_metadata);
-
-	if (!reg_c0)
+	if (!reg_c0 || reg_c0 == MLX5_FS_DEFAULT_FLOW_TAG)
 		return true;
 
 	/* If reg_c0 is not equal to the default flow tag then skb->mark
@@ -633,6 +629,8 @@ bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe,
 	 */
 	skb->mark = 0;
 
+	reg_c1 = be32_to_cpu(cqe->ft_metadata);
+
 	priv = netdev_priv(skb->dev);
 	esw = priv->mdev->priv.eswitch;
 
-- 
2.29.2

[net-next 14/15] net/mlx5: Avoid unnecessary operation

[Saeed Mahameed]

From: Eli Cohen <elic@nvidia.com>

fs_get_obj retrieves the container of fs_parent_node just to pass the
same value as &fs_ns->node. Just pass fs_parent_node to
init_root_tree_recursive() to get exactly the same effect.

Signed-off-by: Eli Cohen <elic@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 | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 66ad599bd488..f5517ea2f6be 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2395,14 +2395,12 @@ static int init_root_tree(struct mlx5_flow_steering *steering,
 			  struct init_tree_node *init_node,
 			  struct fs_node *fs_parent_node)
 {
-	int i;
-	struct mlx5_flow_namespace *fs_ns;
 	int err;
+	int i;
 
-	fs_get_obj(fs_ns, fs_parent_node);
 	for (i = 0; i < init_node->ar_size; i++) {
 		err = init_root_tree_recursive(steering, &init_node->children[i],
-					       &fs_ns->node,
+					       fs_parent_node,
 					       init_node, i);
 		if (err)
 			return err;
-- 
2.29.2

[net-next 15/15] net/mlx5e: Alloc flow spec using kvzalloc instead of kzalloc

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

flow spec is not small and we do allocate it using kvzalloc
in most places of the driver. fix rest of the places
to use kvzalloc to avoid failure in allocation when
memory is too fragmented.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 14 +++++++-------
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         |  4 ++--
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |  4 ++--
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index f81da2dc6af1..3a6095c912f1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -695,7 +695,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 
 	zone_rule->nat = nat;
 
-	spec = kzalloc(sizeof(*spec), GFP_KERNEL);
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec)
 		return -ENOMEM;
 
@@ -737,7 +737,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 
 	zone_rule->attr = attr;
 
-	kfree(spec);
+	kvfree(spec);
 	ct_dbg("Offloaded ct entry rule in zone %d", entry->tuple.zone);
 
 	return 0;
@@ -749,7 +749,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 err_mod_hdr:
 	kfree(attr);
 err_attr:
-	kfree(spec);
+	kvfree(spec);
 	return err;
 }
 
@@ -1684,10 +1684,10 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv,
 	struct mlx5_ct_ft *ft;
 	u32 fte_id = 1;
 
-	post_ct_spec = kzalloc(sizeof(*post_ct_spec), GFP_KERNEL);
+	post_ct_spec = kvzalloc(sizeof(*post_ct_spec), GFP_KERNEL);
 	ct_flow = kzalloc(sizeof(*ct_flow), GFP_KERNEL);
 	if (!post_ct_spec || !ct_flow) {
-		kfree(post_ct_spec);
+		kvfree(post_ct_spec);
 		kfree(ct_flow);
 		return ERR_PTR(-ENOMEM);
 	}
@@ -1822,7 +1822,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv,
 
 	attr->ct_attr.ct_flow = ct_flow;
 	dealloc_mod_hdr_actions(&pre_mod_acts);
-	kfree(post_ct_spec);
+	kvfree(post_ct_spec);
 
 	return rule;
 
@@ -1843,7 +1843,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv,
 err_idr:
 	mlx5_tc_ct_del_ft_cb(ct_priv, ft);
 err_ft:
-	kfree(post_ct_spec);
+	kvfree(post_ct_spec);
 	kfree(ct_flow);
 	netdev_warn(priv->netdev, "Failed to offload ct flow, err %d\n", err);
 	return ERR_PTR(err);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 381a9c8c9da9..34119ce92031 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -60,7 +60,7 @@ static int rx_err_add_rule(struct mlx5e_priv *priv,
 	struct mlx5_flow_spec *spec;
 	int err = 0;
 
-	spec = kzalloc(sizeof(*spec), GFP_KERNEL);
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec)
 		return -ENOMEM;
 
@@ -101,7 +101,7 @@ static int rx_err_add_rule(struct mlx5e_priv *priv,
 	if (err)
 		mlx5_modify_header_dealloc(mdev, modify_hdr);
 out_spec:
-	kfree(spec);
+	kvfree(spec);
 	return err;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 703753ac2e02..a215ccee3e61 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1446,7 +1446,7 @@ esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag)
 	if (!mlx5_eswitch_reg_c1_loopback_supported(esw))
 		return ERR_PTR(-EOPNOTSUPP);
 
-	spec = kzalloc(sizeof(*spec), GFP_KERNEL);
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec)
 		return ERR_PTR(-ENOMEM);
 
@@ -1469,7 +1469,7 @@ esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag)
 	dest.ft = esw->offloads.ft_offloads;
 
 	flow_rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
-	kfree(spec);
+	kvfree(spec);
 
 	if (IS_ERR(flow_rule))
 		esw_warn(esw->dev,
-- 
2.29.2