[pull request][net-next V2 00/16] mlx5 updates 2021-04-13

[pull request][net-next V2 00/16] mlx5 updates 2021-04-13

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

Hi Dave, Jakub,

This patchset includes some updates and cleanup to mlx5 driver.
For more information please see tag log below.

v1->v2: update 1st patch commit message and mlx5 documentation to
include performance numbers and applicability of the metadata settings.

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

Thanks,
Saeed.

---
The following changes since commit 8ef7adc6beb2ef0bce83513dc9e4505e7b21e8c2:

  net: ethernet: ravb: Enable optional refclk (2021-04-12 14:09:59 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2021-04-13

for you to fetch changes up to 5b232ea94c90aa6196321820740e2969ae64e9cb:

  net/mlx5e: Fix RQ creation flow for queues which doesn't support XDP (2021-04-14 11:03:10 -0700)

----------------------------------------------------------------
mlx5-updates-2021-04-13

mlx5 core and netdev driver updates

1) E-Switch updates from Parav,
  1.1) Devlink parameter to control mlx5 metadata enablement for E-Switch
  1.2) Trivial cleanups for E-Switch code
  1.3) Dynamically allocate vport steering namespaces only when required

2) From Jianbo, Use variably sized data structures for Software steering

3) Several minor cleanups

----------------------------------------------------------------
Aya Levin (1):
      net/mlx5e: Fix RQ creation flow for queues which doesn't support XDP

Colin Ian King (1):
      net/mlx5: Fix bit-wise and with zero

Jianbo Liu (1):
      net/mlx5: DR, Use variably sized data structures for different actions

Parav Pandit (9):
      net/mlx5: E-Switch, let user to enable disable metadata
      net/mlx5: E-Switch, Skip querying SF enabled bits
      net/mlx5: E-Switch, Make vport number u16
      net/mlx5: E-Switch Make cleanup sequence mirror of init
      net/mlx5: E-Switch, Convert a macro to a helper routine
      net/mlx5: E-Switch, Move legacy code to a individual file
      net/mlx5: E-Switch, Initialize eswitch acls ns when eswitch is enabled
      net/mlx5: SF, Use device pointer directly
      net/mlx5: SF, Reuse stored hardware function id

Roi Dayan (1):
      net/mlx5: DR, Alloc cmd buffer with kvzalloc() instead of kzalloc()

Wenpeng Liang (3):
      net/mlx5: Add a blank line after declarations
      net/mlx5: Remove return statement exist at the end of void function
      net/mlx5: Replace spaces with tab at the start of a line

 .../device_drivers/ethernet/mellanox/mlx5.rst      |  34 ++
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/devlink.c  |  62 +++
 drivers/net/ethernet/mellanox/mlx5/core/devlink.h  |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c  |   4 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |   5 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   |   2 +-
 .../net/ethernet/mellanox/mlx5/core/esw/legacy.c   | 509 ++++++++++++++++++
 .../net/ethernet/mellanox/mlx5/core/esw/legacy.h   |  22 +
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c  | 595 +++------------------
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |  11 +-
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |  33 +-
 .../net/ethernet/mellanox/mlx5/core/fpga/ipsec.c   |   3 +-
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  |  88 ++-
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.h  |   7 +
 drivers/net/ethernet/mellanox/mlx5/core/lag.c      |   2 -
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  |   1 -
 drivers/net/ethernet/mellanox/mlx5/core/rdma.c     |   1 -
 .../net/ethernet/mellanox/mlx5/core/sf/devlink.c   |   5 +-
 .../net/ethernet/mellanox/mlx5/core/sf/hw_table.c  |  16 +-
 .../mellanox/mlx5/core/steering/dr_action.c        | 242 +++++----
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |   4 +-
 .../ethernet/mellanox/mlx5/core/steering/dr_send.c |  14 +-
 .../mellanox/mlx5/core/steering/dr_table.c         |   4 +-
 .../mellanox/mlx5/core/steering/dr_types.h         | 104 ++--
 include/linux/mlx5/eswitch.h                       |   3 +-
 26 files changed, 996 insertions(+), 778 deletions(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.h

[net-next V2 01/16] net/mlx5: E-Switch, let user to enable disable metadata

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

Currently each packet inserted in eswitch is tagged with a internal
metadata to indicate source vport. Metadata tagging is not always
needed. Metadata insertion is needed for multi-port RoCE, failover
between representors and stacked devices. In many other cases,
metadata enablement is not needed.

Metadata insertion slows down the packet processing rate of the E-switch
when it is in switchdev mode.

Below table show performance gain with metadata disabled for VXLAN
offload rules in both SMFS and DMFS steering mode on ConnectX-5 device.

----------------------------------------------
| steering | metadata | pkt size | rx pps    |
| mode     |          |          | (million) |
----------------------------------------------
| smfs     | disabled | 128Bytes | 42        |
----------------------------------------------
| smfs     | enabled  | 128Bytes | 36        |
----------------------------------------------
| dmfs     | disabled | 128Bytes | 42        |
----------------------------------------------
| dmfs     | enabled  | 128Bytes | 36        |
----------------------------------------------

Hence, allow user to disable metadata using driver specific devlink
parameter. Metadata setting of the eswitch is applicable only for the
switchdev mode.

Example to show and disable metadata before changing eswitch mode:
$ devlink dev param show pci/0000:06:00.0 name esw_port_metadata
pci/0000:06:00.0:
  name esw_port_metadata type driver-specific
    values:
      cmode runtime value true

$ devlink dev param set pci/0000:06:00.0 \
	  name esw_port_metadata value false cmode runtime

$ devlink dev eswitch set pci/0000:06:00.0 mode switchdev

Signed-off-by: Parav Pandit <parav@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Vu Pham <vuhuong@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
changelog:
v1->v2:
 - added performance numbers in commit log
 - updated commit log and documentation for switchdev mode
 - added explicit note on when user can disable metadata in
   documentation
---
 .../device_drivers/ethernet/mellanox/mlx5.rst | 34 ++++++++++
 .../net/ethernet/mellanox/mlx5/core/devlink.c | 62 +++++++++++++++++++
 .../net/ethernet/mellanox/mlx5/core/devlink.h |  1 +
 .../net/ethernet/mellanox/mlx5/core/eswitch.h |  2 +
 .../mellanox/mlx5/core/eswitch_offloads.c     | 30 ++++++---
 5 files changed, 122 insertions(+), 7 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5.rst
index 1b7e32d8a61b..936a10f1942c 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5.rst
@@ -183,6 +183,40 @@ User command examples:
       values:
          cmode driverinit value true
 
+esw_port_metadata: Eswitch port metadata state
+----------------------------------------------
+When applicable, disabling Eswitch metadata can increase packet rate
+up to 20% depending on the use case and packet sizes.
+
+Eswitch port metadata state controls whether to internally tag packets with
+metadata. Metadata tagging must be enabled for multi-port RoCE, failover
+between representors and stacked devices.
+By default metadata is enabled on the supported devices in E-switch.
+Metadata is applicable only for E-switch in switchdev mode and
+users may disable it when NONE of the below use cases will be in use:
+1. HCA is in Dual/multi-port RoCE mode.
+2. VF/SF representor bonding (Usually used for Live migration)
+3. Stacked devices
+
+When metadata is disabled, the above use cases will fail to initialize if
+users try to enable them.
+
+- Show eswitch port metadata::
+
+    $ devlink dev param show pci/0000:06:00.0 name esw_port_metadata
+      pci/0000:06:00.0:
+        name esw_port_metadata type driver-specific
+          values:
+            cmode runtime value true
+
+- Disable eswitch port metadata::
+
+    $ devlink dev param set pci/0000:06:00.0 name esw_port_metadata value false cmode runtime
+
+- Change eswitch mode to switchdev mode where after choosing the metadata value::
+
+    $ devlink dev eswitch set pci/0000:06:00.0 mode switchdev
+
 mlx5 subfunction
 ================
 mlx5 supports subfunction management using devlink port (see :ref:`Documentation/networking/devlink/devlink-port.rst <devlink_port>`) interface.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index 38c7c44fe883..ee0f5355f120 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -456,6 +456,50 @@ static int mlx5_devlink_large_group_num_validate(struct devlink *devlink, u32 id
 
 	return 0;
 }
+
+static int mlx5_devlink_esw_port_metadata_set(struct devlink *devlink, u32 id,
+					      struct devlink_param_gset_ctx *ctx)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+
+	if (!MLX5_ESWITCH_MANAGER(dev))
+		return -EOPNOTSUPP;
+
+	return mlx5_esw_offloads_vport_metadata_set(dev->priv.eswitch, ctx->val.vbool);
+}
+
+static int mlx5_devlink_esw_port_metadata_get(struct devlink *devlink, u32 id,
+					      struct devlink_param_gset_ctx *ctx)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+
+	if (!MLX5_ESWITCH_MANAGER(dev))
+		return -EOPNOTSUPP;
+
+	ctx->val.vbool = mlx5_eswitch_vport_match_metadata_enabled(dev->priv.eswitch);
+	return 0;
+}
+
+static int mlx5_devlink_esw_port_metadata_validate(struct devlink *devlink, u32 id,
+						   union devlink_param_value val,
+						   struct netlink_ext_ack *extack)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+	u8 esw_mode;
+
+	if (!MLX5_ESWITCH_MANAGER(dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "E-Switch is unsupported");
+		return -EOPNOTSUPP;
+	}
+	esw_mode = mlx5_eswitch_mode(dev);
+	if (esw_mode == MLX5_ESWITCH_OFFLOADS) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "E-Switch must either disabled or non switchdev mode");
+		return -EBUSY;
+	}
+	return 0;
+}
+
 #endif
 
 static int mlx5_devlink_enable_remote_dev_reset_set(struct devlink *devlink, u32 id,
@@ -490,6 +534,12 @@ static const struct devlink_param mlx5_devlink_params[] = {
 			     BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
 			     NULL, NULL,
 			     mlx5_devlink_large_group_num_validate),
+	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_ESW_PORT_METADATA,
+			     "esw_port_metadata", DEVLINK_PARAM_TYPE_BOOL,
+			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			     mlx5_devlink_esw_port_metadata_get,
+			     mlx5_devlink_esw_port_metadata_set,
+			     mlx5_devlink_esw_port_metadata_validate),
 #endif
 	DEVLINK_PARAM_GENERIC(ENABLE_REMOTE_DEV_RESET, BIT(DEVLINK_PARAM_CMODE_RUNTIME),
 			      mlx5_devlink_enable_remote_dev_reset_get,
@@ -519,6 +569,18 @@ static void mlx5_devlink_set_params_init_values(struct devlink *devlink)
 	devlink_param_driverinit_value_set(devlink,
 					   MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM,
 					   value);
+
+	if (MLX5_ESWITCH_MANAGER(dev)) {
+		if (mlx5_esw_vport_match_metadata_supported(dev->priv.eswitch)) {
+			dev->priv.eswitch->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA;
+			value.vbool = true;
+		} else {
+			value.vbool = false;
+		}
+		devlink_param_driverinit_value_set(devlink,
+						   MLX5_DEVLINK_PARAM_ID_ESW_PORT_METADATA,
+						   value);
+	}
 #endif
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
index eff107dad922..7318d44b774b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
@@ -10,6 +10,7 @@ enum mlx5_devlink_param_id {
 	MLX5_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX,
 	MLX5_DEVLINK_PARAM_ID_FLOW_STEERING_MODE,
 	MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM,
+	MLX5_DEVLINK_PARAM_ID_ESW_PORT_METADATA,
 };
 
 struct mlx5_trap_ctx {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index deafb0e03787..c7a73dbd64b4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -314,6 +314,8 @@ int esw_offloads_enable(struct mlx5_eswitch *esw);
 void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw);
 int esw_offloads_init_reps(struct mlx5_eswitch *esw);
 
+bool mlx5_esw_vport_match_metadata_supported(const struct mlx5_eswitch *esw);
+int mlx5_esw_offloads_vport_metadata_set(struct mlx5_eswitch *esw, bool enable);
 u32 mlx5_esw_match_metadata_alloc(struct mlx5_eswitch *esw);
 void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index ab32f685cbb7..1f58e84bdfc6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -2351,8 +2351,7 @@ static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw)
 	mlx5_devcom_unregister_component(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
 }
 
-static bool
-esw_check_vport_match_metadata_supported(const struct mlx5_eswitch *esw)
+bool mlx5_esw_vport_match_metadata_supported(const struct mlx5_eswitch *esw)
 {
 	if (!MLX5_CAP_ESW(esw->dev, esw_uplink_ingress_acl))
 		return false;
@@ -2452,6 +2451,28 @@ static int esw_offloads_metadata_init(struct mlx5_eswitch *esw)
 	return err;
 }
 
+int mlx5_esw_offloads_vport_metadata_set(struct mlx5_eswitch *esw, bool enable)
+{
+	int err = 0;
+
+	down_write(&esw->mode_lock);
+	if (esw->mode != MLX5_ESWITCH_NONE) {
+		err = -EBUSY;
+		goto done;
+	}
+	if (!mlx5_esw_vport_match_metadata_supported(esw)) {
+		err = -EOPNOTSUPP;
+		goto done;
+	}
+	if (enable)
+		esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA;
+	else
+		esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA;
+done:
+	up_write(&esw->mode_lock);
+	return err;
+}
+
 int
 esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw,
 				     struct mlx5_vport *vport)
@@ -2673,9 +2694,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
 	if (err)
 		goto err_metadata;
 
-	if (esw_check_vport_match_metadata_supported(esw))
-		esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA;
-
 	err = esw_offloads_metadata_init(esw);
 	if (err)
 		goto err_metadata;
@@ -2725,7 +2743,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
 err_vport_metadata:
 	esw_offloads_metadata_uninit(esw);
 err_metadata:
-	esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA;
 	mlx5_rdma_disable_roce(esw->dev);
 	mutex_destroy(&esw->offloads.termtbl_mutex);
 	return err;
@@ -2761,7 +2778,6 @@ void esw_offloads_disable(struct mlx5_eswitch *esw)
 	esw_offloads_steering_cleanup(esw);
 	mapping_destroy(esw->offloads.reg_c0_obj_pool);
 	esw_offloads_metadata_uninit(esw);
-	esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA;
 	mlx5_rdma_disable_roce(esw->dev);
 	mutex_destroy(&esw->offloads.termtbl_mutex);
 	esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE;
-- 
2.30.2

Re: [net-next V2 01/16] net/mlx5: E-Switch, let user to enable disable metadata

[Jakub Kicinski]

On Wed, 14 Apr 2021 11:05:50 -0700 Saeed Mahameed wrote:
> From: Parav Pandit <parav@nvidia.com>
> 
> Currently each packet inserted in eswitch is tagged with a internal
> metadata to indicate source vport. Metadata tagging is not always
> needed. Metadata insertion is needed for multi-port RoCE, failover
> between representors and stacked devices. In many other cases,
> metadata enablement is not needed.
> 
> Metadata insertion slows down the packet processing rate of the E-switch
> when it is in switchdev mode.
> 
> Below table show performance gain with metadata disabled for VXLAN
> offload rules in both SMFS and DMFS steering mode on ConnectX-5 device.
> 
> ----------------------------------------------
> | steering | metadata | pkt size | rx pps    |
> | mode     |          |          | (million) |
> ----------------------------------------------
> | smfs     | disabled | 128Bytes | 42        |
> ----------------------------------------------
> | smfs     | enabled  | 128Bytes | 36        |
> ----------------------------------------------
> | dmfs     | disabled | 128Bytes | 42        |
> ----------------------------------------------
> | dmfs     | enabled  | 128Bytes | 36        |
> ----------------------------------------------
> 
> Hence, allow user to disable metadata using driver specific devlink
> parameter. Metadata setting of the eswitch is applicable only for the
> switchdev mode.
> 
> Example to show and disable metadata before changing eswitch mode:
> $ devlink dev param show pci/0000:06:00.0 name esw_port_metadata
> pci/0000:06:00.0:
>   name esw_port_metadata type driver-specific
>     values:
>       cmode runtime value true
> 
> $ devlink dev param set pci/0000:06:00.0 \
> 	  name esw_port_metadata value false cmode runtime
> 
> $ devlink dev eswitch set pci/0000:06:00.0 mode switchdev
> 
> Signed-off-by: Parav Pandit <parav@nvidia.com>
> Reviewed-by: Roi Dayan <roid@nvidia.com>
> Reviewed-by: Vu Pham <vuhuong@nvidia.com>
> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Acked-by: Jakub Kicinski <kuba@kernel.org>

Re: [net-next V2 01/16] net/mlx5: E-Switch, let user to enable disable metadata

[patchwork-bot+netdevbpf]

Hello:

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

On Wed, 14 Apr 2021 11:05:50 -0700 you wrote:
> From: Parav Pandit <parav@nvidia.com>
> 
> Currently each packet inserted in eswitch is tagged with a internal
> metadata to indicate source vport. Metadata tagging is not always
> needed. Metadata insertion is needed for multi-port RoCE, failover
> between representors and stacked devices. In many other cases,
> metadata enablement is not needed.
> 
> [...]

Here is the summary with links:
  - [net-next,V2,01/16] net/mlx5: E-Switch, let user to enable disable metadata
    https://git.kernel.org/netdev/net-next/c/7bf481d7e75a
  - [net-next,V2,02/16] net/mlx5: E-Switch, Skip querying SF enabled bits
    https://git.kernel.org/netdev/net-next/c/7d5ae4789192
  - [net-next,V2,03/16] net/mlx5: E-Switch, Make vport number u16
    https://git.kernel.org/netdev/net-next/c/6308a5f06be0
  - [net-next,V2,04/16] net/mlx5: E-Switch Make cleanup sequence mirror of init
    https://git.kernel.org/netdev/net-next/c/13795553a84d
  - [net-next,V2,05/16] net/mlx5: E-Switch, Convert a macro to a helper routine
    https://git.kernel.org/netdev/net-next/c/b16f2bb6b6ca
  - [net-next,V2,06/16] net/mlx5: E-Switch, Move legacy code to a individual file
    https://git.kernel.org/netdev/net-next/c/b55b35382e51
  - [net-next,V2,07/16] net/mlx5: E-Switch, Initialize eswitch acls ns when eswitch is enabled
    https://git.kernel.org/netdev/net-next/c/57b92bdd9e14
  - [net-next,V2,08/16] net/mlx5: SF, Use device pointer directly
    https://git.kernel.org/netdev/net-next/c/6e74e6ea1b64
  - [net-next,V2,09/16] net/mlx5: SF, Reuse stored hardware function id
    https://git.kernel.org/netdev/net-next/c/a74ed24c437e
  - [net-next,V2,10/16] net/mlx5: DR, Use variably sized data structures for different actions
    https://git.kernel.org/netdev/net-next/c/9dac2966c531
  - [net-next,V2,11/16] net/mlx5: DR, Alloc cmd buffer with kvzalloc() instead of kzalloc()
    https://git.kernel.org/netdev/net-next/c/b7f86258a264
  - [net-next,V2,12/16] net/mlx5: Fix bit-wise and with zero
    https://git.kernel.org/netdev/net-next/c/82c3ba31c370
  - [net-next,V2,13/16] net/mlx5: Add a blank line after declarations
    https://git.kernel.org/netdev/net-next/c/02f47c04c36c
  - [net-next,V2,14/16] net/mlx5: Remove return statement exist at the end of void function
    https://git.kernel.org/netdev/net-next/c/9dee115bc147
  - [net-next,V2,15/16] net/mlx5: Replace spaces with tab at the start of a line
    https://git.kernel.org/netdev/net-next/c/31450b435fe6
  - [net-next,V2,16/16] net/mlx5e: Fix RQ creation flow for queues which doesn't support XDP
    https://git.kernel.org/netdev/net-next/c/5b232ea94c90

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

[net-next V2 02/16] net/mlx5: E-Switch, Skip querying SF enabled bits

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

With vhca events, SF state is queried through the VHCA events. Device no
longer expects SF bitmap in the query eswitch functions command.

Hence, remove it to simplify the code.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 6cf04a366f99..b3bc82e419b6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1390,15 +1390,9 @@ const u32 *mlx5_esw_query_functions(struct mlx5_core_dev *dev)
 {
 	int outlen = MLX5_ST_SZ_BYTES(query_esw_functions_out);
 	u32 in[MLX5_ST_SZ_DW(query_esw_functions_in)] = {};
-	u16 max_sf_vports;
 	u32 *out;
 	int err;
 
-	max_sf_vports = mlx5_sf_max_functions(dev);
-	/* Device interface is array of 64-bits */
-	if (max_sf_vports)
-		outlen += DIV_ROUND_UP(max_sf_vports, BITS_PER_TYPE(__be64)) * sizeof(__be64);
-
 	out = kvzalloc(outlen, GFP_KERNEL);
 	if (!out)
 		return ERR_PTR(-ENOMEM);
-- 
2.30.2

[net-next V2 03/16] net/mlx5: E-Switch, Make vport number u16

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

Vport number is 16-bit field in hardware. Make it u16.

Move location of vport in the structure so that it reduces a hole
in the structure.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h          | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 3 ++-
 include/linux/mlx5/eswitch.h                               | 3 +--
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index c7a73dbd64b4..a4b9f78bf4d6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -152,7 +152,6 @@ enum mlx5_eswitch_vport_event {
 
 struct mlx5_vport {
 	struct mlx5_core_dev    *dev;
-	int                     vport;
 	struct hlist_head       uc_list[MLX5_L2_ADDR_HASH_SIZE];
 	struct hlist_head       mc_list[MLX5_L2_ADDR_HASH_SIZE];
 	struct mlx5_flow_handle *promisc_rule;
@@ -174,6 +173,7 @@ struct mlx5_vport {
 		u32 max_rate;
 	} qos;
 
+	u16 vport;
 	bool                    enabled;
 	enum mlx5_eswitch_vport_event enabled_events;
 	struct devlink_port *dl_port;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 1f58e84bdfc6..bbb707117296 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -986,12 +986,13 @@ static void mlx5_eswitch_del_send_to_vport_meta_rules(struct mlx5_eswitch *esw)
 static int
 mlx5_eswitch_add_send_to_vport_meta_rules(struct mlx5_eswitch *esw)
 {
-	int num_vfs, vport_num, rule_idx = 0, err = 0;
 	struct mlx5_flow_destination dest = {};
 	struct mlx5_flow_act flow_act = {0};
+	int num_vfs, rule_idx = 0, err = 0;
 	struct mlx5_flow_handle *flow_rule;
 	struct mlx5_flow_handle **flows;
 	struct mlx5_flow_spec *spec;
+	u16 vport_num;
 
 	num_vfs = esw->esw_funcs.num_vfs;
 	flows = kvzalloc(num_vfs * sizeof(*flows), GFP_KERNEL);
diff --git a/include/linux/mlx5/eswitch.h b/include/linux/mlx5/eswitch.h
index 429a710c5a99..9cf1da2883c6 100644
--- a/include/linux/mlx5/eswitch.h
+++ b/include/linux/mlx5/eswitch.h
@@ -152,8 +152,7 @@ mlx5_eswitch_vport_match_metadata_enabled(const struct mlx5_eswitch *esw)
 };
 
 static inline u32
-mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw,
-					  int vport_num)
+mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw, u16 vport_num)
 {
 	return 0;
 };
-- 
2.30.2

[net-next V2 04/16] net/mlx5: E-Switch Make cleanup sequence mirror of init

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

Make cleanup sequence mirror of init sequence for cleaning up reps
and freeing vports.

Also when reps initialization fails, there is no need to perform reps
cleanup.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index b3bc82e419b6..9009574372fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1856,7 +1856,6 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
 abort:
 	if (esw->work_queue)
 		destroy_workqueue(esw->work_queue);
-	esw_offloads_cleanup_reps(esw);
 	kfree(esw->vports);
 	kfree(esw);
 	return err;
@@ -1871,7 +1870,6 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
 
 	esw->dev->priv.eswitch = NULL;
 	destroy_workqueue(esw->work_queue);
-	esw_offloads_cleanup_reps(esw);
 	mutex_destroy(&esw->state_lock);
 	WARN_ON(!xa_empty(&esw->offloads.vhca_map));
 	xa_destroy(&esw->offloads.vhca_map);
@@ -1879,6 +1877,7 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
 	mlx5e_mod_hdr_tbl_destroy(&esw->offloads.mod_hdr);
 	mutex_destroy(&esw->offloads.encap_tbl_lock);
 	mutex_destroy(&esw->offloads.decap_tbl_lock);
+	esw_offloads_cleanup_reps(esw);
 	kfree(esw->vports);
 	kfree(esw);
 }
-- 
2.30.2

[net-next V2 05/16] net/mlx5: E-Switch, Convert a macro to a helper routine

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

Convert ESW_ALLOWED macro to a helper routine so that it can be used in
other eswitch files.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c | 32 +++++++++----------
 .../net/ethernet/mellanox/mlx5/core/eswitch.h |  5 +++
 2 files changed, 20 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 9009574372fc..6a70e385beb8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1443,8 +1443,6 @@ static void mlx5_eswitch_clear_vf_vports_info(struct mlx5_eswitch *esw)
 }
 
 /* Public E-Switch API */
-#define ESW_ALLOWED(esw) ((esw) && MLX5_ESWITCH_MANAGER((esw)->dev))
-
 int mlx5_eswitch_load_vport(struct mlx5_eswitch *esw, u16 vport_num,
 			    enum mlx5_eswitch_vport_event enabled_events)
 {
@@ -1713,7 +1711,7 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs)
 {
 	int ret;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return 0;
 
 	down_write(&esw->mode_lock);
@@ -1773,7 +1771,7 @@ void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw, bool clear_vf)
 
 void mlx5_eswitch_disable(struct mlx5_eswitch *esw, bool clear_vf)
 {
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return;
 
 	down_write(&esw->mode_lock);
@@ -2023,7 +2021,7 @@ int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
 	int other_vport = 1;
 	int err = 0;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 	if (IS_ERR(evport))
 		return PTR_ERR(evport);
@@ -2111,7 +2109,7 @@ int mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 	u8 set_flags = 0;
 	int err = 0;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 
 	if (vlan || qos)
@@ -2140,7 +2138,7 @@ int mlx5_eswitch_set_vport_spoofchk(struct mlx5_eswitch *esw,
 	bool pschk;
 	int err = 0;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 	if (IS_ERR(evport))
 		return PTR_ERR(evport);
@@ -2248,7 +2246,7 @@ int mlx5_eswitch_set_vepa(struct mlx5_eswitch *esw, u8 setting)
 	if (!esw)
 		return -EOPNOTSUPP;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 
 	mutex_lock(&esw->state_lock);
@@ -2269,7 +2267,7 @@ int mlx5_eswitch_get_vepa(struct mlx5_eswitch *esw, u8 *setting)
 	if (!esw)
 		return -EOPNOTSUPP;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 
 	if (esw->mode != MLX5_ESWITCH_LEGACY)
@@ -2285,7 +2283,7 @@ int mlx5_eswitch_set_vport_trust(struct mlx5_eswitch *esw,
 	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
 	int err = 0;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 	if (IS_ERR(evport))
 		return PTR_ERR(evport);
@@ -2369,7 +2367,7 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, u16 vport,
 	bool max_rate_supported;
 	int err = 0;
 
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return -EPERM;
 	if (IS_ERR(evport))
 		return PTR_ERR(evport);
@@ -2534,7 +2532,7 @@ u8 mlx5_eswitch_mode(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eswitch *esw = dev->priv.eswitch;
 
-	return ESW_ALLOWED(esw) ? esw->mode : MLX5_ESWITCH_NONE;
+	return mlx5_esw_allowed(esw) ? esw->mode : MLX5_ESWITCH_NONE;
 }
 EXPORT_SYMBOL_GPL(mlx5_eswitch_mode);
 
@@ -2544,7 +2542,7 @@ mlx5_eswitch_get_encap_mode(const struct mlx5_core_dev *dev)
 	struct mlx5_eswitch *esw;
 
 	esw = dev->priv.eswitch;
-	return ESW_ALLOWED(esw) ? esw->offloads.encap :
+	return mlx5_esw_allowed(esw) ? esw->offloads.encap :
 		DEVLINK_ESWITCH_ENCAP_MODE_NONE;
 }
 EXPORT_SYMBOL(mlx5_eswitch_get_encap_mode);
@@ -2590,7 +2588,7 @@ bool mlx5_esw_hold(struct mlx5_core_dev *mdev)
 	struct mlx5_eswitch *esw = mdev->priv.eswitch;
 
 	/* e.g. VF doesn't have eswitch so nothing to do */
-	if (!ESW_ALLOWED(esw))
+	if (!mlx5_esw_allowed(esw))
 		return true;
 
 	if (down_read_trylock(&esw->mode_lock) != 0)
@@ -2607,7 +2605,7 @@ void mlx5_esw_release(struct mlx5_core_dev *mdev)
 {
 	struct mlx5_eswitch *esw = mdev->priv.eswitch;
 
-	if (ESW_ALLOWED(esw))
+	if (mlx5_esw_allowed(esw))
 		up_read(&esw->mode_lock);
 }
 
@@ -2619,7 +2617,7 @@ void mlx5_esw_get(struct mlx5_core_dev *mdev)
 {
 	struct mlx5_eswitch *esw = mdev->priv.eswitch;
 
-	if (ESW_ALLOWED(esw))
+	if (mlx5_esw_allowed(esw))
 		atomic64_inc(&esw->user_count);
 }
 
@@ -2631,7 +2629,7 @@ void mlx5_esw_put(struct mlx5_core_dev *mdev)
 {
 	struct mlx5_eswitch *esw = mdev->priv.eswitch;
 
-	if (ESW_ALLOWED(esw))
+	if (mlx5_esw_allowed(esw))
 		atomic64_dec_if_positive(&esw->user_count);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index a4b9f78bf4d6..a61a33c36fa8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -521,6 +521,11 @@ const u32 *mlx5_esw_query_functions(struct mlx5_core_dev *dev);
 #define esw_debug(dev, format, ...)				\
 	mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__)
 
+static inline bool mlx5_esw_allowed(const struct mlx5_eswitch *esw)
+{
+	return esw && MLX5_ESWITCH_MANAGER(esw->dev);
+}
+
 /* The returned number is valid only when the dev is eswitch manager. */
 static inline u16 mlx5_eswitch_manager_vport(struct mlx5_core_dev *dev)
 {
-- 
2.30.2

[net-next V2 06/16] net/mlx5: E-Switch, Move legacy code to a individual file

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

Currently eswitch offers two modes. Legacy and offloads.
Offloads code is already in its own file eswitch_offloads.c

However eswitch.c contains the eswitch legacy code and common
infrastructure  code.

To enable future extensions and to better manage generic common eswitch
infrastructure code, move the legacy code to its own legacy.c file.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/Makefile  |   2 +-
 .../ethernet/mellanox/mlx5/core/esw/legacy.c  | 509 +++++++++++++++++
 .../ethernet/mellanox/mlx5/core/esw/legacy.h  |  22 +
 .../net/ethernet/mellanox/mlx5/core/eswitch.c | 511 +-----------------
 .../net/ethernet/mellanox/mlx5/core/eswitch.h |   2 +
 5 files changed, 539 insertions(+), 507 deletions(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.h

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 8bde58379ac6..a1223e904190 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -50,7 +50,7 @@ mlx5_core-$(CONFIG_MLX5_TC_CT)	     += en/tc_ct.o
 # Core extra
 #
 mlx5_core-$(CONFIG_MLX5_ESWITCH)   += eswitch.o eswitch_offloads.o eswitch_offloads_termtbl.o \
-				      ecpf.o rdma.o
+				      ecpf.o rdma.o esw/legacy.o
 mlx5_core-$(CONFIG_MLX5_ESWITCH)   += esw/acl/helper.o \
 				      esw/acl/egress_lgcy.o esw/acl/egress_ofld.o \
 				      esw/acl/ingress_lgcy.o esw/acl/ingress_ofld.o \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
new file mode 100644
index 000000000000..8ab1224653a4
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
@@ -0,0 +1,509 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2021 Mellanox Technologies Ltd */
+
+#include <linux/etherdevice.h>
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/mlx5_ifc.h>
+#include <linux/mlx5/vport.h>
+#include <linux/mlx5/fs.h>
+#include "esw/acl/lgcy.h"
+#include "esw/legacy.h"
+#include "mlx5_core.h"
+#include "eswitch.h"
+#include "fs_core.h"
+
+enum {
+	LEGACY_VEPA_PRIO = 0,
+	LEGACY_FDB_PRIO,
+};
+
+static int esw_create_legacy_vepa_table(struct mlx5_eswitch *esw)
+{
+	struct mlx5_flow_table_attr ft_attr = {};
+	struct mlx5_core_dev *dev = esw->dev;
+	struct mlx5_flow_namespace *root_ns;
+	struct mlx5_flow_table *fdb;
+	int err;
+
+	root_ns = mlx5_get_fdb_sub_ns(dev, 0);
+	if (!root_ns) {
+		esw_warn(dev, "Failed to get FDB flow namespace\n");
+		return -EOPNOTSUPP;
+	}
+
+	/* num FTE 2, num FG 2 */
+	ft_attr.prio = LEGACY_VEPA_PRIO;
+	ft_attr.max_fte = 2;
+	ft_attr.autogroup.max_num_groups = 2;
+	fdb = mlx5_create_auto_grouped_flow_table(root_ns, &ft_attr);
+	if (IS_ERR(fdb)) {
+		err = PTR_ERR(fdb);
+		esw_warn(dev, "Failed to create VEPA FDB err %d\n", err);
+		return err;
+	}
+	esw->fdb_table.legacy.vepa_fdb = fdb;
+
+	return 0;
+}
+
+static void esw_destroy_legacy_fdb_table(struct mlx5_eswitch *esw)
+{
+	esw_debug(esw->dev, "Destroy FDB Table\n");
+	if (!esw->fdb_table.legacy.fdb)
+		return;
+
+	if (esw->fdb_table.legacy.promisc_grp)
+		mlx5_destroy_flow_group(esw->fdb_table.legacy.promisc_grp);
+	if (esw->fdb_table.legacy.allmulti_grp)
+		mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp);
+	if (esw->fdb_table.legacy.addr_grp)
+		mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp);
+	mlx5_destroy_flow_table(esw->fdb_table.legacy.fdb);
+
+	esw->fdb_table.legacy.fdb = NULL;
+	esw->fdb_table.legacy.addr_grp = NULL;
+	esw->fdb_table.legacy.allmulti_grp = NULL;
+	esw->fdb_table.legacy.promisc_grp = NULL;
+	atomic64_set(&esw->user_count, 0);
+}
+
+static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_table_attr ft_attr = {};
+	struct mlx5_core_dev *dev = esw->dev;
+	struct mlx5_flow_namespace *root_ns;
+	struct mlx5_flow_table *fdb;
+	struct mlx5_flow_group *g;
+	void *match_criteria;
+	int table_size;
+	u32 *flow_group_in;
+	u8 *dmac;
+	int err = 0;
+
+	esw_debug(dev, "Create FDB log_max_size(%d)\n",
+		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
+
+	root_ns = mlx5_get_fdb_sub_ns(dev, 0);
+	if (!root_ns) {
+		esw_warn(dev, "Failed to get FDB flow namespace\n");
+		return -EOPNOTSUPP;
+	}
+
+	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
+	if (!flow_group_in)
+		return -ENOMEM;
+
+	table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
+	ft_attr.max_fte = table_size;
+	ft_attr.prio = LEGACY_FDB_PRIO;
+	fdb = mlx5_create_flow_table(root_ns, &ft_attr);
+	if (IS_ERR(fdb)) {
+		err = PTR_ERR(fdb);
+		esw_warn(dev, "Failed to create FDB Table err %d\n", err);
+		goto out;
+	}
+	esw->fdb_table.legacy.fdb = fdb;
+
+	/* Addresses group : Full match unicast/multicast addresses */
+	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
+		 MLX5_MATCH_OUTER_HEADERS);
+	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
+	dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, outer_headers.dmac_47_16);
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
+	/* Preserve 2 entries for allmulti and promisc rules*/
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 3);
+	eth_broadcast_addr(dmac);
+	g = mlx5_create_flow_group(fdb, flow_group_in);
+	if (IS_ERR(g)) {
+		err = PTR_ERR(g);
+		esw_warn(dev, "Failed to create flow group err(%d)\n", err);
+		goto out;
+	}
+	esw->fdb_table.legacy.addr_grp = g;
+
+	/* Allmulti group : One rule that forwards any mcast traffic */
+	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
+		 MLX5_MATCH_OUTER_HEADERS);
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 2);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 2);
+	eth_zero_addr(dmac);
+	dmac[0] = 0x01;
+	g = mlx5_create_flow_group(fdb, flow_group_in);
+	if (IS_ERR(g)) {
+		err = PTR_ERR(g);
+		esw_warn(dev, "Failed to create allmulti flow group err(%d)\n", err);
+		goto out;
+	}
+	esw->fdb_table.legacy.allmulti_grp = g;
+
+	/* Promiscuous group :
+	 * One rule that forward all unmatched traffic from previous groups
+	 */
+	eth_zero_addr(dmac);
+	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
+		 MLX5_MATCH_MISC_PARAMETERS);
+	MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port);
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 1);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1);
+	g = mlx5_create_flow_group(fdb, flow_group_in);
+	if (IS_ERR(g)) {
+		err = PTR_ERR(g);
+		esw_warn(dev, "Failed to create promisc flow group err(%d)\n", err);
+		goto out;
+	}
+	esw->fdb_table.legacy.promisc_grp = g;
+
+out:
+	if (err)
+		esw_destroy_legacy_fdb_table(esw);
+
+	kvfree(flow_group_in);
+	return err;
+}
+
+static void esw_destroy_legacy_vepa_table(struct mlx5_eswitch *esw)
+{
+	esw_debug(esw->dev, "Destroy VEPA Table\n");
+	if (!esw->fdb_table.legacy.vepa_fdb)
+		return;
+
+	mlx5_destroy_flow_table(esw->fdb_table.legacy.vepa_fdb);
+	esw->fdb_table.legacy.vepa_fdb = NULL;
+}
+
+static int esw_create_legacy_table(struct mlx5_eswitch *esw)
+{
+	int err;
+
+	memset(&esw->fdb_table.legacy, 0, sizeof(struct legacy_fdb));
+	atomic64_set(&esw->user_count, 0);
+
+	err = esw_create_legacy_vepa_table(esw);
+	if (err)
+		return err;
+
+	err = esw_create_legacy_fdb_table(esw);
+	if (err)
+		esw_destroy_legacy_vepa_table(esw);
+
+	return err;
+}
+
+static void esw_cleanup_vepa_rules(struct mlx5_eswitch *esw)
+{
+	if (esw->fdb_table.legacy.vepa_uplink_rule)
+		mlx5_del_flow_rules(esw->fdb_table.legacy.vepa_uplink_rule);
+
+	if (esw->fdb_table.legacy.vepa_star_rule)
+		mlx5_del_flow_rules(esw->fdb_table.legacy.vepa_star_rule);
+
+	esw->fdb_table.legacy.vepa_uplink_rule = NULL;
+	esw->fdb_table.legacy.vepa_star_rule = NULL;
+}
+
+static void esw_destroy_legacy_table(struct mlx5_eswitch *esw)
+{
+	esw_cleanup_vepa_rules(esw);
+	esw_destroy_legacy_fdb_table(esw);
+	esw_destroy_legacy_vepa_table(esw);
+}
+
+#define MLX5_LEGACY_SRIOV_VPORT_EVENTS (MLX5_VPORT_UC_ADDR_CHANGE | \
+					MLX5_VPORT_MC_ADDR_CHANGE | \
+					MLX5_VPORT_PROMISC_CHANGE)
+
+int esw_legacy_enable(struct mlx5_eswitch *esw)
+{
+	struct mlx5_vport *vport;
+	int ret, i;
+
+	ret = esw_create_legacy_table(esw);
+	if (ret)
+		return ret;
+
+	mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs)
+		vport->info.link_state = MLX5_VPORT_ADMIN_STATE_AUTO;
+
+	ret = mlx5_eswitch_enable_pf_vf_vports(esw, MLX5_LEGACY_SRIOV_VPORT_EVENTS);
+	if (ret)
+		esw_destroy_legacy_table(esw);
+	return ret;
+}
+
+void esw_legacy_disable(struct mlx5_eswitch *esw)
+{
+	struct esw_mc_addr *mc_promisc;
+
+	mlx5_eswitch_disable_pf_vf_vports(esw);
+
+	mc_promisc = &esw->mc_promisc;
+	if (mc_promisc->uplink_rule)
+		mlx5_del_flow_rules(mc_promisc->uplink_rule);
+
+	esw_destroy_legacy_table(esw);
+}
+
+static int _mlx5_eswitch_set_vepa_locked(struct mlx5_eswitch *esw,
+					 u8 setting)
+{
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *flow_rule;
+	struct mlx5_flow_spec *spec;
+	int err = 0;
+	void *misc;
+
+	if (!setting) {
+		esw_cleanup_vepa_rules(esw);
+		return 0;
+	}
+
+	if (esw->fdb_table.legacy.vepa_uplink_rule)
+		return 0;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	/* Uplink rule forward uplink traffic to FDB */
+	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
+	MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_UPLINK);
+
+	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
+	MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
+
+	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest.ft = esw->fdb_table.legacy.fdb;
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	flow_rule = mlx5_add_flow_rules(esw->fdb_table.legacy.vepa_fdb, spec,
+					&flow_act, &dest, 1);
+	if (IS_ERR(flow_rule)) {
+		err = PTR_ERR(flow_rule);
+		goto out;
+	} else {
+		esw->fdb_table.legacy.vepa_uplink_rule = flow_rule;
+	}
+
+	/* Star rule to forward all traffic to uplink vport */
+	memset(&dest, 0, sizeof(dest));
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
+	dest.vport.num = MLX5_VPORT_UPLINK;
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	flow_rule = mlx5_add_flow_rules(esw->fdb_table.legacy.vepa_fdb, NULL,
+					&flow_act, &dest, 1);
+	if (IS_ERR(flow_rule)) {
+		err = PTR_ERR(flow_rule);
+		goto out;
+	} else {
+		esw->fdb_table.legacy.vepa_star_rule = flow_rule;
+	}
+
+out:
+	kvfree(spec);
+	if (err)
+		esw_cleanup_vepa_rules(esw);
+	return err;
+}
+
+int mlx5_eswitch_set_vepa(struct mlx5_eswitch *esw, u8 setting)
+{
+	int err = 0;
+
+	if (!esw)
+		return -EOPNOTSUPP;
+
+	if (!mlx5_esw_allowed(esw))
+		return -EPERM;
+
+	mutex_lock(&esw->state_lock);
+	if (esw->mode != MLX5_ESWITCH_LEGACY) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	err = _mlx5_eswitch_set_vepa_locked(esw, setting);
+
+out:
+	mutex_unlock(&esw->state_lock);
+	return err;
+}
+
+int mlx5_eswitch_get_vepa(struct mlx5_eswitch *esw, u8 *setting)
+{
+	if (!esw)
+		return -EOPNOTSUPP;
+
+	if (!mlx5_esw_allowed(esw))
+		return -EPERM;
+
+	if (esw->mode != MLX5_ESWITCH_LEGACY)
+		return -EOPNOTSUPP;
+
+	*setting = esw->fdb_table.legacy.vepa_uplink_rule ? 1 : 0;
+	return 0;
+}
+
+int esw_legacy_vport_acl_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
+{
+	int ret;
+
+	/* Only non manager vports need ACL in legacy mode */
+	if (mlx5_esw_is_manager_vport(esw, vport->vport))
+		return 0;
+
+	ret = esw_acl_ingress_lgcy_setup(esw, vport);
+	if (ret)
+		goto ingress_err;
+
+	ret = esw_acl_egress_lgcy_setup(esw, vport);
+	if (ret)
+		goto egress_err;
+
+	return 0;
+
+egress_err:
+	esw_acl_ingress_lgcy_cleanup(esw, vport);
+ingress_err:
+	return ret;
+}
+
+void esw_legacy_vport_acl_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
+{
+	if (mlx5_esw_is_manager_vport(esw, vport->vport))
+		return;
+
+	esw_acl_egress_lgcy_cleanup(esw, vport);
+	esw_acl_ingress_lgcy_cleanup(esw, vport);
+}
+
+int mlx5_esw_query_vport_drop_stats(struct mlx5_core_dev *dev,
+				    struct mlx5_vport *vport,
+				    struct mlx5_vport_drop_stats *stats)
+{
+	u64 rx_discard_vport_down, tx_discard_vport_down;
+	struct mlx5_eswitch *esw = dev->priv.eswitch;
+	u64 bytes = 0;
+	int err = 0;
+
+	if (esw->mode != MLX5_ESWITCH_LEGACY)
+		return 0;
+
+	mutex_lock(&esw->state_lock);
+	if (!vport->enabled)
+		goto unlock;
+
+	if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_counter))
+		mlx5_fc_query(dev, vport->egress.legacy.drop_counter,
+			      &stats->rx_dropped, &bytes);
+
+	if (vport->ingress.legacy.drop_counter)
+		mlx5_fc_query(dev, vport->ingress.legacy.drop_counter,
+			      &stats->tx_dropped, &bytes);
+
+	if (!MLX5_CAP_GEN(dev, receive_discard_vport_down) &&
+	    !MLX5_CAP_GEN(dev, transmit_discard_vport_down))
+		goto unlock;
+
+	err = mlx5_query_vport_down_stats(dev, vport->vport, 1,
+					  &rx_discard_vport_down,
+					  &tx_discard_vport_down);
+	if (err)
+		goto unlock;
+
+	if (MLX5_CAP_GEN(dev, receive_discard_vport_down))
+		stats->rx_dropped += rx_discard_vport_down;
+	if (MLX5_CAP_GEN(dev, transmit_discard_vport_down))
+		stats->tx_dropped += tx_discard_vport_down;
+
+unlock:
+	mutex_unlock(&esw->state_lock);
+	return err;
+}
+
+int mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
+				u16 vport, u16 vlan, u8 qos)
+{
+	u8 set_flags = 0;
+	int err = 0;
+
+	if (!mlx5_esw_allowed(esw))
+		return -EPERM;
+
+	if (vlan || qos)
+		set_flags = SET_VLAN_STRIP | SET_VLAN_INSERT;
+
+	mutex_lock(&esw->state_lock);
+	if (esw->mode != MLX5_ESWITCH_LEGACY) {
+		if (!vlan)
+			goto unlock; /* compatibility with libvirt */
+
+		err = -EOPNOTSUPP;
+		goto unlock;
+	}
+
+	err = __mlx5_eswitch_set_vport_vlan(esw, vport, vlan, qos, set_flags);
+
+unlock:
+	mutex_unlock(&esw->state_lock);
+	return err;
+}
+
+int mlx5_eswitch_set_vport_spoofchk(struct mlx5_eswitch *esw,
+				    u16 vport, bool spoofchk)
+{
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
+	bool pschk;
+	int err = 0;
+
+	if (!mlx5_esw_allowed(esw))
+		return -EPERM;
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
+
+	mutex_lock(&esw->state_lock);
+	if (esw->mode != MLX5_ESWITCH_LEGACY) {
+		err = -EOPNOTSUPP;
+		goto unlock;
+	}
+	pschk = evport->info.spoofchk;
+	evport->info.spoofchk = spoofchk;
+	if (pschk && !is_valid_ether_addr(evport->info.mac))
+		mlx5_core_warn(esw->dev,
+			       "Spoofchk in set while MAC is invalid, vport(%d)\n",
+			       evport->vport);
+	if (evport->enabled && esw->mode == MLX5_ESWITCH_LEGACY)
+		err = esw_acl_ingress_lgcy_setup(esw, evport);
+	if (err)
+		evport->info.spoofchk = pschk;
+
+unlock:
+	mutex_unlock(&esw->state_lock);
+	return err;
+}
+
+int mlx5_eswitch_set_vport_trust(struct mlx5_eswitch *esw,
+				 u16 vport, bool setting)
+{
+	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
+	int err = 0;
+
+	if (!mlx5_esw_allowed(esw))
+		return -EPERM;
+	if (IS_ERR(evport))
+		return PTR_ERR(evport);
+
+	mutex_lock(&esw->state_lock);
+	if (esw->mode != MLX5_ESWITCH_LEGACY) {
+		err = -EOPNOTSUPP;
+		goto unlock;
+	}
+	evport->info.trusted = setting;
+	if (evport->enabled)
+		esw_vport_change_handle_locked(evport);
+
+unlock:
+	mutex_unlock(&esw->state_lock);
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.h
new file mode 100644
index 000000000000..e0820bb72b57
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2021 Mellanox Technologies Ltd */
+
+#ifndef __MLX5_ESW_LEGACY_H__
+#define __MLX5_ESW_LEGACY_H__
+
+#define MLX5_LEGACY_SRIOV_VPORT_EVENTS (MLX5_VPORT_UC_ADDR_CHANGE | \
+					MLX5_VPORT_MC_ADDR_CHANGE | \
+					MLX5_VPORT_PROMISC_CHANGE)
+
+struct mlx5_eswitch;
+
+int esw_legacy_enable(struct mlx5_eswitch *esw);
+void esw_legacy_disable(struct mlx5_eswitch *esw);
+
+int esw_legacy_vport_acl_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport);
+void esw_legacy_vport_acl_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport);
+
+int mlx5_esw_query_vport_drop_stats(struct mlx5_core_dev *dev,
+				    struct mlx5_vport *vport,
+				    struct mlx5_vport_drop_stats *stats);
+#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 6a70e385beb8..3dfcd4ec1556 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -36,6 +36,7 @@
 #include <linux/mlx5/vport.h>
 #include <linux/mlx5/fs.h>
 #include "esw/acl/lgcy.h"
+#include "esw/legacy.h"
 #include "mlx5_core.h"
 #include "lib/eq.h"
 #include "eswitch.h"
@@ -61,9 +62,6 @@ struct vport_addr {
 	bool mc_promisc;
 };
 
-static void esw_destroy_legacy_fdb_table(struct mlx5_eswitch *esw);
-static void esw_cleanup_vepa_rules(struct mlx5_eswitch *esw);
-
 static int mlx5_eswitch_check(const struct mlx5_core_dev *dev)
 {
 	if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH)
@@ -278,226 +276,6 @@ esw_fdb_set_vport_promisc_rule(struct mlx5_eswitch *esw, u16 vport)
 	return __esw_fdb_set_vport_rule(esw, vport, true, mac_c, mac_v);
 }
 
-enum {
-	LEGACY_VEPA_PRIO = 0,
-	LEGACY_FDB_PRIO,
-};
-
-static int esw_create_legacy_vepa_table(struct mlx5_eswitch *esw)
-{
-	struct mlx5_flow_table_attr ft_attr = {};
-	struct mlx5_core_dev *dev = esw->dev;
-	struct mlx5_flow_namespace *root_ns;
-	struct mlx5_flow_table *fdb;
-	int err;
-
-	root_ns = mlx5_get_fdb_sub_ns(dev, 0);
-	if (!root_ns) {
-		esw_warn(dev, "Failed to get FDB flow namespace\n");
-		return -EOPNOTSUPP;
-	}
-
-	/* num FTE 2, num FG 2 */
-	ft_attr.prio = LEGACY_VEPA_PRIO;
-	ft_attr.max_fte = 2;
-	ft_attr.autogroup.max_num_groups = 2;
-	fdb = mlx5_create_auto_grouped_flow_table(root_ns, &ft_attr);
-	if (IS_ERR(fdb)) {
-		err = PTR_ERR(fdb);
-		esw_warn(dev, "Failed to create VEPA FDB err %d\n", err);
-		return err;
-	}
-	esw->fdb_table.legacy.vepa_fdb = fdb;
-
-	return 0;
-}
-
-static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw)
-{
-	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
-	struct mlx5_flow_table_attr ft_attr = {};
-	struct mlx5_core_dev *dev = esw->dev;
-	struct mlx5_flow_namespace *root_ns;
-	struct mlx5_flow_table *fdb;
-	struct mlx5_flow_group *g;
-	void *match_criteria;
-	int table_size;
-	u32 *flow_group_in;
-	u8 *dmac;
-	int err = 0;
-
-	esw_debug(dev, "Create FDB log_max_size(%d)\n",
-		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
-
-	root_ns = mlx5_get_fdb_sub_ns(dev, 0);
-	if (!root_ns) {
-		esw_warn(dev, "Failed to get FDB flow namespace\n");
-		return -EOPNOTSUPP;
-	}
-
-	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
-	if (!flow_group_in)
-		return -ENOMEM;
-
-	table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
-	ft_attr.max_fte = table_size;
-	ft_attr.prio = LEGACY_FDB_PRIO;
-	fdb = mlx5_create_flow_table(root_ns, &ft_attr);
-	if (IS_ERR(fdb)) {
-		err = PTR_ERR(fdb);
-		esw_warn(dev, "Failed to create FDB Table err %d\n", err);
-		goto out;
-	}
-	esw->fdb_table.legacy.fdb = fdb;
-
-	/* Addresses group : Full match unicast/multicast addresses */
-	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
-		 MLX5_MATCH_OUTER_HEADERS);
-	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
-	dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, outer_headers.dmac_47_16);
-	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
-	/* Preserve 2 entries for allmulti and promisc rules*/
-	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 3);
-	eth_broadcast_addr(dmac);
-	g = mlx5_create_flow_group(fdb, flow_group_in);
-	if (IS_ERR(g)) {
-		err = PTR_ERR(g);
-		esw_warn(dev, "Failed to create flow group err(%d)\n", err);
-		goto out;
-	}
-	esw->fdb_table.legacy.addr_grp = g;
-
-	/* Allmulti group : One rule that forwards any mcast traffic */
-	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
-		 MLX5_MATCH_OUTER_HEADERS);
-	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 2);
-	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 2);
-	eth_zero_addr(dmac);
-	dmac[0] = 0x01;
-	g = mlx5_create_flow_group(fdb, flow_group_in);
-	if (IS_ERR(g)) {
-		err = PTR_ERR(g);
-		esw_warn(dev, "Failed to create allmulti flow group err(%d)\n", err);
-		goto out;
-	}
-	esw->fdb_table.legacy.allmulti_grp = g;
-
-	/* Promiscuous group :
-	 * One rule that forward all unmatched traffic from previous groups
-	 */
-	eth_zero_addr(dmac);
-	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
-		 MLX5_MATCH_MISC_PARAMETERS);
-	MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port);
-	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 1);
-	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1);
-	g = mlx5_create_flow_group(fdb, flow_group_in);
-	if (IS_ERR(g)) {
-		err = PTR_ERR(g);
-		esw_warn(dev, "Failed to create promisc flow group err(%d)\n", err);
-		goto out;
-	}
-	esw->fdb_table.legacy.promisc_grp = g;
-
-out:
-	if (err)
-		esw_destroy_legacy_fdb_table(esw);
-
-	kvfree(flow_group_in);
-	return err;
-}
-
-static void esw_destroy_legacy_vepa_table(struct mlx5_eswitch *esw)
-{
-	esw_debug(esw->dev, "Destroy VEPA Table\n");
-	if (!esw->fdb_table.legacy.vepa_fdb)
-		return;
-
-	mlx5_destroy_flow_table(esw->fdb_table.legacy.vepa_fdb);
-	esw->fdb_table.legacy.vepa_fdb = NULL;
-}
-
-static void esw_destroy_legacy_fdb_table(struct mlx5_eswitch *esw)
-{
-	esw_debug(esw->dev, "Destroy FDB Table\n");
-	if (!esw->fdb_table.legacy.fdb)
-		return;
-
-	if (esw->fdb_table.legacy.promisc_grp)
-		mlx5_destroy_flow_group(esw->fdb_table.legacy.promisc_grp);
-	if (esw->fdb_table.legacy.allmulti_grp)
-		mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp);
-	if (esw->fdb_table.legacy.addr_grp)
-		mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp);
-	mlx5_destroy_flow_table(esw->fdb_table.legacy.fdb);
-
-	esw->fdb_table.legacy.fdb = NULL;
-	esw->fdb_table.legacy.addr_grp = NULL;
-	esw->fdb_table.legacy.allmulti_grp = NULL;
-	esw->fdb_table.legacy.promisc_grp = NULL;
-	atomic64_set(&esw->user_count, 0);
-}
-
-static int esw_create_legacy_table(struct mlx5_eswitch *esw)
-{
-	int err;
-
-	memset(&esw->fdb_table.legacy, 0, sizeof(struct legacy_fdb));
-	atomic64_set(&esw->user_count, 0);
-
-	err = esw_create_legacy_vepa_table(esw);
-	if (err)
-		return err;
-
-	err = esw_create_legacy_fdb_table(esw);
-	if (err)
-		esw_destroy_legacy_vepa_table(esw);
-
-	return err;
-}
-
-static void esw_destroy_legacy_table(struct mlx5_eswitch *esw)
-{
-	esw_cleanup_vepa_rules(esw);
-	esw_destroy_legacy_fdb_table(esw);
-	esw_destroy_legacy_vepa_table(esw);
-}
-
-#define MLX5_LEGACY_SRIOV_VPORT_EVENTS (MLX5_VPORT_UC_ADDR_CHANGE | \
-					MLX5_VPORT_MC_ADDR_CHANGE | \
-					MLX5_VPORT_PROMISC_CHANGE)
-
-static int esw_legacy_enable(struct mlx5_eswitch *esw)
-{
-	struct mlx5_vport *vport;
-	int ret, i;
-
-	ret = esw_create_legacy_table(esw);
-	if (ret)
-		return ret;
-
-	mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs)
-		vport->info.link_state = MLX5_VPORT_ADMIN_STATE_AUTO;
-
-	ret = mlx5_eswitch_enable_pf_vf_vports(esw, MLX5_LEGACY_SRIOV_VPORT_EVENTS);
-	if (ret)
-		esw_destroy_legacy_table(esw);
-	return ret;
-}
-
-static void esw_legacy_disable(struct mlx5_eswitch *esw)
-{
-	struct esw_mc_addr *mc_promisc;
-
-	mlx5_eswitch_disable_pf_vf_vports(esw);
-
-	mc_promisc = &esw->mc_promisc;
-	if (mc_promisc->uplink_rule)
-		mlx5_del_flow_rules(mc_promisc->uplink_rule);
-
-	esw_destroy_legacy_table(esw);
-}
-
 /* E-Switch vport UC/MC lists management */
 typedef int (*vport_addr_action)(struct mlx5_eswitch *esw,
 				 struct vport_addr *vaddr);
@@ -919,7 +697,7 @@ static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw,
 				(promisc_all || promisc_mc));
 }
 
-static void esw_vport_change_handle_locked(struct mlx5_vport *vport)
+void esw_vport_change_handle_locked(struct mlx5_vport *vport)
 {
 	struct mlx5_core_dev *dev = vport->dev;
 	struct mlx5_eswitch *esw = dev->priv.eswitch;
@@ -1170,56 +948,20 @@ static void node_guid_gen_from_mac(u64 *node_guid, const u8 *mac)
 	((u8 *)node_guid)[0] = mac[5];
 }
 
-static int esw_vport_create_legacy_acl_tables(struct mlx5_eswitch *esw,
-					      struct mlx5_vport *vport)
-{
-	int ret;
-
-	/* Only non manager vports need ACL in legacy mode */
-	if (mlx5_esw_is_manager_vport(esw, vport->vport))
-		return 0;
-
-	ret = esw_acl_ingress_lgcy_setup(esw, vport);
-	if (ret)
-		goto ingress_err;
-
-	ret = esw_acl_egress_lgcy_setup(esw, vport);
-	if (ret)
-		goto egress_err;
-
-	return 0;
-
-egress_err:
-	esw_acl_ingress_lgcy_cleanup(esw, vport);
-ingress_err:
-	return ret;
-}
-
 static int esw_vport_setup_acl(struct mlx5_eswitch *esw,
 			       struct mlx5_vport *vport)
 {
 	if (esw->mode == MLX5_ESWITCH_LEGACY)
-		return esw_vport_create_legacy_acl_tables(esw, vport);
+		return esw_legacy_vport_acl_setup(esw, vport);
 	else
 		return esw_vport_create_offloads_acl_tables(esw, vport);
 }
 
-static void esw_vport_destroy_legacy_acl_tables(struct mlx5_eswitch *esw,
-						struct mlx5_vport *vport)
-
-{
-	if (mlx5_esw_is_manager_vport(esw, vport->vport))
-		return;
-
-	esw_acl_egress_lgcy_cleanup(esw, vport);
-	esw_acl_ingress_lgcy_cleanup(esw, vport);
-}
-
 static void esw_vport_cleanup_acl(struct mlx5_eswitch *esw,
 				  struct mlx5_vport *vport)
 {
 	if (esw->mode == MLX5_ESWITCH_LEGACY)
-		esw_vport_destroy_legacy_acl_tables(esw, vport);
+		esw_legacy_vport_acl_cleanup(esw, vport);
 	else
 		esw_vport_destroy_offloads_acl_tables(esw, vport);
 }
@@ -2103,205 +1845,6 @@ int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
 	return err;
 }
 
-int mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
-				u16 vport, u16 vlan, u8 qos)
-{
-	u8 set_flags = 0;
-	int err = 0;
-
-	if (!mlx5_esw_allowed(esw))
-		return -EPERM;
-
-	if (vlan || qos)
-		set_flags = SET_VLAN_STRIP | SET_VLAN_INSERT;
-
-	mutex_lock(&esw->state_lock);
-	if (esw->mode != MLX5_ESWITCH_LEGACY) {
-		if (!vlan)
-			goto unlock; /* compatibility with libvirt */
-
-		err = -EOPNOTSUPP;
-		goto unlock;
-	}
-
-	err = __mlx5_eswitch_set_vport_vlan(esw, vport, vlan, qos, set_flags);
-
-unlock:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
-int mlx5_eswitch_set_vport_spoofchk(struct mlx5_eswitch *esw,
-				    u16 vport, bool spoofchk)
-{
-	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
-	bool pschk;
-	int err = 0;
-
-	if (!mlx5_esw_allowed(esw))
-		return -EPERM;
-	if (IS_ERR(evport))
-		return PTR_ERR(evport);
-
-	mutex_lock(&esw->state_lock);
-	if (esw->mode != MLX5_ESWITCH_LEGACY) {
-		err = -EOPNOTSUPP;
-		goto unlock;
-	}
-	pschk = evport->info.spoofchk;
-	evport->info.spoofchk = spoofchk;
-	if (pschk && !is_valid_ether_addr(evport->info.mac))
-		mlx5_core_warn(esw->dev,
-			       "Spoofchk in set while MAC is invalid, vport(%d)\n",
-			       evport->vport);
-	if (evport->enabled && esw->mode == MLX5_ESWITCH_LEGACY)
-		err = esw_acl_ingress_lgcy_setup(esw, evport);
-	if (err)
-		evport->info.spoofchk = pschk;
-
-unlock:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
-static void esw_cleanup_vepa_rules(struct mlx5_eswitch *esw)
-{
-	if (esw->fdb_table.legacy.vepa_uplink_rule)
-		mlx5_del_flow_rules(esw->fdb_table.legacy.vepa_uplink_rule);
-
-	if (esw->fdb_table.legacy.vepa_star_rule)
-		mlx5_del_flow_rules(esw->fdb_table.legacy.vepa_star_rule);
-
-	esw->fdb_table.legacy.vepa_uplink_rule = NULL;
-	esw->fdb_table.legacy.vepa_star_rule = NULL;
-}
-
-static int _mlx5_eswitch_set_vepa_locked(struct mlx5_eswitch *esw,
-					 u8 setting)
-{
-	struct mlx5_flow_destination dest = {};
-	struct mlx5_flow_act flow_act = {};
-	struct mlx5_flow_handle *flow_rule;
-	struct mlx5_flow_spec *spec;
-	int err = 0;
-	void *misc;
-
-	if (!setting) {
-		esw_cleanup_vepa_rules(esw);
-		return 0;
-	}
-
-	if (esw->fdb_table.legacy.vepa_uplink_rule)
-		return 0;
-
-	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
-	if (!spec)
-		return -ENOMEM;
-
-	/* Uplink rule forward uplink traffic to FDB */
-	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
-	MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_UPLINK);
-
-	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
-	MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
-
-	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-	dest.ft = esw->fdb_table.legacy.fdb;
-	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
-	flow_rule = mlx5_add_flow_rules(esw->fdb_table.legacy.vepa_fdb, spec,
-					&flow_act, &dest, 1);
-	if (IS_ERR(flow_rule)) {
-		err = PTR_ERR(flow_rule);
-		goto out;
-	} else {
-		esw->fdb_table.legacy.vepa_uplink_rule = flow_rule;
-	}
-
-	/* Star rule to forward all traffic to uplink vport */
-	memset(&dest, 0, sizeof(dest));
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
-	dest.vport.num = MLX5_VPORT_UPLINK;
-	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
-	flow_rule = mlx5_add_flow_rules(esw->fdb_table.legacy.vepa_fdb, NULL,
-					&flow_act, &dest, 1);
-	if (IS_ERR(flow_rule)) {
-		err = PTR_ERR(flow_rule);
-		goto out;
-	} else {
-		esw->fdb_table.legacy.vepa_star_rule = flow_rule;
-	}
-
-out:
-	kvfree(spec);
-	if (err)
-		esw_cleanup_vepa_rules(esw);
-	return err;
-}
-
-int mlx5_eswitch_set_vepa(struct mlx5_eswitch *esw, u8 setting)
-{
-	int err = 0;
-
-	if (!esw)
-		return -EOPNOTSUPP;
-
-	if (!mlx5_esw_allowed(esw))
-		return -EPERM;
-
-	mutex_lock(&esw->state_lock);
-	if (esw->mode != MLX5_ESWITCH_LEGACY) {
-		err = -EOPNOTSUPP;
-		goto out;
-	}
-
-	err = _mlx5_eswitch_set_vepa_locked(esw, setting);
-
-out:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
-int mlx5_eswitch_get_vepa(struct mlx5_eswitch *esw, u8 *setting)
-{
-	if (!esw)
-		return -EOPNOTSUPP;
-
-	if (!mlx5_esw_allowed(esw))
-		return -EPERM;
-
-	if (esw->mode != MLX5_ESWITCH_LEGACY)
-		return -EOPNOTSUPP;
-
-	*setting = esw->fdb_table.legacy.vepa_uplink_rule ? 1 : 0;
-	return 0;
-}
-
-int mlx5_eswitch_set_vport_trust(struct mlx5_eswitch *esw,
-				 u16 vport, bool setting)
-{
-	struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport);
-	int err = 0;
-
-	if (!mlx5_esw_allowed(esw))
-		return -EPERM;
-	if (IS_ERR(evport))
-		return PTR_ERR(evport);
-
-	mutex_lock(&esw->state_lock);
-	if (esw->mode != MLX5_ESWITCH_LEGACY) {
-		err = -EOPNOTSUPP;
-		goto unlock;
-	}
-	evport->info.trusted = setting;
-	if (evport->enabled)
-		esw_vport_change_handle_locked(evport);
-
-unlock:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
 static u32 calculate_vports_min_rate_divider(struct mlx5_eswitch *esw)
 {
 	u32 fw_max_bw_share = MLX5_CAP_QOS(esw->dev, max_tsar_bw_share);
@@ -2406,50 +1949,6 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, u16 vport,
 	return err;
 }
 
-static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
-					       struct mlx5_vport *vport,
-					       struct mlx5_vport_drop_stats *stats)
-{
-	struct mlx5_eswitch *esw = dev->priv.eswitch;
-	u64 rx_discard_vport_down, tx_discard_vport_down;
-	u64 bytes = 0;
-	int err = 0;
-
-	if (esw->mode != MLX5_ESWITCH_LEGACY)
-		return 0;
-
-	mutex_lock(&esw->state_lock);
-	if (!vport->enabled)
-		goto unlock;
-
-	if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_counter))
-		mlx5_fc_query(dev, vport->egress.legacy.drop_counter,
-			      &stats->rx_dropped, &bytes);
-
-	if (vport->ingress.legacy.drop_counter)
-		mlx5_fc_query(dev, vport->ingress.legacy.drop_counter,
-			      &stats->tx_dropped, &bytes);
-
-	if (!MLX5_CAP_GEN(dev, receive_discard_vport_down) &&
-	    !MLX5_CAP_GEN(dev, transmit_discard_vport_down))
-		goto unlock;
-
-	err = mlx5_query_vport_down_stats(dev, vport->vport, 1,
-					  &rx_discard_vport_down,
-					  &tx_discard_vport_down);
-	if (err)
-		goto unlock;
-
-	if (MLX5_CAP_GEN(dev, receive_discard_vport_down))
-		stats->rx_dropped += rx_discard_vport_down;
-	if (MLX5_CAP_GEN(dev, transmit_discard_vport_down))
-		stats->tx_dropped += tx_discard_vport_down;
-
-unlock:
-	mutex_unlock(&esw->state_lock);
-	return err;
-}
-
 int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
 				 u16 vport_num,
 				 struct ifla_vf_stats *vf_stats)
@@ -2517,7 +2016,7 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
 	vf_stats->broadcast =
 		MLX5_GET_CTR(out, received_eth_broadcast.packets);
 
-	err = mlx5_eswitch_query_vport_drop_stats(esw->dev, vport, &stats);
+	err = mlx5_esw_query_vport_drop_stats(esw->dev, vport, &stats);
 	if (err)
 		goto free_out;
 	vf_stats->rx_dropped = stats.rx_dropped;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index a61a33c36fa8..b289d756a7e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -814,6 +814,8 @@ void mlx5_esw_put(struct mlx5_core_dev *dev);
 int mlx5_esw_try_lock(struct mlx5_eswitch *esw);
 void mlx5_esw_unlock(struct mlx5_eswitch *esw);
 
+void esw_vport_change_handle_locked(struct mlx5_vport *vport);
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
-- 
2.30.2

[net-next V2 07/16] net/mlx5: E-Switch, Initialize eswitch acls ns when eswitch is enabled

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

Currently eswitch flow steering (FS) namespace of vport's ingress and
egress ACL are enabled when FS layer is initialized. This is done even
when eswitch is diabled. This demands that total eswitch ports to be
known to FS layer without eswitch in use.

Given the FS core is not dependent on eswitch, make namespace init and
cleanup routines as helper routines to be invoked only when eswitch is
needed.

With this change, ingress and egress ACL namespaces are created only
when eswitch legacy/offloads mode is enabled.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/eswitch.c | 53 +++++++++--
 .../net/ethernet/mellanox/mlx5/core/fs_core.c | 88 ++++++++-----------
 .../net/ethernet/mellanox/mlx5/core/fs_core.h |  7 ++
 3 files changed, 93 insertions(+), 55 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 3dfcd4ec1556..1bb229ecd43b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1367,6 +1367,47 @@ static void mlx5_esw_mode_change_notify(struct mlx5_eswitch *esw, u16 mode)
 	blocking_notifier_call_chain(&esw->n_head, 0, &info);
 }
 
+static int mlx5_esw_acls_ns_init(struct mlx5_eswitch *esw)
+{
+	struct mlx5_core_dev *dev = esw->dev;
+	int total_vports;
+	int err;
+
+	total_vports = mlx5_eswitch_get_total_vports(dev);
+
+	if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) {
+		err = mlx5_fs_egress_acls_init(dev, total_vports);
+		if (err)
+			return err;
+	} else {
+		esw_warn(dev, "engress ACL is not supported by FW\n");
+	}
+
+	if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) {
+		err = mlx5_fs_ingress_acls_init(dev, total_vports);
+		if (err)
+			goto err;
+	} else {
+		esw_warn(dev, "ingress ACL is not supported by FW\n");
+	}
+	return 0;
+
+err:
+	if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support))
+		mlx5_fs_egress_acls_cleanup(dev);
+	return err;
+}
+
+static void mlx5_esw_acls_ns_cleanup(struct mlx5_eswitch *esw)
+{
+	struct mlx5_core_dev *dev = esw->dev;
+
+	if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support))
+		mlx5_fs_ingress_acls_cleanup(dev);
+	if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support))
+		mlx5_fs_egress_acls_cleanup(dev);
+}
+
 /**
  * mlx5_eswitch_enable_locked - Enable eswitch
  * @esw:	Pointer to eswitch
@@ -1395,14 +1436,12 @@ int mlx5_eswitch_enable_locked(struct mlx5_eswitch *esw, int mode, int num_vfs)
 		return -EOPNOTSUPP;
 	}
 
-	if (!MLX5_CAP_ESW_INGRESS_ACL(esw->dev, ft_support))
-		esw_warn(esw->dev, "ingress ACL is not supported by FW\n");
-
-	if (!MLX5_CAP_ESW_EGRESS_ACL(esw->dev, ft_support))
-		esw_warn(esw->dev, "engress ACL is not supported by FW\n");
-
 	mlx5_eswitch_get_devlink_param(esw);
 
+	err = mlx5_esw_acls_ns_init(esw);
+	if (err)
+		return err;
+
 	mlx5_eswitch_update_num_of_vfs(esw, num_vfs);
 
 	esw_create_tsar(esw);
@@ -1438,6 +1477,7 @@ int mlx5_eswitch_enable_locked(struct mlx5_eswitch *esw, int mode, int num_vfs)
 		mlx5_rescan_drivers(esw->dev);
 
 	esw_destroy_tsar(esw);
+	mlx5_esw_acls_ns_cleanup(esw);
 	return err;
 }
 
@@ -1506,6 +1546,7 @@ void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw, bool clear_vf)
 		mlx5_rescan_drivers(esw->dev);
 
 	esw_destroy_tsar(esw);
+	mlx5_esw_acls_ns_cleanup(esw);
 
 	if (clear_vf)
 		mlx5_eswitch_clear_vf_vports_info(esw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 0216bd63a42d..f74d2c834037 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2229,17 +2229,21 @@ struct mlx5_flow_namespace *mlx5_get_flow_vport_acl_namespace(struct mlx5_core_d
 {
 	struct mlx5_flow_steering *steering = dev->priv.steering;
 
-	if (!steering || vport >= mlx5_eswitch_get_total_vports(dev))
+	if (!steering)
 		return NULL;
 
 	switch (type) {
 	case MLX5_FLOW_NAMESPACE_ESW_EGRESS:
+		if (vport >= steering->esw_egress_acl_vports)
+			return NULL;
 		if (steering->esw_egress_root_ns &&
 		    steering->esw_egress_root_ns[vport])
 			return &steering->esw_egress_root_ns[vport]->ns;
 		else
 			return NULL;
 	case MLX5_FLOW_NAMESPACE_ESW_INGRESS:
+		if (vport >= steering->esw_ingress_acl_vports)
+			return NULL;
 		if (steering->esw_ingress_root_ns &&
 		    steering->esw_ingress_root_ns[vport])
 			return &steering->esw_ingress_root_ns[vport]->ns;
@@ -2571,43 +2575,11 @@ static void cleanup_root_ns(struct mlx5_flow_root_namespace *root_ns)
 	clean_tree(&root_ns->ns.node);
 }
 
-static void cleanup_egress_acls_root_ns(struct mlx5_core_dev *dev)
-{
-	struct mlx5_flow_steering *steering = dev->priv.steering;
-	int i;
-
-	if (!steering->esw_egress_root_ns)
-		return;
-
-	for (i = 0; i < mlx5_eswitch_get_total_vports(dev); i++)
-		cleanup_root_ns(steering->esw_egress_root_ns[i]);
-
-	kfree(steering->esw_egress_root_ns);
-	steering->esw_egress_root_ns = NULL;
-}
-
-static void cleanup_ingress_acls_root_ns(struct mlx5_core_dev *dev)
-{
-	struct mlx5_flow_steering *steering = dev->priv.steering;
-	int i;
-
-	if (!steering->esw_ingress_root_ns)
-		return;
-
-	for (i = 0; i < mlx5_eswitch_get_total_vports(dev); i++)
-		cleanup_root_ns(steering->esw_ingress_root_ns[i]);
-
-	kfree(steering->esw_ingress_root_ns);
-	steering->esw_ingress_root_ns = NULL;
-}
-
 void mlx5_cleanup_fs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_flow_steering *steering = dev->priv.steering;
 
 	cleanup_root_ns(steering->root_ns);
-	cleanup_egress_acls_root_ns(dev);
-	cleanup_ingress_acls_root_ns(dev);
 	cleanup_root_ns(steering->fdb_root_ns);
 	steering->fdb_root_ns = NULL;
 	kfree(steering->fdb_sub_ns);
@@ -2852,10 +2824,9 @@ static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering, int vpo
 	return PTR_ERR_OR_ZERO(prio);
 }
 
-static int init_egress_acls_root_ns(struct mlx5_core_dev *dev)
+int mlx5_fs_egress_acls_init(struct mlx5_core_dev *dev, int total_vports)
 {
 	struct mlx5_flow_steering *steering = dev->priv.steering;
-	int total_vports = mlx5_eswitch_get_total_vports(dev);
 	int err;
 	int i;
 
@@ -2871,7 +2842,7 @@ static int init_egress_acls_root_ns(struct mlx5_core_dev *dev)
 		if (err)
 			goto cleanup_root_ns;
 	}
-
+	steering->esw_egress_acl_vports = total_vports;
 	return 0;
 
 cleanup_root_ns:
@@ -2882,10 +2853,24 @@ static int init_egress_acls_root_ns(struct mlx5_core_dev *dev)
 	return err;
 }
 
-static int init_ingress_acls_root_ns(struct mlx5_core_dev *dev)
+void mlx5_fs_egress_acls_cleanup(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_steering *steering = dev->priv.steering;
+	int i;
+
+	if (!steering->esw_egress_root_ns)
+		return;
+
+	for (i = 0; i < steering->esw_egress_acl_vports; i++)
+		cleanup_root_ns(steering->esw_egress_root_ns[i]);
+
+	kfree(steering->esw_egress_root_ns);
+	steering->esw_egress_root_ns = NULL;
+}
+
+int mlx5_fs_ingress_acls_init(struct mlx5_core_dev *dev, int total_vports)
 {
 	struct mlx5_flow_steering *steering = dev->priv.steering;
-	int total_vports = mlx5_eswitch_get_total_vports(dev);
 	int err;
 	int i;
 
@@ -2901,7 +2886,7 @@ static int init_ingress_acls_root_ns(struct mlx5_core_dev *dev)
 		if (err)
 			goto cleanup_root_ns;
 	}
-
+	steering->esw_ingress_acl_vports = total_vports;
 	return 0;
 
 cleanup_root_ns:
@@ -2912,6 +2897,21 @@ static int init_ingress_acls_root_ns(struct mlx5_core_dev *dev)
 	return err;
 }
 
+void mlx5_fs_ingress_acls_cleanup(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_steering *steering = dev->priv.steering;
+	int i;
+
+	if (!steering->esw_ingress_root_ns)
+		return;
+
+	for (i = 0; i < steering->esw_ingress_acl_vports; i++)
+		cleanup_root_ns(steering->esw_ingress_root_ns[i]);
+
+	kfree(steering->esw_ingress_root_ns);
+	steering->esw_ingress_root_ns = NULL;
+}
+
 static int init_egress_root_ns(struct mlx5_flow_steering *steering)
 {
 	int err;
@@ -2974,16 +2974,6 @@ int mlx5_init_fs(struct mlx5_core_dev *dev)
 			if (err)
 				goto err;
 		}
-		if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) {
-			err = init_egress_acls_root_ns(dev);
-			if (err)
-				goto err;
-		}
-		if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) {
-			err = init_ingress_acls_root_ns(dev);
-			if (err)
-				goto err;
-		}
 	}
 
 	if (MLX5_CAP_FLOWTABLE_SNIFFER_RX(dev, ft_support)) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index b24a9849c45e..e577a2c424af 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -129,6 +129,8 @@ struct mlx5_flow_steering {
 	struct mlx5_flow_root_namespace	*rdma_rx_root_ns;
 	struct mlx5_flow_root_namespace	*rdma_tx_root_ns;
 	struct mlx5_flow_root_namespace	*egress_root_ns;
+	int esw_egress_acl_vports;
+	int esw_ingress_acl_vports;
 };
 
 struct fs_node {
@@ -287,6 +289,11 @@ int mlx5_flow_namespace_set_mode(struct mlx5_flow_namespace *ns,
 int mlx5_init_fs(struct mlx5_core_dev *dev);
 void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
 
+int mlx5_fs_egress_acls_init(struct mlx5_core_dev *dev, int total_vports);
+void mlx5_fs_egress_acls_cleanup(struct mlx5_core_dev *dev);
+int mlx5_fs_ingress_acls_init(struct mlx5_core_dev *dev, int total_vports);
+void mlx5_fs_ingress_acls_cleanup(struct mlx5_core_dev *dev);
+
 #define fs_get_obj(v, _node)  {v = container_of((_node), typeof(*v), node); }
 
 #define fs_list_for_each_entry(pos, root)		\
-- 
2.30.2

[net-next V2 08/16] net/mlx5: SF, Use device pointer directly

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

At many places in the code, device pointer is directly available. Make
use of it, instead of accessing it from the table.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/sf/hw_table.c    | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c
index c9bddde04047..ec53c11c8344 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c
@@ -67,8 +67,8 @@ int mlx5_sf_hw_table_sf_alloc(struct mlx5_core_dev *dev, u32 usr_sfnum)
 		goto exist_err;
 	}
 
-	hw_fn_id = mlx5_sf_sw_to_hw_id(table->dev, sw_id);
-	err = mlx5_cmd_alloc_sf(table->dev, hw_fn_id);
+	hw_fn_id = mlx5_sf_sw_to_hw_id(dev, sw_id);
+	err = mlx5_cmd_alloc_sf(dev, hw_fn_id);
 	if (err)
 		goto err;
 
@@ -80,7 +80,7 @@ int mlx5_sf_hw_table_sf_alloc(struct mlx5_core_dev *dev, u32 usr_sfnum)
 	return sw_id;
 
 vhca_err:
-	mlx5_cmd_dealloc_sf(table->dev, hw_fn_id);
+	mlx5_cmd_dealloc_sf(dev, hw_fn_id);
 err:
 	table->sfs[i].allocated = false;
 exist_err:
@@ -93,8 +93,8 @@ static void _mlx5_sf_hw_id_free(struct mlx5_core_dev *dev, u16 id)
 	struct mlx5_sf_hw_table *table = dev->priv.sf_hw_table;
 	u16 hw_fn_id;
 
-	hw_fn_id = mlx5_sf_sw_to_hw_id(table->dev, id);
-	mlx5_cmd_dealloc_sf(table->dev, hw_fn_id);
+	hw_fn_id = mlx5_sf_sw_to_hw_id(dev, id);
+	mlx5_cmd_dealloc_sf(dev, hw_fn_id);
 	table->sfs[id].allocated = false;
 	table->sfs[id].pending_delete = false;
 }
@@ -123,7 +123,7 @@ void mlx5_sf_hw_table_sf_deferred_free(struct mlx5_core_dev *dev, u16 id)
 		goto err;
 	state = MLX5_GET(query_vhca_state_out, out, vhca_state_context.vhca_state);
 	if (state == MLX5_VHCA_STATE_ALLOCATED) {
-		mlx5_cmd_dealloc_sf(table->dev, hw_fn_id);
+		mlx5_cmd_dealloc_sf(dev, hw_fn_id);
 		table->sfs[id].allocated = false;
 	} else {
 		table->sfs[id].pending_delete = true;
@@ -216,7 +216,7 @@ int mlx5_sf_hw_table_create(struct mlx5_core_dev *dev)
 		return 0;
 
 	table->vhca_nb.notifier_call = mlx5_sf_hw_vhca_event;
-	return mlx5_vhca_event_notifier_register(table->dev, &table->vhca_nb);
+	return mlx5_vhca_event_notifier_register(dev, &table->vhca_nb);
 }
 
 void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev)
@@ -226,7 +226,7 @@ void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev)
 	if (!table)
 		return;
 
-	mlx5_vhca_event_notifier_unregister(table->dev, &table->vhca_nb);
+	mlx5_vhca_event_notifier_unregister(dev, &table->vhca_nb);
 	/* Dealloc SFs whose firmware event has been missed. */
 	mlx5_sf_hw_dealloc_all(table);
 }
-- 
2.30.2

[net-next V2 09/16] net/mlx5: SF, Reuse stored hardware function id

[Saeed Mahameed]

From: Parav Pandit <parav@nvidia.com>

SF's hardware function id is already stored in mlx5_sf. Reuse it,
instead of querying the hw table.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index 60a6328a9ca0..52226d9b9a6d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -270,15 +270,14 @@ static int mlx5_sf_add(struct mlx5_core_dev *dev, struct mlx5_sf_table *table,
 {
 	struct mlx5_eswitch *esw = dev->priv.eswitch;
 	struct mlx5_sf *sf;
-	u16 hw_fn_id;
 	int err;
 
 	sf = mlx5_sf_alloc(table, new_attr->sfnum, extack);
 	if (IS_ERR(sf))
 		return PTR_ERR(sf);
 
-	hw_fn_id = mlx5_sf_sw_to_hw_id(dev, sf->id);
-	err = mlx5_esw_offloads_sf_vport_enable(esw, &sf->dl_port, hw_fn_id, new_attr->sfnum);
+	err = mlx5_esw_offloads_sf_vport_enable(esw, &sf->dl_port, sf->hw_fn_id,
+						new_attr->sfnum);
 	if (err)
 		goto esw_err;
 	*new_port_index = sf->port_index;
-- 
2.30.2

[net-next V2 10/16] net/mlx5: DR, Use variably sized data structures for different actions

[Saeed Mahameed]

From: Jianbo Liu <jianbol@nvidia.com>

mlx5dr_action is a generally used data structure, and there is an
union for different types of actions in it. The size of mlx5dr_action
is about 72 bytes, but for those actions with fewer fields, most of
the allocated memory is wasted.
Remove this union, and mlx5dr_action becomes a generic action header.
Then actions are dynamically allocated with needed memory, the data
for each action is stored right after the header.

Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c   | 242 ++++++++++--------
 .../mellanox/mlx5/core/steering/dr_send.c     |   8 +-
 .../mellanox/mlx5/core/steering/dr_table.c    |   4 +-
 .../mellanox/mlx5/core/steering/dr_types.h    | 104 ++++----
 4 files changed, 199 insertions(+), 159 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 28a7971cac6a..949879cf2092 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -313,8 +313,8 @@ static int dr_action_handle_cs_recalc(struct mlx5dr_domain *dmn,
 		 * table, since there is an *assumption* that in such case FW
 		 * will recalculate the CS.
 		 */
-		if (dest_action->dest_tbl.is_fw_tbl) {
-			*final_icm_addr = dest_action->dest_tbl.fw_tbl.rx_icm_addr;
+		if (dest_action->dest_tbl->is_fw_tbl) {
+			*final_icm_addr = dest_action->dest_tbl->fw_tbl.rx_icm_addr;
 		} else {
 			mlx5dr_dbg(dmn,
 				   "Destination FT should be terminating when modify TTL is used\n");
@@ -326,8 +326,8 @@ static int dr_action_handle_cs_recalc(struct mlx5dr_domain *dmn,
 		/* If destination is vport we will get the FW flow table
 		 * that recalculates the CS and forwards to the vport.
 		 */
-		ret = mlx5dr_domain_cache_get_recalc_cs_ft_addr(dest_action->vport.dmn,
-								dest_action->vport.caps->num,
+		ret = mlx5dr_domain_cache_get_recalc_cs_ft_addr(dest_action->vport->dmn,
+								dest_action->vport->caps->num,
 								final_icm_addr);
 		if (ret) {
 			mlx5dr_err(dmn, "Failed to get FW cs recalc flow table\n");
@@ -369,6 +369,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 	action_domain = dr_action_get_action_domain(dmn->type, nic_dmn->ste_type);
 
 	for (i = 0; i < num_actions; i++) {
+		struct mlx5dr_action_dest_tbl *dest_tbl;
 		struct mlx5dr_action *action;
 		int max_actions_type = 1;
 		u32 action_type;
@@ -382,37 +383,38 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 			break;
 		case DR_ACTION_TYP_FT:
 			dest_action = action;
-			if (!action->dest_tbl.is_fw_tbl) {
-				if (action->dest_tbl.tbl->dmn != dmn) {
+			dest_tbl = action->dest_tbl;
+			if (!dest_tbl->is_fw_tbl) {
+				if (dest_tbl->tbl->dmn != dmn) {
 					mlx5dr_err(dmn,
 						   "Destination table belongs to a different domain\n");
 					goto out_invalid_arg;
 				}
-				if (action->dest_tbl.tbl->level <= matcher->tbl->level) {
+				if (dest_tbl->tbl->level <= matcher->tbl->level) {
 					mlx5_core_warn_once(dmn->mdev,
 							    "Connecting table to a lower/same level destination table\n");
 					mlx5dr_dbg(dmn,
 						   "Connecting table at level %d to a destination table at level %d\n",
 						   matcher->tbl->level,
-						   action->dest_tbl.tbl->level);
+						   dest_tbl->tbl->level);
 				}
 				attr.final_icm_addr = rx_rule ?
-					action->dest_tbl.tbl->rx.s_anchor->chunk->icm_addr :
-					action->dest_tbl.tbl->tx.s_anchor->chunk->icm_addr;
+					dest_tbl->tbl->rx.s_anchor->chunk->icm_addr :
+					dest_tbl->tbl->tx.s_anchor->chunk->icm_addr;
 			} else {
 				struct mlx5dr_cmd_query_flow_table_details output;
 				int ret;
 
 				/* get the relevant addresses */
-				if (!action->dest_tbl.fw_tbl.rx_icm_addr) {
+				if (!action->dest_tbl->fw_tbl.rx_icm_addr) {
 					ret = mlx5dr_cmd_query_flow_table(dmn->mdev,
-									  action->dest_tbl.fw_tbl.type,
-									  action->dest_tbl.fw_tbl.id,
+									  dest_tbl->fw_tbl.type,
+									  dest_tbl->fw_tbl.id,
 									  &output);
 					if (!ret) {
-						action->dest_tbl.fw_tbl.tx_icm_addr =
+						dest_tbl->fw_tbl.tx_icm_addr =
 							output.sw_owner_icm_root_1;
-						action->dest_tbl.fw_tbl.rx_icm_addr =
+						dest_tbl->fw_tbl.rx_icm_addr =
 							output.sw_owner_icm_root_0;
 					} else {
 						mlx5dr_err(dmn,
@@ -422,50 +424,50 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 					}
 				}
 				attr.final_icm_addr = rx_rule ?
-					action->dest_tbl.fw_tbl.rx_icm_addr :
-					action->dest_tbl.fw_tbl.tx_icm_addr;
+					dest_tbl->fw_tbl.rx_icm_addr :
+					dest_tbl->fw_tbl.tx_icm_addr;
 			}
 			break;
 		case DR_ACTION_TYP_QP:
 			mlx5dr_info(dmn, "Domain doesn't support QP\n");
 			goto out_invalid_arg;
 		case DR_ACTION_TYP_CTR:
-			attr.ctr_id = action->ctr.ctr_id +
-				action->ctr.offeset;
+			attr.ctr_id = action->ctr->ctr_id +
+				action->ctr->offeset;
 			break;
 		case DR_ACTION_TYP_TAG:
-			attr.flow_tag = action->flow_tag;
+			attr.flow_tag = action->flow_tag->flow_tag;
 			break;
 		case DR_ACTION_TYP_TNL_L2_TO_L2:
 			break;
 		case DR_ACTION_TYP_TNL_L3_TO_L2:
-			attr.decap_index = action->rewrite.index;
-			attr.decap_actions = action->rewrite.num_of_actions;
+			attr.decap_index = action->rewrite->index;
+			attr.decap_actions = action->rewrite->num_of_actions;
 			attr.decap_with_vlan =
 				attr.decap_actions == WITH_VLAN_NUM_HW_ACTIONS;
 			break;
 		case DR_ACTION_TYP_MODIFY_HDR:
-			attr.modify_index = action->rewrite.index;
-			attr.modify_actions = action->rewrite.num_of_actions;
-			recalc_cs_required = action->rewrite.modify_ttl &&
+			attr.modify_index = action->rewrite->index;
+			attr.modify_actions = action->rewrite->num_of_actions;
+			recalc_cs_required = action->rewrite->modify_ttl &&
 					     !mlx5dr_ste_supp_ttl_cs_recalc(&dmn->info.caps);
 			break;
 		case DR_ACTION_TYP_L2_TO_TNL_L2:
 		case DR_ACTION_TYP_L2_TO_TNL_L3:
-			attr.reformat_size = action->reformat.reformat_size;
-			attr.reformat_id = action->reformat.reformat_id;
+			attr.reformat_size = action->reformat->reformat_size;
+			attr.reformat_id = action->reformat->reformat_id;
 			break;
 		case DR_ACTION_TYP_VPORT:
-			attr.hit_gvmi = action->vport.caps->vhca_gvmi;
+			attr.hit_gvmi = action->vport->caps->vhca_gvmi;
 			dest_action = action;
 			if (rx_rule) {
 				/* Loopback on WIRE vport is not supported */
-				if (action->vport.caps->num == WIRE_PORT)
+				if (action->vport->caps->num == WIRE_PORT)
 					goto out_invalid_arg;
 
-				attr.final_icm_addr = action->vport.caps->icm_address_rx;
+				attr.final_icm_addr = action->vport->caps->icm_address_rx;
 			} else {
-				attr.final_icm_addr = action->vport.caps->icm_address_tx;
+				attr.final_icm_addr = action->vport->caps->icm_address_tx;
 			}
 			break;
 		case DR_ACTION_TYP_POP_VLAN:
@@ -477,7 +479,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 			if (attr.vlans.count == MLX5DR_MAX_VLANS)
 				return -EINVAL;
 
-			attr.vlans.headers[attr.vlans.count++] = action->push_vlan.vlan_hdr;
+			attr.vlans.headers[attr.vlans.count++] = action->push_vlan->vlan_hdr;
 			break;
 		default:
 			goto out_invalid_arg;
@@ -530,17 +532,37 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 	return -EINVAL;
 }
 
+static unsigned int action_size[DR_ACTION_TYP_MAX] = {
+	[DR_ACTION_TYP_TNL_L2_TO_L2] = sizeof(struct mlx5dr_action_reformat),
+	[DR_ACTION_TYP_L2_TO_TNL_L2] = sizeof(struct mlx5dr_action_reformat),
+	[DR_ACTION_TYP_TNL_L3_TO_L2] = sizeof(struct mlx5dr_action_rewrite),
+	[DR_ACTION_TYP_L2_TO_TNL_L3] = sizeof(struct mlx5dr_action_reformat),
+	[DR_ACTION_TYP_FT]           = sizeof(struct mlx5dr_action_dest_tbl),
+	[DR_ACTION_TYP_CTR]          = sizeof(struct mlx5dr_action_ctr),
+	[DR_ACTION_TYP_TAG]          = sizeof(struct mlx5dr_action_flow_tag),
+	[DR_ACTION_TYP_MODIFY_HDR]   = sizeof(struct mlx5dr_action_rewrite),
+	[DR_ACTION_TYP_VPORT]        = sizeof(struct mlx5dr_action_vport),
+	[DR_ACTION_TYP_PUSH_VLAN]    = sizeof(struct mlx5dr_action_push_vlan),
+};
+
 static struct mlx5dr_action *
 dr_action_create_generic(enum mlx5dr_action_type action_type)
 {
 	struct mlx5dr_action *action;
+	int extra_size;
+
+	if (action_type < DR_ACTION_TYP_MAX)
+		extra_size = action_size[action_type];
+	else
+		return NULL;
 
-	action = kzalloc(sizeof(*action), GFP_KERNEL);
+	action = kzalloc(sizeof(*action) + extra_size, GFP_KERNEL);
 	if (!action)
 		return NULL;
 
 	action->action_type = action_type;
 	refcount_set(&action->refcount, 1);
+	action->data = action + 1;
 
 	return action;
 }
@@ -559,10 +581,10 @@ mlx5dr_action_create_dest_table_num(struct mlx5dr_domain *dmn, u32 table_num)
 	if (!action)
 		return NULL;
 
-	action->dest_tbl.is_fw_tbl = true;
-	action->dest_tbl.fw_tbl.dmn = dmn;
-	action->dest_tbl.fw_tbl.id = table_num;
-	action->dest_tbl.fw_tbl.type = FS_FT_FDB;
+	action->dest_tbl->is_fw_tbl = true;
+	action->dest_tbl->fw_tbl.dmn = dmn;
+	action->dest_tbl->fw_tbl.id = table_num;
+	action->dest_tbl->fw_tbl.type = FS_FT_FDB;
 	refcount_inc(&dmn->refcount);
 
 	return action;
@@ -579,7 +601,7 @@ mlx5dr_action_create_dest_table(struct mlx5dr_table *tbl)
 	if (!action)
 		goto dec_ref;
 
-	action->dest_tbl.tbl = tbl;
+	action->dest_tbl->tbl = tbl;
 
 	return action;
 
@@ -624,12 +646,12 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 		case DR_ACTION_TYP_VPORT:
 			hw_dests[i].vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;
 			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
-			hw_dests[i].vport.num = dest_action->vport.caps->num;
-			hw_dests[i].vport.vhca_id = dest_action->vport.caps->vhca_gvmi;
+			hw_dests[i].vport.num = dest_action->vport->caps->num;
+			hw_dests[i].vport.vhca_id = dest_action->vport->caps->vhca_gvmi;
 			if (reformat_action) {
 				reformat_req = true;
 				hw_dests[i].vport.reformat_id =
-					reformat_action->reformat.reformat_id;
+					reformat_action->reformat->reformat_id;
 				ref_actions[num_of_ref++] = reformat_action;
 				hw_dests[i].vport.flags |= MLX5_FLOW_DEST_VPORT_REFORMAT_ID;
 			}
@@ -637,10 +659,10 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 
 		case DR_ACTION_TYP_FT:
 			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-			if (dest_action->dest_tbl.is_fw_tbl)
-				hw_dests[i].ft_id = dest_action->dest_tbl.fw_tbl.id;
+			if (dest_action->dest_tbl->is_fw_tbl)
+				hw_dests[i].ft_id = dest_action->dest_tbl->fw_tbl.id;
 			else
-				hw_dests[i].ft_id = dest_action->dest_tbl.tbl->table_id;
+				hw_dests[i].ft_id = dest_action->dest_tbl->tbl->table_id;
 			break;
 
 		default:
@@ -657,8 +679,8 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 				      hw_dests,
 				      num_of_dests,
 				      reformat_req,
-				      &action->dest_tbl.fw_tbl.id,
-				      &action->dest_tbl.fw_tbl.group_id);
+				      &action->dest_tbl->fw_tbl.id,
+				      &action->dest_tbl->fw_tbl.group_id);
 	if (ret)
 		goto free_action;
 
@@ -667,11 +689,11 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 	for (i = 0; i < num_of_ref; i++)
 		refcount_inc(&ref_actions[i]->refcount);
 
-	action->dest_tbl.is_fw_tbl = true;
-	action->dest_tbl.fw_tbl.dmn = dmn;
-	action->dest_tbl.fw_tbl.type = FS_FT_FDB;
-	action->dest_tbl.fw_tbl.ref_actions = ref_actions;
-	action->dest_tbl.fw_tbl.num_of_ref_actions = num_of_ref;
+	action->dest_tbl->is_fw_tbl = true;
+	action->dest_tbl->fw_tbl.dmn = dmn;
+	action->dest_tbl->fw_tbl.type = FS_FT_FDB;
+	action->dest_tbl->fw_tbl.ref_actions = ref_actions;
+	action->dest_tbl->fw_tbl.num_of_ref_actions = num_of_ref;
 
 	kfree(hw_dests);
 
@@ -696,10 +718,10 @@ mlx5dr_action_create_dest_flow_fw_table(struct mlx5dr_domain *dmn,
 	if (!action)
 		return NULL;
 
-	action->dest_tbl.is_fw_tbl = 1;
-	action->dest_tbl.fw_tbl.type = ft->type;
-	action->dest_tbl.fw_tbl.id = ft->id;
-	action->dest_tbl.fw_tbl.dmn = dmn;
+	action->dest_tbl->is_fw_tbl = 1;
+	action->dest_tbl->fw_tbl.type = ft->type;
+	action->dest_tbl->fw_tbl.id = ft->id;
+	action->dest_tbl->fw_tbl.dmn = dmn;
 
 	refcount_inc(&dmn->refcount);
 
@@ -715,7 +737,7 @@ mlx5dr_action_create_flow_counter(u32 counter_id)
 	if (!action)
 		return NULL;
 
-	action->ctr.ctr_id = counter_id;
+	action->ctr->ctr_id = counter_id;
 
 	return action;
 }
@@ -728,7 +750,7 @@ struct mlx5dr_action *mlx5dr_action_create_tag(u32 tag_value)
 	if (!action)
 		return NULL;
 
-	action->flow_tag = tag_value & 0xffffff;
+	action->flow_tag->flow_tag = tag_value & 0xffffff;
 
 	return action;
 }
@@ -794,8 +816,8 @@ dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
 		if (ret)
 			return ret;
 
-		action->reformat.reformat_id = reformat_id;
-		action->reformat.reformat_size = data_sz;
+		action->reformat->reformat_id = reformat_id;
+		action->reformat->reformat_size = data_sz;
 		return 0;
 	}
 	case DR_ACTION_TYP_TNL_L2_TO_L2:
@@ -811,28 +833,28 @@ dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
 							  data, data_sz,
 							  hw_actions,
 							  ACTION_CACHE_LINE_SIZE,
-							  &action->rewrite.num_of_actions);
+							  &action->rewrite->num_of_actions);
 		if (ret) {
 			mlx5dr_dbg(dmn, "Failed creating decap l3 action list\n");
 			return ret;
 		}
 
-		action->rewrite.chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool,
-							       DR_CHUNK_SIZE_8);
-		if (!action->rewrite.chunk) {
+		action->rewrite->chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool,
+								DR_CHUNK_SIZE_8);
+		if (!action->rewrite->chunk) {
 			mlx5dr_dbg(dmn, "Failed allocating modify header chunk\n");
 			return -ENOMEM;
 		}
 
-		action->rewrite.data = (void *)hw_actions;
-		action->rewrite.index = (action->rewrite.chunk->icm_addr -
+		action->rewrite->data = (void *)hw_actions;
+		action->rewrite->index = (action->rewrite->chunk->icm_addr -
 					 dmn->info.caps.hdr_modify_icm_addr) /
 					 ACTION_CACHE_LINE_SIZE;
 
 		ret = mlx5dr_send_postsend_action(dmn, action);
 		if (ret) {
 			mlx5dr_dbg(dmn, "Writing decap l3 actions to ICM failed\n");
-			mlx5dr_icm_free_chunk(action->rewrite.chunk);
+			mlx5dr_icm_free_chunk(action->rewrite->chunk);
 			return ret;
 		}
 		return 0;
@@ -867,7 +889,7 @@ struct mlx5dr_action *mlx5dr_action_create_push_vlan(struct mlx5dr_domain *dmn,
 	if (!action)
 		return NULL;
 
-	action->push_vlan.vlan_hdr = vlan_hdr_h;
+	action->push_vlan->vlan_hdr = vlan_hdr_h;
 	return action;
 }
 
@@ -898,7 +920,7 @@ mlx5dr_action_create_packet_reformat(struct mlx5dr_domain *dmn,
 	if (!action)
 		goto dec_ref;
 
-	action->reformat.dmn = dmn;
+	action->reformat->dmn = dmn;
 
 	ret = dr_action_create_reformat_action(dmn,
 					       data_sz,
@@ -1104,17 +1126,17 @@ dr_action_modify_check_set_field_limitation(struct mlx5dr_action *action,
 					    const __be64 *sw_action)
 {
 	u16 sw_field = MLX5_GET(set_action_in, sw_action, field);
-	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
 
 	if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_A) {
-		action->rewrite.allow_rx = 0;
+		action->rewrite->allow_rx = 0;
 		if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_TX) {
 			mlx5dr_dbg(dmn, "Unsupported field %d for RX/FDB set action\n",
 				   sw_field);
 			return -EINVAL;
 		}
 	} else if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_B) {
-		action->rewrite.allow_tx = 0;
+		action->rewrite->allow_tx = 0;
 		if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_RX) {
 			mlx5dr_dbg(dmn, "Unsupported field %d for TX/FDB set action\n",
 				   sw_field);
@@ -1122,7 +1144,7 @@ dr_action_modify_check_set_field_limitation(struct mlx5dr_action *action,
 		}
 	}
 
-	if (!action->rewrite.allow_rx && !action->rewrite.allow_tx) {
+	if (!action->rewrite->allow_rx && !action->rewrite->allow_tx) {
 		mlx5dr_dbg(dmn, "Modify SET actions not supported on both RX and TX\n");
 		return -EINVAL;
 	}
@@ -1135,7 +1157,7 @@ dr_action_modify_check_add_field_limitation(struct mlx5dr_action *action,
 					    const __be64 *sw_action)
 {
 	u16 sw_field = MLX5_GET(set_action_in, sw_action, field);
-	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
 
 	if (sw_field != MLX5_ACTION_IN_FIELD_OUT_IP_TTL &&
 	    sw_field != MLX5_ACTION_IN_FIELD_OUT_IPV6_HOPLIMIT &&
@@ -1153,7 +1175,7 @@ static int
 dr_action_modify_check_copy_field_limitation(struct mlx5dr_action *action,
 					     const __be64 *sw_action)
 {
-	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
 	u16 sw_fields[2];
 	int i;
 
@@ -1162,14 +1184,14 @@ dr_action_modify_check_copy_field_limitation(struct mlx5dr_action *action,
 
 	for (i = 0; i < 2; i++) {
 		if (sw_fields[i] == MLX5_ACTION_IN_FIELD_METADATA_REG_A) {
-			action->rewrite.allow_rx = 0;
+			action->rewrite->allow_rx = 0;
 			if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_TX) {
 				mlx5dr_dbg(dmn, "Unsupported field %d for RX/FDB set action\n",
 					   sw_fields[i]);
 				return -EINVAL;
 			}
 		} else if (sw_fields[i] == MLX5_ACTION_IN_FIELD_METADATA_REG_B) {
-			action->rewrite.allow_tx = 0;
+			action->rewrite->allow_tx = 0;
 			if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_RX) {
 				mlx5dr_dbg(dmn, "Unsupported field %d for TX/FDB set action\n",
 					   sw_fields[i]);
@@ -1178,7 +1200,7 @@ dr_action_modify_check_copy_field_limitation(struct mlx5dr_action *action,
 		}
 	}
 
-	if (!action->rewrite.allow_rx && !action->rewrite.allow_tx) {
+	if (!action->rewrite->allow_rx && !action->rewrite->allow_tx) {
 		mlx5dr_dbg(dmn, "Modify copy actions not supported on both RX and TX\n");
 		return -EINVAL;
 	}
@@ -1190,7 +1212,7 @@ static int
 dr_action_modify_check_field_limitation(struct mlx5dr_action *action,
 					const __be64 *sw_action)
 {
-	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
 	u8 action_type;
 	int ret;
 
@@ -1239,7 +1261,7 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 {
 	const struct mlx5dr_ste_action_modify_field *hw_dst_action_info;
 	const struct mlx5dr_ste_action_modify_field *hw_src_action_info;
-	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
 	int ret, i, hw_idx = 0;
 	__be64 *sw_action;
 	__be64 hw_action;
@@ -1249,8 +1271,8 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 
 	*modify_ttl = false;
 
-	action->rewrite.allow_rx = 1;
-	action->rewrite.allow_tx = 1;
+	action->rewrite->allow_rx = 1;
+	action->rewrite->allow_tx = 1;
 
 	for (i = 0; i < num_sw_actions; i++) {
 		sw_action = &sw_actions[i];
@@ -1358,13 +1380,13 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 	if (ret)
 		goto free_hw_actions;
 
-	action->rewrite.chunk = chunk;
-	action->rewrite.modify_ttl = modify_ttl;
-	action->rewrite.data = (u8 *)hw_actions;
-	action->rewrite.num_of_actions = num_hw_actions;
-	action->rewrite.index = (chunk->icm_addr -
-				 dmn->info.caps.hdr_modify_icm_addr) /
-				 ACTION_CACHE_LINE_SIZE;
+	action->rewrite->chunk = chunk;
+	action->rewrite->modify_ttl = modify_ttl;
+	action->rewrite->data = (u8 *)hw_actions;
+	action->rewrite->num_of_actions = num_hw_actions;
+	action->rewrite->index = (chunk->icm_addr -
+				  dmn->info.caps.hdr_modify_icm_addr) /
+				  ACTION_CACHE_LINE_SIZE;
 
 	ret = mlx5dr_send_postsend_action(dmn, action);
 	if (ret)
@@ -1399,7 +1421,7 @@ mlx5dr_action_create_modify_header(struct mlx5dr_domain *dmn,
 	if (!action)
 		goto dec_ref;
 
-	action->rewrite.dmn = dmn;
+	action->rewrite->dmn = dmn;
 
 	ret = dr_action_create_modify_action(dmn,
 					     actions_sz,
@@ -1451,8 +1473,8 @@ mlx5dr_action_create_dest_vport(struct mlx5dr_domain *dmn,
 	if (!action)
 		return NULL;
 
-	action->vport.dmn = vport_dmn;
-	action->vport.caps = vport_cap;
+	action->vport->dmn = vport_dmn;
+	action->vport->caps = vport_cap;
 
 	return action;
 }
@@ -1464,44 +1486,44 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 
 	switch (action->action_type) {
 	case DR_ACTION_TYP_FT:
-		if (action->dest_tbl.is_fw_tbl)
-			refcount_dec(&action->dest_tbl.fw_tbl.dmn->refcount);
+		if (action->dest_tbl->is_fw_tbl)
+			refcount_dec(&action->dest_tbl->fw_tbl.dmn->refcount);
 		else
-			refcount_dec(&action->dest_tbl.tbl->refcount);
+			refcount_dec(&action->dest_tbl->tbl->refcount);
 
-		if (action->dest_tbl.is_fw_tbl &&
-		    action->dest_tbl.fw_tbl.num_of_ref_actions) {
+		if (action->dest_tbl->is_fw_tbl &&
+		    action->dest_tbl->fw_tbl.num_of_ref_actions) {
 			struct mlx5dr_action **ref_actions;
 			int i;
 
-			ref_actions = action->dest_tbl.fw_tbl.ref_actions;
-			for (i = 0; i < action->dest_tbl.fw_tbl.num_of_ref_actions; i++)
+			ref_actions = action->dest_tbl->fw_tbl.ref_actions;
+			for (i = 0; i < action->dest_tbl->fw_tbl.num_of_ref_actions; i++)
 				refcount_dec(&ref_actions[i]->refcount);
 
 			kfree(ref_actions);
 
-			mlx5dr_fw_destroy_md_tbl(action->dest_tbl.fw_tbl.dmn,
-						 action->dest_tbl.fw_tbl.id,
-						 action->dest_tbl.fw_tbl.group_id);
+			mlx5dr_fw_destroy_md_tbl(action->dest_tbl->fw_tbl.dmn,
+						 action->dest_tbl->fw_tbl.id,
+						 action->dest_tbl->fw_tbl.group_id);
 		}
 		break;
 	case DR_ACTION_TYP_TNL_L2_TO_L2:
-		refcount_dec(&action->reformat.dmn->refcount);
+		refcount_dec(&action->reformat->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_TNL_L3_TO_L2:
-		mlx5dr_icm_free_chunk(action->rewrite.chunk);
-		refcount_dec(&action->reformat.dmn->refcount);
+		mlx5dr_icm_free_chunk(action->rewrite->chunk);
+		refcount_dec(&action->rewrite->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_L2_TO_TNL_L2:
 	case DR_ACTION_TYP_L2_TO_TNL_L3:
-		mlx5dr_cmd_destroy_reformat_ctx((action->reformat.dmn)->mdev,
-						action->reformat.reformat_id);
-		refcount_dec(&action->reformat.dmn->refcount);
+		mlx5dr_cmd_destroy_reformat_ctx((action->reformat->dmn)->mdev,
+						action->reformat->reformat_id);
+		refcount_dec(&action->reformat->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_MODIFY_HDR:
-		mlx5dr_icm_free_chunk(action->rewrite.chunk);
-		kfree(action->rewrite.data);
-		refcount_dec(&action->rewrite.dmn->refcount);
+		mlx5dr_icm_free_chunk(action->rewrite->chunk);
+		kfree(action->rewrite->data);
+		refcount_dec(&action->rewrite->dmn->refcount);
 		break;
 	default:
 		break;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index 8a6a56f9dc4e..24acced415d3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -572,12 +572,12 @@ int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn,
 	struct postsend_info send_info = {};
 	int ret;
 
-	send_info.write.addr = (uintptr_t)action->rewrite.data;
-	send_info.write.length = action->rewrite.num_of_actions *
+	send_info.write.addr = (uintptr_t)action->rewrite->data;
+	send_info.write.length = action->rewrite->num_of_actions *
 				 DR_MODIFY_ACTION_SIZE;
 	send_info.write.lkey = 0;
-	send_info.remote_addr = action->rewrite.chunk->mr_addr;
-	send_info.rkey = action->rewrite.chunk->rkey;
+	send_info.remote_addr = action->rewrite->chunk->mr_addr;
+	send_info.rkey = action->rewrite->chunk->rkey;
 
 	ret = dr_postsend_icm_data(dmn, &send_info);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
index b599b6beb5b9..30ae3cda6d2e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
@@ -29,7 +29,7 @@ int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl,
 			last_htbl = tbl->rx.s_anchor;
 
 		tbl->rx.default_icm_addr = action ?
-			action->dest_tbl.tbl->rx.s_anchor->chunk->icm_addr :
+			action->dest_tbl->tbl->rx.s_anchor->chunk->icm_addr :
 			tbl->rx.nic_dmn->default_icm_addr;
 
 		info.type = CONNECT_MISS;
@@ -53,7 +53,7 @@ int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl,
 			last_htbl = tbl->tx.s_anchor;
 
 		tbl->tx.default_icm_addr = action ?
-			action->dest_tbl.tbl->tx.s_anchor->chunk->icm_addr :
+			action->dest_tbl->tbl->tx.s_anchor->chunk->icm_addr :
 			tbl->tx.nic_dmn->default_icm_addr;
 
 		info.type = CONNECT_MISS;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 4af0e4e6a13c..462673947f3c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -806,53 +806,71 @@ struct mlx5dr_ste_action_modify_field {
 	u8 l4_type;
 };
 
+struct mlx5dr_action_rewrite {
+	struct mlx5dr_domain *dmn;
+	struct mlx5dr_icm_chunk *chunk;
+	u8 *data;
+	u16 num_of_actions;
+	u32 index;
+	u8 allow_rx:1;
+	u8 allow_tx:1;
+	u8 modify_ttl:1;
+};
+
+struct mlx5dr_action_reformat {
+	struct mlx5dr_domain *dmn;
+	u32 reformat_id;
+	u32 reformat_size;
+};
+
+struct mlx5dr_action_dest_tbl {
+	u8 is_fw_tbl:1;
+	union {
+		struct mlx5dr_table *tbl;
+		struct {
+			struct mlx5dr_domain *dmn;
+			u32 id;
+			u32 group_id;
+			enum fs_flow_table_type type;
+			u64 rx_icm_addr;
+			u64 tx_icm_addr;
+			struct mlx5dr_action **ref_actions;
+			u32 num_of_ref_actions;
+		} fw_tbl;
+	};
+};
+
+struct mlx5dr_action_ctr {
+	u32 ctr_id;
+	u32 offeset;
+};
+
+struct mlx5dr_action_vport {
+	struct mlx5dr_domain *dmn;
+	struct mlx5dr_cmd_vport_cap *caps;
+};
+
+struct mlx5dr_action_push_vlan {
+	u32 vlan_hdr; /* tpid_pcp_dei_vid */
+};
+
+struct mlx5dr_action_flow_tag {
+	u32 flow_tag;
+};
+
 struct mlx5dr_action {
 	enum mlx5dr_action_type action_type;
 	refcount_t refcount;
+
 	union {
-		struct {
-			struct mlx5dr_domain *dmn;
-			struct mlx5dr_icm_chunk *chunk;
-			u8 *data;
-			u16 num_of_actions;
-			u32 index;
-			u8 allow_rx:1;
-			u8 allow_tx:1;
-			u8 modify_ttl:1;
-		} rewrite;
-		struct {
-			struct mlx5dr_domain *dmn;
-			u32 reformat_id;
-			u32 reformat_size;
-		} reformat;
-		struct {
-			u8 is_fw_tbl:1;
-			union {
-				struct mlx5dr_table *tbl;
-				struct {
-					struct mlx5dr_domain *dmn;
-					u32 id;
-					u32 group_id;
-					enum fs_flow_table_type type;
-					u64 rx_icm_addr;
-					u64 tx_icm_addr;
-					struct mlx5dr_action **ref_actions;
-					u32 num_of_ref_actions;
-				} fw_tbl;
-			};
-		} dest_tbl;
-		struct {
-			u32 ctr_id;
-			u32 offeset;
-		} ctr;
-		struct {
-			struct mlx5dr_domain *dmn;
-			struct mlx5dr_cmd_vport_cap *caps;
-		} vport;
-		struct {
-			u32 vlan_hdr; /* tpid_pcp_dei_vid */
-		} push_vlan;
-		u32 flow_tag;
+		void *data;
+		struct mlx5dr_action_rewrite *rewrite;
+		struct mlx5dr_action_reformat *reformat;
+		struct mlx5dr_action_dest_tbl *dest_tbl;
+		struct mlx5dr_action_ctr *ctr;
+		struct mlx5dr_action_vport *vport;
+		struct mlx5dr_action_push_vlan *push_vlan;
+		struct mlx5dr_action_flow_tag *flow_tag;
 	};
 };
 
-- 
2.30.2

[net-next V2 11/16] net/mlx5: DR, Alloc cmd buffer with kvzalloc() instead of kzalloc()

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

The cmd size is 8K so use kvzalloc().

Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c  | 4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index 30b0136b5bc7..461473d31e2e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -287,7 +287,7 @@ int mlx5dr_cmd_create_empty_flow_group(struct mlx5_core_dev *mdev,
 	u32 *in;
 	int err;
 
-	in = kzalloc(inlen, GFP_KERNEL);
+	in = kvzalloc(inlen, GFP_KERNEL);
 	if (!in)
 		return -ENOMEM;
 
@@ -302,7 +302,7 @@ int mlx5dr_cmd_create_empty_flow_group(struct mlx5_core_dev *mdev,
 	*group_id = MLX5_GET(create_flow_group_out, out, group_id);
 
 out:
-	kfree(in);
+	kvfree(in);
 	return err;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index 24acced415d3..c1926d927008 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -406,7 +406,7 @@ static int dr_get_tbl_copy_details(struct mlx5dr_domain *dmn,
 		alloc_size = *num_stes * DR_STE_SIZE;
 	}
 
-	*data = kzalloc(alloc_size, GFP_KERNEL);
+	*data = kvzalloc(alloc_size, GFP_KERNEL);
 	if (!*data)
 		return -ENOMEM;
 
@@ -505,7 +505,7 @@ int mlx5dr_send_postsend_htbl(struct mlx5dr_domain *dmn,
 	}
 
 out_free:
-	kfree(data);
+	kvfree(data);
 	return ret;
 }
 
@@ -562,7 +562,7 @@ int mlx5dr_send_postsend_formatted_htbl(struct mlx5dr_domain *dmn,
 	}
 
 out_free:
-	kfree(data);
+	kvfree(data);
 	return ret;
 }
 
-- 
2.30.2

[net-next V2 12/16] net/mlx5: Fix bit-wise and with zero

[Saeed Mahameed]

From: Colin Ian King <colin.king@canonical.com>

The bit-wise and of the action field with MLX5_ACCEL_ESP_ACTION_DECRYPT
is incorrect as MLX5_ACCEL_ESP_ACTION_DECRYPT is zero and not intended
to be a bit-flag. Fix this by using the == operator as was originally
intended.

Addresses-Coverity: ("Logically dead code")
Fixes: 7dfee4b1d79e ("net/mlx5: IPsec, Refactor SA handle creation and destruction")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
index d43a05e77f67..0b19293cdd74 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
@@ -850,7 +850,7 @@ mlx5_fpga_ipsec_release_sa_ctx(struct mlx5_fpga_ipsec_sa_ctx *sa_ctx)
 		return;
 	}
 
-	if (sa_ctx->fpga_xfrm->accel_xfrm.attrs.action &
+	if (sa_ctx->fpga_xfrm->accel_xfrm.attrs.action ==
 	    MLX5_ACCEL_ESP_ACTION_DECRYPT)
 		ida_free(&fipsec->halloc, sa_ctx->sa_handle);
 
-- 
2.30.2

[net-next V2 13/16] net/mlx5: Add a blank line after declarations

[Saeed Mahameed]

From: Wenpeng Liang <liangwenpeng@huawei.com>

There should be a blank lines after declarations.

Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c    | 4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c
index d5b1eb74d5e5..5cd466ec6492 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c
@@ -392,11 +392,11 @@ static void arfs_may_expire_flow(struct mlx5e_priv *priv)
 {
 	struct arfs_rule *arfs_rule;
 	struct hlist_node *htmp;
+	HLIST_HEAD(del_list);
 	int quota = 0;
 	int i;
 	int j;
 
-	HLIST_HEAD(del_list);
 	spin_lock_bh(&priv->fs.arfs->arfs_lock);
 	mlx5e_for_each_arfs_rule(arfs_rule, htmp, priv->fs.arfs->arfs_tables, i, j) {
 		if (!work_pending(&arfs_rule->arfs_work) &&
@@ -422,10 +422,10 @@ static void arfs_del_rules(struct mlx5e_priv *priv)
 {
 	struct hlist_node *htmp;
 	struct arfs_rule *rule;
+	HLIST_HEAD(del_list);
 	int i;
 	int j;
 
-	HLIST_HEAD(del_list);
 	spin_lock_bh(&priv->fs.arfs->arfs_lock);
 	mlx5e_for_each_arfs_rule(rule, htmp, priv->fs.arfs->arfs_tables, i, j) {
 		hlist_del_init(&rule->hlist);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
index 0b19293cdd74..0bba92cf5dc0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
@@ -1085,6 +1085,7 @@ static int fpga_ipsec_fs_create_fte(struct mlx5_flow_root_namespace *ns,
 	rule->ctx = mlx5_fpga_ipsec_fs_create_sa_ctx(dev, fte, is_egress);
 	if (IS_ERR(rule->ctx)) {
 		int err = PTR_ERR(rule->ctx);
+
 		kfree(rule);
 		return err;
 	}
-- 
2.30.2

[net-next V2 14/16] net/mlx5: Remove return statement exist at the end of void function

[Saeed Mahameed]

From: Wenpeng Liang <liangwenpeng@huawei.com>

void function return statements are not generally useful.

Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/lag.c     | 2 --
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 1 -
 drivers/net/ethernet/mellanox/mlx5/core/rdma.c    | 1 -
 3 files changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
index 127bb92da150..b8748390335f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
@@ -603,8 +603,6 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev)
 	if (err)
 		mlx5_core_err(dev, "Failed to init multipath lag err=%d\n",
 			      err);
-
-	return;
 }
 
 /* Must be called with intf_mutex held */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 19e3e978267e..1f907df5b3a2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -167,7 +167,6 @@ static void irq_set_name(char *name, int vecidx)
 
 	snprintf(name, MLX5_MAX_IRQ_NAME, "mlx5_comp%d",
 		 vecidx - MLX5_IRQ_VEC_COMP_BASE);
-	return;
 }
 
 static int request_irqs(struct mlx5_core_dev *dev, int nvec)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rdma.c b/drivers/net/ethernet/mellanox/mlx5/core/rdma.c
index 8e0dddc6383f..441b5453acae 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/rdma.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/rdma.c
@@ -180,5 +180,4 @@ void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev)
 	mlx5_rdma_del_roce_addr(dev);
 disable_roce:
 	mlx5_nic_vport_disable_roce(dev);
-	return;
 }
-- 
2.30.2

[net-next V2 15/16] net/mlx5: Replace spaces with tab at the start of a line

[Saeed Mahameed]

From: Wenpeng Liang <liangwenpeng@huawei.com>

There should be no spaces at the start of the line.

Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index e58ef8c713e4..34eb1118670f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -52,7 +52,7 @@
 #include "diag/en_rep_tracepoint.h"
 
 #define MLX5E_REP_PARAMS_DEF_LOG_SQ_SIZE \
-        max(0x7, MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE)
+	max(0x7, MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE)
 #define MLX5E_REP_PARAMS_DEF_NUM_CHANNELS 1
 
 static const char mlx5e_rep_driver_name[] = "mlx5e_rep";
-- 
2.30.2

[net-next V2 16/16] net/mlx5e: Fix RQ creation flow for queues which doesn't support XDP

[Saeed Mahameed]

From: Aya Levin <ayal@nvidia.com>

Allow to create an RQ which is not registered as an XDP RQ. For example:
the trap-RQ doesn't register as an XDP RQ.

Fixes: 869c5f926247 ("net/mlx5e: Generalize open RQ")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 2f47608bb9b9..6847e7b909a5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -510,8 +510,9 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 			rq->page_pool = NULL;
 			goto err_free_by_rq_type;
 		}
-		err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq,
-						 MEM_TYPE_PAGE_POOL, rq->page_pool);
+		if (xdp_rxq_info_is_reg(&rq->xdp_rxq))
+			err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq,
+							 MEM_TYPE_PAGE_POOL, rq->page_pool);
 	}
 	if (err)
 		goto err_free_by_rq_type;
-- 
2.30.2