[PATCH net-next V2] net/mlx5e: Allow set_rx_mode on uplink representor

[PATCH net-next V2] net/mlx5e: Allow set_rx_mode on uplink representor

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

set_rx_mode handler was skipped on uplink representor, since uplink
relies on FDB to forward all traffic to it by default, which works
perfectly on a single PF per physical port configuration, as explicit
mac request isn't required, but In case of multi-host and DPU
environments, uplink can only use own mac address, as set_rx_mode
wasn't honored in uplink rep.

Since MPFs (Multi PF switch) requires PFs to request explicit mac
forwarding, this patch enables set_rx_mode on uplink representor to
allow PF mac programming into MPFs table in switchdev mode, allowing
use-cases such as arbitrary mac address forwarding via linux bridge.

Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
---
 - Improve title and commit message -- Kuba.
   v[1]: https://lore.kernel.org/netdev/20260310104841.1862380-1-tariqt@nvidia.com/

 drivers/net/ethernet/mellanox/mlx5/core/en_fs.c   | 10 ++++++++++
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  3 ---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c  |  2 ++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index 9352e2183312..d7556d611155 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -823,8 +823,14 @@ static void mlx5e_destroy_promisc_table(struct mlx5e_flow_steering *fs)
 void mlx5e_fs_set_rx_mode_work(struct mlx5e_flow_steering *fs,
 			       struct net_device *netdev)
 {
+	struct mlx5e_priv *priv = netdev_priv(netdev);
 	struct mlx5e_l2_table *ea = &fs->l2;
 
+	if (mlx5e_is_uplink_rep(priv)) {
+		mlx5e_handle_netdev_addr(fs, netdev);
+		goto update_vport_context;
+	}
+
 	bool rx_mode_enable  = fs->state_destroy;
 	bool promisc_enabled   = rx_mode_enable && (netdev->flags & IFF_PROMISC);
 	bool allmulti_enabled  = rx_mode_enable && (netdev->flags & IFF_ALLMULTI);
@@ -864,6 +870,7 @@ void mlx5e_fs_set_rx_mode_work(struct mlx5e_flow_steering *fs,
 	ea->allmulti_enabled  = allmulti_enabled;
 	ea->broadcast_enabled = broadcast_enabled;
 
+update_vport_context:
 	mlx5e_vport_context_update(fs, netdev);
 }
 
@@ -984,6 +991,9 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_flow_steering *fs,
 	u8 *mc_dmac;
 	u8 *mv_dmac;
 
+	if (!ft)
+		return -EINVAL;
+
 	spec = kvzalloc_obj(*spec);
 	if (!spec)
 		return -ENOMEM;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index f7009da94f0b..3eebdf402129 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4102,9 +4102,6 @@ mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
 
 static void mlx5e_nic_set_rx_mode(struct mlx5e_priv *priv)
 {
-	if (mlx5e_is_uplink_rep(priv))
-		return; /* no rx mode for uplink rep */
-
 	queue_work(priv->wq, &priv->set_rx_mode_work);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 1db4ecb2356f..8992f0f7a870 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1369,6 +1369,8 @@ static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv)
 	netdev_unlock(priv->netdev);
 	rtnl_unlock();
 
+	/* clean-up uplink's mpfs mac table */
+	queue_work(priv->wq, &priv->set_rx_mode_work);
 	mlx5e_rep_bridge_cleanup(priv);
 	mlx5e_dcbnl_delete_app(priv);
 	mlx5_notifier_unregister(mdev, &priv->events_nb);
-- 
2.43.0

Re: [PATCH net-next V2] net/mlx5e: Allow set_rx_mode on uplink representor

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 18 Mar 2026 17:54:56 -0700 you wrote:
> From: Saeed Mahameed <saeedm@nvidia.com>
> 
> set_rx_mode handler was skipped on uplink representor, since uplink
> relies on FDB to forward all traffic to it by default, which works
> perfectly on a single PF per physical port configuration, as explicit
> mac request isn't required, but In case of multi-host and DPU
> environments, uplink can only use own mac address, as set_rx_mode
> wasn't honored in uplink rep.
> 
> [...]

Here is the summary with links:
  - [net-next,V2] net/mlx5e: Allow set_rx_mode on uplink representor
    https://git.kernel.org/netdev/net-next/c/609e79253ace

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