Re: [PATCH net-next v3] net/mlx5e: advertise tcp-data-split support

[Nimrod Oren <noren@nvidia.com>]

Thanks for working on this, a few comments below.

On 09/10/2026 5:51, Dimitri Daskalakis wrote:
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
> index 261c466a4d36..c30d751cac2f 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
> @@ -378,6 +378,9 @@ void mlx5e_ethtool_get_ringparam(struct mlx5e_priv *priv,
>  
>  	kernel_param->hds_thresh = 0;
>  	kernel_param->hds_thresh_max = 0;
> +
> +	if (priv->channels.params.packet_merge.type == MLX5E_PACKET_MERGE_SHAMPO)
> +		kernel_param->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_ENABLED;
>  }

With HW-GRO off, tcp-data-split is omitted and appears unsupported.
I think it should report disabled when HW-GRO is supported but inactive.

...

> +	if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_DISABLED) {
> +		NL_SET_ERR_MSG_MOD(extack,
> +				   "TCP-data-split can not be disabled");
> +		return false;

I think the message should emphasize that TCP-data-split can't be
directly/explicitly disabled.

...
> +	/* Disable HW-GRO if it was only kept on for HDS. */
> +	if (dev->cfg->hds_config != kernel_param->tcp_data_split)
> +		netdev_update_features(priv->netdev);

nit: the only transition that needs this is enabled -> auto,
so maybe something like this is preferable:

  if (dev->cfg->hds_config == ETHTOOL_TCP_DATA_SPLIT_ENABLED &&
      kernel_param->tcp_data_split != ETHTOOL_TCP_DATA_SPLIT_ENABLED)

Thanks,
Nimrod