[PATCH] net/mlx5: fix connection tracking state item validation

[Ivan Malov]

Hi,

On Tue, 5 Aug 2025, Khadem Ullah wrote:

> This patch validate a connection tracking state when matching
> 'conntrack is' in rte_flow rules. The conntract possible CT states
> are SYN_RECV, ESTABLISHED, FIN_WAIT, CLOSE_WAIT, LAST_ACK and
> TIME_WAIT. Therefore the maximum possible value to match on
> in rte_flow is TIME_WAIT but mlx5 allowed matching on any values.
>
> This patch validate the CT state item.
> Fixes: aca19061e4b9 ('net/mlx5: validate connection tracking item')
> Cc: stable at dpdk.org
>
> Signed-off-by: Khadem Ullah <14pwcse1224 at uetpeshawar.edu.pk>
> ---
> drivers/net/mlx5/mlx5_flow_dv.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
> index 7b9e5018b8..750385cd42 100644
> --- a/drivers/net/mlx5/mlx5_flow_dv.c
> +++ b/drivers/net/mlx5/mlx5_flow_dv.c
> @@ -3290,6 +3290,11 @@ mlx5_flow_dv_validate_item_aso_ct(struct rte_eth_dev *dev,
> 						  NULL,
> 						  "Conflict status bits");
> 	}
> +	if (spec->flags > RTE_FLOW_CONNTRACK_STATE_TIME_WAIT)
> +		return rte_flow_error_set(error, EINVAL,
> +					RTE_FLOW_ERROR_TYPE_ITEM,
> +					NULL,
> +					"Invalid CT state matching \n");

It might be better to enclose the multi-line block in brackets.

Also, is it correct to treat 'flags' like enum 'RTE_FLOW_CONNTRACK_STATE'?
I thought it was following 'RTE_FLOW_CONNTRACK_PKT_STATE' flags instead [1].

[1] https://doc.dpdk.org/api-25.07/rte__flow_8h.html#a7a41946aa03ebca8c432279604265b51

Or am I missing something?

Thank you.

> 	/* State change also needs to be considered. */
> 	*item_flags |= MLX5_FLOW_LAYER_ASO_CT;
> 	return 0;
> -- 
> 2.43.0
>
>