Re: [PATCH v3 net-next 1/6] tcp: Add clean acked data hook

[Eric Dumazet <eric.dumazet@gmail.com>]

On Mon, 2017-12-18 at 13:10 +0200, Ilya Lesokhin wrote:
> Called when a TCP segment is acknowledged.
> Could be used by application protocols who hold additional
> metadata associated with the stream data
> This is required by TLS device offload to release
> metadata associated with acknowledged TLS records.
> 
> Signed-off-by: Boris Pismenny <borisp@mellanox.com>
> Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com>
> ---
>  include/net/inet_connection_sock.h | 2 ++
>  net/ipv4/tcp_input.c               | 3 +++
>  2 files changed, 5 insertions(+)
> 
> diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
> index 8e1bf9ae4a5e..ec405a667a85 100644
> --- a/include/net/inet_connection_sock.h
> +++ b/include/net/inet_connection_sock.h
> @@ -77,6 +77,7 @@ struct inet_connection_sock_af_ops {
>   * @icsk_af_ops		   Operations which are AF_INET{4,6} specific
>   * @icsk_ulp_ops	   Pluggable ULP control hook
>   * @icsk_ulp_data	   ULP private data
> + * @icsk_clean_acked	   Clean acked data hook
>   * @icsk_listen_portaddr_node	hash to the portaddr listener hashtable
>   * @icsk_ca_state:	   Congestion control state
>   * @icsk_retransmits:	   Number of unrecovered [RTO] timeouts
> @@ -102,6 +103,7 @@ struct inet_connection_sock {
>  	const struct inet_connection_sock_af_ops *icsk_af_ops;
>  	const struct tcp_ulp_ops  *icsk_ulp_ops;
>  	void			  *icsk_ulp_data;
> +	void			  (*icsk_clean_acked)(struct sock *sk);
>  	struct hlist_node         icsk_listen_portaddr_node;
>  	unsigned int		  (*icsk_sync_mss)(struct sock *sk, u32 pmtu);
>  	__u8			  icsk_ca_state:6,
> diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
> index 4d55c4b338ee..961abc5be84c 100644
> --- a/net/ipv4/tcp_input.c
> +++ b/net/ipv4/tcp_input.c
> @@ -3592,6 +3592,9 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
>  	if (!prior_packets)
>  		goto no_queue;
>  
> +	if (icsk->icsk_clean_acked)
> +		icsk->icsk_clean_acked(sk);
> +
>  	/* See if we can take anything off of the retransmit queue. */
>  	flag |= tcp_clean_rtx_queue(sk, prior_fack, prior_snd_una, &sack_state);
>  

1) tcp_ack() is already very expensive.

2) Since you do not pass any state here, this looks very suspicious to
me.