Commit 48040793 authored by Yousuk Seung's avatar Yousuk Seung Committed by David S. Miller
Browse files

tcp: add earliest departure time to SCM_TIMESTAMPING_OPT_STATS 



This change adds TCP_NLA_EDT to SCM_TIMESTAMPING_OPT_STATS that reports
the earliest departure time(EDT) of the timestamped skb. By tracking EDT
values of the skb from different timestamps, we can observe when and how
much the value changed. This allows to measure the precise delay
injected on the sender host e.g. by a bpf-base throttler.

Signed-off-by: default avatarYousuk Seung <ysseung@google.com>
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Acked-by: default avatarNeal Cardwell <ncardwell@google.com>
Acked-by: default avatarSoheil Hassas Yeganeh <soheil@google.com>
Acked-by: default avatarYuchung Cheng <ycheng@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent c6886957

include/linux/tcp.h

+2 −1
Original line number Diff line number Diff line
@@ -484,7 +484,8 @@ static inline void tcp_saved_syn_free(struct tcp_sock *tp) 
	tp->saved_syn = NULL;

}



struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk);

struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk,

					       const struct sk_buff *orig_skb);



static inline u16 tcp_mss_clamp(const struct tcp_sock *tp, u16 mss)

{

include/uapi/linux/tcp.h

+1 −0
Original line number Diff line number Diff line
@@ -313,6 +313,7 @@ enum { 
	TCP_NLA_SRTT,		/* smoothed RTT in usecs */

	TCP_NLA_TIMEOUT_REHASH, /* Timeout-triggered rehash attempts */

	TCP_NLA_BYTES_NOTSENT,	/* Bytes in write queue not yet sent */

	TCP_NLA_EDT,		/* Earliest departure time (CLOCK_MONOTONIC) */

};



/* for TCP_MD5SIG socket option */

net/core/skbuff.c

+1 −1
Original line number Diff line number Diff line
@@ -4692,7 +4692,7 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb, 
		if ((sk->sk_tsflags & SOF_TIMESTAMPING_OPT_STATS) &&

		    sk->sk_protocol == IPPROTO_TCP &&

		    sk->sk_type == SOCK_STREAM) {

			skb = tcp_get_timestamping_opt_stats(sk);

			skb = tcp_get_timestamping_opt_stats(sk, orig_skb);

			opt_stats = true;

		} else

#endif

net/ipv4/tcp.c

+5 −1
Original line number Diff line number Diff line
@@ -3501,10 +3501,12 @@ static size_t tcp_opt_stats_get_size(void) 
		nla_total_size(sizeof(u32)) + /* TCP_NLA_SRTT */

		nla_total_size(sizeof(u16)) + /* TCP_NLA_TIMEOUT_REHASH */

		nla_total_size(sizeof(u32)) + /* TCP_NLA_BYTES_NOTSENT */

		nla_total_size_64bit(sizeof(u64)) + /* TCP_NLA_EDT */

		0;

}



struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk)

struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk,

					       const struct sk_buff *orig_skb)

{

	const struct tcp_sock *tp = tcp_sk(sk);

	struct sk_buff *stats;
@@ -3558,6 +3560,8 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk) 
	nla_put_u16(stats, TCP_NLA_TIMEOUT_REHASH, tp->timeout_rehash);

	nla_put_u32(stats, TCP_NLA_BYTES_NOTSENT,

		    max_t(int, 0, tp->write_seq - tp->snd_nxt));

	nla_put_u64_64bit(stats, TCP_NLA_EDT, orig_skb->skb_mstamp_ns,

			  TCP_NLA_PAD);



	return stats;

}

CRA Git | Maintained and supported by SUSTech CRA and CCSE