diff options
| author | David Morley <morleyd@google.com> | 2023-10-06 01:18:40 +0000 |
|---|---|---|
| committer | Paolo Abeni <pabeni@redhat.com> | 2023-10-10 10:02:59 +0200 |
| commit | 95b9a87c6a6b708cccda1f9b7baf9920b80cdabf (patch) | |
| tree | d06f62ced178016f8a07a882290350d1b01ca1a9 /net/ipv4/tcp_input.c | |
| parent | 8cea95b0bd7930367f11e2abceda6e096dd18943 (diff) | |
| download | lwn-95b9a87c6a6b708cccda1f9b7baf9920b80cdabf.tar.gz lwn-95b9a87c6a6b708cccda1f9b7baf9920b80cdabf.zip | |
tcp: record last received ipv6 flowlabel
In order to better estimate whether a data packet has been
retransmitted or is the result of a TLP, we save the last received
ipv6 flowlabel.
To make space for this field we resize the "ato" field in
inet_connection_sock as the current value of TCP_DELACK_MAX can be
fully contained in 8 bits and add a compile_time_assert ensuring this
field is the required size.
v2: addressed kernel bot feedback about dccp_delack_timer()
v3: addressed build error introduced by commit bbf80d713fe7 ("tcp:
derive delack_max from rto_min")
Signed-off-by: David Morley <morleyd@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Tested-by: David Morley <morleyd@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Diffstat (limited to 'net/ipv4/tcp_input.c')
| -rw-r--r-- | net/ipv4/tcp_input.c | 15 |
1 files changed, 15 insertions, 0 deletions
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 4b8f2e74d71d..2ae4b211c125 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -778,6 +778,16 @@ new_measure: tp->rcvq_space.time = tp->tcp_mstamp; } +static void tcp_save_lrcv_flowlabel(struct sock *sk, const struct sk_buff *skb) +{ +#if IS_ENABLED(CONFIG_IPV6) + struct inet_connection_sock *icsk = inet_csk(sk); + + if (skb->protocol == htons(ETH_P_IPV6)) + icsk->icsk_ack.lrcv_flowlabel = ntohl(ip6_flowlabel(ipv6_hdr(skb))); +#endif +} + /* There is something which you must keep in mind when you analyze the * behavior of the tp->ato delayed ack timeout interval. When a * connection starts up, we want to ack as quickly as possible. The @@ -826,6 +836,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb) } } icsk->icsk_ack.lrcvtime = now; + tcp_save_lrcv_flowlabel(sk, skb); tcp_ecn_check_ce(sk, skb); @@ -4519,6 +4530,9 @@ static void tcp_rcv_spurious_retrans(struct sock *sk, const struct sk_buff *skb) if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq && sk_rethink_txhash(sk)) NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDUPLICATEDATAREHASH); + + /* Save last flowlabel after a spurious retrans. */ + tcp_save_lrcv_flowlabel(sk, skb); } static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb) @@ -4835,6 +4849,7 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) u32 seq, end_seq; bool fragstolen; + tcp_save_lrcv_flowlabel(sk, skb); tcp_ecn_check_ce(sk, skb); if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) { |