diff options
| author | Philo Lu <lulie@linux.alibaba.com> | 2024-04-21 12:20:09 +0800 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2024-04-25 08:52:12 +0100 |
| commit | 2bf90a57f0e682872c5cfb66ffa45e432bb9c7ae (patch) | |
| tree | 3e523116c5e7de8d0a426fa022f7a4c13966e453 | |
| parent | 14b5fb2145caeb909a1cd57d9cd5e0c3cd005642 (diff) | |
| download | net-next-2bf90a57f0e682872c5cfb66ffa45e432bb9c7ae.tar.gz | |
tcp: update sacked after tracepoint in __tcp_retransmit_skb
Marking TCP_SKB_CB(skb)->sacked with TCPCB_EVER_RETRANS after the
traceopint (trace_tcp_retransmit_skb), then we can get the
retransmission efficiency by counting skbs w/ and w/o TCPCB_EVER_RETRANS
mark in this tracepoint.
We have discussed to achieve this with BPF_SOCK_OPS in [0], and using
tracepoint is thought to be a better solution.
[0]
https://lore.kernel.org/all/20240417124622.35333-1-lulie@linux.alibaba.com/
Signed-off-by: Philo Lu <lulie@linux.alibaba.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
| -rw-r--r-- | net/ipv4/tcp_output.c | 11 |
1 files changed, 6 insertions, 5 deletions
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 99a1d88f7f47b9..ce59e4499b66df 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3418,11 +3418,6 @@ start: err = tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC); } - /* To avoid taking spuriously low RTT samples based on a timestamp - * for a transmit that never happened, always mark EVER_RETRANS - */ - TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS; - if (BPF_SOCK_OPS_TEST_FLAG(tp, BPF_SOCK_OPS_RETRANS_CB_FLAG)) tcp_call_bpf_3arg(sk, BPF_SOCK_OPS_RETRANS_CB, TCP_SKB_CB(skb)->seq, segs, err); @@ -3432,6 +3427,12 @@ start: } else if (err != -EBUSY) { NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPRETRANSFAIL, segs); } + + /* To avoid taking spuriously low RTT samples based on a timestamp + * for a transmit that never happened, always mark EVER_RETRANS + */ + TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS; + return err; } |