diff options
| author |   Eric Dumazet <edumazet@google.com> | 2019-09-30 15:44:44 -0700 |
|---|---|---|
| committer |   David S. Miller <davem@davemloft.net> | 2019-10-01 21:40:49 -0400 |
| commit | 3256a2d6ab1f71f9a1bd2d7f6f18eb8108c48d17 (patch) | |
| tree | d2a73ff8ac5783d93545bb2bd497067c6243a1fa /net | |
| parent | vsock: Fix a lockdep warning in __vsock_release() (diff) | |
| download | linux-dev-3256a2d6ab1f71f9a1bd2d7f6f18eb8108c48d17.tar.xz linux-dev-3256a2d6ab1f71f9a1bd2d7f6f18eb8108c48d17.zip | |
tcp: adjust rto_base in retransmits_timed_out()
The cited commit exposed an old retransmits_timed_out() bug
which assumed it could call tcp_model_timeout() with
TCP_RTO_MIN as rto_base for all states.
But flows in SYN_SENT or SYN_RECV state uses a different
RTO base (1 sec instead of 200 ms, unless BPF choses
another value)
This caused a reduction of SYN retransmits from 6 to 4 with
the default /proc/sys/net/ipv4/tcp_syn_retries value.
Fixes: a41e8a88b06e ("tcp: better handle TCP_USER_TIMEOUT in SYN_SENT state")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Marek Majkowski <marek@cloudflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to '')
| -rw-r--r-- | net/ipv4/tcp_timer.c | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 40de2d2364a1..05be564414e9 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -198,8 +198,13 @@ static bool retransmits_timed_out(struct sock *sk, return false; start_ts = tcp_sk(sk)->retrans_stamp; - if (likely(timeout == 0)) - timeout = tcp_model_timeout(sk, boundary, TCP_RTO_MIN); + if (likely(timeout == 0)) { + unsigned int rto_base = TCP_RTO_MIN; + + if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) + rto_base = tcp_timeout_init(sk); + timeout = tcp_model_timeout(sk, boundary, rto_base); + } return (s32)(tcp_time_stamp(tcp_sk(sk)) - start_ts - timeout) >= 0; } |