Message ID | 20240627024227.3040278-1-ncardwell.sw@gmail.com (mailing list archive) |
---|---|
State | Accepted |
Commit | a6458ab7fd4f427d4f6f54380453ad255b7fde83 |
Delegated to: | Netdev Maintainers |
Headers | show |
Series | [net] UPSTREAM: tcp: fix DSACK undo in fast recovery to call tcp_try_to_open() | expand |
Hello: This patch was applied to netdev/net.git (main) by David S. Miller <davem@davemloft.net>: On Wed, 26 Jun 2024 22:42:27 -0400 you wrote: > From: Neal Cardwell <ncardwell@google.com> > > In some production workloads we noticed that connections could > sometimes close extremely prematurely with ETIMEDOUT after > transmitting only 1 TLP and RTO retransmission (when we would normally > expect roughly tcp_retries2 = TCP_RETR2 = 15 RTOs before a connection > closes with ETIMEDOUT). > > [...] Here is the summary with links: - [net] UPSTREAM: tcp: fix DSACK undo in fast recovery to call tcp_try_to_open() https://git.kernel.org/netdev/net/c/a6458ab7fd4f You are awesome, thank you!
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2e39cb881e209..e67cbeeeb95b4 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3077,7 +3077,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, return; if (tcp_try_undo_dsack(sk)) - tcp_try_keep_open(sk); + tcp_try_to_open(sk, flag); tcp_identify_packet_loss(sk, ack_flag); if (icsk->icsk_ca_state != TCP_CA_Recovery) {