summaryrefslogtreecommitdiffstats
path: root/net
diff options
context:
space:
mode:
authorYuchung Cheng2019-01-09 03:14:28 +0100
committerGreg Kroah-Hartman2019-01-22 21:40:30 +0100
commitd7fe54c17aa3995a62345946d6ff8ddc03c17d2a (patch)
treed5d1d148b26a3ca16b7db009ba2716b093886a75 /net
parentpacket: Do not leak dev refcounts on error exit (diff)
downloadkernel-qcow2-linux-d7fe54c17aa3995a62345946d6ff8ddc03c17d2a.tar.gz
kernel-qcow2-linux-d7fe54c17aa3995a62345946d6ff8ddc03c17d2a.tar.xz
kernel-qcow2-linux-d7fe54c17aa3995a62345946d6ff8ddc03c17d2a.zip
tcp: change txhash on SYN-data timeout
[ Upstream commit c5715b8fabfca0ef85903f8bad2189940ed41cc8 ] Previously upon SYN timeouts the sender recomputes the txhash to try a different path. However this does not apply on the initial timeout of SYN-data (active Fast Open). Therefore an active IPv6 Fast Open connection may incur one second RTO penalty to take on a new path after the second SYN retransmission uses a new flow label. This patch removes this undesirable behavior so Fast Open changes the flow label just like the regular connections. This also helps avoid falsely disabling Fast Open on the sender which triggers after two consecutive SYN timeouts on Fast Open. Signed-off-by: Yuchung Cheng <ycheng@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'net')
-rw-r--r--net/ipv4/tcp_timer.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 57eae8d70ba1..b1b5a648def6 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -224,7 +224,7 @@ static int tcp_write_timeout(struct sock *sk)
if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) {
if (icsk->icsk_retransmits) {
dst_negative_advice(sk);
- } else if (!tp->syn_data && !tp->syn_fastopen) {
+ } else {
sk_rethink_txhash(sk);
}
retry_until = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries;