aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorYuchung Cheng2017-12-07 11:33:31 -0800
committerDavid S. Miller2017-12-08 14:14:11 -0500
commitcd1fc85b4399d47e3d6626301741ba8c38cd475a (patch)
tree86125c235e1df3c21686d6c4625f47d1b6ce20e3
parent0ce294d88457bccd7f9991f883fec80022a1ddbd (diff)
tcp: always evaluate losses in RACK upon undo
When sender detects spurious retransmission, all packets marked lost are remarked to be in-flight. However some may be considered lost based on its timestamps in RACK. This patch forces RACK to re-evaluate, which may be skipped previously if the ACK does not advance RACK timestamp. Signed-off-by: Yuchung Cheng <ycheng@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Priyaranjan Jha <priyarjha@google.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-rw-r--r--net/ipv4/tcp_input.c1
1 files changed, 1 insertions, 0 deletions
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 075c559570e6..9550cc42de2d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2329,6 +2329,7 @@ static void tcp_undo_cwnd_reduction(struct sock *sk, bool unmark_loss)
}
tp->snd_cwnd_stamp = tcp_jiffies32;
tp->undo_marker = 0;
+ tp->rack.advanced = 1; /* Force RACK to re-exam losses */
}
static inline bool tcp_may_undo(const struct tcp_sock *tp)