tcp/dccp: drop SYN packets if accept queue is full
authorEric Dumazet <edumazet@google.com>
Wed, 26 Oct 2016 16:27:57 +0000 (09:27 -0700)
committerJianxin Pan <jianxin.pan@amlogic.com>
Tue, 31 Oct 2017 08:04:32 +0000 (01:04 -0700)
commit3347f3d676d364d3b8b6f289068be011efade4e4
tree1a15e3fd21ed21785642084e6a18bc63e35ae7c6
parentd76db689ad6faec6f35bac90566f7eb57aa2c557
tcp/dccp: drop SYN packets if accept queue is full

PD#152538: tcp/dccp: drop SYN packets if accept queue is full

tcp/dccp: drop SYN packets if accept queue is full

Per listen(fd, backlog) rules, there is really no point accepting a SYN,
sending a SYNACK, and dropping the following ACK packet if accept queue
is full, because application is not draining accept queue fast enough.

This behavior is fooling TCP clients that believe they established a
flow, while there is nothing at server side. They might then send about
10 MSS (if using IW10) that will be dropped anyway while server is under
stress.

Change-Id: I674728eac69b418255522afc8d345995b8891a7e
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
include/net/inet_connection_sock.h
net/dccp/ipv4.c
net/dccp/ipv6.c
net/ipv4/tcp_input.c