aboutsummaryrefslogtreecommitdiffstats
path: root/net/ipv4/tcp_nv.c
diff options
context:
space:
mode:
authorYuchung Cheng <ycheng@google.com>2017-08-03 20:38:51 -0700
committerDavid S. Miller <davem@davemloft.net>2017-08-06 21:25:10 -0700
commit4faf783998b8cb88294e9df89032f473f8771b78 (patch)
tree0ad450f18cbf6c23f23e9f181a4d4ee5c6a99b7a /net/ipv4/tcp_nv.c
parentnet: systemport: Support 64bit statistics (diff)
downloadlinux-dev-4faf783998b8cb88294e9df89032f473f8771b78.tar.xz
linux-dev-4faf783998b8cb88294e9df89032f473f8771b78.zip
tcp: fix cwnd undo in Reno and HTCP congestion controls
Using ssthresh to revert cwnd is less reliable when ssthresh is bounded to 2 packets. This patch uses an existing variable in TCP "prior_cwnd" that snapshots the cwnd right before entering fast recovery and RTO recovery in Reno. This fixes the issue discussed in netdev thread: "A buggy behavior for Linux TCP Reno and HTCP" https://www.spinics.net/lists/netdev/msg444955.html Suggested-by: Neal Cardwell <ncardwell@google.com> Reported-by: Wei Sun <unlcsewsun@gmail.com> Signed-off-by: Yuchung Cheng <ncardwell@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/tcp_nv.c')
0 files changed, 0 insertions, 0 deletions