path: root/net/ipv4/tcp_bbr.c
diff options
authorNeal Cardwell <ncardwell@google.com>2018-05-01 21:45:41 -0400
committerDavid S. Miller <davem@davemloft.net>2018-05-02 11:12:32 -0400
commite6e6a278b1eaffa19d42186bfacd1ffc15a50b3f (patch)
tree3e42446d622d43a07493f1b110bd9986345eff95 /net/ipv4/tcp_bbr.c
parentnet: ethernet: ti: cpsw: fix packet leaking in dual_mac mode (diff)
tcp_bbr: fix to zero idle_restart only upon S/ACKed data
Previously the bbr->idle_restart tracking was zeroing out the bbr->idle_restart bit upon ACKs that did not SACK or ACK anything, e.g. receiving incoming data or receiver window updates. In such situations BBR would forget that this was a restart-from-idle situation, and if the min_rtt had expired it would unnecessarily enter PROBE_RTT (even though we were actually restarting from idle but had merely forgotten that fact). The fix is simple: we need to remember we are restarting from idle until we receive a S/ACK for some data (a S/ACK for the first flight of data we send as we are restarting). This commit is a stable candidate for kernels back as far as 4.9. Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control") Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: Priyaranjan Jha <priyarjha@google.com> Signed-off-by: Yousuk Seung <ysseung@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/tcp_bbr.c')
1 files changed, 3 insertions, 1 deletions
diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
index 158d105e76da..58e2f479ffb4 100644
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -806,7 +806,9 @@ static void bbr_update_min_rtt(struct sock *sk, const struct rate_sample *rs)
- bbr->idle_restart = 0;
+ /* Restart after idle ends only once we process a new S/ACK for data */
+ if (rs->delivered > 0)
+ bbr->idle_restart = 0;
static void bbr_update_model(struct sock *sk, const struct rate_sample *rs)