diff options
author | Florin Coras <fcoras@cisco.com> | 2022-01-06 11:58:24 -0800 |
---|---|---|
committer | Florin Coras <fcoras@cisco.com> | 2022-01-06 16:38:24 -0800 |
commit | c8144356338995a277593bc08315d3b1ddcc8cb3 (patch) | |
tree | 4eed6c04bf1e924e9d7312efe20af34db429c88b /src/vnet/tcp/tcp_input.c | |
parent | ed5be47009aca45008ba9d44e76ac51ec1f7f156 (diff) |
tcp: update snd_congestion only during congestion
If running without sacks, if snd_una does not cover snd_congestion fast
recovery can be missed but the two heuristics from RFC6582 should avoid
that.
Also snd_congestion was used as a means of inferring if the connection
recently exited congestion while setting the persist timer but that does
not always work correctly if not congested.
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I94d4ac738cdd4f7f23f62e97dd63059de1cd4af9
Diffstat (limited to 'src/vnet/tcp/tcp_input.c')
-rw-r--r-- | src/vnet/tcp/tcp_input.c | 9 |
1 files changed, 0 insertions, 9 deletions
diff --git a/src/vnet/tcp/tcp_input.c b/src/vnet/tcp/tcp_input.c index 11491a68257..64b9334cbea 100644 --- a/src/vnet/tcp/tcp_input.c +++ b/src/vnet/tcp/tcp_input.c @@ -802,15 +802,6 @@ tcp_cc_update (tcp_connection_t * tc, tcp_rate_sample_t * rs) /* If a cumulative ack, make sure dupacks is 0 */ tc->rcv_dupacks = 0; - - /* When dupacks hits the threshold we only enter fast retransmit if - * cumulative ack covers more than snd_congestion. Should snd_una - * wrap this test may fail under otherwise valid circumstances. - * Therefore, proactively update snd_congestion when wrap detected. */ - if (PREDICT_FALSE - (seq_leq (tc->snd_congestion, tc->snd_una - tc->bytes_acked) - && seq_gt (tc->snd_congestion, tc->snd_una))) - tc->snd_congestion = tc->snd_una - 1; } /** |