We correctly avoid doing that at the beginning of tcp_prepare_flags(),
but we might clear the flag later on if we actually end up sending a
"flag" segment.
Make sure we don't, otherwise we might delay window updates after a
zero-window condition significantly, and significantly affect
throughput.
In some cases, we're forcing peers to send zero-window probes or
keep-alive segments.
Signed-off-by: Stefano Brivio
Reviewed-by: David Gibson
---
tcp.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tcp.c b/tcp.c
index 8eeef4c..9bf7b8b 100644
--- a/tcp.c
+++ b/tcp.c
@@ -1279,7 +1279,8 @@ int tcp_prepare_flags(const struct ctx *c, struct tcp_tap_conn *conn,
th->fin = !!(flags & FIN);
if (th->ack) {
- if (SEQ_GE(conn->seq_ack_to_tap, conn->seq_from_tap))
+ if (SEQ_GE(conn->seq_ack_to_tap, conn->seq_from_tap) &&
+ conn->wnd_to_tap)
conn_flag(c, conn, ~ACK_TO_TAP_DUE);
else
conn_flag(c, conn, ACK_TO_TAP_DUE);
--
2.43.0