Hi, What is the initial value of snd_ssthresh for linux 2.4? I guess when snd_cwnd grows cross this value, snd_cwnd is governed by cong avoid algorithm and increases 1 segment every RTT. Can anyone clarifies this piece of code? void tcp_update_metrics(struct sock *sk) { .... if (tp->snd_ssthresh >= 0xFFFF) { /* Slow start still did not finish. */ }else if (tp->snd_cwnd > tp->snd_ssthresh && tp->ca_state == TCP_CA_Open) { /* Cong. avoidance phase, cwnd is reliable. */ } In cong avoidance, snd_cwnd increases at a rate of one seg in a RTT. so for each ack, does it increase like this: tp->snd_cwnd += 1 / tp->snd_cwnd and snd_cwnd increase 1 only when snd_cwnd acks recieved? Where the RTT reflects itself here? Thanks, - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html