Hello, list!
I am looking at the code that handles duplicate acks from peer. Please
tell me if I'm wrong.
dupacks are counted here
static void tcp_add_reno_sack(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
tp->sacked_out++; **********
and the comparison to the fast-rexmit threshold is done here:
/* Trick#4: It is still not OK... But will it be useful to delay
* recovery more?
*/
packets_out = tcp_get_pcount(&tp->packets_out);
if (packets_out <= tp->reordering &&
tcp_get_pcount(&tp->sacked_out) >= max_t(__u32, packets_out/2,
sysctl_tcp_reordering) && *******
where sysctl_tcp_reordering holds the threshold and is 3 if using default values
Is this correct?
Additionally, could anybody point me to a full reference of how the
"Linux NewReno/SACK/FACK/ECN state machine" works?
Thanx in advance,
Eduardo
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
