Messages in this thread | | | Date | Wed, 11 Nov 1998 17:31:37 +0100 (CET) | From | Andrea Arcangeli <> | Subject | Re: TCP, congestion window & retransmission |
| |
On Tue, 10 Nov 1998, David S. Miller wrote:
>traces I had while I did most of my TCP output engine rewrite. I >always kept forgetting to go back and see if it was correct to do slow >start or not during timeout based recovery.
According to the RFC2001 in case of duplicate acks or timeout TCP has to do slow-start until TCP has not the congestion window open to the value of sstresh (set to max(cwnd/2,1) before -> at timeout time). When the congestion window is >= sstrash, TCP must go in congestion avoidance incrementing cwnd of segsize*segsize/cwnd and at most incrementing of 1 unit every RTT (this way cwnd increase liner and not exponential).
Andrea Arcangeli
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |