On Wed, Jul 26, 2006 at 04:10:58PM +1200, Ian McDonald wrote: > On 7/26/06, Ma Lin <gavin.malin@xxxxxxxxx> wrote: > >My planned change in CCID2 is something to do with the sequence > >number, the non-cumulative ACK, and the moving of congestion window. > >Which files am I suppose to look at? Changing the dynamics of cwnd should be relatively easy. I purpousely pulled out a helper method "change_cwnd" in ccid2 which enables easy logging & modification of the window. Also, I think I added "congestion_event()" or similar, which is called once when there are losses in an RTT. Screwing with the ACK code is a nightmare. That's where I found most of the bugs. Note that the ack vector is "parsed" twice [which is a design bug: 1) ackvec.c in option parsing, and 2) in ccid2.c upon reception of an ack]. - : send the line "unsubscribe dccp" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
