- From: "Brian F. G. Bidulock" <bidulock AT openss7.org>
- To: Torbjörn Andersson <torbjorn.andersson AT kau.se>
- Cc: discussion AT sctp.de
- Subject: Re: [SCTP-Discussion] Initial slowstart behavior of SCTP
- Date: Wed Dec 4 09:42:02 2002
- Dsn-notification-to: <bidulock AT openss7.org>
- List-id: SCTP Discussions <discussion.sctp.de>
- Organization: http://www.openss7.org/
Torbjörn,
RFC 2960 and in particular the IG clearly permits up to 1*MTU-1
slop (in a single message) over the current cwnd. Otherwise,
cwnd will not grow (it takes acknolwedged bytes beyond the cwnd
to grow cwnd). Any implementation is correct to send up to
1*MTU slop over the current congestion window.
--brian
On Wed, 04 Dec 2002, Torbjörn Andersson wrote:
>
I have not received any answer on my previous question in my previous
>
mail and I wondering why?
>
>
I speculate that your implementation does this:
>
>
The flowcontroll allows up to 2*MTU bytes outstanding.
>
This is checked before each packet is sent and after two packets
>
there is only, in my case, 200 bytes unfilled in the cwnd.
>
This allows one new packet to be sent, but only with a payload
>
of 200 bytes but this limit is not checked by the packetizer, which
>
sends a full packet, with MTU bytes.
>
>
What do you think of this? Do you have another explanation for this ?
>
>
/Regards
>
>
Torbjörn Andersson
>
University of Karlstad, Sweden
>
>
_______________________________________________
>
discussion mailing list
>
discussion AT sctp.de
>
http://www.sctp.de/mailman/listinfo/discussion
--
Brian F. G. Bidulock ¦ The reasonable man adapts himself to the ¦
bidulock AT openss7.org
¦ world; the unreasonable one persists in ¦
http://www.openss7.org/ ¦ trying to adapt the world to himself. ¦
¦ Therefore all progress depends on the ¦
¦ unreasonable man. -- George Bernard Shaw ¦
Archive powered by MHonArc 2.6.19+.