[PATCH] Added NoDelay config option and nodelay subsystem option

Rick Jones rick_jones2 at hp.com
Fri Feb 1 08:28:31 EST 2002

> > The reason to have the socket buffer only bwdelay is so that when that
> > first tripple hits the third response cannot go into the socket buffer
> > and hit TCP until the fifth(?) request arrives, so the fourth response
> > ACK's both the four request and the fifth request. (It does this
> > because it was sent by the app to TCP after the fifth request arrived
> > because the fifth request acked the second response) I think I have
> > that right though I am in a maze of twist passages :)
> ...and do not want to go there either.  Yes, you might achieve what you
> propose with a carefully selected socket buffer, but after a while you
> will have to send a window update or some other little silly packet and
> you're toast again...  You'll probably find a way around that too, but
> I'd rather enable TCP_NODELAY and be done with it, I think.  :-)

typcial application programmer attitude :)

I was doing the back flips there to deal with the initite speeds and
such and to guarantee things in the proper sequence, from those
experiements I ran yesterday, it appears that timing is pretty-much key.

what are the request and reponse sizes involved in sftp anyhow?

any ideas yet what sort of bandwidth delay products are going to be

rick jones
Wisdom Teeth are impacted, people are affected by the effects of events.
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...

More information about the openssh-unix-dev mailing list