Date: Thu, 27 Sep 2001 10:14:59 -0700 From: jayanth <jayanth@yahoo-inc.com> To: tsuchiya@flab.fujitsu.co.jp Cc: jayanth@yahoo-inc.com, net@FreeBSD.ORG, silby@silby.com Subject: Re: TCP performance question Message-ID: <20010927101459.A26671@yahoo-inc.com> In-Reply-To: <200109270652.PAA03232@const.kawasaki.flab.fujitsu.co.jp>; from tsuchiya@flab.fujitsu.co.jp on Thu, Sep 27, 2001 at 03:52:15PM %2B0900 References: <200109270652.PAA03232@const.kawasaki.flab.fujitsu.co.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
Probably add a new flag TF_IDLE that is true, if there is more data to send when the connection was idle. This way the next time around the idle variable will be true if the TF_IDLE flag is true and if we can empty the socket buffer all the data will be sent. Will send you a patch soon. jayanth tsuchiya@flab.fujitsu.co.jp (tsuchiya@flab.fujitsu.co.jp) wrote: > > > This issue is a combination of mbuf cluster size and the > > TF_MORETOCOME flag. > > > if (len) { > > if (len == tp->t_maxseg) > > goto send; > > if (!(tp->t_flags & TF_MORETOCOME) && > > (idle || tp->t_flags & TF_NODELAY) && > > (tp->t_flags & TF_NOPUSH) == 0 && > > len + off >= so->so_snd.sb_cc) > > goto send; > > When I changed the condition, the problem we had did not occur. I am wondering > what is the right fix. > Yoshi > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010927101459.A26671>