From owner-freebsd-hackers Thu May 29 13:44:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA08273 for hackers-outgoing; Thu, 29 May 1997 13:44:34 -0700 (PDT) Received: from friley01.res.iastate.edu (friley01.res.iastate.edu [129.186.78.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA08267 for ; Thu, 29 May 1997 13:44:29 -0700 (PDT) Received: from friley01.res.iastate.edu (localhost [127.0.0.1]) by friley01.res.iastate.edu (8.8.5/8.8.5) with ESMTP id PAA04544 for ; Thu, 29 May 1997 15:44:30 -0500 (CDT) Message-Id: <199705292044.PAA04544@friley01.res.iastate.edu> X-Mailer: exmh version 2.0gamma 1/27/96 To: hackers@freebsd.org Subject: Re: TCP/IP bug? Unnecessary fragmentation... In-reply-to: Your message of Thu, 29 May 1997 13:17:17 -0700. <97May29.131731pdt.177489@crevenia.parc.xerox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 29 May 1997 15:44:30 -0500 From: Chris Csanady Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Another oddity is that if you close the connection immediately, for the >same data sizes that it does this odd fragmentation (100 < S < 208), >the FIN is piggybacked on the 2nd data segment. This doesn't happen >with any other data size. I have some graphs, and I believe that it does it at many other sizes too. The performance hit is _much_ more that just mbuf/cluster handling should impose it would seem. It seems to do it up around 2000, 2100(?), 4000... Is this a generic 44BSD problem, or is it FreeBSD specific? Chris > > Bill > >200 bytes: >20:07:36.948660 sundae.1156 > crevenia.discard: S 3914627240:3914627240(0) win 16384 (D F) >20:07:36.950164 crevenia.discard > sundae.1156: S 703872000:703872000(0) ack 3 914627241 win 4096 >20:07:36.950390 sundae.1156 > crevenia.discard: . ack 1 win 17520 (DF) >20:07:36.950722 sundae.1156 > crevenia.discard: P 1:101(100) ack 1 win 17520 ( DF) >20:07:36.951096 sundae.1156 > crevenia.discard: FP 101:201(100) ack 1 win 1752 0 (DF) >20:07:36.951753 crevenia.discard > sundae.1156: . ack 202 win 3896 >20:07:36.965928 crevenia.discard > sundae.1156: F 1:1(0) ack 202 win 4096 >20:07:36.966140 sundae.1156 > crevenia.discard: . ack 2 win 17520 (DF) > >1024 bytes: >20:07:57.229256 sundae.1159 > crevenia.discard: S 3918727601:3918727601(0) win 16384 (D F) >20:07:57.230772 crevenia.discard > sundae.1159: S 706688000:706688000(0) ack 3 918727602 win 4096 >20:07:57.231011 sundae.1159 > crevenia.discard: . ack 1 win 17520 (DF) >20:07:57.231796 sundae.1159 > crevenia.discard: P 1:1025(1024) ack 1 win 17520 (DF) >20:07:57.232110 sundae.1159 > crevenia.discard: F 1025:1025(0) ack 1 win 17520 (DF) >20:07:57.233398 crevenia.discard > sundae.1159: . ack 1026 win 3072 >20:07:57.242112 crevenia.discard > sundae.1159: F 1:1(0) ack 1026 win 4096 >20:07:57.242355 sundae.1159 > crevenia.discard: . ack 2 win 17520 (DF) > >2048 bytes: >20:08:02.212852 sundae.1160 > crevenia.discard: S 3919846017:3919846017(0) win 16384 (D F) >20:08:02.214375 crevenia.discard > sundae.1160: S 707392000:707392000(0) ack 3 919846018 win 4096 >20:08:02.214608 sundae.1160 > crevenia.discard: . ack 1 win 17520 (DF) >20:08:02.215719 sundae.1160 > crevenia.discard: . 1:1461(1460) ack 1 win 17520 (DF) >20:08:02.216294 sundae.1160 > crevenia.discard: P 1461:2049(588) ack 1 win 175 20 (DF) >20:08:02.216629 sundae.1160 > crevenia.discard: F 2049:2049(0) ack 1 win 17520 (DF) >20:08:02.217980 crevenia.discard > sundae.1160: . ack 2050 win 2048 >20:08:02.229945 crevenia.discard > sundae.1160: . ack 2050 win 4096 >20:08:02.231652 crevenia.discard > sundae.1160: F 1:1(0) ack 2050 win 4096 >20:08:02.231843 sundae.1160 > crevenia.discard: . ack 2 win 17520 (DF)