From owner-freebsd-hackers Sat Nov 28 13:30:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA16129 for freebsd-hackers-outgoing; Sat, 28 Nov 1998 13:30:43 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from cs.tamu.edu (clavin.cs.tamu.edu [128.194.130.106]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA16124 for ; Sat, 28 Nov 1998 13:30:41 -0800 (PST) (envelope-from gurudatt@cs.tamu.edu) Received: from dilbert.cs.tamu.edu (IDENT:2146@dilbert [128.194.133.100]) by cs.tamu.edu (8.9.1/8.9.1) with ESMTP id PAA00891 for ; Sat, 28 Nov 1998 15:29:52 -0600 (CST) Received: from localhost by dilbert.cs.tamu.edu (8.9.1/8.9.1) with SMTP id PAA05279 for ; Sat, 28 Nov 1998 15:29:32 -0600 (CST) X-Authentication-Warning: dilbert.cs.tamu.edu: gurudatt owned process doing -bs Date: Sat, 28 Nov 1998 15:29:32 -0600 (CST) From: Gurudatt Shenoy X-Sender: gurudatt@dilbert To: FreeBSD Hackers Subject: Packet Delay calculation Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I'm trying to see how much delay a packet typically suffers in the FreeBSD tcp/ip stack. This is what I thought of doing: trace the path of data flow from the place a socket write call is made to the place where the data is actually passed on to the network interface; place calls to microtime() at these locations and see the time difference between them. Could anyone point out potential problems with this approach? Any better alternatives? I'm rather new to this kind of thing and would appreciate some feedback. Thanks, Guru To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message