From owner-freebsd-hackers Tue Nov 7 12:26:19 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id MAA01871 for hackers-outgoing; Tue, 7 Nov 1995 12:26:19 -0800 Received: from who.cdrom.com (who.cdrom.com [192.216.222.3]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id MAA01866 for ; Tue, 7 Nov 1995 12:26:17 -0800 Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211]) by who.cdrom.com (8.6.12/8.6.11) with ESMTP id MAA26894 for ; Tue, 7 Nov 1995 12:24:11 -0800 Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA18215; Tue, 7 Nov 1995 13:19:04 -0700 From: Terry Lambert Message-Id: <199511072019.NAA18215@phaeton.artisoft.com> Subject: Re: %time spend in ip packet processing To: guido@gvr.win.tue.nl (Guido van Rooij) Date: Tue, 7 Nov 1995 13:19:04 -0700 (MST) Cc: hackers@FreeBSD.org In-Reply-To: <199511071915.UAA26991@gvr.win.tue.nl> from "Guido van Rooij" at Nov 7, 95 08:15:06 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 670 Sender: owner-hackers@FreeBSD.org Precedence: bulk > How can I see how much time the kernel spends in IP pakcet processing? > top's output of %interrup doesnt include it so it seems. This > is on a 2.0R system. If you can get a time stamp facility and add funtion exit as well as entry tags (or do stack hacking like I did with -Gh's __peneter() calls in MSVC), then you can do kernel block profiling. Sounds like block profiling is what you need -- not statistical profiling, which is what the gprof stuff provides. There's a paper on ftp.sage.usenix.org on profiling this way. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.