From owner-freebsd-hackers Fri May 16 14:34:20 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA20970 for hackers-outgoing; Fri, 16 May 1997 14:34:20 -0700 (PDT) Received: from caipfs.rutgers.edu (root@caipfs.rutgers.edu [128.6.155.100]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA20962 for ; Fri, 16 May 1997 14:34:12 -0700 (PDT) Received: from jenolan.caipgeneral (jenolan.rutgers.edu [128.6.111.5]) by caipfs.rutgers.edu (8.8.5/8.8.5) with SMTP id RAA04202; Fri, 16 May 1997 17:31:45 -0400 (EDT) Received: by jenolan.caipgeneral (SMI-8.6/SMI-SVR4) id RAA00859; Fri, 16 May 1997 17:30:11 -0400 Date: Fri, 16 May 1997 17:30:11 -0400 Message-Id: <199705162130.RAA00859@jenolan.caipgeneral> From: "David S. Miller" To: kuku@gilberto.physik.RWTH-Aachen.DE CC: hutton@isi.edu, hackers@FreeBSD.ORG In-reply-to: <199705161601.SAA01821@gil.physik.rwth-aachen.de> (message from Christoph Kukulies on Fri, 16 May 1997 18:01:41 +0200 (MEST)) Subject: Re: throughtput measurements for fast ethernet Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk From: Christoph Kukulies Date: Fri, 16 May 1997 18:01:41 +0200 (MEST) Interesting. a) I don't know how efficient the bzero() is (inline? #idef KERNEL?) over a statementwise zeroing of a 20 byte structure and why this. It's just not something you do in a time critical path. b) Could you elaborate to a mundane how TCP latency is defined? I know the term 'interrupt latency' being defined as the time from the occurence of an interrupt to the first statement serving the interrupt. Check out lat_tcp.c from lmbench for one perspective on how this can be defined. ---------------------------------------------//// Yow! 11.26 MB/s remote host TCP bandwidth & //// 199 usec remote TCP latency over 100Mb/s //// ethernet. Beat that! //// -----------------------------------------////__________ o David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><