From owner-freebsd-hackers@FreeBSD.ORG Sun Aug 17 14:16:30 2003 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CFA7937B401; Sun, 17 Aug 2003 14:16:30 -0700 (PDT) Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2219943F75; Sun, 17 Aug 2003 14:16:30 -0700 (PDT) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.12.9/8.12.9) id h7HLGRYO026554; Sun, 17 Aug 2003 16:16:27 -0500 (CDT) (envelope-from dan) Date: Sun, 17 Aug 2003 16:16:27 -0500 From: Dan Nelson To: Pawel Jakub Dawidek Message-ID: <20030817211627.GC2653@dan.emsphone.com> References: <20030815111856.GN395@garage.freebsd.pl> <20030817011545.GW395@garage.freebsd.pl> <20030817015029.GA2653@dan.emsphone.com> <20030817020236.GX395@garage.freebsd.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030817020236.GX395@garage.freebsd.pl> X-OS: FreeBSD 5.1-CURRENT X-message-flag: Outlook Error User-Agent: Mutt/1.5.4i cc: freebsd-hackers@freebsd.org cc: Robert Watson cc: Buckie Subject: Re: GEOM Gate. X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 17 Aug 2003 21:16:31 -0000 In the last episode (Aug 17), Pawel Jakub Dawidek said: > On Sat, Aug 16, 2003 at 08:50:30PM -0500, Dan Nelson wrote: > +> What kind of hardware were you using? 2.5MB/sec NFS sounds > +> abysmal. > > I don't think it is a hardware problem. > > Run this test on 5.1-CURRENT with: > > options INVARIANTS > options INVARIANT_SUPPORT > options WITNESS > > and without any network and NFS optimization. Yes, another test system (P5 MMX/233 laptop) drops its NFS throughput from 4MB/sec to 2.5MB/sec when built with those flags. So I'm assuming you're doing these benchmarks on a comparable system? I think this just demonstrates that you should not run benchmarks with all your debugging flags enabled :) Most people will not be running production systems with WITNESS, and parts of the kernel that bog down under the heavy load of WITNESS may work just fine on a regular kernal config. -- Dan Nelson dnelson@allantgroup.com