From owner-freebsd-hackers Wed Sep 25 22:40:58 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id WAA18601 for hackers-outgoing; Wed, 25 Sep 1996 22:40:58 -0700 (PDT) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id WAA18554; Wed, 25 Sep 1996 22:40:54 -0700 (PDT) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.7.6/CET-v2.1) with SMTP id FAA12346; Thu, 26 Sep 1996 05:40:45 GMT Date: Thu, 26 Sep 1996 14:40:45 +0900 (JST) From: Michael Hancock To: dyson@FreeBSD.ORG cc: Jake Hamby , hackers@FreeBSD.ORG Subject: Re: Anyone tried HURD yet? In-Reply-To: <199609260241.VAA21054@dyson.iquest.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Wed, 25 Sep 1996, John S. Dyson wrote: > > > > If anyone else has tried HURD, I'd be interested in your opinions. I'm > > also curious if anyone is seriously using Lites. As an aside, I currently > > have _five_ OS's on four partitions of two SCSI hard drives of a single > > 486: Windows 95, NT, Linux, FreeBSD, and Solaris/x86, _all_ accessible > > from BootEasy. Adding HURD would make a grand total of six. Is this some > > sort of world record? :-) > > > I had recently tried out Lites (during one of my benchmarking runs.) It > is certainly an achievement in technology (IMO.) However, the performance > of certain ops was quite slow. I passed around some performance numbers > to -core, but have lost my harddrive with the results. > > Please don't take my performance comments as putting down the Mach or > Lites projects -- otherwise they are very interesting -- but OS perf > under certain circumstances was very far behind FreeBSD/Linux. There > are many reasons for using the Mach based OSes, but perf is probably > not going to be one of them. In particular I remember fork/exec as > being maybe 10x-20x slower... Perhaps someone in -core could forward > my results that I had measured? These came off a Chorus page. How did they compare to these numbers? 7 microseconds Interrupt Latency 8 microseconds Context Switch on i486 at 50 Mhz 28 microseconds light-weight RPC on i486 at 50 Mhz See, http://www.chorus.com/Products/Datasheets/nucleusv3.html Regards, Mike Hancock