From owner-freebsd-hackers Wed Nov 5 14:06:49 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA12540 for hackers-outgoing; Wed, 5 Nov 1997 14:06:49 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA12531 for ; Wed, 5 Nov 1997 14:06:42 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.7/8.8.5) id RAA08809; Wed, 5 Nov 1997 17:06:17 -0500 (EST) From: "John S. Dyson" Message-Id: <199711052206.RAA08809@dyson.iquest.net> Subject: Re: FreeBSD shines..[Fwd: Re: semaphore speed] In-Reply-To: from =?ISO-8859-1?Q?Dag=2DErling_Coidan_Sm=F8rgrav?= at "Nov 5, 97 09:31:19 pm" To: dag-erli@ifi.uio.no (Dag-Erling Coidan Smørgrav) Date: Wed, 5 Nov 1997 17:06:17 -0500 (EST) Cc: mark@vmunix.com, freebsd-hackers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Dag-Erling Coidan Smørgrav said: > > [long discussion about benchmark results in Linux/FreeBSD deleted] > > [Tons of boring and meaningless benchmark results deleted] > > Has it occurred to any of you that your benchmark results are > irrelevant unless you run the Linux and FreeBSD benchmark on the > *exact* same machine? Only then can we start comparing figures. Even > then, you have to have a comparable degree of customization in each OS > to be able to draw any conclusions; it doesn't make sense to compare a > production FreeBSD installation with a custom kernel and a vanilla > RedHat installation with a generic kernel. > Most of the discussion is associated with the differences in perf between FreeBSD kernels. Frankly, I don't think that the FreeBSD-current and RH results show that much difference to even be concerned about. Microbenchmarks aren't that valuable for real world performance comparisons anyway. Since kernel developers on the mailing list have been a party to the discussions, I don't see a problem. I don't agree with gloating about the numbers, but they are odd. -- John dyson@freebsd.org jdyson@nc.com