From owner-freebsd-hackers  Sun Apr 14 14:27:37 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id OAA15880
          for hackers-outgoing; Sun, 14 Apr 1996 14:27:37 -0700 (PDT)
Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id OAA15875
          for <hackers@freebsd.org>; Sun, 14 Apr 1996 14:27:33 -0700 (PDT)
Received: (from root@localhost) by dyson.iquest.net (8.7.5/8.6.9) id QAA02314; Sun, 14 Apr 1996 16:27:23 -0500 (EST)
From: "John S. Dyson" <toor@dyson.iquest.net>
Message-Id: <199604142127.QAA02314@dyson.iquest.net>
Subject: Re: Unices are created equal, but...
To: nash@mcs.com
Date: Sun, 14 Apr 1996 16:27:23 -0500 (EST)
Cc: hackers@freebsd.org
In-Reply-To: <199604142039.PAA04761@zen.nash.org> from "Alex Nash" at Apr 14, 96 03:39:52 pm
Reply-To: dyson@freebsd.org
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> [ posted to hackers only ]
> 
> The execl throughput test was a complete massacre, with Linux more
> than an order of magnitude faster.  Does anyone familiar with the
> internals of exec know why?
> 
Did you use the old linux shared lib scheme?  It is much faster to
start-up than the new dynamic shared libs.  To compare the kernels,
I suggest linking both benchmarks -static, and you'll find that FreeBSD's
fork/exec performance is a little slower than Linux's (not an order
of magnitude :-)).  I have been working the issue, and think that I have
a few more efficiency improvements in my bag of tricks (running on
my own -current tree.)  I'll probably commit some of them before the
next snap.

John
dyson@freebsd.org