From owner-freebsd-java Sat Apr 8 8: 9: 5 2000 Delivered-To: freebsd-java@freebsd.org Received: from ares.trc.adelaide.edu.au (ares.trc.adelaide.edu.au [129.127.246.5]) by hub.freebsd.org (Postfix) with ESMTP id B49AA37B62D for ; Sat, 8 Apr 2000 08:09:00 -0700 (PDT) (envelope-from glewis@ares.trc.adelaide.edu.au) Received: (from glewis@localhost) by ares.trc.adelaide.edu.au (8.9.3/8.9.3) id AAA17102; Sun, 9 Apr 2000 00:38:50 +0930 (CST) (envelope-from glewis) From: Greg Lewis Message-Id: <200004081508.AAA17102@ares.trc.adelaide.edu.au> Subject: Re: FYI: Object creation test In-Reply-To: <38EF41EB.D419129F@jollem.com> from Ernst de Haan at "Apr 8, 2000 04:27:55 pm" To: Ernst de Haan Date: Sun, 9 Apr 2000 00:38:50 +0930 (CST) Cc: freebsd-java@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL70 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-java@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi Ernst, > FYI: I've tested the fresh JDK 1.2.2 for FreeBSD using some object > creation test program I wrote a while ago. I compared the results with the > JDK 1.1.8 (FreeBSD) and the Blackdown JDK 1.2.2 rc4 (Linux). All tested on > a Pentium II-400, 128 MB memory. The maximum and initial heap sizes were > set to 64M, and 1M objects were created. > > The results: > > --- Number of object creates per second -- > FreeBSD 1.1.8 222061 > Blackdown 1.2.2rc4 163253 (without JIT) > Blackdown 1.2.2rc4 434552 (with sunwjit) > FreeBSD 1.2.2 (pre) 160112 > Sun 1.2.2 Linux 170861 ** FAILED ** Interesting. So we are roughly the same as the Blackdown JDK. I wonder what difference compiling with some agressive optimisation makes. I'm also curious how the open source JITs perform in comparison to sunwjit with the Blackdown JDK. -- Greg Lewis glewis@trc.adelaide.edu.au Computing Officer +61 8 8303 5083 Teletraffic Research Centre To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-java" in the body of the message