From owner-freebsd-java  Tue Sep  5  7:16:44 2000
Delivered-To: freebsd-java@freebsd.org
Received: from ntua.gr (achilles.noc.ntua.gr [147.102.222.210])
	by hub.freebsd.org (Postfix) with ESMTP id 418E237B423
	for <freebsd-java@FreeBSD.ORG>; Tue,  5 Sep 2000 07:16:09 -0700 (PDT)
Received: from netmode.ece.ntua.gr (dolly.netmode.ece.ntua.gr [147.102.13.10])
	by ntua.gr (8.9.3/8.9.3) with ESMTP id RAA13393;
	Tue, 5 Sep 2000 17:16:05 +0300 (EET DST)
Received: by netmode.ece.ntua.gr (Postfix, from userid 410)
	id 3853085C3; Tue,  5 Sep 2000 17:01:22 +0300 (EET DST)
Date: Tue, 5 Sep 2000 17:01:22 +0300
From: Panagiotis Astithas <past@netmode.ntua.gr>
To: SHUDO Kazuyuki <shudoh@muraoka.info.waseda.ac.jp>
Cc: freebsd-java@FreeBSD.ORG
Subject: Re: shuJIT 0.6.5 released
Message-ID: <20000905170122.A28494@netmode.ece.ntua.gr>
Reply-To: past@netmode.ntua.gr
References: <200009051343.WAA12926@cafe.muraoka.info.waseda.ac.jp>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0i
In-Reply-To: <200009051343.WAA12926@cafe.muraoka.info.waseda.ac.jp>; from shudoh@muraoka.info.waseda.ac.jp on Tue, Sep 05, 2000 at 10:43:27PM +0900
X-Organizational-Unit: Network Management and Optimal Design Laboratory
X-Organization: National Technical University of Athens, GREECE
X-Work-Phone: +30-1-772-1-450
X-Work-FAX: +30-1-772-1-452
Sender: owner-freebsd-java@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org

On Tue, Sep 05, 2000 at 10:43:27PM +0900, SHUDO Kazuyuki wrote:
> [Linpack benchmark]	http://www.netlib.org/benchmark/linpackjava/
> 
> This benchmark performs solution of linear equations.
> The method of solution is base on Gaussian elimination.
> 
> This benchmark is small and runs in a short time.
> Therefore JIT compilers (including IBM's) which adopt
> adaptive compilation techniques cannot make a good score.
> But, JIT compilers evaluated this time do not use
> those techniques in their default settings.
> 
> The followings were made on Pentium II 333 MHz
> and Blackdown JDK 1.2.2 FCS.
> 
>   OpenJIT 1.1.13	13.464
>   shuJIT 0.6.5		13.205
>   TYA 1.7v2		11.444
>   OpenJIT 1.1.14	10.899

Regarding the difference in the performance of OpenJIT 1.1.13 vs. 1.1.14,
would you attribute it in the following change?

(from www.openjit.org)
REFINE: Change the default value of max_stack_frames (400 -> 43). This value
depends on platforms, and several JDK 1.1.8 platforms didn't work with the 
old default value (400). We now chose the default value to minimize the 
warm-up time of OpenJIT.  The best value for JDK 1.2.2 (both of Blackdown 
and Sun) is "41" and that for JDK 1.1.8 is "43".

Also on this test, is the highest value the best?

-past


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-java" in the body of the message