Received: from SOUTH-STATION-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA10837; Thu, 15 Feb 96 08:28:55 EST
Received: from liquor.cabi.net by MIT.EDU with SMTP
	id AA12300; Thu, 15 Feb 96 08:28:36 EST
Received: by liquor.cabi.net (8.7.3/8.7.3) id MAA21217; Wed, 14 Feb 1996 12:33:12 -0500
Resent-Date: Wed, 14 Feb 1996 12:33:12 -0500
Date: Wed, 14 Feb 1996 09:20:41 -0800
From: geh@mda.ca (Graeme Hiebert)
Message-Id: <9602141720.AA14725@igneous.mda.ca>
To: java-linux@java.blackdown.org
Subject: Re: Performance issues
X-Sun-Charset: US-ASCII
Resent-Message-Id: <"gig4_.0.P65.hfX8n"@liquor>
Resent-From: java-linux@java.blackdown.org
X-Mailing-List: <java-linux@java.blackdown.org> archive/latest/1144
X-Loop: java-linux@java.blackdown.org
Precedence: list
Resent-Sender: java-linux-request@java.blackdown.org

I've tried j2c, but it doesn't seem to work so well for me.

On the netscape vs. java front, I tried a test compile on my Linux
machine last night.  Unlike the Solaris versions, which ran at nearly
the same speed, I found that under Linux, there was quite a noticeable
difference - it took over eight minutes to compile using Netscape, but
only around seven using java.  Furthermore, Netscape was giving overflow
errors while trying to compile numeric expressions.

I can hardly wait to give Kaffe a shot tonight.

   -g

---
> To: java-linux@java.blackdown.org
> Subject: Re: Performance issues 
> From: "Yuk-Lun Wong" <ywong@hursley.ibm.com>
> 
> One way to *really* speed up your compiles is to compile the
> compiler to native code. Check out j2c on
> 
>     http://www.webcity.co.jp/info/andoh/java/j2c.html
> 
> The current version, although meant to be for the beta2 JDK, compiled
> the 1.0 sun.tools.javac.Main just fine.
> 
> Yuk-Lun.
> 
> 

