Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Received-SPF: pass (nike.apache.org: domain of alexei.fedotov@gmail.com
 designates 74.125.82.177 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:from:date:message-id:subject:to
         :content-type;
        b=s9DCjBcXwVx4Q/On/uMahogTWda5rBPsVxzrffOOnKs72X8+D69chygAaHH8pcH8SU
         V5CkZAE05EeR1jpJyxncVNufDA7sq/vVgcBcT++TGr9XF2zLIIe3cG29F0czMu/OaDQk
         gOAvUa9TY+Plyn+Zuk1uXN97mXZFfSGxCfUG4=
MIME-Version: 1.0
In-Reply-To: <2bf63a68a4e1e8bbbecdc58856090c08@www2-mail.volny.cz>
References: <2bf63a68a4e1e8bbbecdc58856090c08@www2-mail.volny.cz>
From: Alexei Fedotov <alexei.fedotov@gmail.com>
Date: Thu, 11 Nov 2010 23:08:27 +0300
Message-ID: <AANLkTikcoUscbN0D_4PqQojFpNu_1p3B7bOa4To=fLcv@mail.gmail.com>
Subject: Re: Dalvik bytecode version of Harmony's DRLVM feasible as the new
 JVM?
To: dev@harmony.apache.org
Content-Type: multipart/alternative; boundary=0016364d236de5bea90494cc8b80

--0016364d236de5bea90494cc8b80
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hello Paul,
Dalvik VM is APL licensed and runs on x86_64. What is the point to compete?

As years ago I still see an option for DRLVM to serve as a source of high
quality portable modules (e.g. GC, asm utils, logging, crash handler)
reusable by several VM implementations (e.g. Dalvik, QEMU, Parrot, etc)
available as linux packages. Today you are reading this. Tomorrow you are
known as the author of the DRLVM-based IR management module which has just
replaced GIMPLE in gcc trunk. Your way to the glory of zlib is simple - jus=
t
work on accepting your module into as many open source projects, as you can=
,
and it inevitably becomes brilliant and widely used.

We may also create JVM benchmarks and certification testsuites from our
exisitng testbase with common ASF "field of use" restriction: those who
tricked ASF with licenses cannot publish results and never get certified.

--
With best regards / =D1=81 =D0=BD=D0=B0=D0=B8=D0=BB=D1=83=D1=87=D1=88=D0=B8=
=D0=BC=D0=B8 =D0=BF=D0=BE=D0=B6=D0=B5=D0=BB=D0=B0=D0=BD=D0=B8=D1=8F=D0=BC=
=D0=B8,
Alexei Fedotov / =D0=90=D0=BB=D0=B5=D0=BA=D1=81=D0=B5=D0=B9 =D0=A4=D0=B5=D0=
=B4=D0=BE=D1=82=D0=BE=D0=B2,
http://dataved.ru/
+7 916 562 8095


On Thu, Nov 11, 2010 at 6:33 AM, Paul Anderson <p_j_anderson@volny.cz>wrote=
:

> In view of the current problem with the TCK and software patents,
> here's an idea - excuse me if it's come up before or if it's totally
> wacky.
>
> The Dalvik VM was written for phones and similar small devices. For
> servers and desktops, how about making an implementation of the
> DRLVM that runs Dalvik bytecode?
> The Apache name and ethos has great credibility, and a good
> patent-unencumbered VM is desperately needed as an
> Apache-webserver-like de-facto standard.
> For various VM languages, the Android technique of compiling them or
> their interpreters into Java bytecode then transforming to .dex
> could be used in the interim as is already done on Android, followed
> later by native compilation into Dalvik bytecode - so JRuby, JPython
> etc could be adapted quickly to work directly on such a VM, and with
> a bit more effort, even languages that dynamically generate and load
> bytecode.
>
> For the Java language itself, Dalvik-ized Java platform classes
> taken from Harmony could be a separate project with an open TCK
> corresponding to Java 5 and 6, and a Java-to-dex compiler added
> later.
>
> Could it be done? And should it? Maybe James Gosling could advise,
> since he is free!
>
>
>

--0016364d236de5bea90494cc8b80--