Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Re: [9fans] java on dis

37 views
Skip to first unread message

cinap_...@gmx.de

unread,
Apr 28, 2013, 2:01:14 PM4/28/13
to
brucee has done this already:

http://doc.cat-v.org/inferno/java_on_dis/

--
cinap

cinap_...@gmx.de

unread,
Apr 28, 2013, 2:58:11 PM4/28/13
to
yes, didnt read the original proposal where it explicitely links
tho the paper. sorry for the noise and good luck.

--
cinap

OrangeCalx01

unread,
Apr 28, 2013, 11:21:21 AM4/28/13
to
A machine implementing a virtual machine implementing a virtual machine? Oh the latency! Lol. Why not integrate the jvm itself into dis?

Aram Hăvărneanu

unread,
Apr 28, 2013, 12:43:25 PM4/28/13
to
It is not a machine that emulates a machine, but rather a translator
that translates code for one machine to execute on another.

--
Aram Hăvărneanu

Charles Forsyth

unread,
Apr 28, 2013, 2:45:07 PM4/28/13
to
That is the basis for the project proposal: starting by bringing that code up to date.
Also, the Java side included a ton of Sun proprietary code that can't be used directly.
(Probably some of it was open-sourced with Java, but it's still a slog to work it out.)

Dmitry Kabak

unread,
Apr 29, 2013, 1:46:36 AM4/29/13
to
Yep, as said Aram, this will be actually a converter that converts compiled java classes into pure dis code. And there is also 'jvm' program that would be the class loader and linker, thread monitor and will perform other functions of the original JVM. So, I think that execution speed shouldn't be lower than that of limbo.


2013/4/28 <cinap_...@gmx.de>



--
Regards,
Dmitry Kabak

Marc Chantreux

unread,
Apr 29, 2013, 2:58:05 AM4/29/13
to
hello,

On Sun, Apr 28, 2013 at 11:21:21AM -0400, OrangeCalx01 wrote:
> A machine implementing a virtual machine implementing a virtual machine? Oh
> the latency! Lol. Why not integrate the jvm itself into dis?

I really don't know about virtual machines and maybe it's a stupid
question but i'm really currious about it and i really trust the plan9
expertise: why not port inferno on top of the jvm? is this a technical
nonsense? is there some performance/ressources issue or is it just an NIH
manifestation?

regards

Sergey Zhilkin

unread,
Apr 29, 2013, 4:07:47 AM4/29/13
to
There is http://www.vitanuova.com/dist/java.tgz - old java translating software. It's not legaly free.
I think, Charles knows more :)


2013/4/29 Marc Chantreux <kha...@phear.org>



--
С наилучшими пожеланиями
Жилкин Сергей
With best regards
Zhilkin Sergey

Bruce Ellis

unread,
Apr 29, 2013, 5:47:21 AM4/29/13
to
when this thread calms down a bit i'll post some suggestions and hopefully helpful hints - one of which is that there will be tears.

i'll respond to off list technical questions.

brucee

Dmitry Kabak

unread,
Apr 29, 2013, 7:15:47 AM4/29/13
to
Hey, I didn't know that sources exist. But anyway, my work is still really relevant because of the need to update lots of sources to support latest Java and Inferno. And also to make the whole thing open-sourced, as the code you posted contains scaring lines like:

This software is provided SOLELY for experiment and evaluation.
NO PORTION may be distributed in ANY FORM, or used commercially,
without executing appropriate licences with Vita Nuova and possibly,
for some portions, Sun Microsystems Inc. Vita Nuova can advise on the
latter once evaluation is complete


2013/4/29 Sergey Zhilkin <szhi...@gmail.com>



--
Regards,
Dmitry Kabak

Dmitry Kabak

unread,
Apr 29, 2013, 7:22:00 AM4/29/13
to
I'm happy to see the original author of that project here! Looking at sources I've already realized that there will be some tears, but I'm absolutely ready for long hours of really-low-level code, awful debugging and strange and mystical things :)
Looking forward for your questions and suggestions!


2013/4/29 Bruce Ellis <bruce...@gmail.com>



--
Regards,
Dmitry Kabak

erik quanstrom

unread,
Apr 29, 2013, 9:39:34 AM4/29/13
to
On Mon Apr 29 07:16:48 EDT 2013, user...@gmail.com wrote:

> Hey, I didn't know that sources exist. But anyway, my work is still really
> relevant because of the need to update lots of sources to support latest
> Java and Inferno. And also to make the whole thing open-sourced, as the
> code you posted contains scaring lines like:
>
> This software is provided SOLELY for experiment and evaluation.
> NO PORTION may be distributed in ANY FORM, or used commercially,
> without executing appropriate licences with Vita Nuova and possibly,
> for some portions, Sun Microsystems Inc. Vita Nuova can advise on the
> latter once evaluation is complete

the first (open source) license must be choosen. gsoc *requires* open source.

http://www.vitanuova.com/inferno/licence.html

> There is http://www.vitanuova.com/dist/java.tgz - old java translating
> software. It's not legaly free.
> I think, Charles knows more :)

and just to reiterate, if it's not free, it can't be used for gsoc.

- erik

Charles Forsyth

unread,
Apr 29, 2013, 10:02:47 AM4/29/13
to

On 29 April 2013 09:07, Sergey Zhilkin <szhi...@gmail.com> wrote:
. It's not legaly free.

Dmitry already has a copy of the parts that are free.

Charles Forsyth

unread,
Apr 29, 2013, 10:05:35 AM4/29/13
to
On 29 April 2013 15:02, Charles Forsyth <charles...@gmail.com> wrote:
Dmitry already has a copy of the parts that are free.

and so does everyone else: 

>I put it on inferno-j2d.googlecode.com ...

several months ago (it was announced on the inferno list). That's the basis for potential GSoC work.

Charles Forsyth

unread,
Apr 29, 2013, 10:18:59 AM4/29/13
to

On 29 April 2013 15:05, Charles Forsyth <charles...@gmail.com> wrote:
>I put it on inferno-j2d.googlecode.com ...

several months ago (it was announced on the inferno list).

Dmitry had asked about it then, and having looked at it for the first time in years, I realised it had some interesting code to study,
and if anyone was keen enough to try to get it going, there was no reason to be obstructive.
(It's not a trivial task, because I had to remove the Sun code and the JVM has changed, amongst other hurdles.)
Most or all of that Sun code will be in the open-source Java, but I had no time to work that out. I left empty files to help track them down.
0 new messages