Porting to i686-msdosdjgpp

[Eric Bresie]

At the time of phoneME there was also some licensing concerns if it was to
be used on a specific (phone) platform which resulted in a lot of
fragmentation in the JavaME market with each platform having to implement
their own native flavor and consider licensing fees.

I believe the idea with openjdk and the assorted profiles and modularity
 related changes that phoneME (java ME) more or less became OBE open source
jdk.  The rational being rather than maintain a Java and JavaME flavor of
code base, have a single Java based (with modularize compact profile
running on embedded platforms.

From
https://www.electronicdesign.com/technologies/dev-tools/article/21801464/11-myths-about-embedded-java

“ In JDK 8, steps were taken to address this with the introduction of three
compact profiles, the smallest of which only requires a little over 10 MB
of storage. In JDK 9, scheduled for release in Spring 2017, the JDK will be
fully modularized so that application developers can select only the
modules they need to run their application. ”

On a related note
https://minexew.github.io/2021/04/10/phoneme.html

Eric

On Fri, Nov 19, 2021 at 6:05 AM Thomas Stüfe <thomas.stuefe at gmail.com>
wrote:

>
>
> On Fri, Nov 19, 2021 at 12:25 PM Magnus Ihse Bursie <
> magnus.ihse.bursie at oracle.com> wrote:
>
>> On 2021-11-17 14:27, Thomas Stüfe wrote:
>>
>>
>>
>> On Wed, Nov 17, 2021 at 1:57 PM Magnus Ihse Bursie <
>> magnus.ihse.bursie at oracle.com> wrote:
>>
>>>
>>>
>>> On 2021-11-16 07:19, Thomas Stüfe wrote:
>>>
>>>
>>>
>>> On Mon, Nov 15, 2021 at 11:09 PM David Holmes <david.holmes at oracle.com>
>>> wrote:
>>>
>>>> On 16/11/2021 12:06 am, gnufan42 wrote:
>>>> > David Holmes wrote:
>>>> >> I'd say it is technically impossible to port OpenJDK to DOS as you do
>>>> >> not have any of the necessary operating system support for threads,
>>>> >> synchronization, virtual memory, ....
>>>> >
>>>> >       Well, these difficulties are all overcame by the DJGPP project.
>>>> They use DPMI to let the code runs in 32-bit protected mode, they
>>>> implemented a lot of POSIX functions, including pthread. Otherwise I won't
>>>> be trying.
>>>>
>>>> I'd never heard of DJGPP but the pthread support still seems limited -
>>>> hard to find an accurate current description of what is actually
>>>> supported. So I would not say these difficulties are overcome :) This
>>>> will be an exceedingly complex and challenging project.
>>>>
>>>> Cheers,
>>>> David
>>>>
>>>
>>> I still think that the CVM (JavaME) may be better suited for the task.
>>> From the time I worked with it I remember that it was targeted to
>>> low-memory devices, its C code base was extremely portable, it was very
>>> configurable (important for embedded) etc. We ran it with green threading
>>> (like Loom today) and that worked. OpenJDK OTOH relies on native posix
>>> threads, and there is no easy way around that.
>>>
>>> I seemed to remember that Sun open-sourced JavaME in 2006. But I could
>>> not find the project page.
>>>
>>> https://en.wikipedia.org/wiki/Java_Platform,_Micro_Edition
>>> https://en.wikipedia.org/wiki/PhoneME
>>>
>>> PhoneME website seems defunct now. Does anyone know what happened with
>>> that project?
>>>
>>> Wikipedia has links to archive.org.
>>> https://archive.org/details/phoneme-svn.dump for the source code.
>>>
>>> /Magnus
>>>
>>>
>>>
>> Ah, I missed that. But seems this project is not actively maintained
>> anymore. Pity.
>>
>> Yes, it's a shame. I felt sorry for the source code, just lying rotting
>> there as a SVN dump on archive.org, so I installed subversion, made a
>> few scripts, and converted it to git and put it on Github. [1]
>>
>> That won't bring active development back, but at least the source code is
>> accessible for archaeological reasons. And maybe it'll be enough to get
>> someone to bring it back to life...
>>
>> /Magnus
>>
>> [1] https://github.com/magicus/phoneME
>>
>>
> This is awesome, Magnus! And with full commit history too.
>
> So many memories :)
>
>
> --
Eric Bresie
ebresie at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.java.net/pipermail/porters-dev/attachments/20211121/64f7c034/attachment.htm>