Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Odyssey Web Browser: Solve PPC Webkit Javascript Engine Endianess Problems inherited from Webkit X86! #1

Open
raf1024kb opened this issue Feb 8, 2016 · 12 comments

Comments

@raf1024kb
Copy link

- Brief description and general target of the Bounty:

Webkit Javascript Engine for PPC has a great number of endianess issues, so we need a skilled developer (or a joint team of developers) to perform a hunt search&destroy party of all issues by finding it one by one.

It is a tedious job to do, so it needs to be properly rewarded by a reasonable amount of money, hence the creation of this bounty.

If we Amigans want next release of Odyseey Web Browser being usable with all sites featuring new versions of Javascript, we need:

  1. Skilled developers to get this task assigned. Are you there?

We need also...

  1. Money to keep alive our platform, so donate, donate, donate!

- Here follows technical description taken from aros-exec.org site:


2/2`of application related bounties proposal:


Solve the endianess issue in Odyssey's JavaScript engine to allow using it on big endian CPUs. One of the approaches
can be to bring in the minimal set of changes from Leaopard WebKit repository to Odyssey repository.

The leopard-webkit project: https://sourceforge.net/projects/leopard-webkit/files/600/Sources/

Validation:

  • The non-JIT JavaScript engine for a selected Big Endian system (AmigaOS 3.x, AROS 68k, MorphOS or AmigaOS 4) works
    up to same level as non-JIT JavaScript version for x86 processors.
  • Execution speed is not part of comparison
  • The changes are made in a way allowing easy re-basing with newer versions of WebKit
  • Changes need to be integrated to https://github.com/deadwood-pl/OdysseyWebBrowser and released with license
    compatible with Odyssey
  • A binary of Odyssey for a selected Big Endian system is delivered to general usage

Posted on: 12/8 13:14


http://aros-exec.org/modules/newbb/viewtopic.php?post_id=97679#forumpost97679

@raf1024kb raf1024kb changed the title Solve PPC WebKit Javascript Engine Endianess Problems inherited from Webkit X86! Solve PPC Webkit Javascript Engine Endianess Problems inherited from Webkit X86! Feb 8, 2016
@raf1024kb raf1024kb changed the title Solve PPC Webkit Javascript Engine Endianess Problems inherited from Webkit X86! Odyssey Web Browser: Solve PPC Webkit Javascript Engine Endianess Problems inherited from Webkit X86! Feb 8, 2016
@AbelSoul
Copy link

AbelSoul commented Feb 8, 2016

Hi folks,
Really sorry I haven't been able to contribute so far. I've been tied up with other stuff over the last few weeks and only tried to set up my A1 again over the weekend. Unfortunately it wouldn't boot but I'm hoping I just need a new CMOS battery, which I'll try to pick up this week.
Cheers,Robert

On Monday, 8 February 2016, 12:32, raf1024kb <notifications@github.com> wrote:

Title says all!—
Reply to this email directly or view it on GitHub.

@kas1e
Copy link
Collaborator

kas1e commented Feb 8, 2016

Title says all!

I feel myself dumb , but what you trying to say ?:) Did you solve those welknown endianes issues ? Or someone else already solve that welknown issues in webkit repo and you want our attention to merge them ?:)

@raf1024kb
Copy link
Author

Bountysource needed official issue report on Github to create a new bounty,
so I joined OWB repository and created one.
If you want it assigned to you or to a designed team just tell me and I
will assign it to you.
Else you can edit Issue and assign it to yourself.

Ciao,

Raf
Il 08/feb/2016 14:32, "kas1e" notifications@github.com ha scritto:

Title says all!

I feel myself dumb , but what you trying to say ?:) Did you solve those
welknown endianes issues ? Or someone else already solve that welknown
issues in webkit repo and you want our attention to merge them ?:)


Reply to this email directly or view it on GitHub
#1 (comment)
.

@kas1e
Copy link
Collaborator

kas1e commented Feb 8, 2016 via email

@raf1024kb
Copy link
Author

Well... Anyone should contribute with his own abilities in keeping alive
AmigaOS-like systems and a valuable browser to keep alive our dream.

Deadwood asked if someone was so interested enough in OWB to create a
Bounty in this thread:

http://amigaworld.net/modules/news/article.php?storyid=7623

Well... As long none created any as of today, then I just thought I could
open it by myself!

So I committed it!
That's it.

Was I wrong, Kas1e?

I was mailing chatting with Deadwood in past days and he wanted me to use
Power2People site...
But I prefer use BountySource that is worldwide visible and could attract
also money from big firms like IBM who want to keep alive PPC platform.

By the way Kas1e, did you spot Bountysource IBM bounty for creating LUA
PPC64 version? They offer 5000 US$.

Ciao,

RAF
Il giorno 08/feb/2016 16:36, "kas1e" notifications@github.com ha scritto:

Bountysource needed official issue report on Github to create a new
bounty,
so I joined OWB repository and created one.
If you want it assigned to you or to a designed team just tell me and I
will assign it to you.
Else you can edit Issue and assign it to yourself.

I was just in hope it is something already done , not just about making a
bounty :)


Reply to this email directly or view it on GitHub
#1 (comment)
.

@raf1024kb
Copy link
Author

I updated start comment to better explain nature of Bounty!

@kas1e
Copy link
Collaborator

kas1e commented Feb 9, 2016

Was I wrong, Kas1e?

Probably yes.

Beause for first we should find person who will do so. Without, money will be only collecting dust, and every donor will be angree as nothing will happens.

There is alos no limit is set, so that mean no one will know how much need, and for who. So, my imho it was too fast to create such a bounty.

I was mailing chatting with Deadwood in past days and he wanted me to use
Power2People site...
But I prefer use BountySource that is worldwide visible and could attract
also money from big firms like IBM who want to keep alive PPC platform.

Probably he was right, power2people was better choice, as it 100% prove to be ok and clear enough in money trasnfer. How all that crap works with BountySource no one know from us know, and i can't bet on it. So, from my imho, choice of bounty site was also wrong.

You probably better before find someone, then discuss it with everyone, and only after that, when person is find, and know what to do, and know how much he want, create a bounty.

Imho, you was too fast. That money will colelct dust :)

@raf1024kb
Copy link
Author

I wonder if you Amigan people prefer blabbering and continue crying over
split milk, or rather you want to do some truly concrete to solve our
situations.

Well, I did something concrete by putting money in the face of developers.

Now feel free to blame me for this guilty sin!

Developers in Amigaland are few and in the community we all know who are
the best ones, so it will not difficult to assign bounty task to a good
programmer.

Else we will continue facing a increasing divide with our browser
javascript engine and the rest of the world.


Perhaps speaking of another topic of great interest, I know that in DirOpus
5.92 for MorphOS are still present three main bugs and none wants to solve
it, as MorphOS team rather prefer to improve Ambient GUI Windows Manager to
became similar to Directory Opus, than solving these three trivial bugs.

But at this very moment Directory Opus is a really killer application with
unsurpassed features on any existing computing platform, and we need it
running on MorphOS at its best performance!

I can't ask it to you, Kas1e to commint this MOS Diropus 5.92 bug hunting,
as I know of you as AmigaOS 4 developer and not MorphOS one.

Then sooner or later someone will be forced to start a bounty for that bugs
too...

I am ready for that!

Then we will see if there will be any volunteers to finally solve these
MorphOS Diropus problems when developers could facing proper credits in
their faces.

Do you know if Bountysouce is compatible with Sourceforge issue reports
too?

If answer is positive I could start immediately a new bounty!

Ciao,

Raf
Il 09/feb/2016 09:01, "kas1e" notifications@github.com ha scritto:

Was I wrong, Kas1e?

Probably yes.

Beause for first we should find person who will do so. Without, money will
be only collecting dust, and every donor will be angree as nothing will
happens.

There is alos no limit is set, so that mean no one will know how much
need, and for who. So, my imho it was too fast to create such a bounty.

I was mailing chatting with Deadwood in past days and he wanted me to use
Power2People site...
But I prefer use BountySource that is worldwide visible and could attract
also money from big firms like IBM who want to keep alive PPC platform.

Probably he was right, power2people was better choice, as it 100% prove to
be ok and clear enough in money trasnfer. How all that crap works with
BountySource no one know from us know, and i can't bet on it. So, from my
imho, choice of bounty site was also wrong.

You probably better before find someone, then discuss it with everyone,
and only after that, when person is find, and know what to do, and know how
much he want, create a bounty.

Imho, you was too fast. That money will colelct dust :)


Reply to this email directly or view it on GitHub
#1 (comment)
.

@kas1e
Copy link
Collaborator

kas1e commented Feb 9, 2016 via email

@raf1024kb
Copy link
Author

I just edited again bounty mission info, and added some technical specs on the work to do

@internetzel
Copy link

Here's my current patch from Leopard WebKit for JavaScriptCore.
It's been done on the branch "safari-601-branch" which was originally branched off at trunk revision 186671.

Built with the attached patch (with full optimization on gcc 5.3) on a 32 bit big endian PowerPC machine the JavaScriptCore C-Loop Interpreter passes all tests that I suppose it should pass.

Most chunks in the patch shouldn't be needed for MorphOS but someone who knows how to deal with it, like Fabien Coeurjoly, will easily be able to strip the unneeded parts out.

JavaScriptCore_601.zip

@raf1024kb
Copy link
Author

Thank you Internetzel!

I will pass this info to developers, so they will verify if it is working
properly and then declare you eligible to get bounty.

Ciao,

Raf
Il 01/mar/2016 21:44, "internetzel" notifications@github.com ha scritto:

Here's my current patch from Leopard WebKit for JavaScriptCore.
It's been done on the branch "safari-601-branch" which was originally
branched off at trunk revision 186671.

Built with the attached patch (with full optimization on gcc 5.3) on a 32
bit big endian PowerPC machine the JavaScriptCore C-Loop Interpreter passes
all tests that I suppose it should pass.

Most chunks in the patch shouldn't be needed for MorphOS but someone who
knows how to deal with it, like Fabien Coeurjoly, will easily be able to
strip the unneeded parts out.

JavaScriptCore_601.zip
https://github.com/deadwood-pl/OdysseyWebBrowser/files/153546/JavaScriptCore_601.zip


Reply to this email directly or view it on GitHub
#1 (comment)
.

jaokim added a commit to jaokim/odyssey that referenced this issue Apr 5, 2017
…ser#1).

Added methods that differ from the patch to the current version. Added various #warnings where I found stuff needing more attention -- lots of unimplemented and broken stuff in those areas.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants