RE: [platform-swt-dev] SWT on 64-bit platforms

[Christophe_Cornu@xxxxxxxxxx]

Andrew, JK:
We encourage you and other people interested in 64-bit ports to use the following bugzilla report to discuss the issues and/or solutions you discover. Hopefully this PR will become a useful starter and follow up for 64 bit developers.
http://dev.eclipse.org/bugs/show_bug.cgi?id=27577
I have added to that PR the 9/12/2002 thread between Steve and Tom Tromey.

Chris

"Lah, JK" <jk.lah@xxxxxxxxx> Sent by: platform-swt-dev-admin@xxxxxxxxxxx 03/05/2003 05:23 PM Please respond to platform-swt-dev

To:        "'platform-swt-dev@xxxxxxxxxxx'" <platform-swt-dev@xxxxxxxxxxx>         cc:        "Lah, JK" <jk.lah@xxxxxxxxx>         Subject:        RE: [platform-swt-dev] SWT on 64-bit platforms

Thanks for responding, Andrew.

1. I have spent a few days so far to make changes from 'int' to 'long'.
There are tons of them!  I have a clean compile on the 'C' side, but I have
quite a way to go on the Java side.
2. From what I was able to find out, IBM is committed to support Itanium for
their WebSphere and is expected to release the WebSphere 5.0.1 with Itanium
support in mid April.  I have no details, though.  Are they going go support
Win64, Linux64, or both?  Do they plan to submit the code back to
Eclipse.org?  If anyone has more info, I would appreciate.

Given the possibility that IBM would do the work, my motivation has
decreased, but not eliminated until I find out for sure that they would do
it.

Andrew - I made my changes on the RC1 source.  And I was concentrating on
Linux/motif.  I'd be more than happy to send you my intermediate work (as I
said, the Java side still has tons of warnings), if you are interested.

JK

-----Original Message-----
From: Andrew Haley [mailto:aph@xxxxxxxxxx] 
Sent: Wednesday, March 05, 2003 3:04 AM
To: platform-swt-dev@xxxxxxxxxxx
Cc: Lah, JK
Subject: [platform-swt-dev] SWT on 64-bit platforms

Lah, JK writes:
 > Hi,
 > 
 >  
 > 
 > There was a mail thread on the topic around Sep 12, 2002, and no further
 > discussion since then, as far as I can figure out.
 > 
 >  
 > 
 > What is the current status?  Is anyone working on the project?  Is there
a
 > plan?
 > 
 >  
 > 
 > I have looked into the issue briefly.  So far, I figured the following:
 > 
 >  
 > 
 > - Change type of any Java variable that holds native pointer from "int"
to
 > "long".
 > 
 > - Make corresponding JNI function prototype changes.
 > 
 > - Keep memory management of native side separate from JVM.  In other
words,
 > native code is responsible for freeing memory it allocated.
 > 
 >  
 > 
 > I am sure I don't have my arms around the whole issue, but I have a
serious
 > interest in getting this done one way or other.
 > 
 > Please give me guidance on how to proceed.

I am intending to start working on this soon.  I agree with the basic
plan of extending the native pointers from int to long, but I'm not
sure about moving the memory management to the native side.

I was intending just to sit down and do it.  It may be that we could
co-operate in some way.  Perhaps you could fix the memory management
code in the way that you intend while I'll find and fix all the
pointers and do the appropriate casts.

Andrew.
_______________________________________________
platform-swt-dev mailing list
platform-swt-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/platform-swt-dev
_______________________________________________
platform-swt-dev mailing list
platform-swt-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/platform-swt-dev