Mailing Lists: Apple Mailing Lists

Image of Mac OS face in stamp
 
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Universal JavaApplicationStub - Jar Bundler doesn't give it to me.



We appreciate getting your bugs and feedback on Java previews, and I'd like to explain more of what's going on here:

While the universal JavaApplicationStub that shipped in "Java 1.3.1 and 1.4.2 Release 2", and "J2SE 5.0 Release 3" works just fine now, both on Intel-based Macs and on PowerPC-based Macs, this stub depends on SPI support that is being deprecated.  Generally, our preference is that developers not use this stub, but instead use a later version that does not have this dependancy.

As far as "good" stubs go:

 1) The Xcode 2.2 tools shipped with a universal JavaApplicationStub that is OK going forward.
 2) The latest "J2SE 5.0 Release 4" developer previews (being built for PowerPC-based Macs) have an JavaApplicationStub that is nevertheless supported on Intel-based Macs.  We use a mechansim which detects the PowerPC stub early in the launch process, and routes control into native launch code.  The performance hit at launch time is minimal.

As you can see, it's a bit complex. 

The most straightforward message (and one that works going forward), is "please use the latest stub you have, even if it looks like it is PowerPC-only".

  M McDougall

On Jan 17, 2006, at 11:11 AM, Scott Palmer wrote:

I filed a bug (#4405542) and was told:

Engineering has determined that this issue behaves as intended based on the following information:

Please know that the PowerPC-only JavaApplicationStub will work correctly on Mac OS X for Intel (will run Java natively, not in Rosetta).

Thank you for taking the time to submit this report.
-----

Huh?  So what is it, is the JavaApplicationStub supposed to be PowerPC only, or is it supposed to be universal?  Seems odd that it would not be universal.  I never complained that the PowerPC stub wouldn't work.  But someone hinted that application startup time might be affected if an intel-based Mac had to fudge things to deal with the PowerPC launcher.

e.g.:
On 16-Jun-05, at 6:24 AM, Eric Albert wrote:
In article <email@hidden>,
 "William C. McCain" <email@hidden> wrote:

I guess you are saying that an
Intel machine (or the OS's loader) will "program check" on the foreign
code stub, but the program check handler (or loader) has been tweaked to
"fix it up" by substituting the proper stub.  Presumably, in future
releases of my products I should replace that "thin" stub with the
universal version, just to be a good citizen.

You're right on both counts there.



_______________________________________________
Do not post admin requests to the list. They will be ignored.
Java-dev mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:

This email sent to email@hidden

 _______________________________________________
Do not post admin requests to the list. They will be ignored.
Java-dev mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/java-dev/email@hidden

This email sent to email@hidden

References: 
 >Re: Universal JavaApplicationStub - Jar Bundler doesn't give it to me. (From: Greg Guerin <email@hidden>)
 >Re: Universal JavaApplicationStub - Jar Bundler doesn't give it to me. (From: Hsu <email@hidden>)
 >Re: Universal JavaApplicationStub - Jar Bundler doesn't give it to me. (From: Scott Palmer <email@hidden>)



Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2007 Apple Inc. All rights reserved.