Here are some things I find interesting about Sun's JDK 1.2.2 RC1:
* Their (Sun's) bin/.java_wrapper starts with:
#!/bin/sh
#
# java_wrapper_linux.sh $Revision: 1.14 $ $Date: 1999/08/31 00:11:41 $
All in all this file is exactly what I've checked into the Blackdown
CVS tree:
revision 1.14
date: 1999/08/31 00:11:41; author: kreilede; state: Exp; lines: +12 -8
branches: 1.14.2;
Fixed the annoying "Error: libjava.so not found." problem with links
to jre/bin or when executing ./java from jre/bin.
* jre/README.PRE-RELEASE, jre/README.linux
These are files Steve Byrne wrote for our (Blackdown) JDK 1.2
pre-releases, you'll even find Steves name in jre/README.linux:
README.linux for Java 2 SDK Version 1.2.1 for Linux, pre-release 1
Steve Byrne
29/8/1999
* jre/lib/i386/libawt.so, jre/lib/i386/native_threads/libhpi.so,
jre/lib/i386/green_threads/libhpi.so contain symbols of functions
I've added to the original Sun sources.
* Their native threads VM implements Thread.yield() as nop just like
I did for our JDK. The original code uses sched_yield().
* Their native threads VM uses the a pthread_cancel() based
Thread.interrupt() implementation. That's far different from the
original source code and also a technique I've implemented for the
Blackdown JDK months ago. I guess they're just missing our changes
from the last weeks, these changes improved stability a lot.
enough for now...
AFAIK Sun has all the right to use our code. But todays press release
is a slap in the face for us!
This press release makes it look like all was Suns/Inprises work and
that other (e.g. our) JDKs are just crap.
Juergen Kreileder, Kevin B. Hendricks
--
Blackdown Java-Linux Team
http://www.blackdown.org/java-linux/jdk1.2-status/
----------------------------------------------------------------------
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
