BZ #172: icedtea-6-jre-cacao: cacao badly broken on armhf

Status fields:

Peter Green posted the following bug-report to the Debian bug-tracker:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=688703

The following trivial test program fails with cacao on armhf

import java.math.*;

public class test {
 public static void main(String [] args) {
   System.out.println(3.0);
 }
}

root@debian:/# javac test.java
root@debian:/# java test
3.0
root@debian:/# java -cacao test
Exception in thread "main" java.lang.IllegalArgumentException: disparate values
        at sun.misc.FDBigInt.quoRemIteration(FloatingDecimal.java:2795)
        at sun.misc.FloatingDecimal.dtoa(FloatingDecimal.java:833)
        at sun.misc.FloatingDecimal.<init>(FloatingDecimal.java:460)
        at java.lang.Double.toString(Double.java:196)
        at java.lang.String.valueOf(String.java:2985)
        at java.io.PrintStream.print(PrintStream.java:606)
        at java.io.PrintStream.println(PrintStream.java:745)
        at test.main(test.java:5)
root@debian:/#

armhf can never have worked. This will require some register allocator tweaking
*shudder*.

(In reply to comment #1)
> armhf can never have worked. This will require some register allocator tweaking
> *shudder*.

After a failed attempt at fixing the register allocation yesterday, I have to say that
it's somewhat surprising that this huge behemoth of a register allocator does even work
at all. It's just so unapproachable :(.

I'm thoroughly confused. I have a working cross-compiled CACAO/Classpath running in a
quantal armhf chroot. How is this possible?

Progress being made, looking very good. Interim check-ins at
<https://bitbucket.org/Ringdingcoder/cacao-dev/commits/all/tip/branch%28%22armhf%22%29>.

http://mips.complang.tuwien.ac.at/hg/cacao/rev/5c93a7b6bcd5