values types for Java

[Adam Megacz]

Andrew Haley <aph@redhat.com> writes:
> J.-D. Choi, M. Gupta, M. Serrano, V. C. Sreedhar, and S. Midki.
> Escape analysis for Java.  In Object-Oriented Programming, Systems,
> Languages, and Applications (OOPSLA), 1999. Languages (POPL), 2002.
> http://www.research.ibm.com/jalapeno/publication.html#oopsla99_escape

> I would like to add this to gcj.

Two problems for use with GCJ:

1. They use interprocedural analysis, which is not compatible with
   separate compilation.  In a JVM, the compiler can trace through the
   implementations of all methods which are called; gcj cannot do that
   at compile time -- so you have no way of knowing if calling
   foo.bar() causes foo to stash a reference to itself somewhere else.
   A JVM can simply inspect the implementation of the bar() method.

2. This is overly conservative compared to Per's suggestion (I'm not
   sure if you're proposing it as a replacement for stack allocation
   or just an intermediate step).

  - a

-- 
"Education is not filling a bucket but lighting a fire." -- WB Yeats