Hello Folks, Are any of the below due to go into the OpenJDK?
Interface injection http://openjdk.java.net/projects/mlvm/subprojects.html#InterfaceInjection Continuations and stack introspection http://openjdk.java.net/projects/mlvm/subprojects.html#StackContinuations Tail calls and tail recursion http://openjdk.java.net/projects/mlvm/subprojects.html#TailCalls Tuples and value-oriented types http://openjdk.java.net/projects/mlvm/subprojects.html#TupleSignatures My googling reveals that two of these have been implemented, and dynamic class updates. The language I use - Scala - would benefit greatly from their inclusion. Hence why I enquire. I understand that the main issue would be requiring the other JVM vendors to also implement the bytecodes, so it's not only technically difficulty per se that inhibits someone from implementing it, i.e., someone could write it but if the other JVM vendors don't agree to it the effort would be wasted ... or is this incorrect? I've heard that the JVM is still faster than the CLR despite not having these features. An extra confirmation would be great. Unfortunately, I don't think I'd be able to implement any of the above myself, just saying in case the thought crossed someone's mind ;). I'd definitely give a dollar though *. On this point are there any other docs that offer an introduction to the JVM (other VMs). Currently I'm reading Christian Wimmer's ph.d. thesis http://www.ssw.uni-linz.ac.at/Teaching/PhDTheses/Wimmer/, which offers a very good introduction, and if that goes well http://www.ssw.uni-linz.ac.at/Teaching/PhDTheses/Kotzmann/ by Thomas Kotzmann. The wiki is quite deep, I'll leave that for another day. * http://mail.openjdk.java.net/pipermail/jdk7-dev/2009-June/000706.html —Mohamed
_______________________________________________ mlvm-dev mailing list [email protected] http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
