SSA; loop optimisations

Unfortunately it seems that JKit does not currently implement any SSA (Single Static Assignment) transformation. This is annoying, because SSA would make the code movement optimisations I am looking at easier.

On another note, I read [1] this week. This paper looks at what benefit there would be in doing traditional loop transformations in a Java compiler, given that JVM typically do some optimisation in there JIT compiler anyway. The authors find that these optimisations are generally worthwhile, which suggests that my optimisations are also likely to be worthwhile. Traditional loop transformations like this will probably be my first step in working towards moving pure function calls out of loops. I am working with IR (JKIL) though, rather than bytecode.

  1. Simon Hammond and David Lacey, “Loop Transformations in the Ahead-of-Time Optimization of Java Bytecode,” in Proceedings of the 15th International Conference on Compiler Construction, ed. Alan Mycroft and Andreas Zeller (Springer Verlag, 2006)

Tags: , , , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: