Implementing typed intermediate languages

Abstract

Recent advances in compiler technology have demonstrated the benefits of using strongly typed intermediate languages to compile richly typed source languages (e.g., ML). A typepreserving compiler can use types to guide advanced optimizations and to help generate provably secure mobile code. Types, unfortunately, are very hard to represent and manipulate efficiently; a naive implementation can easily add exponential overhead to the compilation and execution of a program. This paper describes our experience with implementing the FLINT typed intermediate language in the SML/NJ production compiler.We observe that a type-preserving compiler will not scale to handle large types unless all of its type-preserving stages preserve the asymptotic time and space usage in representing and manipulating types. We present a series of novel techniques for achieving this property and give empirical evidence of their effectiveness.

Precision in Practice: A Type-Preserving Java Compiler

“…Standard ML of New Jersey is an interactive runtime system and compiler based on a strongly-typed intermediate language called FLINT [18]. We extended the FLINT language of version 110.30 and implemented a new front end for Java class files.…”

“…The techniques of Shao, et al made the FLINT typed IL practical enough to use in a production compiler [18]. Although different type structures arise in our Java encodings, the techniques are quite successful.…”

The Wizard of TILT: Efficient?, Convenient, and Abstract Type Representations

“…If we decide to implement type constructors as an ADT [3], then we are able to make representation choices but lose the ability to pattern match. Code interfacing with the ADT is stylistically less appealing and we lose some compile-time checking (such as checks for redundant or nonexhaustive matches).…”

“…Their results show that these changes made compilation much faster [3]! How then, do we account for the drastically differing results between the superficially similar TILT IL and FLINT implementations?…”

Principled Scavenging

“…We have started implementing our type-safe GCs in the FLINT system [16], however, making the implementation realistic still involves solving the remaining problems (e.g., breadth-first copying, remembered sets, and data structures with cycles, which we still cannot support satisfactorily) thus is beyond the scope of this paper. Nevertheless, we believe our current contributions constitute a significant step towards the goal of providing a practical type-safe garbage collector.…”