Threaded Code Generation with a Meta-Tracing JIT Compiler

Abstract

Language implementation frameworks such as RPython and Truffle/Graal are effective tools for creating a high-performance language with lower effort than implementing from scratch. The two frameworks support only a single JIT compilation strategy, tracebased compilation and method-based compilation, but they have its own advantages and disadvantages. We proposed a meta-hybrid JIT compiler framework to take advantages of the two strategies as a language implementation framework. We also implemented a proof-of-concept framework called BacCaml.As a next step, in this position paper, we propose a new approach to realize a method-based baseline JIT compiler along with a tracebased JIT compilation. We aim to use it for further speed-up by preventing the path-divergence problem, which causes serious slowdown. We also show how to implement the baseline JIT compiler with minimal changes on top of RPython.

Two-level Just-in-Time Compilation with One Interpreter and One Engine

“…The generic interpreter is converted into an interpreter that includes two definitions -one for baseline JIT compilation and another is for tracing JIT compilation. To remove the task of manually writing redundant definitions in the methodtraversal interpreter [15], we internally generate both the method-traversal interpreter (for the baseline JIT compilation) and a normal interpreter (for tracing JIT compilation) from the generic interpreter definition.…”

“…We do not create two separated compilers in adaptive RPython but indeed generate two different interpreters -the one for the first level of compilation and another is for the second level. As first-level compilation, we support threaded code [3,13] in a meta-tracing compiler (we call this technique threaded code generation [15]). The second one is RPython's original tracing JIT compilation.…”

“…However, we currently handle branching guards only and leave type optimization to the last JIT tier 2. Note that threaded code generation[15] yields call and guard operations.In contrast to normal tracing JIT compilation, it can reduce the code size and compilation time by 80 % and 60 %.…”