Interpreter

Interpretation

Machine Code translation executed at execution time, so slow

A program that translates and executes code line by line, rather than translating the entire code into machine language at once like a compiler, taking each line of source code as input, translating it, executing it immediately, and then outputting the result

Interpreter Notion

Interpreter Property

Modern CPUs (especially Zen 2/3 and similar) benefit from Out-of-Order execution, Branch Prediction, Superscalar pipelines, and other features, making even well-implemented interpreters very fast; most of the benefits of JIT can also be implemented in interpreters

JIT: so you want to be faster than an interpreter on modern CPUs…

Since my previous blog entry about JIT compiler for PostgreSQL, sadly not much happened due to a lack of time, but still some things were done (biggest improvement was the port to ARM64, a few optimizations, implementing more opcodes…). But I am often asking myself how to really beat the interpreter… And on “modern” CPUs, with a well written interpreter, that’s far more complicated than many would imagine. So in order to explain all this and show how I am planning to improve performance (possibly of the interpreter itself too, thus making this endeavor self-defeating), let’s first talk about…