K2 Compiler Performance Benchmarks and How to Measure Them on Your Projects
With the Kotlin 2.0.0 release drawing ever closer, the K2 compiler is now available for you to try! In this blog post, we explore the performance of the K2 compiler in various projects and give you the tools to collect your own performance statistics.
The road to the Stable version of the K2 compiler has been a long one. Since we made the decision to rewrite the compiler from scratch, we’ve added a new type inference algorithm, new JVM and JS IR (Intermediate Representation) backends, and finally, a new frontend.
The fundamental change in the frontend is the use of one unified data structure that contains more semantic information. This change makes compilation more efficient and helps IntelliJ IDEA analyze your Kotlin code.
The driving force behind this initiative was the desire to speed up the development of new language features, unify all platforms that Kotlin supports, and improve performance for Kotlin developers.
The Kotlin project that we used to run performance tests is available on GitHub. If you’d like to use it to collect your own statistics from your Kotlin projects, see Collect your own measurements.
