https://hgpu.org/?p=17996
Accelerating Interpreted Programming Languages on GPUs with Just-In-Time Compilation and Runtime Optimisations