https://hgpu.org/?p=26766
Can We Run in Parallel? Automating Loop Parallelization for TornadoVM