You're missing the point of GrallVm: Write the compiler in Java, not CPP. That greatly relaxes the curve and cross sectional skillet required to implement this

You're missing the point of GrallVm: Write the compiler in Java, not CPP. That greatly relaxes the curve and cross sectional skillet required to implement this function: byte [] compileToNative(byte[] byteCode);