Out of the box LLVM optimizations won't do much if you put work in preoptimizing. I think this is even mentioned in docs. IIRC they say that the generic optimiz

Out of the box LLVM optimizations won't do much if you put work in preoptimizing. I think this is even mentioned in docs. IIRC they say that the generic optimizations only really work on naive IR generation and that more mature projects will probably not find them useful and instead create their own transforms - if you're mainly using the JIT then your approach seems the best. Running the pre-bundled optimizations is really a brute force approach to optimization. It works great for just getting things going but you outgrow it pretty quick.