Probably don't need to implement because LLVM automatically does this but if possible extended branchless optimisation
Probably don't need to implement because LLVM automatically does this but if possible extended branchless optimisation