andjo403 added a comment. so it is possible that the large improvement in compile time is due to we broke the pipeline and due to this less optimization was possible/executed. Repository: rL LLVM https://reviews.llvm.org/D38154