LLVM optimizations are the overwhelming majority of the compilation bottleneck for us over at Feldera. We blogged about some of the challenges we faced here: ht

LLVM optimizations are the overwhelming majority of the compilation bottleneck for us over at Feldera. We blogged about some of the challenges we faced here: https://www.feldera.com/blog/cutting-down-rust-compile-times-from-30-to-2-minutes-with-one-thousand-crates https://www.feldera.com/blog/cutting-down-rust-compile-times... We almost definitely need to build a JIT in the future to avoid this problem.