A few months ago someone wrote a much faster -O0 backend for LLVM, though it seems it didn't get much attention upstream: https://discourse.llvm.org/t/tpde-llvm-10-20x-faster-llvm-o0...

Discussion: https://news.ycombinator.com/item?id=45072481

There are also codegen projects that don't use LLVM IR that are faster like Cranelift: https://github.com/bytecodealliance/wasmtime/tree/main/crane...