Get the latest tech news

TPDE-LLVM: Faster LLVM -O0 Back-End

5 years ago, @nikic wrote: I can’t say a 10% improvement is making LLVM fast again, we would need a 10x improvement for it to deserve that label. We recently open-sourced TPDE and our fast LLVM baseline back-end (TPDE-LLVM), which is 10-20x faster than the LLVM -O0 back-end with similar runtime performance and 10-30% larger code size. We support a typical subset of LLVM-IR and only target x86-64 and AArch64. Posting this here, as this might be interesting for the LLVM community – questions/c...

Rust code works in principle, but some popular crates use illegal (=not natively supported) vector types, which are not implemented yet. Besides some more IR features, our current plans include: DWARF support (we already have some prototype) and better register allocation than “spill everything”. If someone provides sufficient motivation, other big items are non-ELF platforms, non-small-PIC code models, and other targets.

Get the Android app