The binutils BFD linker is the most feature complete linker we have. Unfortunately it has one mis-feature, it's relative slowness. Emerging
new programming paradigms, for better or worse, amplify this problem, so much so that in recent years multiple new linker
implementations came about. They are all faster in one or another way, but always lack in feature parity, so they can't easily be used as
full replacement.
Instead of implementing yet another linker, when we already have a perfectly fine one, I set out to do something about the speed of it. This
presentation is about that status of that work