I was just wondering when benchmark testing (
Bencher) would become stable. Right now, I still have to use nightly rust or else
Bencher isn’t usable (and
I can’t answer your question, but check out criterion, which provides benchmarking on stable.
I’ve been slowly migrating everything over to Criterion. It’s a little bit more of a pain to use than the standard benchmark harness, but it works on stable, is configurable and is more principled in how it goes about benchmarking. I wrote
critcmp to deal with some of the UX issues I’ve experienced with Criterion.
It may be better to ask in internals about it:
but the whole situation looks weird. From that thread it seems like it’s getting pushed back waiting for something better to come along, but nothing better is being planned.