Switching to it in Warp produced a very significant speedup on real files.https://github.com/facebook/warp/pull/5
That's to be expected as warp will call it often and it will stay in the cache. For a program that calls it less often the other methods may be faster.
Basically, it depends on the use case.Glad to see it's helped in warp. Any idea how it is comparing to clang with this improvement?
