Those are benchmarks which took way too long or short to execute a single iteration or benchmarks which changed in time anyway because of previous fixes.
I renamed those benchmarks so that they are now treated as "new" benchmarks.
The first is copied from https://github.com/apple/swift/pull/13930's
contribution (with a minor bug fix applied). The second is an
adaptation that tries to avoid creating copies and operate using
indices directly.
