Categorically-speaking, this forum may not be the best place to ask this question. However, rust tends to attract the best coders, so perhaps somebody knows the answer to this question.
When using Intel CPUs, I would pass
sse4.2, and friends into the rustc's compile arguments. With the new apple silicon, such features do not seem to exist (at least, with the same name). What would be the equivalent optimizations in the aarch64-apple-darwin cpu context? Without this, many of rust's cryptographic crates suffer significant performance penalties due to non-explicit cpu optimizations