🚀 The feature, motivation and pitch

Currently, the vectorize_with_alignment could handle arbitrary elements, but this could also cause some overhead.
To further improve the performance, we can add some branches to it, so eg

  // Fast path identical to the one added for the write version above.
  bool can_vec = ((addr & (WIDTH - 1)) == 0) && ((len & (VEC_SIZE - 1)) == 0);
  if (can_vec) {
    vec_op
  }

  // arbitrary num of elements supported logic now
  ...

I will have a pr for this soon

Alternatives

No response

Additional context

No response

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.