I see the intel compiler 17.0 (in Release configuration) translate C-runtime (like "strlen" or "memcpy") as inline functions used SSE/SSE2.
1) Is it possible to take a look on these function - I sure they are pretty good as samples - maybe somebody knows?
2) What is the efficiency of using SSE2 for such trivial as "strlen"?
- Intel® Advanced Vector Extensions (Intel® AVX)
- Intel® Streaming SIMD Extensions
- Parallel Computing