2 ms·
A wild guess on what he is trying to do: - Use Vector Processor instead of Scalar Processor, SIMD(SSE/AVX/etc) has 16 128-bit registers. In the wikipedia entry
by fabianhjr 10y ago
A wild guess on what he is trying to do:
- Use Vector Processor instead of Scalar Processor, SIMD(SSE/AVX/etc) has 16 128-bit registers. In the wikipedia entry of SSE it says that SSE2 has support for 16 chars.
Things that don't add up:
- This should be a compiler optimization
- This makes something readable unnecessarily complex
- s is mutating
- You could eliminate the inner loop and s