Agner's CPU blog

I posted this question on stackoverflow: https://stackoverflow.com/questions/77707238/looking-for-an-efficient-function-to-find-an-index-of-max-element-in-simd-vector. The answer claims, among others, that using "legacy BSF instruction (slow on AMD), instead of using GNU C __builtin_ctz on non-MSVC"...

Is there an obvious performance penalty in using array<Vec32uc, 8> instead of ContainerV<Vec32uc, 8>? One reason for this choice is https://godbolt.org/ not having vector_containers.h available.