Replacing a 32-bit loop counter with 64-bit introduces crazy performance deviations with _mm_popcnt_u64 on Intel CPUs
Replacing a 32-bit loop counter with 64-bit introduces crazy performance deviations with _mm_popcnt_u64 on Intel CPUs技术背景在使用 _mm_popcnt_u64 指令时,将 32 位循环计数器替换为 64 位会在 Intel CPU 上引入疯狂的性能偏差。这主要是由于 popcnt