https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97759
--- Comment #10 from gcc-bugs at marehr dot dialup.fu-berlin.de --- And maybe a related question: I know that an arithmetic implementation might auto-vectorize, but would a popcount implementation do that too? Since AVX512_BITALG (https://software.intel.com/sites/landingpage/IntrinsicsGuide/#text=popcnt&expand=4350) retro-introduced popcount for smaller vector lengths, what is with instruction sets like AVX2 and lower?
