[furyofantares]

> I think the optimizer recognizes modulo 2 and converts it to bitwise AND.

You don't have to guess, you could turn or O3 or look at the disassembly.

[lcampbell]

Interestingly, not only do both versions emit the same assembly, but clang both autovectorizes and unrolls the loop:

https://godbolt.org/z/qxsKWfz9s

[caspper69]

gcc does it even when it's not optimizing