A typical solution

Here is a function that processes an array:

1
2
3
4
5
6
7
8

auto process_values(const std::vector<uint8_t> &vec)
{
    return std::count_if(
        vec.begin(),
        vec.end(),
        [](uint8_t x) { return x % 2 == 0; }
    );
}

Let's look at the generated assembly:

 1
 2
 3
 4
 5
 6
 7

process_values():
    vpbroadcastb    xmm1, byte ptr [rip + .LCPI0_1]
    vmovd   xmm2, dword ptr [rsi + rax]
    vpandn  xmm2, xmm2, xmm1
    add     rax, 4
    cmp     r8, rax
    jne     .LBB0_6

The compiler generates vectorized code that processes four values at a time.