64 bit multiply with 128 bit result, compiles to native amd64 x86_64 instruction with gcc 9.4 -O3

mul64.c

typedef struct { uint64_t lo, hi; } mul64_t;

mul64_t mul64(uint64_t a, uint64_t b)
{
    mul64_t ret;
    __uint128_t c = (__uint128_t) a * (__uint128_t) b;
    ret.hi = c >> 64;
    ret.lo = c;
    return ret;
}

/*
mul64(unsigned long, unsigned long):
        mov     rax, rdi
        mul     rsi
        ret
*/