Why lib func in java is running faster than customized?

Question

I did some investigation for Integer.bitCount recently. I found a interesting result that Integer.bitCount is much faster than my own func, even the code is the same.

I thought is due to JIT, but I checked the document, and found the JIT is based on runtime strategy. It confused me.

public static void main(String[] args) {
    long sum = 0;
    long start, end;
    start = System.currentTimeMillis();
    for (int i = Integer.MIN_VALUE; i != Integer.MAX_VALUE; i++) {
        sum += bitCount(i);
        //sum += Integer.bitCount(i);
    }
    end = System.currentTimeMillis();
    System.out.println(sum);
    System.out.println(end - start);
}

private static int bitCount(int i) {
    i = i - ((i >>> 1) & 0x55555555);
    i = (i & 0x33333333) + ((i >>> 2) & 0x33333333);
    i = (i + (i >>> 4)) & 0x0f0f0f0f;
    i = i + (i >>> 8);
    i = i + (i >>> 16);
    return i & 0x3f;
}

// for bitCount result

68719476736
8715

// for Integer.bitCount result

68719476736
1892

Answer 1

Your benchmark is not accurate. But disregarding that, one reason is because Integer#bitCount is marked as @HotSpotIntrinsicCandidate. This means that the HotSpot JVM can replace the method body with assembly code to improve performance. From the annotation's source code:

The @HotSpotIntrinsicCandidate annotation is specific to the HotSpot Virtual Machine. It indicates that an annotated method may be (but is not guaranteed to be) intrinsified by the HotSpot VM. A method is intrinsified if the HotSpot VM replaces the annotated method with hand-written assembly and/or hand-written compiler IR -- a compiler intrinsic -- to improve performance. The @HotSpotIntrinsicCandidate annotation is internal to the Java libraries and is therefore not supposed to have any relevance for application code.

Try disabling intrinsics and running your test again; you should see a significant slowdown.