Is there any optimization for thread safety in for loop of Java?

Kidsunbo :

I have a snippet of code that change a counter in two threads. It's not thread safe because I didn't put any atomic variable or lock in the code. It gives the right result as I expected if the code only run once, but I want to run it for several times, so I put the code into a for loop. And the question is that only the first or the first two loops will generate the result I expect. For the rest of the loops, the results are always 0, which seems to be thread safe. Is there any inner operator in Java Virtual Machine resulting such thing?

I have tried change the number of loops, and the first one or two are always what I expect, but the others are 0 no matter how many loops there are.

Counter:

private static class Counter {
    private int count;

    public void increase() {
        count++;
    }

    public void decrease() {
        count--;
    }

    public int getCount() {
        return count;
    }
}

Person:

// This is just a thread to increase and decrease the counter for many times.
private static class Person extends Thread {
    private Counter c;

    public Person(Counter c) {
        this.c = c;
    }

    @Override
    public void run() {
        for (int i = 0; i < 100000; i++) {
            c.increase();
            c.decrease();
        }
    }
}

Main method:

public static void main(String[] args) throws InterruptedException {
    for (int i = 0; i < 10; i++) {
        Counter c = new Counter();
        Person p1 = new Person(c);
        Person p2 = new Person(c);
        p1.start();
        p2.start();
        p1.join();
        p2.join();
        System.out.println("run "+i+": "+c.getCount());        
   }
}

Output:

run 0: 243
run 1: 12
run 2: 0
run 3: 0
run 4: 0
run 5: 0
run 6: 0
run 7: 0
run 8: 0
run 9: 0

I don't know why the rest of the results are always 0. But I guess it's about the optimization of JVM. Is it right that the JVM optimizes the code when some loops have been done, and it omits the rest loops and always gives 0 as answer?

SirFartALot :

I think the JVM is optimizing here like you said.

I added some outputs with timings to your question, which clearly show, that optimization happens there.

public static void main(String[] args) throws InterruptedException {

    for (int i = 0; i < 10; i++) {
        final long startTime = System.currentTimeMillis();
        Counter c = new Counter();
        Person p1 = new Person(c);
        Person p2 = new Person(c);
        p1.start();
        p2.start();
        p1.join();
        p2.join();
        final long endTime = System.currentTimeMillis();
        System.out.println(String.format("run %s: %s (%s ms)", i, c.getCount(), endTime - startTime));        
   }
}

Results:

run 0: 1107 (8 ms)
run 1: 1 (1 ms)
run 2: 0 (2 ms)
run 3: 0 (0 ms)
run 4: 0 (0 ms)
run 5: 0 (0 ms)
run 6: 0 (1 ms)
run 7: 0 (0 ms)
run 8: 0 (0 ms)
run 9: 0 (0 ms)

The first iterations the program needs a lot of time, wheras in later execution nearly no time at all is used.

Seems to be legit to suspect optimazation for this behaviour.

Using a volatile int count:

run 0: 8680 (15 ms)
run 1: 6943 (12 ms)
run 2: 446 (7 ms)
run 3: -398 (7 ms)
run 4: 431 (8 ms)
run 5: -5489 (6 ms)
run 6: 237 (7 ms)
run 7: 122 (7 ms)
run 8: -87 (7 ms)
run 9: 112 (7 ms)