Explanation of Java BetterSorry

BetterSorry.java

public final int getAndIncrement(int i) {
    while (true) {
        int current = get(i); 
        int next = current + 1;
        if (compareAndSet(i, current, next))
            return current;
    }
}
//So this is how the AtomicIntegerArray implements getAndIncrement

//What makes it safe is that it get's the value, creates the next value
//and if during that time current has not changed, set's it to next.

//Think of compareAndSet as

public final int compareAndSet(int i, int current, int next) {
    if(get(i) == current) {
        set(i) = next;
    }
}

//But the reason it is slow is that it can get stuck in this loop for some time

//So to limit that, we only loop "a little"

//The simplest way of doing this is what I submitted for BetterSorry

    public void incVal(int i) {
        while (true) {
            value[i] += 1;
            return;
        }
    }

    public void decVal(int i) {
        while (true) {
            value[i] -= 1;
            return;
        }
    }

    public boolean swap(int i, int j) {
	if (value[i] == 0) {
	    return false;
	}
        decVal(i);
        incVal(j);
	return true;
    }


// Notice I only looped once and immediately returned


// Ideally, you would change the while true to do the same thing as getAndIncrement
// but at most loop some small number of times
// maybe 2 or 3 times
// So


    public void incVal(int i) {
        int current = get(i); 
        int next = current + 1;
        if (value.compareAndSet(i, current, next))
            return;
        int current = get(i); 
        int next = current + 1;
        if (value.compareAndSet(i, current, next))
            return;
        value[i] += 1;
    }

    public void decVal(int i) {
        int current = get(i); 
        int next = current - 1;
        if (value.compareAndSet(i, current, next))
            return current;
        int current = get(i); 
        int next = current - 1;
        if (value.compareAndSet(i, current, next))
            return current;
        value[i] -= 1;
    }


// This code assumes that the optimal solution is to check 2 times and the third time to just do it
// Ultimately you would repeat that code
// more times for more safety, less speed and less times for less safety, more speed
// DO NOT use a for loop because keeping track of the counter of the for loop takes time
// You would just want to past the code manually  multiple times if you wanted more safety