Infinite loop in release mode

Question

When I run the following code in debug mode, it'll successfully finish and exit. However, if I run the following code in release mode, it'll get stuck in an infinite loop and never finish.

static void Main(string[] args)
{
    bool stop = false;

    new Thread(() =>
    {
        Thread.Sleep(1000);
        stop = true;
        Console.WriteLine("Set \"stop\" to true.");

    }).Start();

    Console.WriteLine("Entering loop.");

    while (!stop)
    {
    }

    Console.WriteLine("Done.");
}

Which optimization is causing it to get stuck in an infinite loop?

Answer 1

My guess would be processor caching of the stop variable on the main thread. In debug mode the memory model is stricter because the debugger needs to be able to provide a sensible view of the variable's state across all threads.

Try making a field and marking it as volatile:

volatile bool stop = false;

static void Main(string[] args)
{

    new Thread(() =>
    {
        Thread.Sleep(1000);
        stop = true;
        Console.WriteLine("Set \"stop\" to true.");

    }).Start();

    Console.WriteLine("Entering loop.");

    while (!stop)
    {
    }

    Console.WriteLine("Done.");
}

Answer 2

Because it's not thread safe you update the main thread variable stop inside the child thread. It will always be unpredictable. To work with any situation like this, have a look on this article.

The volatile keyword instructs the compiler to generate an acquire-fence on every read from that field, and a release-fence on every write to that field. An acquire-fence prevents other reads/writes from being moved before the fence; a release-fence prevents other reads/writes from being moved after the fence. These “half-fences” are faster than full fences because they give the runtime and hardware more scope for optimization.