Esta pregunta ya tiene una respuesta aquí:

¿Por qué la transmisión en paralelo con lambda en el inicializador estático causa un punto muerto? 3 respuestas

El siguiente código conduce a un punto muerto (en mi PC):

public class Test {
    static {
        final int SUM = IntStream.range(0, 100)
                .parallel()
                .reduce((n, m) -> n + m)
                .getAsInt();
    }

    public static void main(String[] args) {
        System.out.println("Finished");
    }
}

Pero si reemplazo areduce argumento lambda con clase anónima no conduce a un punto muerto:

public class Test {
    static {
        final int SUM = IntStream.range(0, 100)
                .parallel()
                .reduce(new IntBinaryOperator() {
                    @Override
                    public int applyAsInt(int n, int m) {
                        return n + m;
                    }
                })
                .getAsInt();
    }

    public static void main(String[] args) {
        System.out.println("Finished");
    }
}

¿Podría explicar esa situación?

PD

Encontré ese código (un poco diferente al anterior):

public class Test {
    static {
        final int SUM = IntStream.range(0, 100)
                .parallel()
                .reduce(new IntBinaryOperator() {
                    @Override
                    public int applyAsInt(int n, int m) {
                        return sum(n, m);
                    }
                })
                .getAsInt();
    }

    private static int sum(int n, int m) {
        return n + m;
    }

    public static void main(String[] args) {
        System.out.println("Finished");
    }
}

works no es estable. En la mayoría de los casos se cuelga pero a veces termina con éxito:

Realmente no puedo entender por qué este comportamiento no es estable. En realidad, vuelvo a probar el primer fragmento de código y el comportamiento es el mismo. Entonces, el último código es igual al primero.

Para comprender qué hilos se usan, agregué después de "iniciar sesión":

public class Test {
    static {
        final int SUM = IntStream.range(0, 100)
                .parallel()
                .reduce((n, m) -> {
                    System.out.println(Thread.currentThread().getName());
                    return (n + m);
                })
                .getAsInt();
    }

    public static void main(String[] args) {
        System.out.println("Finished");
    }
}

Para el caso en que la aplicación finalice correctamente, veo el siguiente registro:

main
main
main
main
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
main
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
main
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
ForkJoinPool.commonPool-worker-1
Finished

PD. 2

I Undestand que reduce es suficiente operaciones complejas. Encontré un ejemplo más simple para mostrar ese problema:

public class Test {
    static {
        System.out.println("static initializer: " + Thread.currentThread().getName());

        final long SUM = IntStream.range(0, 2)
                .parallel()
                .mapToObj(i -> {
                    System.out.println("map: " + Thread.currentThread().getName() + " " + i);
                    return i;
                })
                .count();
    }

    public static void main(String[] args) {
        System.out.println("Finished");
    }
}

para caso feliz (caso raro) veo el siguiente resultado:

static initializer: main
map: main 1
map: main 0
Finished

ejemplo de caso feliz para rango de transmisión extendido:

static initializer: main
map: main 2
map: main 3
map: ForkJoinPool.commonPool-worker-2 4
map: ForkJoinPool.commonPool-worker-1 1
map: ForkJoinPool.commonPool-worker-3 0
Finished

ejemplo de caso que conduce a un punto muerto:

static initializer: main
map: main 1

También conduce a un punto muerto pero no para cada inicio.