【发布时间】:2017-01-11 13:50:08
【问题描述】:
我正在尝试理解 ThreadPoolExecutor 类。我发现该类中声明了一些最终变量,但无法理解它们的用途。
private final AtomicInteger ctl = new AtomicInteger(ctlOf(RUNNING, 0));
private static final int COUNT_BITS = Integer.SIZE - 3; //29
private static final int CAPACITY = (1 << COUNT_BITS) - 1; //536870911 00011111111111111111111111111111
// RUN_STATE is stored in the high-order bits
private static final int RUNNING = -1 << COUNT_BITS; //-536870912 11100000000000000000000000000000
private static final int SHUTDOWN = 0 << COUNT_BITS; //0 00000000000000000000000000000000
private static final int STOP = 1 << COUNT_BITS; //536870912 00100000000000000000000000000000
private static final int TIDYING = 2 << COUNT_BITS; //1073741824 01000000000000000000000000000000
private static final int TERMINATED = 3 << COUNT_BITS; //1610612736 01100000000000000000000000000000
以上是最终变量及其二进制和十进制值。
然后我找到了两种使用这些变量的方法:
private static int runStateOf(int c) { return c & ~CAPACITY; } // RUN_STATE & ~CAPACITY = RUN_STATE
private static int workerCountOf(int c) { return c & CAPACITY; } // RUN_STATE & CAPACITY = 0
private static int ctlOf(int rs, int wc) { return rs | wc; }
方法前面的 cmets 是我观察到的输出。
现在在ThreadPoolExecutor#execute(runnable) 方法中,
它正在使用If fewer than corePoolSize threads are running的语句进行以下计算
int c = ctl.get();
if (workerCountOf(c) < corePoolSize)
我试图理解,在这种情况下,workerCountOf(c) 的值可能大于 corePoolSize。 如您所见,ctl 的初始值为 RUNNING。
此外,还有一些方法可以原子地增加和减少 ctl 值,
private boolean compareAndIncrementWorkerCount(int expect) {
return ctl.compareAndSet(expect, expect + 1);
}
private boolean compareAndDecrementWorkerCount(int expect) {
return ctl.compareAndSet(expect, expect - 1);
}
现在假设有 5 个线程正在运行,所以 ctl = RUNNING + 5,
即便如此workerCountOf(ctl.get()) = 0,
作为((RUNNING+5) & CAPACITY) = 0。
谁能解释一下创建这些最终变量的原因及其用途?
workerCountOf() 方法实际上如何返回没有正在运行的线程?
我一定是错过了什么。
谢谢
【问题讨论】:
标签: java multithreading threadpoolexecutor