阻塞队列与我们平常接触的普通队列(LinkedList或ArrayList等)的最大不同点,在于阻塞队列支持阻塞添加和阻塞删除方法。
- 阻塞添加
所谓的阻塞添加是指当阻塞队列元素已满时,队列会阻塞加入元素的线程,直队列元素不满时才重新唤醒线程执行元素加入操作。
- 阻塞删除
阻塞删除是指在队列元素为空时,删除队列元素的线程将被阻塞,直到队列不为空再执行删除操作(一般都会返回被删除的元素)
阻塞队列 (BlockingQueue)是Java util.concurrent包下重要的数据结构,BlockingQueue提供了线程安全的队列访问方式:当阻塞队列进行插入数据时,如果队列已满,线程将会阻塞等待直到队列非满;从阻塞队列取数据时,如果队列已空,线程将会阻塞等待直到队列非空。并发包下很多高级同步类的实现都是基于BlockingQueue实现的
BlockingQueue 具有 4 组不同的方法用于插入、移除以及对队列中的元素进行检查。如果请求的操作不能得到立即执行的话,每个方法的表现也不同。这些方法如下:
BlockingQueue 的实现有很多,本文只简单介绍下ArrayBlockingQueue 和 LinkedBlockingQueue的阻塞原理。
1.ArrayBlockingQueue
我们利用阻塞队列来实现生产者-消费者模式,先看一个小例子:AB工人生产面包,123同学消费面包:
package com.cjian.threadpool.blockingqueue;
import java.util.concurrent.ArrayBlockingQueue;
public class Producer implements Runnable{
//生产的面包放到这
private ArrayBlockingQueue<Bread> breadQueue;
public Producer(ArrayBlockingQueue<Bread> breadQueue) {
this.breadQueue = breadQueue;
}
@Override
public void run() {
while(true){
try {
//每三秒生产一个面包
Thread.sleep(3000);
breadQueue.put(new Bread());
System.out.println(Thread.currentThread().getName()+"生产了一个面包");
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
}
package com.cjian.threadpool.blockingqueue;
import java.util.concurrent.ArrayBlockingQueue;
public class Consumer implements Runnable{
//从这里消费面包
private ArrayBlockingQueue<Bread> breadQueue;
public Consumer(ArrayBlockingQueue<Bread> breadQueue) {
this.breadQueue = breadQueue;
}
@Override
public void run() {
while(true){
try {
//每一秒消费一个面包
Thread.sleep(1000);
breadQueue.take();
System.out.println(Thread.currentThread().getName()+"消费一个面包");
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
}
测试:
package com.cjian.threadpool.blockingqueue;
import java.util.concurrent.ArrayBlockingQueue;
public class Test {
public static void main(String[] args) {
ArrayBlockingQueue<Bread> breadQueue = new ArrayBlockingQueue<>(10);
new Thread(new Producer(breadQueue),"A").start();
new Thread(new Producer(breadQueue),"B").start();
new Thread(new Consumer(breadQueue),"1").start();
new Thread(new Consumer(breadQueue),"2").start();
new Thread(new Consumer(breadQueue),"3").start();
}
}
A生产了一个面包
3消费一个面包
B生产了一个面包
1消费一个面包
A生产了一个面包
1消费一个面包
B生产了一个面包
2消费一个面包
B生产了一个面包
A生产了一个面包
3消费一个面包
1消费一个面包
一起来探究下,底层实现:
//Object数组
final Object[] items;
/** items index for next take, poll, peek or remove */
//下次取元素的下标
int takeIndex;
/** items index for next put, offer, or add */
//下次插入的下标
int putIndex;
/** Number of elements in the queue */
//队列元素大小
int count;
/*
* Concurrency control uses the classic two-condition algorithm
* found in any textbook.
*/
/** Main lock guarding all access */
//只有一把锁,读写共用一把锁
final ReentrantLock lock;
/** Condition for waiting takes */
//获取元素的条件
private final Condition notEmpty;
/** Condition for waiting puts */
//插入元素的条件
private final Condition notFull;
//指定容量
public ArrayBlockingQueue(int capacity) {
this(capacity, false);
}
//指定容量大小,指定是否公平锁
public ArrayBlockingQueue(int capacity, boolean fair) {
if (capacity <= 0)
throw new IllegalArgumentException();
this.items = new Object[capacity];
lock = new ReentrantLock(fair);
notEmpty = lock.newCondition();
notFull = lock.newCondition();
}
//指定容量,公平性,以及可以对传进来的集合初始化,放入队列中
public ArrayBlockingQueue(int capacity, boolean fair,
Collection<? extends E> c) {
this(capacity, fair);
final ReentrantLock lock = this.lock;
lock.lock(); // Lock only for visibility, not mutual exclusion
try {
int i = 0;
try {
for (E e : c) {
checkNotNull(e);
items[i++] = e;
}
} catch (ArrayIndexOutOfBoundsException ex) {
throw new IllegalArgumentException();
}
count = i;
putIndex = (i == capacity) ? 0 : i;
} finally {
lock.unlock();
}
}
既然用到了阻塞队列,应该都是在使用它的可阻塞特性吧,所以,接下来我们分析put 和take是如何实现的阻塞功能的,其他方法也都很简单,就不分析了
put
public void put(E e) throws InterruptedException {
//为空,爆空指针
checkNotNull(e);
final ReentrantLock lock = this.lock;
lock.lockInterruptibly();
try {
while (count == items.length)//如果队列元素大小与数组大小相等,说明满了,阻塞当前线程
notFull.await();
//插入元素
enqueue(e);
} finally {
lock.unlock();
}
}
private void enqueue(E x) {
final Object[] items = this.items;
//直接放入putIndex下标位置
items[putIndex] = x;
if (++putIndex == items.length)//如果已经是最后一位了,下次插入需要从0下标开始
putIndex = 0;
//自增1
count++;
//唤醒取元素的线程
notEmpty.signal();
}
take
public E take() throws InterruptedException {
final ReentrantLock lock = this.lock;
lock.lockInterruptibly();
try {
while (count == 0)//count==0,说明队列没有元素,直接阻塞
notEmpty.await();
//取出元素
return dequeue();
} finally {
lock.unlock();
}
}
private E dequeue() {
final Object[] items = this.items;
@SuppressWarnings("unchecked")
//取出队首的元素,先进先出
E x = (E) items[takeIndex];
items[takeIndex] = null;
if (++takeIndex == items.length)//如果这次取出的元素是数组的最后一位,takeIndex置为0
takeIndex = 0;
count--;
if (itrs != null)
itrs.elementDequeued();
//有元素被取出,唤醒插入时被阻塞的线程
notFull.signal();
return x;
}
由此可见,阻塞的功能就是通过ReentrantLock 锁的条件(condition)来实现的。
remove
public boolean remove(Object o) {
if (o == null) return false;
final Object[] items = this.items;
final ReentrantLock lock = this.lock;
lock.lock();
try {
if (count > 0) {
final int putIndex = this.putIndex;
int i = takeIndex;
do {
if (o.equals(items[i])) {//遍历队列中的元素
removeAt(i);
return true;
}
if (++i == items.length)
i = 0;
} while (i != putIndex);
}
return false;
} finally {
lock.unlock();
}
}
总结下ArrayBlockingQueue:
数据结构:静态数组,容量固定
锁:ReentrantLock,存取是同一把锁
阻塞:notEmpty 出队:当count==0 时,阻塞当前线程; notFull 入队:当count大小等于数据大小时,阻塞当前线程
入队:从队首开始添加,并记录putIndex(到队位时置为0),并唤醒notEmpty
出队:从队首开始取,记录takeIndex (到队位时置为0),唤醒notFull
先进先出,读写互相排斥
2.LinkedBlockingQueue
private final int capacity; // 队列容量,如果构造时未指定则为Integer.MAX_VALUE
//使用AtomicInteger来统计队列中元素数量
private final AtomicInteger count = new AtomicInteger(0);
/**
* Head of linked list.
* Invariant: head.item == null
*/
private transient Node<E> head; // 队列的头元素 值为null
/**
* Tail of linked list.
* Invariant: last.next == null
*/
private transient Node<E> last; // 队列的尾元素 它的下一个节点为null
/** Lock held by take, poll, etc */
private final ReentrantLock takeLock = new ReentrantLock(); // 出队锁
/** Wait queue for waiting takes */
private final Condition notEmpty = takeLock.newCondition(); // 取出线程condition
/** Lock held by put, offer, etc */
private final ReentrantLock putLock = new ReentrantLock(); // 入队锁
/** Wait queue for waiting puts */
private final Condition notFull = putLock.newCondition(); // 添加线程condition
构造函数:
public LinkedBlockingQueue() {
this(Integer.MAX_VALUE);
}
public LinkedBlockingQueue(int capacity) {
if (capacity <= 0) throw new IllegalArgumentException();
this.capacity = capacity;
last = head = new Node<E>(null);
}
public LinkedBlockingQueue(Collection<? extends E> c) {
this(Integer.MAX_VALUE);
final ReentrantLock putLock = this.putLock;
putLock.lock(); // Never contended, but necessary for visibility
try {
int n = 0;
for (E e : c) {
if (e == null)
throw new NullPointerException();
if (n == capacity)
throw new IllegalStateException("Queue full");
enqueue(new Node<E>(e));
++n;
}
count.set(n);
} finally {
putLock.unlock();
}
}
put
public void put(E e) throws InterruptedException {
if (e == null) throw new NullPointerException();//先检查添加值是否为null
int c = -1; // 必须使用局部变量来表示队列元素数量,负数表示操作失败
Node<E> node = new Node(e); //先创建新的节点
final ReentrantLock putLock = this.putLock; //使用putLock来保证线程安全
final AtomicInteger count = this.count;
putLock.lockInterruptibly();
try {
while (count.get() == capacity) {//当队列已满,添加线程阻塞
notFull.await();
}
enqueue(node); // 调用enqueue方法添加到队尾
c = count.getAndIncrement(); //调用AtomicInteger的getAndIncrement()是数量加1
if (c + 1 < capacity)//添加成功后判断是否可以继续添加,队列未满
notFull.signal(); //唤醒添加线程
} finally {
putLock.unlock();
}
if (c == 0) // 添加后如果队列中只有一个元素,唤醒一个取出线程,使用取出锁
signalNotEmpty();
}
private void enqueue(Node<E> node) {
last = last.next = node; //将新的节点添加到队尾,并变成新的尾节点
}
take
public E take() throws InterruptedException {
E x;
int c = -1;
final AtomicInteger count = this.count;
final ReentrantLock takeLock = this.takeLock;//使用takeLock保证线程安全
takeLock.lockInterruptibly();
try {
while (count.get() == 0) {//当队列为空,取出线程阻塞
notEmpty.await();
}
x = dequeue(); //掉用dequeue方法从队头取出元素
c = count.getAndDecrement(); //调用AtomicInteger的getAndDecrement()将count值减1
if (c > 1)//判断如果当前队列之前元素的数量大于1,唤醒取出线程
notEmpty.signal();
} finally {
takeLock.unlock();
}
if (c == capacity)//之前队列元素数量为容量值,取出一个,只能唤醒一个添加线程
signalNotFull();
return x;
}
private E dequeue() {
Node<E> h = head; //队列的头结点是值为null的节点
Node<E> first = h.next; //返回头节点之后的第一个节点
h.next = h; // help GC
//因为创建节点时创建了一个新的对象,所以需要GC,即需要将头节点的后继节点指向自身,帮助GC
head = first;//将新的头节点置为将要删除的第一个节点
E x = first.item; //将节点的值赋给x
first.item = null;//将节点值置为null,变为新的头节点
return x;//返回取出的值
}
private void signalNotFull() {
final ReentrantLock putLock = this.putLock;
putLock.lock();
try {
notFull.signal();
} finally {
putLock.unlock();
}
}
remove
public boolean remove(Object o) {
if (o == null) return false;
fullyLock(); //此时将入队锁和出队锁全部锁住来保证线程安全
try {
for (Node<E> trail = head, p = trail.next;
p != null;
trail = p, p = p.next) {// 循环遍历查找值相等的元素
if (o.equals(p.item)) {
unlink(p, trail);//调用unlink删除此节点
return true;//操作成功返回true
}
}
return false;
} finally {
fullyUnlock();
}
}
void fullyLock() {
putLock.lock();
takeLock.lock();
}
链表Node 可以指定容量,默认Integer.MAX_VALUE,内部Node存储元素
锁分离: 存取互不排斥,操作的是不同的node对象: takeLock 取Node节点保证前驱后继不会乱
putLock 存node节点保证前驱后继不会乱
阻塞 同ArrayBlockingQueue
入队:队尾入队 记录last节点 指向刚入队的
出队:队首出队 记录head节点 然后head往后移
删除元素的时候两个锁一起加
先进先出