读写锁

一个资源可以被多个读操作访问，或者被一个写操作访问，但是不能同时存在多个写操作（读写互斥，写写互斥）

• ReentrantReadWriteLock 是读写锁的实现类

读写锁的缺点:

• 读写锁的实现是基于AQS的，所以读写锁的实现是比较复杂的，性能也不如synchronized
• 造成锁饥饿问题，写锁的优先级高于读锁，如果读锁的线程很多，写锁的线程很少，那么写锁的线程可能一直获取不到锁，导致锁饥饿问题
• 读的时候不能写，写的时候可以读,但是要合并.

锁降级

• 锁降级是指将写锁降级为读锁
• 锁降级的目的是为了避免死锁
• 锁降级的过程是先获取写锁，然后获取读锁，再释放写锁,再释放读锁.

一个演示锁降级的例子:

public class LockDowngrade {
    private ReentrantReadWriteLock lock = new ReentrantReadWriteLock();
    private ReentrantReadWriteLock.ReadLock readLock = lock.readLock();
    private ReentrantReadWriteLock.WriteLock writeLock = lock.writeLock();

    public void read() {
        readLock.lock();
        System.out.println(Thread.currentThread().getName() + " 获取读锁");
        writeLock.lock();
        System.out.println(Thread.currentThread().getName() + " 获取写锁");
        writeLock.unlock();
        System.out.println(Thread.currentThread().getName() + " 释放写锁");
        readLock.unlock();
        System.out.println(Thread.currentThread().getName() + " 释放读锁");
    }

    public void write() {
        writeLock.lock();
        System.out.println(Thread.currentThread().getName() + " 获取写锁");
        readLock.lock();
        System.out.println(Thread.currentThread().getName() + " 获取读锁");
        readLock.unlock();
        System.out.println(Thread.currentThread().getName() + " 释放读锁");
        writeLock.unlock();
        System.out.println(Thread.currentThread().getName() + " 释放写锁");
    }

    public static void main(String[] args) {
        LockDowngrade lockDowngrade = new LockDowngrade();
        new Thread(() -> lockDowngrade.read(), "Thread-1").start();
        new Thread(() -> lockDowngrade.write(), "Thread-2").start();
    }
}