ArrayBlockingQueue源码中为什么方法要用局部变量引用类变量

Question

代码：
类变量:
private final E[] items;
private final ReentrantLock lock;
方法：
public void put(E o) throws InterruptedException {
if (o == null) throw new NullPointerException();
[b] final E[] items = this.items;
final ReentrantLock lock = this.lock;[/b]
lock.lockInterruptibly();
try {
try {
while (count == items.length)
notFull.await();
} catch (InterruptedException ie) {
notFull.signal(); // propagate to non-interrupted thread
throw ie;
}
insert(o);
} finally {
lock.unlock();
}
}
看黑体字部分 put方法中局部变量items和lock的使用有什么意义?

Answer 1

首先在JDK 7中，这段代码变成这样子了：
[code="java"]
final Object[] items;

public void put(E e) throws InterruptedException {
checkNotNull(e);
final ReentrantLock lock = this.lock;
lock.lockInterruptibly();
try {
while (count == items.length)
notFull.await();
insert(e);
} finally {
lock.unlock();
}
}
[/code]
然后做个实验：
[code="java"]
final Object[] items = new Object[10];

public void test() {
if(items.length == 0) {
}

int i = items.length;
}

public void test2() {
final Object[] items = this.items;
if(items.length == 0) {
}

int i = items.length;
}
[/code]
然后javap一下，javap -p -c -s Test >> Test.log，得到如下代码：
[code="java"]
public void test();
Signature: ()V
Code:
0: aload_0

1: getfield #3 // Field items:[Ljava/lang/Object;
4: arraylength

5: ifne 8
8: aload_0

9: getfield #3 // Field items:[Ljava/lang/Object;
12: arraylength

13: istore_1

14: return

public void test2();
Signature: ()V
Code:
0: aload_0

1: getfield #3 // Field items:[Ljava/lang/Object;
4: astore_1

5: aload_1 //load 局部变量 items
6: arraylength

7: ifne 10
10: aload_1

// 这里少了getfield，因为aload_1 load的就是items
11: arraylength

12: istore_2

13: return

[/code]

两种写法唯一的区别，是getfield指令，getfield在对象内相对来说开销是比较廉价的，但前者（test方法）显然在代码可读性上，高出很多，如果不存在大量的实例变量引用，性能可以忽略不计，估计这也正是为什么JDK7采用这种简单的写法的原因吧。