单个生产者单个消费者无锁队列c++实现

阅读更多

keyword: lockfree、lock-free、queue、buffer、无锁

相关文章：

无锁环形缓冲（Wait-free ring buffer） 

http://aigo.iteye.com/blog/1913518

 

方案：用atomic实现

boost提供了一种标准的所谓Wait-free ring buffer，官方文档上叫buffer，但是其实现其实还是queue，因为push和pop的时候没有size参数，就是说其内部是具体的元素对象，不是类似字节流那种闭环的buffer。

另外，他其实还是用到了轻量级的线程同步atomic，但比lock的开销少至少一个数量级。
下面例子是boost官方文档上的，这里只是提供了思路，没有boost也行，std::atomic也能实现

Wait-free ring buffer

http://www.boost.org/doc/libs/1_60_0/doc/html/atomic/usage_examples.html#boost_atomic.usage_examples.example_ringbuffer.implementation

实现：

 

#include 

template
class ringbuffer {
public:
  ringbuffer() : head_(0), tail_(0) {}

  bool push(const T & value)
  {
    size_t head = head_.load(boost::memory_order_relaxed);
    size_t next_head = next(head);
    if (next_head == tail_.load(boost::memory_order_acquire))
      return false;
    ring_[head] = value;
    head_.store(next_head, boost::memory_order_release);
    return true;
  }
  bool pop(T & value)
  {
    size_t tail = tail_.load(boost::memory_order_relaxed);
    if (tail == head_.load(boost::memory_order_acquire))
      return false;
    value = ring_[tail];
    tail_.store(next(tail), boost::memory_order_release);
    return true;
  }
private:
  size_t next(size_t current)
  {
    return (current + 1) % Size;
  }
  T ring_[Size];
  boost::atomic head_, tail_;
};

  调用：

 

ringbuffer r;

// try to insert an element
if (r.push(42)) { /* succeeded */ }
else { /* buffer full */ }

// try to retrieve an element
int value;
if (r.pop(value)) { /* succeeded */ }
else { /* buffer empty */ }

 

如果不需要无锁，只需要一个队列，还可以使用boost线程的API：circular_buffer

http://www.boost.org/doc/libs/1_60_0/doc/html/circular_buffer.html

 

 

 

误解：用volatile实现lock-free

之前网上不少关于用volatile来实现无锁的例子，实际上是错误的。
volatile无法实现线程安全，即使是一读一写。volatile设计的目的就不是为线程安全考虑的，在特殊的硬件和编译器上volatile才有数据同步的作用。

著名C++专家Herb Sutters对无锁问题的相关论述：

原文：http://www.drdobbs.com/parallel/volatile-vs-volatile/212701484

Volatile is orthogonal to what you use to implement atomics. In C++ it tells the compiler that certain it is not safe to perform optimizations with that variable. Herb Sutters lays it out: 

To safely write lock-free code that communicates between threads without using locks, prefer to use ordered atomic variables: Java/.NET volatile, C++0x atomic, and C-compatible atomic_T.

 

To safely communicate with special hardware or other memory that has unusual semantics, use unoptimizable variables: ISO C/C++ volatile. Remember that reads and writes of these variables are not necessarily atomic, however.

 

 Finally, to express a variable that both has unusual semantics and has any or all of the atomicity and/or ordering guarantees needed for lock-free coding, only the ISO C++0x draft Standard provides a direct way to spell it: volatile atomic.

 

 

下面这个例子是用volatile实现无锁的例子，这个仅限于特定硬件！如果要实现无锁，还不如参考linux kernel中的fifo，它连volatile都没用。

CSDN上某博主的文章：http://blog.csdn.net/chenjiayi_yun/article/details/8945841 

限制一个线程读，一个线程写，不加锁的队列，使用单链表实现，测试环境：centos 5.9 

#include 
#include 


template
struct qnode
{
    struct qnode *next;
    QElmType data;
};
template
class queue
{
public:
        queue() {init();}
        ~queue() {destroy();}

        bool init()
        {
                m_front = m_rear = new qnode;
                if (!m_front)
                        return false;
                m_front->next = 0;
                return true;
        }
        void destroy()
        {
                while (m_front)
                {
                        m_rear = m_front->next;
                        delete m_front;
                        m_front = m_rear;
                }
        }
        bool push(QElmType e)
        {
                struct qnode *p = new qnode;
                if (!p)
                        return false;
                p->next = 0;
                m_rear->next = p;
                m_rear->data = e;
                m_rear = p;
                return true;
        }
        bool pop(QElmType *e)
        {
                if (m_front == m_rear)
                        return false;


                struct qnode *p = m_front;
                *e = p->data;
                m_front = p->next;
                delete p;
                return true;
        }
private:
  struct qnode * volatile m_front, * volatile m_rear;
};


queue g_q;


void *thread1(void * l)
{
        int i = 0;
        while (1)
        {
                g_q.push(i);
                i++;
                usleep(::rand()%1000);
        }
        return 0;
}
void *thread2(void * l)
{
        int i;
        while (1)
        {
                if (g_q.pop(&i))
                        std::cout << i << std::endl;
                //else
                        //std::cout << "can not pop" << std::endl;
                usleep(::rand()%1000);
        }
        return 0;
}


int main(int argc, char* argv[])
{
        pthread_t t1,t2;
        pthread_create(&t1, 0, thread1, 0);
        pthread_create(&t2, 0, thread2, 0);
        char ch;
        while (1)
        {
                std::cin >> ch;
                if (ch == 'q')
                        break;
        }
       return 0;
}

 

代码比较简单，只实现2个线程间的无锁。

这个无锁队列主要是使用两个volatile 的指针来判断是否还有任务（ volatile m_front,  volatile m_rear）。

只能实现两个线程间的无锁队列，一个是工作者线程一个是提供任务线程，因为不能让两个或以上的线程来修改指针m_front 和指针m_rear，否者会出现问题。

这个无锁队列的实现是基于m_front指针的修改是由一个线程来完成的，m_rear的修改是由另一个线程来完成的，不会出现同时两个线程修改同一个指针。