QWaitConditioin的思考1

16:50:392017-03-31

本文章为原创，如若引用请指明出处，否则追责。

以 生产、消费者的例子说明

#include <QtCore/QCoreApplication>
#include <QWaitCondition>
#include <QThread>
#include <QMutex>
#include <iostream>

const int DataSize = 100;
const int BufferSize = 1;
char buffer[BufferSize];

QWaitCondition bufferIsNotFull;
QWaitCondition bufferIsNotEmpty;
QMutex mutex;
int usedSpace;

class Producer : public QThread
{
protected:
    void run()
    {
        for (int i = 0; i < DataSize; ++i)
        {
            mutex.lock();//A
            while (usedSpace == BufferSize)
            {
                bufferIsNotFull.wait(&mutex);//B
            }
            std::cerr<<"P";//C
            ++usedSpace;
            bufferIsNotEmpty.wakeAll();
            mutex.unlock();
        }
    }//…G
};

class Consumer : public QThread
{
protected:
    void run()
    {
        for (int i = 0; i < DataSize; ++i)
        {
            mutex.lock();//D
            while (usedSpace == 0)
            {
                bufferIsNotEmpty.wait(&mutex);//E
            }
            std::cerr<<"C";//F
            --usedSpace;
            bufferIsNotFull.wakeAll();
            mutex.unlock();
        }
        std::cerr<<std::endl;
    }//…K
};

int main(int argc, char *argv[])
{
   Producer producer;
   Consumer consumer;
   producer.start();
   consumer.start();
   producer.wait();
   consumer.wait();
return 0;
}

根据QWaitCondition::wait(lockedMutex)的说明:Releases the lockedMutex and waits on the wait condition. The lockedMutex must be initially locked by the calling thread.

在调用wait之前应将lockedMutex锁住 见上面代码的B和A处，以及E、D处的调用关系。

之所以先执行Releases the lockedMutex 是因为，调用wait(lockedMutex)后程序直接跳转到G（执行完B）或K处（执行完D），而不再执行C、F处的代码，所以要release the lockedmutex

否则不能保证mutex的lock、unlock的成对执行。