QWaitCondition Class

The QWaitCondition class provides a condition variable for synchronizing threads. More...

Header: #include <QWaitCondition>
qmake: QT += core

Note: All functions in this class are thread-safe.

Public Functions

QWaitCondition()
~QWaitCondition()
void notify_all()
void notify_one()
bool wait(QMutex *lockedMutex, QDeadlineTimer deadline = QDeadlineTimer(QDeadlineTimer::Forever))
bool wait(QMutex *lockedMutex, unsigned long time)
bool wait(QReadWriteLock *lockedReadWriteLock, QDeadlineTimer deadline = QDeadlineTimer(QDeadlineTimer::Forever))
bool wait(QReadWriteLock *lockedReadWriteLock, unsigned long time)
void wakeAll()
void wakeOne()

Detailed Description

QWaitCondition allows a thread to tell other threads that some sort of condition has been met. One or many threads can block waiting for a QWaitCondition to set a condition with wakeOne() or wakeAll(). Use wakeOne() to wake one randomly selected thread or wakeAll() to wake them all.

For example, let's suppose that we have three tasks that should be performed whenever the user presses a key. Each task could be split into a thread, each of which would have a run() body like this:

 forever {
     mutex.lock();
     keyPressed.wait(&mutex);
     do_something();
     mutex.unlock();
 }

Here, the keyPressed variable is a global variable of type QWaitCondition.

A fourth thread would read key presses and wake the other three threads up every time it receives one, like this:

 forever {
     getchar();
     keyPressed.wakeAll();
 }

The order in which the three threads are woken up is undefined. Also, if some of the threads are still in do_something() when the key is pressed, they won't be woken up (since they're not waiting on the condition variable) and so the task will not be performed for that key press. This issue can be solved using a counter and a QMutex to guard it. For example, here's the new code for the worker threads:

 forever {
     mutex.lock();
     keyPressed.wait(&mutex);
     ++count;
     mutex.unlock();

     do_something();

     mutex.lock();
     --count;
     mutex.unlock();
 }

Here's the code for the fourth thread:

 forever {
     getchar();

     mutex.lock();
     // Sleep until there are no busy worker threads
     while (count > 0) {
         mutex.unlock();
         sleep(1);
         mutex.lock();
     }
     keyPressed.wakeAll();
     mutex.unlock();
 }

The mutex is necessary because the results of two threads attempting to change the value of the same variable simultaneously are unpredictable.

Wait conditions are a powerful thread synchronization primitive. The Wait Conditions Example example shows how to use QWaitCondition as an alternative to QSemaphore for controlling access to a circular buffer shared by a producer thread and a consumer thread.

See also QMutex, QSemaphore, QThread, and Wait Conditions Example.

Member Function Documentation

QWaitCondition::QWaitCondition()

Constructs a new wait condition object.

QWaitCondition::~QWaitCondition()

Destroys the wait condition object.

void QWaitCondition::notify_all()

This function is provided for STL compatibility. It is equivalent to wakeAll().

This function was introduced in Qt 5.8.

void QWaitCondition::notify_one()

This function is provided for STL compatibility. It is equivalent to wakeOne().

This function was introduced in Qt 5.8.

bool QWaitCondition::wait(QMutex *lockedMutex, QDeadlineTimer deadline = QDeadlineTimer(QDeadlineTimer::Forever))

Releases the lockedMutex and waits on the wait condition. The lockedMutex must be initially locked by the calling thread. If lockedMutex is not in a locked state, the behavior is undefined. If lockedMutex is a recursive mutex, this function returns immediately. The lockedMutex will be unlocked, and the calling thread will block until either of these conditions is met:

  • Another thread signals it using wakeOne() or wakeAll(). This function will return true in this case.
  • the deadline given by deadline is reached. If deadline is QDeadlineTimer::Forever (the default), then the wait will never timeout (the event must be signalled). This function will return false if the wait timed out.

The lockedMutex will be returned to the same locked state. This function is provided to allow the atomic transition from the locked state to the wait state.

This function was introduced in Qt 5.12.

See also wakeOne() and wakeAll().

bool QWaitCondition::wait(QMutex *lockedMutex, unsigned long time)

This is an overloaded function.

bool QWaitCondition::wait(QReadWriteLock *lockedReadWriteLock, QDeadlineTimer deadline = QDeadlineTimer(QDeadlineTimer::Forever))

Releases the lockedReadWriteLock and waits on the wait condition. The lockedReadWriteLock must be initially locked by the calling thread. If lockedReadWriteLock is not in a locked state, this function returns immediately. The lockedReadWriteLock must not be locked recursively, otherwise this function will not release the lock properly. The lockedReadWriteLock will be unlocked, and the calling thread will block until either of these conditions is met:

  • Another thread signals it using wakeOne() or wakeAll(). This function will return true in this case.
  • the deadline given by deadline is reached. If deadline is QDeadlineTimer::Forever (the default), then the wait will never timeout (the event must be signalled). This function will return false if the wait timed out.

The lockedReadWriteLock will be returned to the same locked state. This function is provided to allow the atomic transition from the locked state to the wait state.

This function was introduced in Qt 5.12.

See also wakeOne() and wakeAll().

bool QWaitCondition::wait(QReadWriteLock *lockedReadWriteLock, unsigned long time)

This is an overloaded function.

void QWaitCondition::wakeAll()

Wakes all threads waiting on the wait condition. The order in which the threads are woken up depends on the operating system's scheduling policies and cannot be controlled or predicted.

See also wakeOne().

void QWaitCondition::wakeOne()

Wakes one thread waiting on the wait condition. The thread that is woken up depends on the operating system's scheduling policies, and cannot be controlled or predicted.

If you want to wake up a specific thread, the solution is typically to use different wait conditions and have different threads wait on different conditions.

See also wakeAll().