GVKun编程网logo

C++11并行编程-条件变量(condition_variable)详细说明(c++条件变量使用)

3

此处将为大家介绍关于C++11并行编程-条件变量(condition_variable)详细说明的详细内容,并且为您解答有关c++条件变量使用的相关问题,此外,我们还将为您介绍关于androidCon

此处将为大家介绍关于C++11并行编程-条件变量(condition_variable)详细说明的详细内容,并且为您解答有关c++条件变量使用的相关问题,此外,我们还将为您介绍关于android ConditionVariable、android.os.ConditionVariable的实例源码、c – std :: condition_variable :: wait_until的实现、c – `std :: condition_variable :: wait_for`经常调用谓词的有用信息。

本文目录一览:

C++11并行编程-条件变量(condition_variable)详细说明(c++条件变量使用)

C++11并行编程-条件变量(condition_variable)详细说明(c++条件变量使用)

<condition_variable >头文件主要包含有类和函数相关的条件变量。

包括相关类 std::condition_variable和 std::condition_variable_any,还有枚举类型std::cv_status。另外还包含函数 std::notify_all_at_thread_exit(),以下分别介绍一下以上几种类型。

std::condition_variable 类介绍

std::condition_variable是条件变量,很多其它有关条件变量的定义參考维基百科。Linux下使用 Pthread库中的 pthread_cond_*() 函数提供了与条件变量相关的功能, Windows 则參考 MSDN

当 std::condition_variable对象的某个wait 函数被调用的时候,它使用 std::unique_lock(通过 std::mutex) 来锁住当前线程。

当前线程会一直被堵塞。直到另外一个线程在同样的 std::condition_variable 对象上调用了 notification 函数来唤醒当前线程。

std::condition_variable 对象通常使用 std::unique_lock<std::mutex> 来等待,假设须要使用另外的 lockable 类型,能够使用std::condition_variable_any类。本文后面会讲到 std::condition_variable_any 的使用方法。

#include <iostream>                // std::cout
#include <thread>                // std::thread
#include <mutex>                // std::mutex,std::unique_lock
#include <condition_variable>    // std::condition_variable

std::mutex mtx; // 全局相互排斥锁.
std::condition_variable cv; // 全局条件变量.
bool ready = false; // 全局标志位.

void do_print_id(int id)
{
    std::unique_lock <std::mutex> lck(mtx);
    while (!ready) // 假设标志位不为 true,则等待...
        cv.wait(lck); // 当前线程被堵塞,当全局标志位变为 true 之后,// 线程被唤醒,继续往下运行打印线程编号id.
    std::cout << "thread " << id << ‘\n‘;
}

void go()
{
    std::unique_lock <std::mutex> lck(mtx);
    ready = true; // 设置全局标志位为 true.
    cv.notify_all(); // 唤醒全部线程.
}

int main()
{
    std::thread threads[10];
    // spawn 10 threads:
    for (int i = 0; i < 10; ++i)
        threads[i] = std::thread(do_print_id,i);

    std::cout << "10 threads ready to race...\n";
    go(); // go!

  for (auto & th:threads)
        th.join();

    return 0;
}

结果:

10 threads ready to race...
thread 1
thread 0
thread 2
thread 3
thread 4
thread 5
thread 6
thread 7
thread 8
thread 9

std::condition_variable 的拷贝构造函数被禁用,仅仅提供了默认构造函数。

看看 std::condition_variable 的各个成员函数

std::condition_variable::wait() 介绍:

std::condition_variable提供了两种 wait() 函数。

void wait (unique_lock<mutex>& lck);

template <class Predicate>
void wait (unique_lock<mutex>& lck,Predicate pred);

当前线程调用 wait() 后将被堵塞(此时当前线程应该获得了锁(mutex),最好还是设获得锁 lck),直到另外某个线程调用 notify_* 唤醒了当前线程

线程被堵塞时,该函数会自己主动调用 lck.unlock() 释放锁,使得其它被堵塞在锁竞争上的线程得以继续运行。另外,一旦当前线程获得通知(notified,一般是另外某个线程调用 notify_* 唤醒了当前线程),wait()函数也是自己主动调用 lck.lock()使得lck的状态和 wait 函数被调用时同样

在另外一种情况下(即设置了 Predicate)。仅仅有当 pred 条件为false 时调用 wait() 才会堵塞当前线程。而且在收到其它线程的通知后仅仅有当 pred 为 true 时才会被解除堵塞

因此另外一种情况相似以下代码:

#include <iostream>                // std::cout
#include <thread>                // std::thread,std::this_thread::yield
#include <mutex>                // std::mutex,std::unique_lock
#include <condition_variable>    // std::condition_variable

std::mutex mtx;
std::condition_variable cv;

int cargo = 0;
bool shipment_available()
{
    return cargo != 0;
}

// 消费者线程.
void consume(int n)
{
    for (int i = 0; i < n; ++i) {
        std::unique_lock <std::mutex> lck(mtx);
        cv.wait(lck,shipment_available);
        std::cout << cargo << ‘\n‘;
        cargo = 0;
    }
}

int main()
{
    std::thread consumer_thread(consume,10); // 消费者线程.

    // 主线程为生产者线程,生产 10 个物品.
    for (int i = 0; i < 10; ++i) {
        while (shipment_available())
            std::this_thread::yield();
        std::unique_lock <std::mutex> lck(mtx);
        cargo = i + 1;
        cv.notify_one();
    }

    consumer_thread.join();

    return 0;
}
1
2
3
4
5
6
7
8
9
10

std::condition_variable::wait_for() 介绍

template <class Rep,class Period>
  cv_status wait_for (unique_lock<mutex>& lck,const chrono::duration<Rep,Period>& rel_time);

template <class Rep,class Period,class Predicate>
       bool wait_for (unique_lock<mutex>& lck,Period>& rel_time,Predicate pred);

std::condition_variable::wait() 相似,只是 wait_for能够指定一个时间段,在当前线程收到通知或者指定的时间 rel_time 超时之前。该线程都会处于堵塞状态。而一旦超时或者收到了其它线程的通知,wait_for返回,剩下的处理步骤和 wait()相似。

另外,wait_for 的重载版本号的最后一个參数pred表示 wait_for的预測条件。仅仅有当 pred条件为false时调用 wait()才会堵塞当前线程,而且在收到其它线程的通知后仅仅有当 pred为 true时才会被解除堵塞,因此相当于例如以下代码:

return wait_until (lck,chrono::steady_clock::Now() + rel_time,std::move(pred));

请看以下的样例(參考),以下的样例中,主线程等待th线程输入一个值。然后将th线程从终端接收的值打印出来。在th线程接受到值之前,主线程一直等待。每一个一秒超时一次,并打印一个 ".":  

#include <iostream>           // std::cout
#include <thread>             // std::thread
#include <chrono>             // std::chrono::seconds
#include <mutex>              // std::mutex,std::unique_lock
#include <condition_variable> // std::condition_variable,std::cv_status

std::condition_variable cv;

int value;

void do_read_value()
{
    std::cin >> value;
    cv.notify_one();
}

int main ()
{
    std::cout << "Please,enter an integer (I‘ll be printing dots): \n";
    std::thread th(do_read_value);

    std::mutex mtx;
    std::unique_lock<std::mutex> lck(mtx);
    while (cv.wait_for(lck,std::chrono::seconds(1)) == std::cv_status::timeout) {
        std::cout << ‘.‘;
        std::cout.flush();
    }

    std::cout << "You entered: " << value << ‘\n‘;

    th.join();
    return 0;
}

std::condition_variable::wait_until 介绍

template <class Clock,class Duration>
  cv_status wait_until (unique_lock<mutex>& lck,const chrono::time_point<Clock,Duration>& abs_time);

template <class Clock,class Duration,class Predicate>
       bool wait_until (unique_lock<mutex>& lck,Duration>& abs_time,Predicate pred);

与 std::condition_variable::wait_for 相似,可是wait_until能够指定一个时间点,在当前线程收到通知或者指定的时间点 abs_time超时之前,该线程都会处于堵塞状态。而一旦超时或者收到了其它线程的通知,wait_until返回。剩下的处理步骤和 wait_until() 相似。

另外,wait_until的重载版本号的最后一个參数 pred表示 wait_until 的预測条件。仅仅有当 pred 条件为 false时调用 wait()才会堵塞当前线程,而且在收到其它线程的通知后仅仅有当pred为 true时才会被解除堵塞,因此相当于例如以下代码:

while (!pred())
  if ( wait_until(lck,abs_time) == cv_status::timeout)
    return pred();
return true;

std::condition_variable::notify_one() 介绍

唤醒某个等待(wait)线程。假设当前没有等待线程,则该函数什么也不做,假设同一时候存在多个等待线程,则唤醒某个线程是不确定的(unspecified)

请看下例(參考):

#include <iostream>                // std::cout
#include <thread>                // std::thread
#include <mutex>                // std::mutex,std::unique_lock
#include <condition_variable>    // std::condition_variable

std::mutex mtx;
std::condition_variable cv;

int cargo = 0; // shared value by producers and consumers

void consumer()
{
    std::unique_lock < std::mutex > lck(mtx);
    while (cargo == 0)
        cv.wait(lck);
    std::cout << cargo << ‘\n‘;
    cargo = 0;
}

void producer(int id)
{
    std::unique_lock < std::mutex > lck(mtx);
    cargo = id;
    cv.notify_one();
}

int main()
{
    std::thread consumers[10],producers[10];

    // spawn 10 consumers and 10 producers:
    for (int i = 0; i < 10; ++i) {
        consumers[i] = std::thread(consumer);
        producers[i] = std::thread(producer,i + 1);
    }

    // join them back:
    for (int i = 0; i < 10; ++i) {
        producers[i].join();
        consumers[i].join();
    }

    return 0;
}

std::condition_variable::notify_all() 介绍

唤醒全部的等待(wait)线程。假设当前没有等待线程,则该函数什么也不做。请看以下的样例:

#include <iostream>                // std::cout
#include <thread>                // std::thread
#include <mutex>                // std::mutex,i);

    std::cout << "10 threads ready to race...\n";
    go(); // go!

  for (auto & th:threads)
        th.join();

    return 0;
}

std::condition_variable_any 介绍

与 std::condition_variable相似。仅仅只是std::condition_variable_any的 wait 函数能够接受不论什么 lockable參数,而 std::condition_variable仅仅能接受 std::unique_lock<std::mutex>类型的參数,除此以外,和std::condition_variable差点儿全然一样。

std::cv_status枚举类型介绍

cv_status::no_timeout wait_for 或者wait_until没有超时,即在规定的时间段内线程收到了通知。

cv_status::timeout  wait_for 或者 wait_until 超时。
std::notify_all_at_thread_exit

函数原型为:

void notify_all_at_thread_exit (condition_variable& cond,unique_lock<mutex> lck);

当调用该函数的线程退出时,全部在 cond 条件变量上等待的线程都会收到通知。

请看下例(參考):

#include <iostream>           // std::cout
#include <thread>             // std::thread
#include <mutex>              // std::mutex,std::unique_lock
#include <condition_variable> // std::condition_variable

std::mutex mtx;
std::condition_variable cv;
bool ready = false;

void print_id (int id) {
  std::unique_lock<std::mutex> lck(mtx);
  while (!ready) cv.wait(lck);
  // ...
  std::cout << "thread " << id << ‘\n‘;
}

void go() {
  std::unique_lock<std::mutex> lck(mtx);
  std::notify_all_at_thread_exit(cv,std::move(lck));
  ready = true;
}

int main ()
{
  std::thread threads[10];
  // spawn 10 threads:
  for (int i=0; i<10; ++i)
    threads[i] = std::thread(print_id,i);
  std::cout << "10 threads ready to race...\n";

  std::thread(go).detach();   // go!

  for (auto& th : threads) th.join();

  return 0;
}

<condition_variable> 头文件里的两个条件变量类(std::condition_variablestd::condition_variable_any)、枚举类型(std::cv_status)、以及辅助函数(std::notify_all_at_thread_exit())都已经介绍完  

 

从wait函数执行流程来看下条件变量的使用方法和原理:

https://blog.csdn.net/liu3612162/article/details/88343266

 

参考链接:

https://www.cnblogs.com/bhlsheji/p/5035018.html

https://www.cnblogs.com/wangshaowei/p/9593201.html

android ConditionVariable

android ConditionVariable

frameworks/base/core/java/android/os/ConditionVariable.java

  1 /*
  2  * copyright (C) 2006 The Android Open Source Project
  3  *
  4  * Licensed under the Apache License, Version 2.0 (the "License");
  5  * you may not use this file except in compliance with the License.
  6  * You may obtain a copy of the License at
  7  *
  8  *      http://www.apache.org/licenses/LICENSE-2.0
  9  *
 10  * Unless required by applicable law or agreed to in writing, software
 11  * distributed under the License is distributed on an "AS IS" BASIS,
 12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 13  * See the License for the specific language governing permissions and
 14  * limitations under the License.
 15  */
 16 
 17 package android.os;
 18 
 19 /**
 20  * Class that implements the condition variable locking paradigm.
 21  *
 22  * <p>
 23  * This differs from the built-in java.lang.Object wait() and notify()
 24  * in that this class contains the condition to wait on itself.  That means
 25  * open(), close() and block() are sticky.  If open() is called before block(),
 26  * block() will not block, and instead return immediately.
 27  *
 28  * <p>
 29  * This class uses itself as the object to wait on, so if you wait()
 30  * or notify() on a ConditionVariable, the results are undefined.
 31  */
 32 public class ConditionVariable
 33 {
 34     private volatile boolean mCondition;
 35 
 36     /**
 37      * Create the ConditionVariable in the default closed state.
 38      */
 39     public ConditionVariable()
 40     {
 41         mCondition = false;
 42     }
 43 
 44     /**
 45      * Create the ConditionVariable with the given state.
 46      * 
 47      * <p>
 48      * Pass true for opened and false for closed.
 49      */
 50     public ConditionVariable(boolean state)
 51     {
 52         mCondition = state;
 53     }
 54 
 55     /**
 56      * Open the condition, and release all threads that are blocked.
 57      *
 58      * <p>
 59      * Any threads that later approach block() will not block unless close()
 60      * is called.
 61      */
 62     public void open()
 63     {
 64         synchronized (this) {
 65             boolean old = mCondition;
 66             mCondition = true;
 67             if (!old) {
 68                 this.notifyAll();
 69             }
 70         }
 71     }
 72 
 73     /**
 74      * Reset the condition to the closed state.
 75      *
 76      * <p>
 77      * Any threads that call block() will block until someone calls open.
 78      */
 79     public void close()
 80     {
 81         synchronized (this) {
 82             mCondition = false;
 83         }
 84     }
 85 
 86     /**
 87      * Block the current thread until the condition is opened.
 88      *
 89      * <p>
 90      * If the condition is already opened, return immediately.
 91      */
 92     public void block()
 93     {
 94         synchronized (this) {
 95             while (!mCondition) {
 96                 try {
 97                     this.wait();
 98                 }
 99                 catch (InterruptedException e) {
100                 }
101             }
102         }
103     }
104 
105     /**
106      * Block the current thread until the condition is opened or until
107      * timeoutMs milliseconds have passed.
108      *
109      * <p>
110      * If the condition is already opened, return immediately.
111      *
112      * @param timeoutMs the maximum time to wait in milliseconds.
113      *
114      * @return true if the condition was opened, false if the call returns
115      * because of the timeout.
116      */
117     public boolean block(long timeoutMs)
118     {
119         // Object.wait(0) means wait forever, to mimic this, we just
120         // call the other block() method in that case.  It simplifies
121         // this code for the common case.
122         if (timeoutMs != 0) {
123             synchronized (this) {
124                 long Now = SystemClock.elapsedRealtime();
125                 long end = Now + timeoutMs;
126                 while (!mCondition && Now < end) {
127                     try {
128                         this.wait(end-Now);
129                     }
130                     catch (InterruptedException e) {
131                     }
132                     Now = SystemClock.elapsedRealtime();
133                 }
134                 return mCondition;
135             }
136         } else {
137             this.block();
138             return true;
139         }
140     }
141 }

 

android.os.ConditionVariable的实例源码

android.os.ConditionVariable的实例源码

项目:airgram    文件:AudioTrack.java   
@H_301_6@
/**
 * Creates an audio track using the specified audio capabilities and stream type.
 *
 * @param audioCapabilities The current audio playback capabilities.
 * @param streamType The type of audio stream for the underlying {@link android.media.AudioTrack}.
 */
public AudioTrack(AudioCapabilities audioCapabilities,int streamType) {
  this.audioCapabilities = audioCapabilities;
  this.streamType = streamType;
  releasingConditionVariable = new ConditionVariable(true);
  if (Util.SDK_INT >= 18) {
    try {
      getLatencyMethod =
          android.media.AudioTrack.class.getmethod("getLatency",(Class<?>[]) null);
    } catch (NoSuchMethodException e) {
      // There's no guarantee this method exists. Do nothing.
    }
  }
  if (Util.SDK_INT >= 23) {
    audioTrackUtil = new AudioTrackUtilV23();
  } else if (Util.SDK_INT >= 19) {
    audioTrackUtil = new AudioTrackUtilV19();
  } else {
    audioTrackUtil = new AudioTrackUtil();
  }
  playheadOffsets = new long[MAX_PLAYHEAD_OFFSET_COUNT];
  volume = 1.0f;
  startMediaTimeState = START_NOT_SET;
}

c – std :: condition_variable :: wait_until的实现

c – std :: condition_variable :: wait_until的实现

我正在阅读std :: condition_variable :: wait_until的libstdc implementation,这里是源代码:

template<typename _Clock,typename _Duration>
  cv_status
  wait_until(unique_lock<mutex>& __lock,const chrono::time_point<_Clock,_Duration>& __atime)
  {
    // DR 887 - Sync unkNown clock to kNown clock.
    const typename _Clock::time_point __c_entry = _Clock::Now();
    const __clock_t::time_point __s_entry = __clock_t::Now();
    const auto __delta = __atime - __c_entry;
    const auto __s_atime = __s_entry + __delta;

    return __wait_until_impl(__lock,__s_atime);
  }

template<typename _Clock,typename _Duration,typename _Predicate>
  bool
  wait_until(unique_lock<mutex>& __lock,_Duration>& __atime,_Predicate __p)
  {
    while (!__p())
      if (wait_until(__lock,__atime) == cv_status::timeout)
        return __p();
    return true;
  }

第二个函数调用循环中的第一个函数.它将执行时钟同步操作.如果我们调用第二个函数,同步操作可能会运行多次.每次都需要同步时钟吗?我认为代码可以改进在第二个功能中只通过同步时钟一次.对吧?

解决方法

是的,不.

是的,可以优化此代码,以便在循环时不操作time_points.但是,我不确定这是否真的有必要.

考虑是什么使得谓词wait_until循环.

当notified时,它检查谓词以查看是否有工作要做.

>如果谓词返回true,即条件变量保护的条件为真,则wait_until返回true.
>如果超时过去,wait_until将返回谓词的值,该值通常为false(否则我们会预期condition_variable已被通知).

这只留下一个循环实际循环的情况:当通知condition_variable时,谓词返回false.

这被称为spurious wakeup,并不是典型的情况,所以它并不值得优化.

c – `std :: condition_variable :: wait_for`经常调用谓词

c – `std :: condition_variable :: wait_for`经常调用谓词

请考虑以下codesnippet:
#include <iostream>
#include <condition_variable>
#include <chrono>
#include <mutex>

int main () {
  std::mutex y;
  std::condition_variable x;
  std::unique_lock<std::mutex>lock{y};
  int i = 0;
  auto increment = [&] {++i; return false;};
  using namespace std::chrono_literals;

  //lock 5s if increment returns false
  //let's see how often was increment called?
  x.wait_for(lock,5s,increment);
  std::cout << i << std::endl;

  //compare this with a simple loop:
  //how often can my system call increment in 5s?
  auto const end = std::chrono::system_clock::Now() + 5s;
  i = 0;
  while (std::chrono::system_clock::Now() < end) {
    increment();
  }
  std::cout << i;
}

据我所知wait_for,在wait_for之后我应该是O(1)(让我们假设虚假解锁很少见).

但是,我明白了
对于内核4.17.14,Intel(R)Core(TM)i7-6700 cpu @ 3.40GHz,i~ = 3e8,
对于内核3.10.0,Intel(R)Xeon(R)cpu E5-2630 v4 @ 2.20GHz,i~ = 8e6.

这听起来很有趣,所以我通过比较一个运行5秒的简单循环来检查. i的结果大致相同,只有5-10%的差异.

题:
wait_for在做什么?它是否按预期工作,我只是理解cppreference错误,或者我搞砸了?

第二,(可选)问题:我的这个巨大差异来自哪里?

附加信息:
(gcc7.3,gcc8.2,clang6.0),flags:-O3 –std = c 17都可以得到可比较的结果.

解决方法

libstdc有一个不幸的编译和看似没有pthread工作的能力,但它不能正常运行.

请参阅此libstdc错误:https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58929

您需要在编译命令中添加“-pthread”.

今天关于C++11并行编程-条件变量(condition_variable)详细说明c++条件变量使用的介绍到此结束,谢谢您的阅读,有关android ConditionVariable、android.os.ConditionVariable的实例源码、c – std :: condition_variable :: wait_until的实现、c – `std :: condition_variable :: wait_for`经常调用谓词等更多相关知识的信息可以在本站进行查询。

本文标签: