(精华)2020年12月23日 .NET Core 多线程底层详解(互斥锁)

tech2024-11-29  10

互斥锁:Monitor 和 mutex

定义:private static readonly object Lock = new object();

使用:Monitor.Enter(Lock); //todo Monitor.Exit(Lock);

作用:将会锁住代码块的内容,并阻止其他线程进入该代码块,直到该代码块运行完成,释放该锁。

注意:定义的锁对象应该是 私有的,静态的,只读的,引用类型的对象,这样可以防止外部改变锁对象

Monitor有TryEnter的功能,可以防止出现死锁的问题,lock没有

定义:private static readonly Mutex mutex = new Mutex();

使用:mutex.WaitOne(); //todo mutex.ReleaseMutex();

作用:将会锁住代码块的内容,并阻止其他线程进入该代码块,直到该代码块运行完成,释放该锁。

注意:定义的锁对象应该是 私有的,静态的,只读的,引用类型的对象,这样可以防止外部改变锁对象

Mutex本身是可以系统级别的,所以是可以跨越进程的

Monitor 测试实现:

using System; using System.Text; using System.Threading; using System.Threading.Tasks; namespace Sample5_2_monitor_lock { class Program { private static int _TaskNum = 3; private static Task[] _Tasks; private static StringBuilder _StrBlder; private const int RUN_LOOP = 50; private static void Work1(int TaskID) { int i = 0; string log = ""; bool lockToken = false; while (i < RUN_LOOP) { log = String.Format("Time: {0} Task : #{1} Value: {2} =====\n", DateTime.Now.TimeOfDay, TaskID, i); i++; try { lockToken = false; Monitor.Enter(_StrBlder, ref lockToken); _StrBlder.Append(log); } finally { if (lockToken) Monitor.Exit(_StrBlder); } } } private static void Work2(int TaskID) { int i = 0; string log = ""; bool lockToken = false; while (i < RUN_LOOP) { log = String.Format("Time: {0} Task : #{1} Value: {2} *****\n", DateTime.Now.TimeOfDay, TaskID, i); i++; try { lockToken = false; Monitor.Enter(_StrBlder, ref lockToken); _StrBlder.Append(log); } finally { if (lockToken) Monitor.Exit(_StrBlder); } } } private static void Work3(int TaskID) { int i = 0; string log = ""; bool lockToken = false; while (i < RUN_LOOP) { log = String.Format("Time: {0} Task : #{1} Value: {2} ~~~~~\n", DateTime.Now.TimeOfDay, TaskID, i); i++; try { lockToken = false; Monitor.Enter(_StrBlder, ref lockToken); _StrBlder.Append(log); } finally { if (lockToken) Monitor.Exit(_StrBlder); } } } static void Main(string[] args) { _Tasks = new Task[_TaskNum]; _StrBlder = new StringBuilder(); _Tasks[0] = Task.Factory.StartNew((num) => { var taskid = (int)num; Work1(taskid); }, 0); _Tasks[1] = Task.Factory.StartNew((num) => { var taskid = (int)num; Work2(taskid); }, 1); _Tasks[2] = Task.Factory.StartNew((num) => { var taskid = (int)num; Work3(taskid); }, 2); var finalTask = Task.Factory.ContinueWhenAll(_Tasks, (tasks) => { Task.WaitAll(_Tasks); Console.WriteLine("=========================================================="); Console.WriteLine("All Phase is completed"); Console.WriteLine("=========================================================="); Console.WriteLine(_StrBlder); }); try { finalTask.Wait(); } catch (AggregateException aex) { Console.WriteLine("Task failed And Canceled" + aex.ToString()); } finally { } Console.ReadLine(); } } }

锁超时的使用:

其中主要使用的是 Monitor.TryEnter(),函数,其中多了一个设置超时时间的参数。 代码中让每个锁的超时Timer为2秒,在Work1中挺顿5秒,这样造成了Work2和Work3的超时。

using System; using System.Text; using System.Threading; using System.Threading.Tasks; namespace Sample5_3_monitor_lock_timeout { class Program { private static int _TaskNum = 3; private static Task[] _Tasks; private static StringBuilder _StrBlder; private const int RUN_LOOP = 50; private static void Work1(int TaskID) { int i = 0; string log = ""; bool lockToken = false; while (i < RUN_LOOP) { log = String.Format("Time: {0} Task : #{1} Value: {2} =====\n", DateTime.Now.TimeOfDay, TaskID, i); i++; try { lockToken = false; Monitor.TryEnter(_StrBlder, 2000, ref lockToken); if (!lockToken) { Console.WriteLine("Work1 TIMEOUT!! Will throw Exception"); throw new TimeoutException("Work1 TIMEOUT!!"); } System.Threading.Thread.Sleep(5000); _StrBlder.Append(log); } finally { if (lockToken) Monitor.Exit(_StrBlder); } } } private static void Work2(int TaskID) { int i = 0; string log = ""; bool lockToken = false; while (i < RUN_LOOP) { log = String.Format("Time: {0} Task : #{1} Value: {2} *****\n", DateTime.Now.TimeOfDay, TaskID, i); i++; try { lockToken = false; Monitor.TryEnter(_StrBlder, 2000, ref lockToken); if (!lockToken) { Console.WriteLine("Work2 TIMEOUT!! Will throw Exception"); throw new TimeoutException("Work2 TIMEOUT!!"); } _StrBlder.Append(log); } finally { if (lockToken) Monitor.Exit(_StrBlder); } } } private static void Work3(int TaskID) { int i = 0; string log = ""; bool lockToken = false; while (i < RUN_LOOP) { log = String.Format("Time: {0} Task : #{1} Value: {2} ~~~~~\n", DateTime.Now.TimeOfDay, TaskID, i); i++; try { lockToken = false; Monitor.TryEnter(_StrBlder, 2000, ref lockToken); if (!lockToken) { Console.WriteLine("Work3 TIMEOUT!! Will throw Exception"); throw new TimeoutException("Work3 TIMEOUT!!"); } _StrBlder.Append(log); } finally { if (lockToken) Monitor.Exit(_StrBlder); } } } static void Main(string[] args) { _Tasks = new Task[_TaskNum]; _StrBlder = new StringBuilder(); _Tasks[0] = Task.Factory.StartNew((num) => { var taskid = (int)num; Work1(taskid); }, 0); _Tasks[1] = Task.Factory.StartNew((num) => { var taskid = (int)num; Work2(taskid); }, 1); _Tasks[2] = Task.Factory.StartNew((num) => { var taskid = (int)num; Work3(taskid); }, 2); var finalTask = Task.Factory.ContinueWhenAll(_Tasks, (tasks) => { Task.WaitAll(_Tasks); Console.WriteLine("=========================================================="); Console.WriteLine("All Phase is completed"); Console.WriteLine("=========================================================="); Console.WriteLine(_StrBlder); }); try { finalTask.Wait(); } catch (AggregateException aex) { Console.WriteLine("Task failed And Canceled" + aex.ToString()); } finally { } Console.ReadLine(); } } }

mutex 测试实现 :

1、initiallyOwned表示创建mutex的线程是否拥有该互斥体。true表示创建线程拥有互斥锁,只有在创建线程中调用ReleaseMutex释放后,其他等待线程才能参与抢夺互斥体的活动。false表示互斥锁体于与空闲状态,其他等待互斥锁的线程立即参与到抢夺互斥锁的活动中去。

2、在上面程序中如果创建mutex时使用true参数,故在启动其他线程后必须执行mutex.ReleaseMutex(),如果不释放mutex,则其他线程将一直等待下去。使用ture,相当于一创建就使用waitone()

3、mutex.WaitOne()与mutex.ReleaseMutex()要像 { } 一样配对使用,否则将出现 “由于出现被放弃的 mutex,等待过程结束” 的异常

4、mutex与monitor相比,没有暂时释放的功能;因此mutex一经释放,原释放资源的线程也将重新参与新一轮对mutex的争夺过程。

using (var mutex = new Mutex(false, "name")) { try { mutex.WaitOne(); //do something } catch(Exception ex) { throw ex; } finally { mutex.ReleaseMutex(); } }

简单例子:实现递归

//递归里的锁变量不相干 public static class RecursiveSample { private static readonly Mutex Lock = new Mutex(); private static int _counterA = 0; private static int _counterB = 0; public static void IncrementCounterA() { // 获取锁 Lock.WaitOne(); try { _counterA++; _counterB++; } finally { // 释放锁 Lock.ReleaseMutex(); } } public static void IncrementCounterB() { // 获取锁 Lock.WaitOne(); try { _counterA++; _counterB++; } finally { // 释放锁 Lock.ReleaseMutex(); } } public static void IncrementCounter() { // 获取锁 Lock.WaitOne(); try { IncrementCounterA(); IncrementCounterB(); } finally { // 释放锁 Lock.ReleaseMutex(); } } }
最新回复(0)