Muokkaa

Jaa


ThreadStaticAttribute Class

Definition

Indicates that the value of a static field is unique for each thread.

public ref class ThreadStaticAttribute : Attribute
[System.AttributeUsage(System.AttributeTargets.Field, Inherited=false)]
public class ThreadStaticAttribute : Attribute
[System.AttributeUsage(System.AttributeTargets.Field, Inherited=false)]
[System.Serializable]
public class ThreadStaticAttribute : Attribute
[System.AttributeUsage(System.AttributeTargets.Field, Inherited=false)]
[System.Serializable]
[System.Runtime.InteropServices.ComVisible(true)]
public class ThreadStaticAttribute : Attribute
[<System.AttributeUsage(System.AttributeTargets.Field, Inherited=false)>]
type ThreadStaticAttribute = class
    inherit Attribute
[<System.AttributeUsage(System.AttributeTargets.Field, Inherited=false)>]
[<System.Serializable>]
type ThreadStaticAttribute = class
    inherit Attribute
[<System.AttributeUsage(System.AttributeTargets.Field, Inherited=false)>]
[<System.Serializable>]
[<System.Runtime.InteropServices.ComVisible(true)>]
type ThreadStaticAttribute = class
    inherit Attribute
Public Class ThreadStaticAttribute
Inherits Attribute
Inheritance
ThreadStaticAttribute
Attributes

Examples

The following example instantiates a random number generator, creates ten threads in addition to the main thread, and then generates two million random numbers in each thread. It uses the ThreadStaticAttribute attribute to calculate the sum and the count of random numbers per thread. It also defines two additional per-thread fields, previous and abnormal, that allows it to detect corruption of the random number generator.

using System;
using System.Threading;

public class Example
{
   [ThreadStatic] static double previous = 0.0;
   [ThreadStatic] static double sum = 0.0;
   [ThreadStatic] static int calls = 0;
   [ThreadStatic] static bool abnormal;
   static int totalNumbers = 0;
   static CountdownEvent countdown;
   private static Object lockObj;
   Random rand;
   
   public Example()
   { 
      rand = new Random();
      lockObj = new Object();
      countdown = new CountdownEvent(1);
   } 

   public static void Main()
   {
      Example ex = new Example();
      Thread.CurrentThread.Name = "Main";
      ex.Execute();
      countdown.Wait();
      Console.WriteLine("{0:N0} random numbers were generated.", totalNumbers);
   }

   private void Execute()
   {   
      for (int threads = 1; threads <= 10; threads++)
      {
         Thread newThread = new Thread(new ThreadStart(this.GetRandomNumbers));
         countdown.AddCount();
         newThread.Name = threads.ToString();
         newThread.Start();
      }
      this.GetRandomNumbers();
   }

   private void GetRandomNumbers()
   {
      double result = 0.0;

      for (int ctr = 0; ctr < 2000000; ctr++)
      {
         lock (lockObj) {
            result = rand.NextDouble();
            calls++;
            Interlocked.Increment(ref totalNumbers);
            // We should never get the same random number twice.
            if (result == previous) {
               abnormal = true;
               break;
            }
            else {
               previous = result;
               sum += result;
            }   
         }
      }
      // get last result
      if (abnormal)
         Console.WriteLine("Result is {0} in {1}", previous, Thread.CurrentThread.Name);
       
      Console.WriteLine("Thread {0} finished random number generation.", Thread.CurrentThread.Name);
      Console.WriteLine("Sum = {0:N4}, Mean = {1:N4}, n = {2:N0}\n", sum, sum/calls, calls);        
      countdown.Signal();
   }
}
// The example displays output similar to the following:
//    Thread 1 finished random number generation.
//    Sum = 1,000,556.7483, Mean = 0.5003, n = 2,000,000
//    
//    Thread 6 finished random number generation.
//    Sum = 999,704.3865, Mean = 0.4999, n = 2,000,000
//    
//    Thread 2 finished random number generation.
//    Sum = 999,680.8904, Mean = 0.4998, n = 2,000,000
//    
//    Thread 10 finished random number generation.
//    Sum = 999,437.5132, Mean = 0.4997, n = 2,000,000
//    
//    Thread 8 finished random number generation.
//    Sum = 1,000,663.7789, Mean = 0.5003, n = 2,000,000
//    
//    Thread 4 finished random number generation.
//    Sum = 999,379.5978, Mean = 0.4997, n = 2,000,000
//    
//    Thread 5 finished random number generation.
//    Sum = 1,000,011.0605, Mean = 0.5000, n = 2,000,000
//    
//    Thread 9 finished random number generation.
//    Sum = 1,000,637.4556, Mean = 0.5003, n = 2,000,000
//    
//    Thread Main finished random number generation.
//    Sum = 1,000,676.2381, Mean = 0.5003, n = 2,000,000
//    
//    Thread 3 finished random number generation.
//    Sum = 999,951.1025, Mean = 0.5000, n = 2,000,000
//    
//    Thread 7 finished random number generation.
//    Sum = 1,000,844.5217, Mean = 0.5004, n = 2,000,000
//    
//    22,000,000 random numbers were generated.
open System
open System.Threading

type Example() =
    [<ThreadStatic; DefaultValue>] 
    static val mutable private previous : double

    [<ThreadStatic; DefaultValue>] 
    static val mutable private sum : double
    
    [<ThreadStatic; DefaultValue>] 
    static val mutable private calls : int

    [<ThreadStatic; DefaultValue>] 
    static val mutable private abnormal : bool
   
    static let mutable totalNumbers = 0
    static let countdown = new CountdownEvent(1)
    static let lockObj = obj ()
    let rand = Random()


    member this.Execute() =
        for threads = 1 to 10 do
            let newThread = new Thread(ThreadStart this.GetRandomNumbers)
            countdown.AddCount()
            newThread.Name <- threads.ToString()
            newThread.Start()
        this.GetRandomNumbers()
        countdown.Wait()
        printfn $"{totalNumbers:N0} random numbers were generated."

    member _.GetRandomNumbers() =
        let mutable i = 0
        while i < 2000000 do
            lock lockObj (fun () ->
                let result = rand.NextDouble()
                Example.calls <- Example.calls + 1
                Interlocked.Increment &totalNumbers |> ignore
                // We should never get the same random number twice.
                if result = Example.previous then
                    Example.abnormal <- true
                    i <- 2000001 // break
                else
                    Example.previous <- result
                    Example.sum <- Example.sum + result )
            i <- i + 1
        // get last result
        if Example.abnormal then
            printfn $"Result is {Example.previous} in {Thread.CurrentThread.Name}"
        
        printfn $"Thread {Thread.CurrentThread.Name} finished random number generation."
        printfn $"Sum = {Example.sum:N4}, Mean = {Example.sum / float Example.calls:N4}, n = {Example.calls:N0}\n"
        countdown.Signal() |> ignore

let ex = Example()
Thread.CurrentThread.Name <- "Main"
ex.Execute()

// The example displays output similar to the following:
//    Thread 1 finished random number generation.
//    Sum = 1,000,556.7483, Mean = 0.5003, n = 2,000,000
//    
//    Thread 6 finished random number generation.
//    Sum = 999,704.3865, Mean = 0.4999, n = 2,000,000
//    
//    Thread 2 finished random number generation.
//    Sum = 999,680.8904, Mean = 0.4998, n = 2,000,000
//    
//    Thread 10 finished random number generation.
//    Sum = 999,437.5132, Mean = 0.4997, n = 2,000,000
//    
//    Thread 8 finished random number generation.
//    Sum = 1,000,663.7789, Mean = 0.5003, n = 2,000,000
//    
//    Thread 4 finished random number generation.
//    Sum = 999,379.5978, Mean = 0.4997, n = 2,000,000
//    
//    Thread 5 finished random number generation.
//    Sum = 1,000,011.0605, Mean = 0.5000, n = 2,000,000
//    
//    Thread 9 finished random number generation.
//    Sum = 1,000,637.4556, Mean = 0.5003, n = 2,000,000
//    
//    Thread Main finished random number generation.
//    Sum = 1,000,676.2381, Mean = 0.5003, n = 2,000,000
//    
//    Thread 3 finished random number generation.
//    Sum = 999,951.1025, Mean = 0.5000, n = 2,000,000
//    
//    Thread 7 finished random number generation.
//    Sum = 1,000,844.5217, Mean = 0.5004, n = 2,000,000
//    
//    22,000,000 random numbers were generated.
Imports System.Threading

Public Class Example
   <ThreadStatic> Shared previous As Double = 0.0
   <ThreadStatic> Shared sum As Double = 0.0
   <ThreadStatic> Shared calls As Integer = 0
   <ThreadStatic> Shared abnormal As Boolean
   Shared totalNumbers As Integer = 0
   Shared countdown As CountdownEvent
   Private Shared lockObj As Object
   Dim rand As Random

   Public Sub New()
      rand = New Random()
      lockObj = New Object()
      countdown = New CountdownEvent(1)
   End Sub

   Public Shared Sub Main()
      Dim ex As New Example()
      Thread.CurrentThread.Name = "Main"
      ex.Execute()
      countdown.Wait()
      Console.WriteLine("{0:N0} random numbers were generated.", totalNumbers)
   End Sub

   Private Sub Execute()
      For threads As Integer = 1 To 10
         Dim newThread As New Thread(New ThreadStart(AddressOf GetRandomNumbers))
         countdown.AddCount()
         newThread.Name = threads.ToString()
         newThread.Start()
      Next
      Me.GetRandomNumbers()
   End Sub

   Private Sub GetRandomNumbers()
      Dim result As Double = 0.0
      
       
      For ctr As Integer = 1 To 2000000
         SyncLock lockObj
            result = rand.NextDouble()
            calls += 1
            Interlocked.Increment(totalNumbers)
            ' We should never get the same random number twice.
            If result = previous Then
               abnormal = True
               Exit For
            Else
               previous = result
               sum += result
            End If   
         End SyncLock
      Next
      ' Get last result.
      If abnormal Then
         Console.WriteLine("Result is {0} in {1}", previous, Thread.CurrentThread.Name)
      End If       
      
      Console.WriteLine("Thread {0} finished random number generation.", Thread.CurrentThread.Name)
      Console.WriteLine("Sum = {0:N4}, Mean = {1:N4}, n = {2:N0}", sum, sum/calls, calls)
      Console.WriteLine()        
      countdown.Signal()
   End Sub
End Class
' The example displays output similar to the following:
'    Thread 1 finished random number generation.
'    Sum = 1,000,556.7483, Mean = 0.5003, n = 2,000,000
'    
'    Thread 6 finished random number generation.
'    Sum = 999,704.3865, Mean = 0.4999, n = 2,000,000
'    
'    Thread 2 finished random number generation.
'    Sum = 999,680.8904, Mean = 0.4998, n = 2,000,000
'    
'    Thread 10 finished random number generation.
'    Sum = 999,437.5132, Mean = 0.4997, n = 2,000,000
'    
'    Thread 8 finished random number generation.
'    Sum = 1,000,663.7789, Mean = 0.5003, n = 2,000,000
'    
'    Thread 4 finished random number generation.
'    Sum = 999,379.5978, Mean = 0.4997, n = 2,000,000
'    
'    Thread 5 finished random number generation.
'    Sum = 1,000,011.0605, Mean = 0.5000, n = 2,000,000
'    
'    Thread 9 finished random number generation.
'    Sum = 1,000,637.4556, Mean = 0.5003, n = 2,000,000
'    
'    Thread Main finished random number generation.
'    Sum = 1,000,676.2381, Mean = 0.5003, n = 2,000,000
'    
'    Thread 3 finished random number generation.
'    Sum = 999,951.1025, Mean = 0.5000, n = 2,000,000
'    
'    Thread 7 finished random number generation.
'    Sum = 1,000,844.5217, Mean = 0.5004, n = 2,000,000
'    
'    22,000,000 random numbers were generated.

The example uses the lock statement in C#, the lock function in F#, and the SyncLock construct in Visual Basic to synchronize access to the random number generator. This prevents corruption of the random number generator, which typically results in its returning a value of zero for all subsequent calls.

The example also uses the CountdownEvent class to ensure that each thread has finished generating random numbers before it displays the total number of calls. Otherwise, if the main thread completes execution before the additional threads that it spawns, it displays an inaccurate value for the total number of method calls.

Remarks

A static field marked with ThreadStaticAttribute is not shared between threads. Each executing thread has a separate instance of the field, and independently sets and gets values for that field. If the field is accessed on a different thread, it will contain a different value.

Note that in addition to applying the ThreadStaticAttribute attribute to a field, you must also define it as a static field (in C# or F#) or a Shared field (in Visual Basic).

Note

Do not specify initial values for fields marked with ThreadStaticAttribute, because such initialization occurs only once, when the class constructor executes, and therefore affects only one thread. If you do not specify an initial value, you can rely on the field being initialized to its default value if it is a value type, or to null if it is a reference type.

Use this attribute as it is, and do not derive from it.

For more information about using attributes, see Attributes.

Constructors

ThreadStaticAttribute()

Initializes a new instance of the ThreadStaticAttribute class.

Properties

TypeId

When implemented in a derived class, gets a unique identifier for this Attribute.

(Inherited from Attribute)

Methods

Equals(Object)

Returns a value that indicates whether this instance is equal to a specified object.

(Inherited from Attribute)
GetHashCode()

Returns the hash code for this instance.

(Inherited from Attribute)
GetType()

Gets the Type of the current instance.

(Inherited from Object)
IsDefaultAttribute()

When overridden in a derived class, indicates whether the value of this instance is the default value for the derived class.

(Inherited from Attribute)
Match(Object)

When overridden in a derived class, returns a value that indicates whether this instance equals a specified object.

(Inherited from Attribute)
MemberwiseClone()

Creates a shallow copy of the current Object.

(Inherited from Object)
ToString()

Returns a string that represents the current object.

(Inherited from Object)

Explicit Interface Implementations

_Attribute.GetIDsOfNames(Guid, IntPtr, UInt32, UInt32, IntPtr)

Maps a set of names to a corresponding set of dispatch identifiers.

(Inherited from Attribute)
_Attribute.GetTypeInfo(UInt32, UInt32, IntPtr)

Retrieves the type information for an object, which can be used to get the type information for an interface.

(Inherited from Attribute)
_Attribute.GetTypeInfoCount(UInt32)

Retrieves the number of type information interfaces that an object provides (either 0 or 1).

(Inherited from Attribute)
_Attribute.Invoke(UInt32, Guid, UInt32, Int16, IntPtr, IntPtr, IntPtr, IntPtr)

Provides access to properties and methods exposed by an object.

(Inherited from Attribute)

Applies to

See also