Threads introduce a tremendous amount of overhead to your process, both in terms of memory consumption and CPU time. 除非有必要，尽量不要使用Threads. 

Run Loop:

A run loop is a piece of infrastructure used to manage events arriving asynchronously on a thread. A run loop works by monitoring one or more event sources for the thread. As events arrive, the system wakes up the thread and dispatches the events to the run loop, which then dispatches them to the handlers you specify. If no events are present and ready to be handled, the run loop puts the thread to sleep.

You are not required to use a run loop with any threads you create but doing so can provide a better experience for the user. Run loops make it possible to create long-lived threads that use a minimal amount of resources. Because a run loop puts its thread to sleep when there is nothing to do, it eliminates the need for polling, which wastes CPU cycles and prevents the processor itself from sleeping and saving power.

To configure a run loop, all you have to do is launch your thread, get a reference to the run loop object, install your event handlers, and tell the run loop to run. The infrastructure provided by both Cocoa and Carbon handles the configuration of the main thread’s run loop for you automatically. If you plan to create long-lived secondary threads, however, you must configure the run loop for those threads yourself.

 

Synchronization Tools:

One of the hazards of threaded programming is resource contention among multiple threads. If multiple threads try to use or modify the same resource at the same time, problems can occur. One way to alleviate the problem is to eliminate the shared resource altogether and make sure each thread has its own distinct set of resources on which to operate. When maintaining completely separate resources is not an option though, you may have to synchronize access to the resource using locks, conditions, atomic operations, and other techniques.

Locks provide a brute force form of protection for code that can be executed by only one thread at a time. The most common type of lock is mutual exclusion lock, also known as a 

mutex. When a thread tries to acquire a mutex that is currently held by another thread, it blocks until the lock is released by the other thread. Several system frameworks provide support for mutex locks, although they are all based on the same underlying technology. In addition, Cocoa provides several variants of the mutex lock to support different types of behavior, such as recursion.

In addition to locks, the system provides support for conditions, which ensure the proper sequencing of tasks within your application. A condition acts as a gatekeeper, blocking a given thread until the condition it represents becomes true. When that happens, the condition releases the thread and allows it to continue. The POSIX layer and Foundation framework both provide direct support for conditions. (If you use operation objects, you can configure dependencies among your operation objects to sequence the execution of tasks, which is very similar to the behavior offered by conditions.)

Although locks and conditions are very common in concurrent design, atomic operations are another way to protect and synchronize access to data. Atomic operations offer a lightweight alternative to locks in situations where you can perform mathematical or logical operations on scalar data types. Atomic operations use special hardware instructions to ensure that modifications to a variable are completed before other threads have a chance to access it.

Design Tips:

Writing thread-creation code manually is tedious and potentially error-prone and you should avoid it whenever possible. Mac OS X and iOS provide implicit support for concurrency through other APIs. Rather than create a thread yourself, consider using asynchronous APIs, GCD, or operation objects to do the work. These technologies do the thread-related work behind the scenes for you and are guaranteed to do it correctly. In addition, technologies such as GCD and operation objects are designed to manage threads much more efficiently than your own code ever could by adjusting the number of active threads based on the current system load.

If your application has a graphical user interface, it is recommended that you receive user-related events and initiate interface updates from your application’s main thread. 

Create a Thread:

There are two ways to create a thread using the  class:

• Use the  class method to spawn the new thread.

• Create a new NSThread object and call its start method. (Supported only in iOS and Mac OS X v10.5 and later.)

If you have an 

NSThread object whose thread is currently running, one way you can send messages to that thread is to use the 

 method of almost any object in your application.

In iOS and Mac OS X v10.5 and later, all objects have the ability to spawn a new thread and use it to execute one of their methods. The 

 method creates a new detached thread and uses the specified method as the entry point for the new thread.  The effect of calling this method is the same as if you called the 

method of 

 with the current object, selector, and parameter object as parameters. The new thread is spawned immediately using the default configuration and begins running. Inside the selector, you must configure the thread just as you would any thread. For example, you would need to set up an autorelease pool (if you were not using garbage collection) and configure the thread’s run loop if you planned to use it. 

Writing Your Thread Entry Routine:

- (void)myThreadMainRoutine

{

NSAutoreleasePool *pool = [[NSAutoreleasePool alloc] init]; // Top-level pool

// Do thread work here.

[pool release];  // Release the objects in the pool.

}

Because the top-level autorelease pool does not release its objects until the thread exits, long-lived threads should create additional autorelease pools to free objects more frequently. For example, a thread that uses a run loop might create and release an autorelease pool each time through that run loop. Releasing objects more frequently prevents your application’s memory footprint from growing too large, which can lead to performance problems. As with any performance-related behavior though, you should measure the actual performance of your code and tune your use of autorelease pools appropriately.

When writing code you want to run on a separate thread, you have two options. The first option is to write the code for a thread as one long task to be performed with little or no interruption, and have the thread exit when it finishes. The second option is put your thread into a loop and have it process requests dynamically as they arrive. The first option requires no special setup for your code; you just start doing the work you want to do. The second option, however, involves setting up your thread’s run loop.

Terminating a Thread:

The recommended way to exit a thread is to let it exit its entry point routine normally. Although Cocoa, POSIX, and Multiprocessing Services offer routines for killing threads directly, the use of such routines is strongly discouraged. Killing a thread prevents that thread from cleaning up after itself. Memory allocated by the thread could potentially be leaked and any other resources currently in use by the thread might not be cleaned up properly, creating potential problems later.

If you anticipate the need to terminate a thread in the middle of an operation, you should design your threads from the outset to respond to a cancel or exit message. For long-running operations, this might mean stopping work periodically and checking to see if such a message arrived. If a message does come in asking the thread to exit, the thread would then have the opportunity to perform any needed cleanup and exit gracefully; otherwise, it could simply go back to work and process the next chunk of data.

One way to respond to cancel messages is to use a run loop input source to receive such messages.

Listing 2-3  Checking for an exit condition during a long job

- (void)threadMainRoutine

{

BOOL moreWorkToDo = YES;

BOOL exitNow = NO;

NSRunLoop* runLoop = [NSRunLoop currentRunLoop];

// Add the exitNow BOOL to the thread dictionary.

NSMutableDictionary* threadDict = [[NSThread currentThread] threadDictionary];

[threadDict setValue:[NSNumber numberWithBool:exitNow] forKey:@"ThreadShouldExitNow"];

// Install an input source.

[self myInstallCustomInputSource];

while (moreWorkToDo && !exitNow)

{

// Do one chunk of a larger body of work here.

// Change the value of the moreWorkToDo Boolean when done.

// Run the run loop but timeout immediately if the input source isn't waiting to fire.

[runLoop runUntilDate:[NSDate date]];

// Check to see if an input source handler changed the exitNow value.

exitNow = [[threadDict valueForKey:@"ThreadShouldExitNow"] boolValue];

}

}

Anatomy of Run Loops:

 A 

run loop is an event processing loop that you use to schedule work and coordinate the receipt of incoming events. The purpose of a run loop is to keep your thread busy when there is work to do and put your thread to sleep when there is none.

Run loop management is not entirely automatic. You must still design your thread’s code to start the run loop at appropriate times and respond to incoming events. Both Cocoa and Core Foundation provide 

run loop objects to help you configure and manage your thread’s run loop. Your application does not need to create these objects explicitly; each thread, including the application’s main thread, has an associated run loop object. Only secondary threads need to run their run loop explicitly, however. In both Carbon and Cocoa applications, the main thread automatically sets up and runs its run loop as part of the general application startup process.

A run loop is very much like its name sounds. It is a loop your thread enters and uses to run event handlers in response to incoming events. Your code provides the control statements used to implement the actual loop portion of the run loop—in other words, your code provides the 

while or 

for loop that drives the run loop. Within your loop, you use a run loop object to "run” the event-processing code that receives events and calls the installed handlers.

A run loop receives events from two different types of sources. 

Input sources deliver asynchronous events, usually messages from another thread or from a different application. 

Timer sources deliver synchronous events, occurring at a scheduled time or repeating interval. Both types of source use an application-specific handler routine to process the event when it arrives.

A 

run loop mode is a collection of input sources and timers to be monitored and a collection of run loop observers to be notified. Each time you run your run loop, you specify (either explicitly or implicitly) a particular “mode” in which to run. During that pass of the run loop, only sources associated with that mode are monitored and allowed to deliver their events. (Similarly, only observers associated with that mode are notified of the run loop’s progress.) Sources associated with other modes hold on to any new events until subsequent passes through the loop in the appropriate mode.

In your code, you identify modes by name. Both Cocoa and Core Foundation define a default mode and several commonly used modes, along with strings for specifying those modes in your code. You can define custom modes by simply specifying a custom string for the mode name. Although the names you assign to custom modes are arbitrary, the contents of those modes are not. You must be sure to add one or more input sources, timers, or run-loop observers to any modes you create for them to be useful.

For secondary threads, you need to decide whether a run loop is necessary, and if it is, configure and start it yourself. You do not need to start a thread’s run loop in all cases. For example, if you use a thread to perform some long-running and predetermined task, you can probably avoid starting the run loop. Run loops are intended for situations where you want more interactivity with the thread. For example, you need to start a run loop if you plan to do any of the following:

• Use ports or custom input sources to communicate with other threads.

• Use timers on the thread.

• Use any of the performSelector… methods in a Cocoa application.

• Keep the thread around to perform periodic tasks.

Using Run Loop Objects:

(1) Getting a Run Loop Object

To get the run loop for the current thread, you use one of the following:

• In a Cocoa application, use the  class method of  to retrieve an NSRunLoop object.

• Use the  function.

Although they are not toll-free bridged types, you can get a CFRunLoopRef opaque type from an NSRunLoop object when needed. The NSRunLoop class defines a  method that returns a CFRunLoopRef type that you can pass to Core Foundation routines. Because both objects refer to the same run loop, you can intermix calls to the NSRunLoop object andCFRunLoopRef opaque type as needed.

(2) Configure the Run Loop

Before you run a run loop on a secondary thread, you must add at least one input source or timer to it. If a run loop does not have any sources to monitor, it exits immediately when you try to run it.

(3) Start the Run Loop

Starting the run loop is necessary only for the secondary threads in your application. A run loop must have at least one input source or timer to monitor. If one is not attached, the run loop exits immediately.

There are several ways to start the run loop, including the following:

• Unconditionally

• With a set time limit

• In a particular mode

Entering your run loop unconditionally is the simplest option, but it is also the least desirable. Running your run loop unconditionally puts the thread into a permanent loop, which gives you very little control over the run loop itself. You can add and remove input sources and timers, but the only way to stop the run loop is to kill it. There is also no way to run the run loop in a custom mode.

Instead of running a run loop unconditionally, it is better to run the run loop with a timeout value. When you use a timeout value, the run loop runs until an event arrives or the allotted time expires. If an event arrives, that event is dispatched to a handler for processing and then the run loop exits. Your code can then restart the run loop to handle the next event. If the allotted time expires instead, you can simply restart the run loop or use the time to do any needed housekeeping.

In addition to a timeout value, you can also run your run loop using a specific mode. Modes and timeout values are not mutually exclusive and can both be used when starting a run loop.

(4) Exiting a Run Loop

There are two ways to make a run loop exit before it has processed an event:

• Configure the run loop to run with a timeout value.

• Tell the run loop to stop.

Using a timeout value is certainly preferred, if you can manage it. Specifying a timeout value lets the run loop finish all of its normal processing, including delivering notifications to run loop observers, before exiting.

Stopping the run loop explicitly with the  function produces a result similar to a timeout. The run loop sends out any remaining run-loop notifications and then exits. The difference is that you can use this technique on run loops you started unconditionally.

Synchronization:

Synchronization helps ensure the correctness of your code, but does so at the expense of performance. The use of synchronization tools introduces delays, even in uncontested cases. Locks and atomic operations generally involve the use of memory barriers and kernel-level synchronization to ensure code is properly protected. And if there is contention for a lock, your threads could block and experience even greater delays.

Using Locks:

(1) NSLock

An 

NSLock object is used to coordinate the operation of multiple threads of execution within the same application. An

NSLock object can be used to mediate access to an application’s global data or to protect a critical section of code, allowing it to run atomically.

You should not use this class to implement a recursive lock. Calling the lock method twice on the same thread will lock up your thread permanently. Use the NSRecursiveLock class to implement recursive locks instead.

Unlocking a lock that is not locked is considered a programmer error and should be fixed in your code. The NSLock class reports such errors by printing an error message to the console when they occur.

(2) Using the @synchronized Directive

The 

@synchronized directive is a convenient way to create mutex locks on the fly in Objective-C code. The 

@synchronizeddirective does what any other mutex lock would do—it prevents different threads from acquiring the same lock at the same time. In this case, however, you do not have to create the mutex or lock object directly.

 

The object passed to the @synchronized directive is a unique identifier used to distinguish the protected block. If you execute the preceding method in two different threads, passing a different object for the anObj parameter on each thread, each would take its lock and continue processing without being blocked by the other. If you pass the same object in both cases, however, one of the threads would acquire the lock first and the other would block until the first thread completed the critical section.

As a precautionary measure, the @synchronized block implicitly adds an exception handler to the protected code. This handler automatically releases the mutex in the event that an exception is thrown. This means that in order to use the@synchronized directive, you must also enable Objective-C exception handling in your code. If you do not want the additional overhead caused by the implicit exception handler, you should consider using the lock classes.

(3) NSRecursiveLock

The  class defines a lock that can be acquired multiple times by the same thread without causing the thread to deadlock. A recursive lock keeps track of how many times it was successfully acquired. Each successful acquisition of the lock must be balanced by a corresponding call to unlock the lock. Only when all of the lock and unlock calls are balanced is the lock actually released so that other threads can acquire it.

As its name implies, this type of lock is commonly used inside a recursive function to prevent the recursion from blocking the thread. You could similarly use it in the non-recursive case to call functions whose semantics demand that they also take the lock.

(4) NSConditionLock

The 

NSConditionLock class defines objects whose locks can be associated with specific, user-defined conditions. Using an

NSConditionLock object, you can ensure that a thread can acquire a lock only if a certain condition is met. Once it has acquired the lock and executed the critical section of code, the thread can relinquish the lock and set the associated condition to something new. The conditions themselves are arbitrary: you define them as needed for your application.

Typically, you use an 

NSConditionLock object when threads need to perform tasks in a specific order, such as when one thread produces data that another consumes. While the producer is executing, the consumer acquires the lock using a condition that is specific to your program. (The condition itself is just an integer value that you define.) When the producer finishes, it unlocks the lock and sets the lock condition to the appropriate integer value to wake the consumer thread, which then proceeds to process the data.

Using Conditions:

Conditions are a special type of lock that you can use to synchronize the order in which operations must proceed. They differ from mutex locks in a subtle way. A thread waiting on a condition remains blocked until that condition is signaled explicitly by another thread.

The semantics for using an NSCondition object are as follows:

1. Lock the condition object.

2. Test a boolean predicate. (This predicate is a boolean flag or other variable in your code that indicates whether it is safe to perform the task protected by the condition.)

3. If the boolean predicate is false, call the condition object’s wait or waitUntilDate: method to block the thread. Upon returning from these methods, go to step 2 to retest your boolean predicate. (Continue waiting and retesting the predicate until it is true.)

4. If the boolean predicate is true, perform the task.

5. Optionally update any predicates (or signal any conditions) affected by your task.

6. When your task is done, unlock the condition object.

线程安全的类和函数:

http://developer.apple.com/library/mac/#documentation/Cocoa/Conceptual/Multithreading/ThreadSafetySummary/ThreadSafetySummary.html