Kotlin?Dispatchers協(xié)程調(diào)度器源碼深入分析
Dispatchers協(xié)程調(diào)度器
CoroutineDispatcher,具有用于調(diào)度任務(wù)的底層執(zhí)行器。ExecutorCoroutineDispatcher的實(shí)例應(yīng)由調(diào)度程序的所有者關(guān)閉。
此類通常用作基于協(xié)程的API和異步API之間的橋梁,異步API需要Executor的實(shí)例。
根據(jù)各種調(diào)度器的繼承關(guān)系,梳理如下繼承結(jié)構(gòu):
CoroutineDispatcher基類將由所有協(xié)程調(diào)度器實(shí)現(xiàn)擴(kuò)展,kotlin官方實(shí)現(xiàn)了以下四種調(diào)度器:
Dispatchers.Default -如果上下文中未指定調(diào)度器或任何其他ContinuationInterceptor,則所有標(biāo)準(zhǔn)構(gòu)建器都使用默認(rèn)值。它使用共享后臺(tái)線程的公共池。對(duì)于消耗CPU資源的計(jì)算密集型協(xié)程來說,這是一個(gè)合適的選擇。
Dispatchers.IO -使用按需創(chuàng)建線程的共享池,用于卸載IO密集型阻塞操作(如文件I/O和阻塞套接字I/O)。
Dispatchers.Unconfined -在當(dāng)前調(diào)用幀中啟動(dòng)協(xié)程執(zhí)行,直到第一次暫停,然后協(xié)程生成器函數(shù)返回。協(xié)程稍后將在相應(yīng)的掛起函數(shù)使用的任何線程中恢復(fù),而不將其限制在任何特定的線程或池中。無約束調(diào)度器通常不應(yīng)在代碼中使用。
HandlerContext -在主線程中調(diào)度任務(wù),android中主線程也就是ui線程,使用該調(diào)度器謹(jǐn)慎ANR異常,不應(yīng)該使用該調(diào)度器調(diào)度阻塞或者耗時(shí)任務(wù)。
可以使用newSingleThreadContext和newFixedThreadPoolContext創(chuàng)建專用線程池。
可以使用asCoroutineDispatcher擴(kuò)展函數(shù)將任意執(zhí)行器轉(zhuǎn)換為調(diào)度器。
Dispatchers.Default
這個(gè)調(diào)度器的類型是DefaultScheduler,一般是做cpu密集計(jì)算型任務(wù),內(nèi)部包含的成員變量IO,也就是對(duì)應(yīng)的Dispatchers.IO調(diào)度器。主要實(shí)現(xiàn)在ExecutorCoroutineDispatcher()
中,代碼如下:
internal object DefaultScheduler : ExperimentalCoroutineDispatcher() { val IO: CoroutineDispatcher = LimitingDispatcher( this, systemProp(IO_PARALLELISM_PROPERTY_NAME, 64.coerceAtLeast(AVAILABLE_PROCESSORS)), "Dispatchers.IO", TASK_PROBABLY_BLOCKING ) //省略。。。 } public open class ExperimentalCoroutineDispatcher( private val corePoolSize: Int, private val maxPoolSize: Int, private val idleWorkerKeepAliveNs: Long, private val schedulerName: String = "CoroutineScheduler" ) : ExecutorCoroutineDispatcher() { public constructor(//省略。。。) override val executor: Executor get() = coroutineScheduler // This is variable for test purposes, so that we can reinitialize from clean state private var coroutineScheduler = createScheduler() override fun dispatch(context: CoroutineContext, block: Runnable): Unit = try { coroutineScheduler.dispatch(block) } catch (e: RejectedExecutionException) { DefaultExecutor.dispatch(context, block) } override fun dispatchYield(context: CoroutineContext, block: Runnable): Unit = try { coroutineScheduler.dispatch(block, tailDispatch = true) } catch (e: RejectedExecutionException) { DefaultExecutor.dispatchYield(context, block) } } //省略。。。 }
Default調(diào)度器其實(shí)沒做什么特別的操作,只是用coroutineScheduler代理實(shí)現(xiàn)了協(xié)程的調(diào)度。
Dispatchers.IO
這個(gè)是LimitingDispatcher類型的,是DefaultScheduler類型的成員變量,而LimitingDispatcher類型又是繼承自ExecutorCoroutineDispatcher的,LimitingDispatcher在它基礎(chǔ)上做了有調(diào)度個(gè)數(shù)限制的排隊(duì)機(jī)制,IO這個(gè)名字代表的IO操作,IO操作又是阻塞線程的操作,線程不能及時(shí)釋放,所以加入了隊(duì)列機(jī)制,防止IO線程爆炸式增長。如下:
internal object DefaultScheduler : ExperimentalCoroutineDispatcher() { val IO: CoroutineDispatcher = LimitingDispatcher( this, systemProp(IO_PARALLELISM_PROPERTY_NAME, 64.coerceAtLeast(AVAILABLE_PROCESSORS)), "Dispatchers.IO", TASK_PROBABLY_BLOCKING ) //省略。。。 } private class LimitingDispatcher( private val dispatcher: ExperimentalCoroutineDispatcher, private val parallelism: Int, private val name: String?, override val taskMode: Int ) : ExecutorCoroutineDispatcher(), TaskContext, Executor { private val queue = ConcurrentLinkedQueue<Runnable>() private val inFlightTasks = atomic(0) private fun dispatch(block: Runnable, tailDispatch: Boolean) { var taskToSchedule = block while (true) { // Commit in-flight tasks slot val inFlight = inFlightTasks.incrementAndGet() // Fast path, if parallelism limit is not reached, dispatch task and return if (inFlight <= parallelism) { dispatcher.dispatchWithContext(taskToSchedule, this, tailDispatch) return } queue.add(taskToSchedule) if (inFlightTasks.decrementAndGet() >= parallelism) { return } taskToSchedule = queue.poll() ?: return } } override fun dispatchYield(context: CoroutineContext, block: Runnable) { dispatch(block, tailDispatch = true) } }
構(gòu)造函數(shù) 傳入了parallelism參數(shù) ,這個(gè)是并發(fā)數(shù)。
dispatchYield方法 實(shí)現(xiàn)是直接調(diào)用的dispatch方法。
dispatch方法:一個(gè)while循環(huán),循環(huán)內(nèi),
- 給inFlightTasks變量加一(這個(gè)變量代表正在調(diào)度中的個(gè)數(shù)),如果inFlightTasks <= parallelism,代表當(dāng)前調(diào)度任務(wù)數(shù)小于最大并發(fā)數(shù),說明可以繼續(xù)向調(diào)度器中調(diào)度任務(wù)
- 否則將任務(wù)加入到隊(duì)列中,接著嘗試將inFlightTasks減一,如果大于并發(fā)數(shù),那么直接結(jié)束;
- 如果小于并發(fā)數(shù),說明剛剛已經(jīng)有任務(wù)結(jié)束了,讓出了并發(fā)數(shù),這個(gè)時(shí)候可以再次嘗試從隊(duì)列中取出任務(wù),從1開始。
override fun afterTask() { var next = queue.poll() // If we have pending tasks in current blocking context, dispatch first if (next != null) { dispatcher.dispatchWithContext(next, this, true) return } inFlightTasks.decrementAndGet() next = queue.poll() ?: return dispatch(next, true) }
afterTask方法
這個(gè)方法是任務(wù)調(diào)度結(jié)束后的回調(diào),這里面首先從隊(duì)列中取出一個(gè)任務(wù),
任務(wù)不為空,讓調(diào)度器調(diào)度這個(gè)任務(wù),結(jié)束;
為空,給調(diào)度任務(wù)數(shù)加一,然后嘗試取出任務(wù),為空返回,不為空,繼續(xù)調(diào)用dispatch方法,整個(gè)流程就串起來了。
整個(gè)流程如下圖所示:
綜上:IO調(diào)度器側(cè)重于調(diào)度任務(wù)數(shù)量的限制,防止IO操作阻塞線程,讓線程數(shù)量爆炸式增長。
Dispatchers.Main
具體的實(shí)現(xiàn)類是HandlerContext
,代碼如下:
HandlerContext(Looper.getMainLooper().asHandler(async = true)) internal class HandlerContext private constructor( private val handler: Handler, private val name: String?, private val invokeImmediately: Boolean ) : HandlerDispatcher(), Delay { //省略。。。 }
主線程中調(diào)度任務(wù),android中主線程也就是ui線程。實(shí)現(xiàn)原理是內(nèi)部持有一個(gè)val handler : Handler = Looper.getMainLooper().asHandler(async = true)
,這個(gè)handler正是主線程的handler。
在調(diào)用dispatch調(diào)度方法的時(shí)候,是使用handler發(fā)送一個(gè)Runnable任務(wù),
override fun dispatch(context: CoroutineContext, block: Runnable) { handler.post(block) }
在delay的時(shí)候,如果當(dāng)前的dispatcher正是HandlerContext,那么實(shí)現(xiàn)是handler發(fā)送一個(gè)延遲了timeMillis
毫秒時(shí)長的Runnable。invokeOnCancellation的擴(kuò)展方法是在協(xié)程被取消的時(shí)候,移除掉該runnable消息。
override fun scheduleResumeAfterDelay(timeMillis: Long, continuation: CancellableContinuation<Unit>) { val block = Runnable { with(continuation) { resumeUndispatched(Unit) } } handler.postDelayed(block, timeMillis.coerceAtMost(MAX_DELAY)) continuation.invokeOnCancellation { handler.removeCallbacks(block) } }
下面這個(gè)方法也比較常看到,就是協(xié)程在調(diào)度continuation的時(shí)候,會(huì)去判斷是不是需要去調(diào)度,不需要的話,直接在當(dāng)前線程執(zhí)行,需要調(diào)度的,需要由dispatcher來重新調(diào)度任務(wù),這樣可能執(zhí)行的線程會(huì)被切換,如果不是主線程的話,、就需要調(diào)度了, 如果是主線程的話立刻執(zhí)行。
override fun isDispatchNeeded(context: CoroutineContext): Boolean { return !invokeImmediately || Looper.myLooper() != handler.looper }
Dispatchers.Unconfined
具體的實(shí)現(xiàn)如下:
internal object Unconfined : CoroutineDispatcher() { //省略。。。 }
isDispatchNeeded直接返回false,代表不需要重新調(diào)度。
override fun isDispatchNeeded(context: CoroutineContext): Boolean = false
dispatchYield沒有被覆寫,直接調(diào)用dispatch方法,用的還是CoroutineDispatcher的實(shí)現(xiàn)。
dispatch的報(bào)錯(cuò)信息顯示,Unconfined調(diào)度器只能在存在YieldContext的時(shí)候調(diào)度,否則就會(huì)報(bào)異常。
//CoroutineDispatcher public open fun dispatchYield(context: CoroutineContext, block: Runnable): Unit = dispatch(context, block) //Unconfined override fun dispatch(context: CoroutineContext, block: Runnable) { // It can only be called by the "yield" function. See also code of "yield" function. val yieldContext = context[YieldContext] if (yieldContext != null) { // report to "yield" that it is an unconfined dispatcher and don't call "block.run()" yieldContext.dispatcherWasUnconfined = true return } throw UnsupportedOperationException("Dispatchers.Unconfined.dispatch function can only be used by the yield function. " + "If you wrap Unconfined dispatcher in your code, make sure you properly delegate " + "isDispatchNeeded and dispatch calls.") }
yied方法:是暫時(shí)讓出工作線程,等待下一次線程調(diào)取恢復(fù)協(xié)程。
yield代碼如下:
public suspend fun yield(): Unit = suspendCoroutineUninterceptedOrReturn sc@ { uCont -> val context = uCont.context context.checkCompletion() val cont = uCont.intercepted() as? DispatchedContinuation<Unit> ?: return@sc Unit if (cont.dispatcher.isDispatchNeeded(context)) { cont.dispatchYield(context, Unit) } else { val yieldContext = YieldContext() cont.dispatchYield(context + yieldContext, Unit) if (yieldContext.dispatcherWasUnconfined) { return@sc if (cont.yieldUndispatched()) COROUTINE_SUSPENDED else Unit } } COROUTINE_SUSPENDED }
如果isDispatchNeeded == true,那么就需要重新將協(xié)程被調(diào)度器調(diào)度一次,線程有可能切換掉;
如果isDispatchNeeded == false,上下文集合需要添加val yieldContext = YieldContext()這個(gè)元素(在上面的Dispatchers.Unconfined
的dispatche方法中,如果有YieldContext元素,將dispatcherWasUnconfined設(shè)置為true,代表yield操作什么都沒有做,需要協(xié)程調(diào)度器用其他方法調(diào)度一次)。
判斷dispatcherWasUnconfined,true:說明Dispatchers.Unconfined什么都沒有做,需要在調(diào)度一次,調(diào)用了yieldUndispatched
方法,這個(gè)方法大概就是讓協(xié)程直接恢復(fù)一次,或者線程調(diào)度一次恢復(fù);
false:說明正在被調(diào)度器調(diào)度,是個(gè)掛起點(diǎn),返回COROUTINE_SUSPENDED值。
不太清楚Dispatchers.Unconfined
這個(gè)調(diào)度器有啥用,有知道的留言下,學(xué)習(xí)學(xué)習(xí)。
協(xié)程調(diào)度器的實(shí)現(xiàn)CoroutineScheduler
調(diào)度過程正真的實(shí)現(xiàn)是CoroutineScheduler
這個(gè)類,上面說的四種調(diào)度器是包裝類,調(diào)度邏輯在CoroutineScheduler中,代碼如下:
internal class CoroutineScheduler( @JvmField val corePoolSize: Int, @JvmField val maxPoolSize: Int, @JvmField val idleWorkerKeepAliveNs: Long = IDLE_WORKER_KEEP_ALIVE_NS, @JvmField val schedulerName: String = DEFAULT_SCHEDULER_NAME ) : Executor, Closeable { //省略。。。 }
構(gòu)造函數(shù)入?yún)?corePoolSize: Int
定義核心線程數(shù),maxPoolSize: Int
定義最大線程數(shù)量
fun dispatch(block: Runnable, taskContext: TaskContext = NonBlockingContext, tailDispatch: Boolean = false) { val task = createTask(block, taskContext) // try to submit the task to the local queue and act depending on the result val currentWorker = currentWorker() val notAdded = currentWorker.submitToLocalQueue(task, tailDispatch) if (notAdded != null) { if (!addToGlobalQueue(notAdded)) { // Global queue is closed in the last step of close/shutdown -- no more tasks should be accepted throw RejectedExecutionException("$schedulerName was terminated") } } val skipUnpark = tailDispatch && currentWorker != null // Checking 'task' instead of 'notAdded' is completely okay if (task.mode == TASK_NON_BLOCKING) { if (skipUnpark) return signalCpuWork() } else { // Increment blocking tasks anyway signalBlockingWork(skipUnpark = skipUnpark) } }
dispatch函數(shù)的實(shí)現(xiàn):
創(chuàng)建task,block如果是Task類型的話,設(shè)置submissionTime變量,submissionTime變量用于延遲執(zhí)行的時(shí)間判斷,以及隊(duì)列排序的時(shí)間順序;設(shè)置taskContext,該變量是task執(zhí)行的協(xié)程上下文。不是Task類型的話,會(huì)創(chuàng)建TaskImp類型的任務(wù)返回,關(guān)鍵是finally中的taskContext.afterTask()
,就是task執(zhí)行完成后需要回調(diào)afterTask通知協(xié)程上下文執(zhí)行完畢了,上面的Dispatchers.IO里面的LimitingDispatcher調(diào)度器就是需要afterTask回調(diào)通知,才能將隊(duì)列中下一個(gè)任務(wù)拋給CoroutineScheduler去執(zhí)行。
internal fun createTask(block: Runnable, taskContext: TaskContext): Task { val nanoTime = schedulerTimeSource.nanoTime() if (block is Task) { block.submissionTime = nanoTime block.taskContext = taskContext return block } return TaskImpl(block, nanoTime, taskContext) } internal class TaskImpl( @JvmField val block: Runnable, submissionTime: Long, taskContext: TaskContext ) : Task(submissionTime, taskContext) { override fun run() { try { block.run() } finally { taskContext.afterTask() } } }
獲取當(dāng)前的工作線程,如果當(dāng)前是工作線程直接返回,不是的話返回空
private fun currentWorker(): Worker? = (Thread.currentThread() as? Worker)?.takeIf {<!--{C}%3C!%2D%2D%20%2D%2D%3E--> it.scheduler == this }
將任務(wù)提交到工作線程的本地隊(duì)列中
private fun Worker?.submitToLocalQueue(task: Task, tailDispatch: Boolean): Task? { if (this == null) return task if (state === WorkerState.TERMINATED) return task if (task.mode == TASK_NON_BLOCKING && state === WorkerState.BLOCKING) { return task } mayHaveLocalTasks = true return localQueue.add(task, fair = tailDispatch) }
返回是空的,說明添加成功了,返回task說明沒有添加成功。
如果線程是中斷狀態(tài),那么直接返回task。 如果任務(wù)是非阻塞的也就是cpu密集型任務(wù),而線程是阻塞的(正在執(zhí)行任務(wù)中),那么不添加任務(wù),直接返回task。 其他情況,添加任務(wù)到隊(duì)列中,mayHaveLocalTasks標(biāo)志位true,代表當(dāng)前線程中有任務(wù)。
沒有添加的話,需要添加到全局隊(duì)列中,globalCpuQueue全局cpu密集型隊(duì)列,globalBlockingQueue全局IO隊(duì)列,根據(jù)任務(wù)類型添加到對(duì)應(yīng)的隊(duì)列中。如果全局隊(duì)列都添加失敗的話,直接拋出異常。
if (notAdded != null) { if (!addToGlobalQueue(notAdded)) { // Global queue is closed in the last step of close/shutdown -- no more tasks should be accepted throw RejectedExecutionException("$schedulerName was terminated") } } val globalCpuQueue = GlobalQueue() val globalBlockingQueue = GlobalQueue() private fun addToGlobalQueue(task: Task): Boolean { return if (task.isBlocking) { globalBlockingQueue.addLast(task) } else { globalCpuQueue.addLast(task) } }
根據(jù)是否是尾部添加和當(dāng)前線程是否是空,決定是否跳過喚醒工作線程的步驟。
val skipUnpark = tailDispatch && currentWorker != null
非阻塞任務(wù):skipUnpark為true,跳過喚醒步驟,否則喚醒cpu密集型線程;阻塞任務(wù):skipUnpark為true,跳過喚醒步驟,喚醒IO線程。
// Checking 'task' instead of 'notAdded' is completely okay if (task.mode == TASK_NON_BLOCKING) { if (skipUnpark) return signalCpuWork() } else { // Increment blocking tasks anyway signalBlockingWork(skipUnpark = skipUnpark) }
看下喚醒步驟的具體實(shí)現(xiàn),大概都是先tryUnpark,喚醒線程,如果沒有喚醒成功,創(chuàng)建一個(gè)新的線程,再次嘗試喚醒。
private fun signalBlockingWork(skipUnpark: Boolean) { // Use state snapshot to avoid thread overprovision val stateSnapshot = incrementBlockingTasks() if (skipUnpark) return if (tryUnpark()) return if (tryCreateWorker(stateSnapshot)) return tryUnpark() // Try unpark again in case there was race between permit release and parking } internal fun signalCpuWork() { if (tryUnpark()) return if (tryCreateWorker()) return tryUnpark() }
看下工作線程的具體實(shí)現(xiàn)吧:
worker繼承自Thread,實(shí)現(xiàn)了run方法,具體是由runWorker()方法實(shí)現(xiàn)的,每個(gè)工作線程都有一個(gè)本地隊(duì)列用于存儲(chǔ)任務(wù),這樣本地有任務(wù)就不用去全局隊(duì)列中去搶資源了,減少鎖競爭。
internal inner class Worker private constructor() : Thread() { //省略。。。 @JvmField val localQueue: WorkQueue = WorkQueue() @JvmField var mayHaveLocalTasks = false override fun run() = runWorker() //省略。。。 }
runWorker() 的實(shí)現(xiàn):
private fun runWorker() { var rescanned = false while (!isTerminated && state != WorkerState.TERMINATED) { val task = findTask(mayHaveLocalTasks) // Task found. Execute and repeat if (task != null) { rescanned = false minDelayUntilStealableTaskNs = 0L executeTask(task) continue } else { mayHaveLocalTasks = false } if (minDelayUntilStealableTaskNs != 0L) { if (!rescanned) { rescanned = true } else { rescanned = false tryReleaseCpu(WorkerState.PARKING) interrupted() LockSupport.parkNanos(minDelayUntilStealableTaskNs) minDelayUntilStealableTaskNs = 0L } continue } tryPark() } tryReleaseCpu(WorkerState.TERMINATED) }
工作線程是用while循環(huán)一直運(yùn)行的,循環(huán)內(nèi):
val task = findTask(mayHaveLocalTasks)
,前面這個(gè)變量mayHaveLocalTasks出現(xiàn)過,在添加task到本地隊(duì)列的時(shí)候,會(huì)置為true,本地隊(duì)列有任務(wù),從本地獲取,沒有就從全局隊(duì)列中獲取,如果還是沒有,從其他線程隊(duì)列中偷取任務(wù)到自己隊(duì)列中:
fun findTask(scanLocalQueue: Boolean): Task? { if (tryAcquireCpuPermit()) return findAnyTask(scanLocalQueue) // If we can't acquire a CPU permit -- attempt to find blocking task val task = if (scanLocalQueue) { localQueue.poll() ?: globalBlockingQueue.removeFirstOrNull() } else { globalBlockingQueue.removeFirstOrNull() } return task ?: trySteal(blockingOnly = true) }
trySteal方法,循環(huán)workers隊(duì)列,遍歷線程本地隊(duì)列,去偷取任務(wù),偷到的話返回任務(wù),沒偷到的話,返回null:
private fun trySteal(blockingOnly: Boolean): Task? { //省略。。。 var currentIndex = nextInt(created) var minDelay = Long.MAX_VALUE repeat(created) { //省略。。。 val worker = workers[currentIndex] if (worker !== null && worker !== this) { val stealResult = if (blockingOnly) { localQueue.tryStealBlockingFrom(victim = worker.localQueue) } else { localQueue.tryStealFrom(victim = worker.localQueue) } if (stealResult == TASK_STOLEN) { return localQueue.poll() } else if (stealResult > 0) { minDelay = min(minDelay, stealResult) } } } minDelayUntilStealableTaskNs = if (minDelay != Long.MAX_VALUE) minDelay else 0 return null }
在偷不到任務(wù)的時(shí)候會(huì)設(shè)置一個(gè)變量,stealResult等于-2,最后minDelayUntilStealableTaskNs 等于0;
internal const val TASK_STOLEN = -1L internal const val NOTHING_TO_STEAL = -2L
在偷取任務(wù)的時(shí)候,如果上個(gè)任務(wù)時(shí)間和這次時(shí)間間隔太短的話,返回下次執(zhí)行的間隔時(shí)間差,minDelayUntilStealableTaskNs設(shè)置為這個(gè)時(shí)間值,大于0。
找到task了,直接執(zhí)行任務(wù)executeTask(task)
,執(zhí)行完成,continue循環(huán),從1開始;
沒找到任務(wù),設(shè)置mayHaveLocalTasks = false
如果minDelayUntilStealableTaskNs不等于0,就是上面的間隔時(shí)間太短的條件觸發(fā),那么讓線程釋放鎖(防止線程執(zhí)行任務(wù)太過密集,等待下次循環(huán)再去調(diào)度任務(wù)),continue循環(huán),從1開始;
上面條件不成立,調(diào)用tryPark(),這個(gè)是和unPark相反的操作,讓線程閑置,放入到線程隊(duì)列中:
private fun tryPark() { if (!inStack()) { parkedWorkersStackPush(this) return } assert { localQueue.size == 0 } workerCtl.value = PARKED // Update value once while (inStack()) { // Prevent spurious wakeups if (isTerminated || state == WorkerState.TERMINATED) break tryReleaseCpu(WorkerState.PARKING) interrupted() // Cleanup interruptions park() } }
首先判斷是否在隊(duì)列中,不在的話,放入線程隊(duì)列中;在隊(duì)列中,將狀態(tài)設(shè)置為PARKED,不斷循環(huán)將釋放線程的cpu占用鎖,嘗試放到隊(duì)列中,park函數(shù)中有可能銷毀工作線程,看線程是否到達(dá)死亡時(shí)間點(diǎn)。
worker工作流程如下圖所示:
總結(jié)
1. Dispatchers的四種調(diào)度器是餓漢式單例對(duì)象,所以一個(gè)進(jìn)程只存在一個(gè)實(shí)例對(duì)象。
2. Dispatchers的四種調(diào)度器中,IO和default是共用的一個(gè)線程池,它的實(shí)現(xiàn)是CoroutineScheduler。
3. CoroutineScheduler線程池,有一個(gè)保存線程的隊(duì)列,有兩種全局任務(wù)隊(duì)列:一個(gè)是IO阻塞型隊(duì)列,一個(gè)是cpu密集型任務(wù)隊(duì)列;Worker線程擁有一個(gè)本地任務(wù)隊(duì)列。
4. Worker線程會(huì)根據(jù)任務(wù)類型,去對(duì)應(yīng)的全局隊(duì)列或者從本地隊(duì)列找任務(wù),找不到會(huì)從其他worker隊(duì)列中偷任務(wù),然后執(zhí)行;worker會(huì)根據(jù)自己的狀態(tài)回到線程隊(duì)列或者銷毀自己。
到此這篇關(guān)于Kotlin Dispatchers協(xié)程調(diào)度器原阿門深入分析的文章就介紹到這了,更多相關(guān)Kotlin Dispatchers內(nèi)容請(qǐng)搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家!
相關(guān)文章
Android基于ImageView繪制的開關(guān)按鈕效果示例
這篇文章主要介紹了Android基于ImageView繪制的開關(guān)按鈕效果,結(jié)合實(shí)例形式分析了Android使用ImageView進(jìn)行按鈕繪制的界面布局、功能實(shí)現(xiàn)及相關(guān)注意事項(xiàng),需要的朋友可以參考下2017-03-03詳解Android 7.0 Settings 加載選項(xiàng)
本篇文章主要介紹了Android 7.0 Settings 加載選項(xiàng),Android 7.0 Settings頂部多了一個(gè)建議選項(xiàng),多了個(gè)側(cè)邊欄,操作更加便捷了,有興趣的可以了解一下。2017-02-02Android實(shí)現(xiàn)數(shù)字跳動(dòng)效果的TextView方法示例
數(shù)字跳動(dòng)效果相信大家應(yīng)該都見過,在開發(fā)加上這種效果后會(huì)讓ui交互看起來非常不錯(cuò),所以下面這篇文章主要給大家介紹了Android實(shí)現(xiàn)數(shù)字跳動(dòng)的TextView的相關(guān)資料,文中給出了詳細(xì)的示例代碼,需要的朋友可以參考學(xué)習(xí),下面來一起看看吧。2017-04-04rxjava+retrofit實(shí)現(xiàn)多圖上傳實(shí)例代碼
本篇文章主要介紹了rxjava+retrofit實(shí)現(xiàn)多圖上傳實(shí)例代碼,小編覺得挺不錯(cuò)的,現(xiàn)在分享給大家,也給大家做個(gè)參考。一起跟隨小編過來看看吧2017-06-06解決Android studio3.6安裝后gradle Download失敗(構(gòu)建不成功)
這篇文章主要介紹了解決Android studio3.6安裝后gradle Download失敗(構(gòu)建不成功),文中通過示例代碼介紹的非常詳細(xì),對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值,需要的朋友們下面隨著小編來一起學(xué)習(xí)學(xué)習(xí)吧2020-03-03Android使用WebView實(shí)現(xiàn)截圖分享功能
這篇文章主要為大家詳細(xì)介紹了Android使用WebView實(shí)現(xiàn)截圖分享功能,具有一定的參考價(jià)值,感興趣的小伙伴們可以參考一下2018-05-05