Github user witgo commented on a diff in the pull request:

    https://github.com/apache/spark/pull/15505#discussion_r103622591
  
    --- Diff: 
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
 ---
    @@ -621,6 +615,80 @@ class CoarseGrainedSchedulerBackend(scheduler: 
TaskSchedulerImpl, val rpcEnv: Rp
       }
     }
     
    -private[spark] object CoarseGrainedSchedulerBackend {
    +private[spark] object CoarseGrainedSchedulerBackend extends Logging {
       val ENDPOINT_NAME = "CoarseGrainedScheduler"
    +
    +  // abort TaskSetManager without exception
    +  private def abortTaskSetManager(
    +      scheduler: TaskSchedulerImpl,
    +      taskId: Long,
    +      msg: => String,
    +      exception: Option[Throwable] = None): Unit = scheduler.synchronized {
    +    scheduler.taskIdToTaskSetManager.get(taskId).foreach { taskSetMgr =>
    +      try {
    +        taskSetMgr.abort(msg, exception)
    +      } catch {
    +        case e: Exception => logError("Exception in error callback", e)
    +      }
    +    }
    +  }
    +
    +  private def isZombieTaskSetManager(
    +    scheduler: TaskSchedulerImpl,
    +    taskId: Long): Unit = scheduler.synchronized {
    +    !scheduler.taskIdToTaskSetManager.get(taskId).exists(_.isZombie)
    +  }
    +
    +  private def getTaskSetManager(
    +    scheduler: TaskSchedulerImpl,
    +    taskId: Long): Option[TaskSetManager] = scheduler.synchronized {
    +    scheduler.taskIdToTaskSetManager.get(taskId)
    +  }
    +
    +  private[scheduler] def prepareSerializedTask(
    +    scheduler: TaskSchedulerImpl,
    +    task: TaskDescription,
    +    abortSet: HashSet[TaskSetManager],
    +    maxRpcMessageSize: Long): ByteBuffer = {
    +    var serializedTask: ByteBuffer = null
    +    if (abortSet.isEmpty || !getTaskSetManager(scheduler, 
task.taskId).exists(_.isZombie)) {
    --- End diff --
    
    This is an optimization code, in most cases abortSet is empty and the 
`!getTaskSetManager(scheduler, task.taskId).exists(_.isZombie)` is not called.
     I changed it into a more readable code.
    
    see 
https://github.com/apache/spark/pull/15505/commits/2f4b3f955cad0fe7546de54737c19500b90ad67d



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

Reply via email to