Fix potential ThreadLocal leaks when using ForkJoinPool

Author: zsxwing

core/src/main/scala/org/apache/spark/rpc/RpcTimeout.scala

@@ -24,7 +24,7 @@ import scala.concurrent.duration._
 import scala.util.control.NonFatal
 
 import org.apache.spark.{SparkConf, SparkException}
-import org.apache.spark.util.Utils
+import org.apache.spark.util.{ThreadUtils, Utils}
 
 /**
  * An exception thrown if RpcTimeout modifies a [[TimeoutException]].
@@ -77,9 +77,7 @@ private[spark] class RpcTimeout(val duration: FiniteDuration, val timeoutProp: S
         throw new SparkException("Exception thrown in awaitResult", t)
     }
     try {
-      // scalastyle:off awaitresult
-      Await.result(future, duration)
-      // scalastyle:on awaitresult
+      ThreadUtils.awaitResult(future, duration)
     } catch addMessageIfTimeout.orElse(wrapAndRethrow)
   }
 }

core/src/main/scala/org/apache/spark/util/ThreadUtils.scala

@@ -19,7 +19,7 @@ package org.apache.spark.util
 
 import java.util.concurrent._
 
-import scala.concurrent.{Await, Awaitable, ExecutionContext, ExecutionContextExecutor}
+import scala.concurrent.{Awaitable, ExecutionContext, ExecutionContextExecutor}
 import scala.concurrent.duration.Duration
 import scala.concurrent.forkjoin.{ForkJoinPool => SForkJoinPool, ForkJoinWorkerThread => SForkJoinWorkerThread}
 import scala.util.control.NonFatal
@@ -180,31 +180,20 @@ private[spark] object ThreadUtils {
 
   // scalastyle:off awaitresult
   /**
-   * Preferred alternative to `Await.result()`. This method wraps and re-throws any exceptions
-   * thrown by the underlying `Await` call, ensuring that this thread's stack trace appears in
-   * logs.
-   */
-  @throws(classOf[SparkException])
-  def awaitResult[T](awaitable: Awaitable[T], atMost: Duration): T = {
-    try {
-      Await.result(awaitable, atMost)
-      // scalastyle:on awaitresult
-    } catch {
-      case NonFatal(t) =>
-        throw new SparkException("Exception thrown in awaitResult: ", t)
-    }
-  }
-
-  /**
-   * Calls `Awaitable.result` directly to avoid using `ForkJoinPool`'s `BlockingContext`, wraps
-   * and re-throws any exceptions with nice stack track.
+   * Preferred alternative to `Await.result()`.
+   *
+   * This method wraps and re-throws any exceptions thrown by the underlying `Await` call, ensuring
+   * that this thread's stack trace appears in logs.
    *
-   * Codes running in the user's thread may be in a thread of Scala ForkJoinPool. As concurrent
-   * executions in ForkJoinPool may see some [[ThreadLocal]] value unexpectedly, this method
-   * basically prevents ForkJoinPool from running other tasks in the current waiting thread.
+   * In addition, it calls `Awaitable.result` directly to avoid using `ForkJoinPool`'s
+   * `BlockingContext`. Codes running in the user's thread may be in a thread of Scala ForkJoinPool.
+   * As concurrent executions in ForkJoinPool may see some [[ThreadLocal]] value unexpectedly, this
+   * method basically prevents ForkJoinPool from running other tasks in the current waiting thread.
+   * In general, we should use this method because many places in Spark use [[ThreadLocal]] and it's
+   * hard to debug when [[ThreadLocal]]s leak to other tasks.
    */
   @throws(classOf[SparkException])
-  def awaitResultInForkJoinSafely[T](awaitable: Awaitable[T], atMost: Duration): T = {
+  def awaitResult[T](awaitable: Awaitable[T], atMost: Duration): T = {
     try {
       // `awaitPermission` is not actually used anywhere so it's safe to pass in null here.
       // See SPARK-13747.
@@ -215,4 +204,5 @@ private[spark] object ThreadUtils {
         throw new SparkException("Exception thrown in awaitResult: ", t)
     }
   }
+  // scalastyle:on awaitresult
 }

scalastyle-config.xml

@@ -200,7 +200,6 @@ This file is divided into 3 sections:
       // scalastyle:off awaitresult
       Await.result(...)
       // scalastyle:on awaitresult
-      If your codes use ThreadLocal and may run in threads created by the user, use ThreadUtils.awaitResultInForkJoinSafely instead.
     ]]></customMessage>
   </check>
 

sql/core/src/main/scala/org/apache/spark/sql/execution/basicPhysicalOperators.scala

@@ -578,7 +578,7 @@ case class SubqueryExec(name: String, child: SparkPlan) extends UnaryExecNode {
   }
 
   override def executeCollect(): Array[InternalRow] = {
-    ThreadUtils.awaitResultInForkJoinSafely(relationFuture, Duration.Inf)
+    ThreadUtils.awaitResult(relationFuture, Duration.Inf)
   }
 }
 

sql/core/src/main/scala/org/apache/spark/sql/execution/exchange/BroadcastExchangeExec.scala

@@ -128,8 +128,7 @@ case class BroadcastExchangeExec(
   }
 
   override protected[sql] def doExecuteBroadcast[T](): broadcast.Broadcast[T] = {
-    ThreadUtils.awaitResultInForkJoinSafely(relationFuture, timeout)
-      .asInstanceOf[broadcast.Broadcast[T]]
+    ThreadUtils.awaitResult(relationFuture, timeout).asInstanceOf[broadcast.Broadcast[T]]
   }
 }