[CORE] Fix performance issue when allScanPartitions size is very large (

#8126)
apache · Dec 9, 2024 · 72fc04e · 72fc04e
1 parent 1100fdf
commit 72fc04e
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/gluten-substrait/src/main/scala/org/apache/gluten/execution/WholeStageTransformer.scala b/gluten-substrait/src/main/scala/org/apache/gluten/execution/WholeStageTransformer.scala
@@ -373,7 +373,7 @@ case class WholeStageTransformer(child: SparkPlan, materializeInput: Boolean = f
        * care of SCAN there won't be any other RDD for SCAN. As a result, genFirstStageIterator
        * rather than genFinalStageIterator will be invoked
        */
-      val allScanPartitions = basicScanExecTransformers.map(_.getPartitions)
+      val allScanPartitions = basicScanExecTransformers.map(_.getPartitions.toIndexedSeq)
       val allScanSplitInfos =
         getSplitInfosFromPartitions(basicScanExecTransformers, allScanPartitions)
       if (GlutenConfig.getConf.enableHdfsViewfs) {