LuciferYang commented on a change in pull request #29975:
URL: https://github.com/apache/spark/pull/29975#discussion_r501476974
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/basicPhysicalOperators.scala
##########
@@ -66,10 +66,22 @@ case class ProjectExec(projectList: Seq[NamedExpression],
child: SparkPlan)
override def doConsume(ctx: CodegenContext, input: Seq[ExprCode], row:
ExprCode): String = {
val exprs = bindReferences[Expression](projectList, child.output)
- val resultVars = exprs.map(_.genCode(ctx))
+ val (subExprsCode, resultVars) = if (conf.subexpressionEliminationEnabled)
{
+ // subexpression elimination
+ val subExprs = ctx.subexpressionEliminationForWholeStageCodegen(exprs)
+ val genVars = ctx.withSubExprEliminationExprs(subExprs.states) {
+ exprs.map(_.genCode(ctx))
+ }
+ (subExprs.codes.mkString("\n"), genVars)
+ } else {
+ ("", exprs.map(_.genCode(ctx)))
+ }
+
// Evaluation of non-deterministic expressions can't be deferred.
val nonDeterministicAttrs =
projectList.filterNot(_.deterministic).map(_.toAttribute)
s"""
Review comment:
manual testing examples like SPARK-32989 , the performance is much
better than before :)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
