andygrove commented on a change in pull request #33624:
URL: https://github.com/apache/spark/pull/33624#discussion_r683595226
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/AdaptiveSparkPlanExec.scala
##########
@@ -536,7 +523,7 @@ case class AdaptiveSparkPlanExec(
case s: ShuffleExchangeLike =>
val newShuffle = applyPhysicalRules(
s.withNewChildren(Seq(optimizedPlan)),
- postStageCreationRules,
+ postStageCreationRules(outputColumnar = false),
Review comment:
I think that we do need to retain the post-stage creation rules. Spark
will insert a RowToColumnar in the adaptive plan for the final query stage but
we need a way for our plugin to replace that with a GPU-specific version and we
can use the extensible post-stage creation rule to do that.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
