marin-ma commented on code in PR #6475:
URL: https://github.com/apache/incubator-gluten/pull/6475#discussion_r1687312260
##########
backends-velox/src/main/scala/org/apache/gluten/backendsapi/velox/VeloxSparkPlanExecApi.scala:
##########
@@ -349,24 +349,30 @@ class VeloxSparkPlanExecApi extends SparkPlanExecApi {
}
def maybeAddAppendBatchesExec(plan: SparkPlan): SparkPlan = {
- if (GlutenConfig.getConf.veloxCoalesceBatchesBeforeShuffle) {
- VeloxAppendBatchesExec(plan,
GlutenConfig.getConf.veloxMinBatchSizeForShuffle)
- } else {
- plan
+ plan match {
+ case shuffle: ColumnarShuffleExchangeExec
+ if !shuffle.useSortBasedShuffle &&
Review Comment:
No matter coalesced or not, sort will convert all inputs into rows, so it
won't benefit from coalesce. The test result shows the coalesce op only brings
extra overhead.
--
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]
