marin-ma commented on code in PR #4872:
URL: https://github.com/apache/incubator-gluten/pull/4872#discussion_r1519250936
##########
cpp/core/jni/JniWrapper.cc:
##########
@@ -830,7 +830,8 @@ JNIEXPORT jlong JNICALL
Java_io_glutenproject_vectorized_ShuffleWriterJniWrapper
jint startPartitionId,
jint pushBufferMaxSize,
jobject partitionPusher,
- jstring partitionWriterTypeJstr) {
+ jstring partitionWriterTypeJstr,
+ jboolean sortBeforeRR) {
Review Comment:
I doubt that the impact of performance loss by adding one project which is
only used to drop one column. The most part of extra overhead should come from
the sort node. By adding this projection, we can also avoid adding argument
`sortBeforeRR` to the several layers in the call stack.
--
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]
