marin-ma commented on code in PR #6022:
URL: https://github.com/apache/incubator-gluten/pull/6022#discussion_r1635832795
##########
backends-velox/src/test/scala/org/apache/gluten/execution/FallbackSuite.scala:
##########
@@ -71,6 +71,23 @@ class FallbackSuite extends VeloxWholeStageTransformerSuite
with AdaptiveSparkPl
collect(plan) { case v: VeloxColumnarToRowExec => v }.size
}
+ test("fallback with shuffle manager") {
+ withSQLConf(GlutenConfig.COLUMNAR_SHUFFLE_ENABLED.key -> "false") {
+ runQueryAndCompare("select c1, count(*) from tmp1 group by c1") {
+ df =>
+ val plan = df.queryExecution.executedPlan
+ val columnarShuffle = find(plan) {
+ case _: ColumnarShuffledJoin => true
Review Comment:
We should check the number of`ColumnarShuffleExchangeExec` and
`ShuffleExchangeExec`. You can use repartition hint to construct the sql
https://spark.apache.org/docs/latest/sql-ref-syntax-qry-select-hints.html#partitioning-hints
e.g. `select /*+ REPARTITION(3, c1) */ * FROM tmp1;`
Then the check should be like:
```
assert(collectColumnarShuffleExchange(plan) == 0)
assert(collectShuffleExchange(plan) == 1)
```
--
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]
