andygrove opened a new issue, #4211:
URL: https://github.com/apache/datafusion-comet/issues/4211
### Describe the bug
I was trying to run TPC-H @ 1TB benchmarks to compare `native_datafusion`
and `native_iceberg_compat` performance. Some queries fail when using
`native_iceberg_compat`.
```
org.apache.spark.SparkException: Job aborted due to stage failure: Task 9 in
stage 18.0 failed 4 times, most recent failure: Lost task 9.3 in stage 18.0
(TID 24604) (240.56.187.86 executor 9): org.apache.spark.SparkException: Number
of rows in each column should be the same, but got [ArrayBuffer(8192, 0)]
at org.apache.comet.vector.NativeUtil.exportBatch(NativeUtil.scala:173)
at
org.apache.comet.vector.NativeUtil.exportBatchToAddresses(NativeUtil.scala:97)
at
org.apache.comet.NativeColumnarToRowConverter.convert(NativeColumnarToRowConverter.scala:90)
at
org.apache.spark.sql.comet.CometNativeColumnarToRowExec.$anonfun$doExecute$3(CometNativeColumnarToRowExec.scala:211)
at scala.collection.Iterator$$anon$10.nextCur(Iterator.scala:594)
at scala.collection.Iterator$$anon$10.hasNext(Iterator.scala:608)
at scala.collection.Iterator$$anon$9.hasNext(Iterator.scala:583)
at
org.apache.spark.shuffle.sort.UnsafeShuffleWriter.write(UnsafeShuffleWriter.java:179)
```
### Steps to reproduce
_No response_
### Expected behavior
_No response_
### Additional context
_No response_
--
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]
