Github user dongjoon-hyun commented on the issue: https://github.com/apache/spark/pull/19943 After minimizing `nextBatch`, it becomes smaller than Parquet's `nextBatch`. But, it's inlined only some cases, but mostly not. It's not helpful. For the other technique, I'll try later. ``` org.apache.spark.sql.execution.datasources.orc.OrcColumnarBatchReader::nextBatch (107 bytes) inline (hot) ```
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org