surnaik opened a new issue, #10502: URL: https://github.com/apache/incubator-gluten/issues/10502
### Backend VL (Velox) ### Bug description I added a repro test case inside - VeloxTPCHSuite.scala -https://github.com/apache/incubator-gluten/blob/main/backends-velox/src/test/scala/org/apache/gluten/execution/VeloxTPCHSuite.scala ``` test("JDK Crash Repro") { sql("select cast(l_orderkey as double) as l_orderkey from lineitem").write.parquet("/root/incubator-gluten/test") withSQLConf( SQLConf.ADAPTIVE_EXECUTION_ENABLED.key -> "false" ) { sql("describe table lineitem") val df = spark.read.parquet("/root/incubator-gluten/test") val lit = org.apache.spark.sql.catalyst.expressions.Literal(Double.NaN) val df1 = df.filter(!col("l_orderkey").equalTo(lit)).filter(col("l_orderkey").gt(0.0)) df1.collect() df1.explain(true) println(df1.explain(true)) } } ``` Issue with the above test case: The above filter condition translates to Not(EqualTo("l_orderkey", "NaN")) - this will eventually go here - https://github.com/facebookincubator/velox/blob/23b1d9ad2927fbd22efe0b5993d4498baa0d11e0/velox/type/Filter.h#L1448 and this will throw a VeloxRuntimeError and this should ideally be caught here by this catch block - https://github.com/facebookincubator/velox/blob/23b1d9ad2927fbd22efe0b5993d4498baa0d11e0/velox/connectors/hive/HiveConnectorUtil.cpp#L891 But this only works in case of OpenJDK 11/17, but fails in every other JDK build - Temurin, Azul, Microsoft, JBR runtime, Huawei, Coretto, Alibaba etc **Expected behaviour** Test should pass **Current behaviour with Temurin JDK** Test fails **Current behaviour with OpenJDK** Test passes: ### Gluten version Gluten-1.3 ### Spark version Spark-3.5.x ### Spark configurations sparkConf .set("spark.plugins", "org.apache.gluten.GlutenPlugin") .set("spark.default.parallelism", "1") .set("spark.memory.offHeap.enabled", "true") .set("spark.memory.offHeap.size", "1024MB") .set("spark.ui.enabled", "false") .set("spark.gluten.ui.enabled", "false") .set("spark.shuffle.manager", "org.apache.spark.shuffle.sort.ColumnarShuffleManager") .set("spark.sql.files.maxPartitionBytes", "1g") .set("spark.sql.shuffle.partitions", "1") .set("spark.gluten.sql.columnar.backend.velox.memInitCapacity", "1m") .set("spark.memory.offHeap.size", "2g") .set("spark.unsafe.exceptionOnMemoryLeak", "true") .set("spark.sql.autoBroadcastJoinThreshold", "-1") .set("spark.gluten.sql.debug", "true") ### System information Gluten Version: 1.3.0 Commit: 646329de0411c32547947777fd4bd7875657f946 CMake Version: 3.28.3 System: Linux-6.12.27-1rodete1-amd64 Arch: x86_64 CPU Name: Model name: AMD EPYC 7B13 BIOS Model name: CPU @ 2.0GHz C++ Compiler: /usr/lib/ccache/c++ C++ Compiler Version: 12.2.0 C Compiler: /usr/lib/ccache/cc C Compiler Version: 12.2.0 CMake Prefix Path: /usr/local;/usr;/;/usr/local/lib/python3.11/dist-packages/cmake/data;/usr/local;/usr/X11R6;/usr/pkg;/opt ### Relevant logs ```bash ``` -- 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]
