Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-22510

Exceptions caused by 64KB JVM bytecode or 64K constant pool entry limit

    Details

    • Type: Umbrella
    • Status: Open
    • Priority: Major
    • Resolution: Unresolved
    • Affects Version/s: 2.2.0
    • Fix Version/s: None
    • Component/s: SQL
    • Labels:

      Description

      Codegen can throw an exception due to the 64KB JVM bytecode or 64K constant pool entry limit.

        Attachments

        1.
        64KB JVM bytecode limit problem with GenerateUnsafeRowJoiner.create() Sub-task Resolved Kazuaki Ishizaki
        2.
        64KB JVM bytecode limit problem with in Sub-task Resolved Kazuaki Ishizaki
        3.
        64KB JVM bytecode limit problem with cast Sub-task Resolved Kazuaki Ishizaki
        4.
        64KB JVM bytecode limit problem with least and greatest Sub-task Resolved Kazuaki Ishizaki
        5.
        64KB JVM bytecode limit problem with concat Sub-task Resolved Kazuaki Ishizaki
        6.
        Coalesce and AtLeastNNonNulls can cause 64KB JVM bytecode limit exception Sub-task Resolved Marco Gaido
        7.
        Filter predicate with many conditions throw stackoverflow error Sub-task Resolved Kazuaki Ishizaki
        8.
        SQL which has large ‘case when’ expressions may cause code generation beyond 64KB Sub-task Resolved Unassigned
        9.
        fix java 64kb compile error for deeply nested expressions Sub-task Resolved Wenchen Fan
        10.
        64KB JVM bytecode limit problem with concat_ws Sub-task Resolved Kazuaki Ishizaki
        11.
        64KB JVM bytecode limit problem with elt Sub-task Resolved Kazuaki Ishizaki
        12.
        Fix 64kb compile error for common expression types Sub-task Resolved Unassigned
        13.
        Create a lot of global variables to reuse an object in generated code Sub-task Resolved Kazuaki Ishizaki
        14.
        64KB JVM bytecode limit problem with FormatString Sub-task Resolved Kazuaki Ishizaki
        15.
        64KB JVM bytecode limit problem with GLM Sub-task Resolved Unassigned
        16.
        Fix 64kb limit for deeply nested expressions under wholestage codegen Sub-task In Progress Liang-Chi Hsieh
        17.
        elt should use splitExpressionsWithCurrentInputs to split expression codes Sub-task Resolved Liang-Chi Hsieh
        18.
        Avoid unnecessary function calls in code generation Sub-task Resolved Marco Gaido
        19.
        Shorten lifetime of global variables used in HashAggregateExec Sub-task Resolved Unassigned
        20.
        Increase default value of spark.sql.codegen.maxFields Sub-task Open Unassigned
        21.
        64KB JVM bytecode limit problem with aggregation Sub-task Open Unassigned
        22.
        64KB JVM bytecode limit problem with filter Sub-task Open Unassigned
        23.
        splitExpression can create too many method calls (generating a Constant Pool limit error) Sub-task Resolved Marco Gaido
        24.
        org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificOrdering" grows beyond 64 KB Sub-task Resolved Liwei Lin
        25.
        Code Generation: Constant Pool Past Limit for Wide/Nested Dataset Sub-task Resolved Kazuaki Ishizaki

          Activity

            People

            • Assignee:
              kiszk Kazuaki Ishizaki
              Reporter:
              smilegator Xiao Li
            • Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

              Dates

              • Created:
                Updated: