Chang Chen created SPARK-54850:
----------------------------------
Summary: Improve extractShuffleIds to find AdaptiveSparkPlanExec
anywhere in plan tree
Key: SPARK-54850
URL: https://issues.apache.org/jira/browse/SPARK-54850
Project: Spark
Issue Type: Improvement
Components: SQL
Affects Versions: 4.1.0
Reporter: Chang Chen
Fix For: 4.1.1
Previously, the method implicitly assumed that if AQE is enabled,
{{AdaptiveSparkPlanExec}} would be at the root of the input. While this holds
true for standard Spark execution (where it is inserted under Command), it may
not be the root in other scenarios, such as when using Gluten, which inserts a
special physical plan for column-to-row transitions
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
