beliefer opened a new pull request, #43449: URL: https://github.com/apache/spark/pull/43449
### What changes were proposed in this pull request? Before https://github.com/apache/spark/pull/39170, Spark only supports insert runtime filter for application side of shuffle join on single-layer. Considered it's not worth to insert more runtime filter if one side of the shuffle join already exists runtime filter, Spark restricts it at https://github.com/apache/spark/blob/7057952f6bc2c5cf97dd408effd1b18bee1cb8f4/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/InjectRuntimeFilter.scala#L346 After https://github.com/apache/spark/pull/39170, Spark supports insert runtime filter for one side of any shuffle join on multi-layer. But the restrictions on multi-layer runtime filter mentioned above looks outdated. ### Why are the changes needed? Release restrictions on multi-layer runtime filter. Expand optimization surface. ### Does this PR introduce _any_ user-facing change? 'No'. New feature. ### How was this patch tested? Test cases updated. ### Was this patch authored or co-authored using generative AI tooling? 'No'. -- 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]
