[SPARK-44236][SQL] Disable WholeStageCodegen when set spark.sql.codegen.factoryMode to NO_CODEGEN
#41779
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…gen.factoryMode` to NO_CODEGEN
|
cc @MaxGekk |
MaxGekk
reviewed
Jun 29, 2023
| @@ -47,8 +47,7 @@ trait CodegenInterpretedPlanTest extends PlanTest { | |||
| super.test(testName + " (codegen path)", testTags: _*)( | |||
| withSQLConf(SQLConf.CODEGEN_FACTORY_MODE.key -> codegenMode) { testFun })(pos) | |||
| super.test(testName + " (interpreted path)", testTags: _*)( | |||
| withSQLConf(SQLConf.CODEGEN_FACTORY_MODE.key -> interpretedMode) { | |||
| withSQLConf(SQLConf.WHOLESTAGE_CODEGEN_ENABLED.key -> "false") { testFun }})(pos) | |||
| withSQLConf(SQLConf.CODEGEN_FACTORY_MODE.key -> interpretedMode) {{ testFun }})(pos) | |||
There was a problem hiding this comment.
Sorry, I didn't get why you reverted #41467 , and why do you need double curly brackets.
There was a problem hiding this comment.
Sorry, I didn't get why you reverted #41467
If when we support set spark.sql.codegen.factoryMode to NO_CODEGEN then disable WholeStageCodegen in this PR(just like set WHOLESTAGE_CODEGEN_ENABLED to false). The change seem unnecessary. This idea come from I collect other codegen config mistakes. The NO_CODEGEN behavior not right at now.
why do you need double curly brackets.
My mistake, I will fix it. Thanks
MaxGekk
reviewed
Jun 29, 2023
| @@ -942,7 +942,8 @@ case class CollapseCodegenStages( | |||
| } | |||
|
|
|||
| def apply(plan: SparkPlan): SparkPlan = { | |||
| if (conf.wholeStageEnabled) { | |||
| if (conf.wholeStageEnabled && CodegenObjectFactoryMode.withName(conf.codegenFactoryMode) | |||
| != CodegenObjectFactoryMode.NO_CODEGEN) { | |||
There was a problem hiding this comment.
@cloud-fan Could you look at this, please.
There was a problem hiding this comment.
Hi, @cloud-fan Can you take a look this, thanks.
cloud-fan
approved these changes
Aug 8, 2023
|
this config is mostly for testing and experimenting, so better to make it easy to use (no need to set whole stage codegen conf if we don't want codegen) |
cloud-fan
pushed a commit
that referenced
this pull request
Aug 8, 2023
…gen.factoryMode` to NO_CODEGEN ### What changes were proposed in this pull request? After #41467 , we fix the `CodegenInterpretedPlanTest ` will execute codeGen even set `spark.sql.codegen.factoryMode` to `NO_CODEGEN`. Before this PR, `spark.sql.codegen.factoryMode` can't disable WholeStageCodegen, many test case want to disable codegen by set `spark.sql.codegen.factoryMode` to `NO_CODEGEN`, but it not work for WholeStageCodegen. So this PR change the `spark.sql.codegen.factoryMode` behavior, when set `NO_CODEGEN`, we will disable `WholeStageCodegen` too. ### Why are the changes needed? Fix the `spark.sql.codegen.factoryMode` config behavior. ### Does this PR introduce _any_ user-facing change? Yes, the config logic changed. ### How was this patch tested? add new test. Closes #41779 from Hisoka-X/SPARK-44236_wholecodegen_disable. Authored-by: Jia Fan <fanjiaeminem@qq.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com> (cherry picked from commit 74fa07c) Signed-off-by: Wenchen Fan <wenchen@databricks.com>
|
Thanks @MaxGekk @cloud-fan |
valentinp17
pushed a commit
to valentinp17/spark
that referenced
this pull request
Aug 24, 2023
…gen.factoryMode` to NO_CODEGEN ### What changes were proposed in this pull request? After apache#41467 , we fix the `CodegenInterpretedPlanTest ` will execute codeGen even set `spark.sql.codegen.factoryMode` to `NO_CODEGEN`. Before this PR, `spark.sql.codegen.factoryMode` can't disable WholeStageCodegen, many test case want to disable codegen by set `spark.sql.codegen.factoryMode` to `NO_CODEGEN`, but it not work for WholeStageCodegen. So this PR change the `spark.sql.codegen.factoryMode` behavior, when set `NO_CODEGEN`, we will disable `WholeStageCodegen` too. ### Why are the changes needed? Fix the `spark.sql.codegen.factoryMode` config behavior. ### Does this PR introduce _any_ user-facing change? Yes, the config logic changed. ### How was this patch tested? add new test. Closes apache#41779 from Hisoka-X/SPARK-44236_wholecodegen_disable. Authored-by: Jia Fan <fanjiaeminem@qq.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com>
ragnarok56
pushed a commit
to ragnarok56/spark
that referenced
this pull request
Mar 2, 2024
…gen.factoryMode` to NO_CODEGEN ### What changes were proposed in this pull request? After apache#41467 , we fix the `CodegenInterpretedPlanTest ` will execute codeGen even set `spark.sql.codegen.factoryMode` to `NO_CODEGEN`. Before this PR, `spark.sql.codegen.factoryMode` can't disable WholeStageCodegen, many test case want to disable codegen by set `spark.sql.codegen.factoryMode` to `NO_CODEGEN`, but it not work for WholeStageCodegen. So this PR change the `spark.sql.codegen.factoryMode` behavior, when set `NO_CODEGEN`, we will disable `WholeStageCodegen` too. ### Why are the changes needed? Fix the `spark.sql.codegen.factoryMode` config behavior. ### Does this PR introduce _any_ user-facing change? Yes, the config logic changed. ### How was this patch tested? add new test. Closes apache#41779 from Hisoka-X/SPARK-44236_wholecodegen_disable. Authored-by: Jia Fan <fanjiaeminem@qq.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
After #41467 , we fix the
CodegenInterpretedPlanTestwill execute codeGen even setspark.sql.codegen.factoryModetoNO_CODEGEN. Before this PR,spark.sql.codegen.factoryModecan't disable WholeStageCodegen, many test case want to disable codegen by setspark.sql.codegen.factoryModetoNO_CODEGEN, but it not work for WholeStageCodegen. So this PR change thespark.sql.codegen.factoryModebehavior, when setNO_CODEGEN, we will disableWholeStageCodegentoo.Why are the changes needed?
Fix the
spark.sql.codegen.factoryModeconfig behavior.Does this PR introduce any user-facing change?
Yes, the config logic changed.
How was this patch tested?
add new test.