[GLUTEN-8851][VL] cuDF: Validate the plan before execution#10889
Merged
jinchengchenghh merged 7 commits intoapache:mainfrom Oct 20, 2025
Merged
[GLUTEN-8851][VL] cuDF: Validate the plan before execution#10889jinchengchenghh merged 7 commits intoapache:mainfrom
jinchengchenghh merged 7 commits intoapache:mainfrom
Conversation
jinchengchenghh
force-pushed
the
validation
branch
from
57ce36c to
39730d8
Compare
zhztheplayer
changed the title
Comment on lines
+41
to
+56
| if (!hasLeaf && VeloxConfig.get.cudfEnableValidation) { | ||
| if ( | ||
| VeloxCudfPlanValidatorJniWrapper.validate( | ||
| transformer.substraitPlan.toProtobuf.toByteArray) | ||
| ) { | ||
| transformer.foreach { | ||
| case _: LeafTransformSupport => | ||
| case t: TransformSupport => | ||
| t.setTagValue(CudfTag.CudfTag, true) | ||
| case _ => | ||
| } | ||
| transformer.setTagValue(CudfTag.CudfTag, true) | ||
| } | ||
| } else { | ||
| transformer.setTagValue(CudfTag.CudfTag, !hasLeaf) | ||
| } |
Member
There was a problem hiding this comment.
A dumb question, after this PR, what happens if user sets spark.gluten.sql.columnar.backend.velox.cudf.enableValidation=false but the plan is actually invalid for cuDF?
Contributor
Author
There was a problem hiding this comment.
Will not check if the pan is fully offloaded to GPU, may introduce extra cudf to velox format conversion. This is not a suggested behavior.
zhztheplayer
approved these changes
Oct 20, 2025
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Initializing task in Task constructor has been merged, so we can get the operator information from task, if validate successfully, set the whole stage transformer cudf tag to true to enable session config kCudfEnabled to true, and set the children operator CUDF tag to true to update the name in Spark UI with "Cudf".
Related issue: #8851