[query] Fix memory leak in combOpFSerialized (and therefore densify)#9028
Merged
danking merged 1 commit intohail-is:masterfrom Jun 30, 2020
Merged
[query] Fix memory leak in combOpFSerialized (and therefore densify)#9028danking merged 1 commit intohail-is:masterfrom
danking merged 1 commit intohail-is:masterfrom
Conversation
The previous implementation, while seeming to be well-abstracted at first, actually had a rather devious property of creating agg states for multiple classes multiple times. I'm still working on figuring out *exactly* the place where our assumptions broke down, but this change definitely fixes the problem, and simplifies the implementation by directly using IR, instead of other compiled functions. This problem was a symptom of a larger issue, which is that the ownership semantics of the current aggregator system is way too complex to be coding against regularly. This all will go away when lowering is complete, in favor of the *much* simpler set of IR nodes that are used in lowering. We may need to address this problem sooner, though. CHANGELOG: Fixed memory leak affecting `Table.annotate` with scans, `hl.experimental.densify`, and `Table.group_by` / `aggregate`.
chrisvittal
approved these changes
Jun 30, 2020
Collaborator
chrisvittal
left a comment
chrisvittal
left a comment
There was a problem hiding this comment.
Thank you for coming up with a solution to this.
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The previous implementation, while seeming to be well-abstracted at
first, actually had a rather devious property of creating agg states
for multiple classes multiple times. I'm still working on figuring out
exactly the place where our assumptions broke down, but this change
definitely fixes the problem, and simplifies the implementation by
directly using IR, instead of other compiled functions.
This problem was a symptom of a larger issue, which is that the
ownership semantics of the current aggregator system is way too complex
to be coding against regularly. This all will go away when lowering is
complete, in favor of the much simpler set of IR nodes that are used
in lowering. We may need to address this problem sooner, though.
CHANGELOG: Fixed memory leak affecting
Table.annotatewith scans,hl.experimental.densify, andTable.group_by/aggregate.