fix(r): Ensure wrapper array stream eagerly releases the wrapped array stream #333
Conversation
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
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #333 +/- ##
==========================================
- Coverage 88.04% 88.03% -0.01%
==========================================
Files 72 68 -4
Lines 11369 11269 -100
==========================================
- Hits 10010 9921 -89
+ Misses 1359 1348 -11 ☔ View full report in Codecov by Sentry. |
paleolimbot
added a commit
to paleolimbot/arrow-nanoarrow
that referenced
this pull request
Dec 8, 2023
…y stream (apache#333) Uncovered when investigating apache/arrow-adbc#1348 and apache/arrow-adbc#1334 . Before, the release callback was getting called by the garbage collector (so no memory leak was reported); however, the order of the array stream finalizer and the user-supplied R finalizer was non-deterministic based on when the garbage collector ran.
paleolimbot
added a commit
to apache/arrow-adbc
that referenced
this pull request
Dec 19, 2023
This PR adds a `.child_count` for all database, connection, and statement objects to ensure that we do not call the C-level release callback while a child object is available. Since very early commits, we had set `R_ExternalPtrProtected()` to the parent object, which meant that if you never called `adbc_XXX_release()`, R would correctly sort out the object dependencies and call the release callbacks in the correct order. One major exception to that was the returned `ArrowArrayStream` wrappers, for which there was no mechanism to add a dependent SEXP until I added it (in nanoarrow 0.2, I think); however, it wasn't actually used for anything except for the stream returned by `read_adbc()`. @nbenn / @krlmlr you are probably the best candidates to give this a review/ensure that it fixes or makes it easier to debug the issues you have been having in adbi! Note that apache/arrow-nanoarrow#333 may have been the primary culprit (I will push that out to CRAN ASAP as a tweak release). Closes apache/arrow-nanoarrow#323 , #1128 . Perhaps related: #1348 . I triggered some extended checks as well (as they include valgrind): https://github.com/paleolimbot/arrow-adbc/actions/runs/7133874110
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.
Uncovered when investigating apache/arrow-adbc#1348 and apache/arrow-adbc#1334 . Before, the release callback was getting called by the garbage collector (so no memory leak was reported); however, the order of the array stream finalizer and the user-supplied R finalizer was non-deterministic based on when the garbage collector ran.