feat(gc): add WeakRef and FinalizationRegistry#702
Merged
Conversation
Implement weak targets, kept objects, finalization cleanup jobs, and error surfacing for queueMicrotask/finalizers. Add JavaScript coverage across interpreter and bytecode behavior plus documentation updates.
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughThis pull request implements ES2021 WeakRef and FinalizationRegistry support and the GC/execution model changes required to support them: queued roots and kept objects in the collector, serialized collection, a dual microtask/finalization queue with queued-root lifecycle, new native WeakRef/FinalizationRegistry types and class-values, engine registration, structuredClone updates, and extensive tests. ChangesWeakRef and FinalizationRegistry (ES2021)
🎯 4 (Complex) | ⏱️ ~60 minutes Possibly Related PRs
Suggested labels
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. Comment |
Contributor
Suite TimingTest Runner (interpreted: 9,965 passed; bytecode: 9,965 passed)
MemoryGC rows aggregate the main thread plus all worker thread-local GCs. Test runner worker shutdown frees thread-local heaps in bulk; that shutdown reclamation is not counted as GC collections or collected objects.
Benchmarks (interpreted: 407; bytecode: 407)
MemoryGC rows aggregate the main thread plus all worker thread-local GCs. Benchmark runner performs explicit between-file collections, so collection and collected-object counts can be much higher than the test runner.
Measured on ubuntu-latest x64. |
Contributor
Benchmark Results407 benchmarks Interpreted: 🟢 352 improved · 🔴 13 regressed · 42 unchanged · avg +12.5% arraybuffer.js — Interp: 🟢 13, 1 unch. · avg +14.2% · Bytecode: 🟢 8, 6 unch. · avg +6.8%
arrays.js — Interp: 🟢 14, 5 unch. · avg +12.7% · Bytecode: 🟢 14, 5 unch. · avg +8.3%
async-await.js — Interp: 🟢 5, 1 unch. · avg +13.7% · Bytecode: 🟢 2, 4 unch. · avg +0.0%
async-generators.js — Interp: 🟢 2 · avg +14.8% · Bytecode: 🔴 1, 1 unch. · avg -23.7%
base64.js — Interp: 🟢 3, 🔴 3, 4 unch. · avg +0.6% · Bytecode: 🟢 2, 🔴 4, 4 unch. · avg +0.4%
classes.js — Interp: 🟢 30, 1 unch. · avg +12.2% · Bytecode: 🟢 18, 13 unch. · avg +5.6%
closures.js — Interp: 🟢 11 · avg +17.3% · Bytecode: 🟢 11 · avg +13.0%
collections.js — Interp: 🟢 12 · avg +18.2% · Bytecode: 🟢 12 · avg +15.1%
csv.js — Interp: 🟢 13 · avg +13.5% · Bytecode: 🟢 13 · avg +15.7%
destructuring.js — Interp: 🟢 20, 2 unch. · avg +13.3% · Bytecode: 🟢 20, 2 unch. · avg +12.3%
fibonacci.js — Interp: 🟢 8 · avg +17.2% · Bytecode: 🟢 8 · avg +21.2%
float16array.js — Interp: 🟢 28, 4 unch. · avg +10.6% · Bytecode: 🟢 29, 3 unch. · avg +10.6%
for-of.js — Interp: 🟢 6, 1 unch. · avg +12.4% · Bytecode: 🟢 7 · avg +16.7%
generators.js — Interp: 🟢 3, 1 unch. · avg +6.3% · Bytecode: 🟢 2, 2 unch. · avg +8.8%
iterators.js — Interp: 🟢 41, 1 unch. · avg +16.1% · Bytecode: 🟢 40, 2 unch. · avg +9.0%
json.js — Interp: 🟢 19, 1 unch. · avg +15.8% · Bytecode: 🟢 20 · avg +10.8%
jsx.jsx — Interp: 🟢 18, 3 unch. · avg +10.2% · Bytecode: 🟢 20, 1 unch. · avg +10.1%
modules.js — Interp: 🟢 8, 1 unch. · avg +24.9% · Bytecode: 🟢 9 · avg +19.3%
numbers.js — Interp: 🟢 11 · avg +14.2% · Bytecode: 🟢 11 · avg +12.2%
objects.js — Interp: 🟢 5, 2 unch. · avg +8.6% · Bytecode: 🟢 7 · avg +11.3%
promises.js — Interp: 🟢 11, 1 unch. · avg +9.8% · Bytecode: 🟢 4, 8 unch. · avg +0.8%
regexp.js — Interp: 🟢 5, 🔴 3, 3 unch. · avg +4.4% · Bytecode: 🟢 5, 🔴 1, 5 unch. · avg +1.7%
strings.js — Interp: 🟢 17, 2 unch. · avg +9.3% · Bytecode: 🟢 19 · avg +10.1%
tsv.js — Interp: 🟢 9 · avg +15.3% · Bytecode: 🟢 8, 1 unch. · avg +13.3%
typed-arrays.js — Interp: 🟢 15, 🔴 2, 5 unch. · avg +0.6% · Bytecode: 🟢 11, 🔴 8, 3 unch. · avg -4.6%
uint8array-encoding.js — Interp: 🟢 15, 🔴 1, 2 unch. · avg +6.1% · Bytecode: 🟢 11, 🔴 1, 6 unch. · avg +4.3%
weak-collections.js — Interp: 🟢 10, 🔴 4, 1 unch. · avg +28.3% · Bytecode: 🟢 10, 🔴 2, 3 unch. · avg -8.3%
Deterministic profile diffDeterministic profile diff: no significant changes. Measured on ubuntu-latest x64. Benchmark ranges compare cached main-branch min/max ops/sec with the PR run; overlapping ranges are treated as unchanged noise. Percentage deltas are secondary context. |
Avoid retaining stale weak unregister-token pointers in FinalizationRegistry cells after GC. Keep the target cell alive for later cleanup and add regression coverage for unregister after token collection.
coderabbitai
Bot
added
new feature
Contributor
test262 Conformance
Areas closest to 100%
Per-test deltas (+79 / -0)Newly passing (79):
Steady-state failures are non-blocking; regressions vs the cached main baseline (lower total pass count, or any PASS → non-PASS transition) fail the conformance gate. Measured on ubuntu-latest x64, bytecode mode. Areas grouped by the first two test262 path components; minimum 25 attempted tests, areas already at 100% excluded. Δ vs main compares against the most recent cached |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@source/units/Goccia.MicrotaskQueue.pas`:
- Around line 150-208: Remove the transient re-rooting in the task execution
path: delete the calls to TGarbageCollector.Instance.AddTempRoot for
ATask.Handler, ATask.Value and Promise at the start of the shown block and
remove the matching TGarbageCollector.Instance.RemoveTempRoot calls in the
finally; these objects are already rooted by AddQueuedRoots/RemoveQueuedRoots
(used by DrainQueue) so the extra AddTempRoot/RemoveTempRoot pairs (which use
set-membership, not ref-counting) must be eliminated to avoid accidentally
clearing an outer root.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: f78980d2-c29a-42a2-8f2f-a7c59404382c
📒 Files selected for processing (37)
docs/built-ins.mddocs/garbage-collector.mddocs/interpreter.mddocs/language-tables.mddocs/language.mddocs/value-system.mdsource/app/GocciaTestRunner.dprsource/units/Goccia.Builtins.GlobalPromise.passource/units/Goccia.Builtins.Globals.passource/units/Goccia.Constants.ConstructorNames.passource/units/Goccia.Engine.passource/units/Goccia.Error.Messages.passource/units/Goccia.Error.Suggestions.passource/units/Goccia.FetchManager.passource/units/Goccia.GarbageCollector.passource/units/Goccia.MicrotaskQueue.passource/units/Goccia.Threading.passource/units/Goccia.Values.ClassValue.passource/units/Goccia.Values.FinalizationRegistryValue.passource/units/Goccia.Values.NativeFunction.passource/units/Goccia.Values.PromiseValue.passource/units/Goccia.Values.WeakRefValue.pastests/built-ins/FinalizationRegistry/constructor.jstests/built-ins/FinalizationRegistry/gc.jstests/built-ins/FinalizationRegistry/prototype/register.jstests/built-ins/FinalizationRegistry/prototype/toStringTag.jstests/built-ins/FinalizationRegistry/prototype/unregister.jstests/built-ins/FinalizationRegistry/subclassing.jstests/built-ins/Object/prototype/toString.jstests/built-ins/WeakRef/constructor.jstests/built-ins/WeakRef/gc.jstests/built-ins/WeakRef/prototype/deref.jstests/built-ins/WeakRef/prototype/toStringTag.jstests/built-ins/WeakRef/subclassing.jstests/built-ins/global-properties/global-this.jstests/built-ins/structuredClone/collections.jstests/language/classes/class-length-property.js
coderabbitai
Bot
added
the
documentation
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
1 participant
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.
Summary
Testing
Verification run:
./fixtures/ffi/build.sh./build.pas testrunner./build.pas loader./build/GocciaTestRunner tests./build/GocciaTestRunner tests --mode=bytecode./format.pas --checkgit diff --check