fix(ci): prove Unity cleanup before lock release by wallstop · Pull Request #270 · Ambiguous-Interactive/DxMessaging

Eli Pinkerton (wallstop) · 2026-07-12T20:13:48Z

Summary

emit fail-closed Unity cleanup proof without masking job results
bound all final cleanup steps and pass proof to lock release
enforce exact acquire -> licensed work -> cleanup -> release contracts across six windows

Verification

focused lifecycle tests: 12/12
Prettier and actionlint
adversarial implementation/re-review: zero remaining issues

Depends on Ambiguous-Interactive/ambiguous-organization-build-lock#25.

Note

Medium Risk
Changes how org Unity locks are released when license return is unproven (depends on ambiguous-organization-build-lock#25); mis-signaling could quarantine runners or leak seats, though return steps remain non-fatal to job outcome.

Overview
Adds fail-closed Unity license cleanup proof so organization lock release can tell a confirmed seat return from an uncertain one, without making the return step fail the job.

The return-unity-license composite now exposes a resource-safe output (defaults false; only set true on exit code 0 or the existing allowlisted nonzero log pattern). Unity workflows give the return step an id, a 5-minute timeout, continue-on-error: true, and pass resource-safe into release-build-lock@v1 on the following Release organization Unity lock step. The same pattern is applied across six licensed windows (unity-tests, benchmarks, perf, release editmode/unitypackage, GameCI experiment); GameCI’s comment now treats the backstop as real using the provisioned editor path instead of documenting a no-op.

ci-aggregate-workflow.test.js locks in acquire → licensed work → return → release ordering and the return action’s non-masking exit 0 contract.

^{Reviewed by Cursor Bugbot for commit 957e109. Bugbot is set up for automated code reviews on this repo. Configure here.}

…ce-lifecycle # Conflicts: # .github/workflows/perf-numbers.yml # .github/workflows/release.yml # .github/workflows/unity-benchmarks.yml # .github/workflows/unity-gameci-experiment.yml # .github/workflows/unity-tests.yml # scripts/__tests__/ci-aggregate-workflow.test.js

Eli Pinkerton (wallstop) · 2026-07-12T20:14:07Z

Copilot review

Please review cleanup proof classification, non-masking timeout behavior, and all lock-window contracts.

Eli Pinkerton (wallstop) · 2026-07-12T20:14:09Z

Cursor (@cursor) review

Please review cleanup proof classification, non-masking timeout behavior, and all lock-window contracts.

Copilot

Pull request overview

This PR strengthens CI’s licensed Unity “lock window” lifecycle by emitting a fail-closed cleanup proof from the return-unity-license composite action and wiring that proof into lock release steps across all Unity-licensed workflows, with contract tests enforcing the acquire → work → cleanup → release ordering.

Changes:

Add a fail-closed resource-safe output to .github/actions/return-unity-license (defaults false, flips true only on proven successful return conditions, and still never fails the job).
Update all six Unity lock windows to: assign an explicit id to the return step, bound it with a 5-minute step timeout, and pass resource-safe into release-build-lock@v1.
Extend scripts/__tests__/ci-aggregate-workflow.test.js with workflow-contract tests for ordering and for return-action proof classifications.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
`scripts/__tests__/ci-aggregate-workflow.test.js`	Adds contract tests enforcing six Unity lock windows’ lifecycle ordering and proof wiring, plus proof-classification assertions for the return action.
`.github/workflows/unity-tests.yml`	Adds `return_unity_license` step id + timeout and forwards `resource-safe` into the lock release step.
`.github/workflows/unity-gameci-experiment.yml`	Updates comment to reflect proof limitations, adds return step id + timeout, forwards `resource-safe` into lock release.
`.github/workflows/unity-benchmarks.yml`	Adds return step id + timeout and forwards `resource-safe` into the lock release step.
`.github/workflows/release.yml`	Adds return step id + timeout in both Unity jobs and forwards `resource-safe` into each lock release step.
`.github/workflows/perf-numbers.yml`	Adds return step id + timeout and forwards `resource-safe` into the lock release step.
`.github/actions/return-unity-license/action.yml`	Introduces `resource-safe` composite output and sets it fail-closed unless return is proven successful.

Copilot · 2026-07-12T20:20:58Z

Copilot review

Please review cleanup proof classification, non-masking timeout behavior, and all lock-window contracts.

Addressed in 2f8866c: the six bounded Return Unity license steps now use continue-on-error: true so a GitHub-enforced timeout cannot mask the licensed job result, and the CI aggregate workflow contract test now pins that invariant alongside the existing cleanup proof and lock-release wiring.

Eli Pinkerton (wallstop) · 2026-07-12T20:28:35Z

Copilot review

Please review exact head: all cleanup timeouts are non-masking, and the GameCI cleanup documentation now matches its provisioned-editor proof behavior.

Eli Pinkerton (wallstop) · 2026-07-12T20:28:37Z

Cursor (@cursor) review

Please review exact head: all cleanup timeouts are non-masking, and the GameCI cleanup documentation now matches its provisioned-editor proof behavior.

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit b2e9793. Configure here.}

Copilot · 2026-07-12T20:30:33Z

Copilot review

Please review exact head: all cleanup timeouts are non-masking, and the GameCI cleanup documentation now matches its provisioned-editor proof behavior.

Reviewed exact head b2e9793: no issues found.

All Unity cleanup backstops remain timeout-bounded with continue-on-error: true, and scripts/__tests__/ci-aggregate-workflow.test.js still enforces that non-masking contract across every lock window.
.github/workflows/unity-gameci-experiment.yml now documents the GameCI backstop correctly: the proof comes from the provisioned editor return step, so resource-safe only flips when that local return proves success/already-absent state.

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

Eli Pinkerton (wallstop) · 2026-07-12T20:40:04Z

Copilot review

Please review exact head: the lifecycle contract now explicitly rejects any missing acquire/work/cleanup/release step before checking order.

Eli Pinkerton (wallstop) · 2026-07-12T20:40:06Z

Cursor (@cursor) review

Please review exact head: the lifecycle contract now explicitly rejects any missing acquire/work/cleanup/release step before checking order.

github-actions · 2026-07-12T20:40:38Z

Dispatch throughput (Unity 6000.3.16f1)

Runner: 13th Gen Intel(R) Core(TM) i9-13900KF, 24C/32T @ 3000MHz; 64GB DDR5@4200; NVIDIA GeForce RTX 3060; Microsoft Windows 11 Pro N (10.0.26200)

Regenerated by CI from commit 957e109 (run logs). These numbers MOVED versus the committed doc; CI commits the refreshed doc directly to the default branch after this PR merges.

Latest CI benchmark run: Unity 6000.3.16f1, commit 957e10983930744638013578ac492c0125c0c512.

Runner: 13th Gen Intel(R) Core(TM) i9-13900KF, 24C/32T @ 3000MHz; 64GB DDR5@4200; NVIDIA GeForce RTX 3060; Microsoft Windows 11 Pro N (10.0.26200)

Dispatch throughput - Standalone (IL2CPP)

Platform: Standalone IL2CPP x64 Release (WindowsPlayer; Unity 6000.3.16f1).

Scenario	Throughput / Wall clock
Empty Bus Dispatch	49.46 M emits/sec
Untargeted Flood (One Handler)	37.62 M emits/sec
Untargeted Flood (Two Handlers, One Priority)	36.80 M emits/sec
Untargeted Flood (Three Handlers, One Priority)	33.61 M emits/sec
Untargeted Flood (Four Handlers, One Priority)	30.62 M emits/sec
Untargeted Flood (Four Handlers, Four Priorities)	31.15 M emits/sec
Untargeted Flood (Sixteen Handlers, One Priority)	18.20 M emits/sec
Untargeted Flood (One Inactive Handler)	39.01 M emits/sec
Untargeted First Dispatch (Cold, Distinct Types)	0.197 ms
Targeted Flood (No Matching Target)	15.77 M emits/sec
Targeted Flood (One Listener)	12.02 M emits/sec
Targeted Flood (Sixteen Listeners)	8.62 M emits/sec
Targeted First Dispatch (Cold, Distinct Types)	0.196 ms
Broadcast Flood (One Handler)	21.50 M emits/sec
Broadcast First Dispatch (Cold, Distinct Types)	0.188 ms
Interceptor Heavy (Four Interceptors)	4.08 M emits/sec
Post-Processing Heavy (Four Post-Processors)	14.54 M emits/sec
Message Bus Construction (1000)	29.923 ms
Registration Token Construction (1000, Prebuilt Handler + Bus)	0.060 ms
Registration Flood (1000 Types, Cold Bus)	598.342 ms
Registration Flood (1000 Types, Warm JIT)	4.413 ms
Untargeted Registration (Marginal, 1000 Same-Type)	0.414 ms
Targeted Registration (Marginal, 1000 Same-Type)	0.461 ms
Broadcast Registration (Marginal, 1000 Same-Type)	0.460 ms
Deregistration Flood (1000 Types, Cold)	1.974 ms
Deregistration Flood (1000 Types, Warm JIT)	1.231 ms

Dispatch throughput - PlayMode (Mono)

Platform: Editor PlayMode Mono x64 Release (WindowsEditor; Unity 6000.3.16f1).

Scenario	Throughput / Wall clock	GC allocs	GC bytes
Empty Bus Dispatch	29.18 M emits/sec	0	0
Untargeted Flood (One Handler)	23.96 M emits/sec	0	0
Untargeted Flood (Two Handlers, One Priority)	22.02 M emits/sec	0	0
Untargeted Flood (Three Handlers, One Priority)	21.13 M emits/sec	0	0
Untargeted Flood (Four Handlers, One Priority)	19.73 M emits/sec	0	0
Untargeted Flood (Four Handlers, Four Priorities)	20.50 M emits/sec	0	0
Untargeted Flood (Sixteen Handlers, One Priority)	9.85 M emits/sec	0	0
Untargeted Flood (One Inactive Handler)	24.97 M emits/sec	0	0
Untargeted First Dispatch (Cold, Distinct Types)	3.402 ms	41	11,318
Targeted Flood (No Matching Target)	25.22 M emits/sec	0	0
Targeted Flood (One Listener)	18.05 M emits/sec	0	0
Targeted Flood (Sixteen Listeners)	8.93 M emits/sec	0	0
Targeted First Dispatch (Cold, Distinct Types)	5.290 ms	42	19,876
Broadcast Flood (One Handler)	16.40 M emits/sec	0	0
Broadcast First Dispatch (Cold, Distinct Types)	3.704 ms	41	19,534
Interceptor Heavy (Four Interceptors)	3.13 M emits/sec	0	0
Post-Processing Heavy (Four Post-Processors)	11.89 M emits/sec	0	0
Message Bus Construction (1000)	31.573 ms	40,011	2,480,736
Registration Token Construction (1000, Prebuilt Handler + Bus)	0.111 ms	1,002	120,096
Registration Flood (1000 Types, Cold Bus)	3137.449 ms	52,145	3,813,528
Registration Flood (1000 Types, Warm JIT)	11.552 ms	30,057	2,489,792
Untargeted Registration (Marginal, 1000 Same-Type)	0.544 ms	2,030	640,484
Targeted Registration (Marginal, 1000 Same-Type)	0.594 ms	2,030	672,484
Broadcast Registration (Marginal, 1000 Same-Type)	0.610 ms	2,030	672,484
Deregistration Flood (1000 Types, Cold)	107.803 ms	42	116,856
Deregistration Flood (1000 Types, Warm JIT)	1.893 ms	42	116,856

Library comparison - throughput (Standalone (IL2CPP))

Technology	Global -> 1 subscriber	Global -> 16 subscribers	Keyed/targeted -> 1 of many	Priority-ordered dispatch	Filtered/intercepted dispatch	Post-processing dispatch	Subscribe/unsubscribe churn	Struct message (no boxing)
DxMessaging	35.80 M emits/sec	18.15 M emits/sec	11.19 M emits/sec	30.75 M emits/sec	9.02 M emits/sec	15.64 M emits/sec	0.91 M emits/sec	36.18 M emits/sec
MessagePipe	86.34 M emits/sec	17.34 M emits/sec	10.68 M emits/sec	N/A	74.28 M emits/sec	N/A	2.06 M emits/sec	94.37 M emits/sec
UniRx MessageBroker	5.11 M emits/sec	2.75 M emits/sec	N/A	N/A	N/A	N/A	0.75 M emits/sec	5.13 M emits/sec
Zenject SignalBus	2.53 M emits/sec	1.35 M emits/sec	N/A	N/A	N/A	N/A	1.64 M emits/sec	2.45 M emits/sec
Unity Atoms	124.52 M emits/sec	47.95 M emits/sec	134.37 M emits/sec	N/A	N/A	N/A	13.28 M emits/sec	N/A
ScriptableObject channel	136.02 M emits/sec	27.60 M emits/sec	192.81 M emits/sec	N/A	N/A	N/A	35.29 M emits/sec	184.47 M emits/sec
UnityEvent	93.71 M emits/sec	11.31 M emits/sec	102.88 M emits/sec	N/A	N/A	N/A	3.56 M emits/sec	104.22 M emits/sec
C# event	294.11 M emits/sec	53.62 M emits/sec	75.66 M emits/sec	N/A	N/A	N/A	12.99 M emits/sec	288.13 M emits/sec
Unity SendMessage	8.28 M emits/sec	1.05 M emits/sec	8.19 M emits/sec	N/A	N/A	N/A	N/A	N/A

Library comparison - GC allocations per 10k ops (PlayMode (Mono))

Technology	Global -> 1 subscriber	Global -> 16 subscribers	Keyed/targeted -> 1 of many	Priority-ordered dispatch	Filtered/intercepted dispatch	Post-processing dispatch	Subscribe/unsubscribe churn	Struct message (no boxing)
DxMessaging	0	0	0	0	0	0	80,000	0
MessagePipe	0	0	0	N/A	0	N/A	20,000	0
UniRx MessageBroker	0	0	N/A	N/A	N/A	N/A	150,000	0
Zenject SignalBus	20,000	20,000	N/A	N/A	N/A	N/A	70,000	20,000
Unity Atoms	0	0	0	N/A	N/A	N/A	0	N/A
ScriptableObject channel	0	0	0	N/A	N/A	N/A	0	0
UnityEvent	0	0	0	N/A	N/A	N/A	50,000	0
C# event	0	0	0	N/A	N/A	N/A	0	0
Unity SendMessage	10,000	10,000	10,000	N/A	N/A	N/A	N/A	N/A

Library comparison - GC allocated bytes per 10k ops (PlayMode (Mono))

Technology	Global -> 1 subscriber	Global -> 16 subscribers	Keyed/targeted -> 1 of many	Priority-ordered dispatch	Filtered/intercepted dispatch	Post-processing dispatch	Subscribe/unsubscribe churn	Struct message (no boxing)
DxMessaging	0	0	0	0	0	0	7,560,000	0
MessagePipe	0	0	0	N/A	0	N/A	560,000	0
UniRx MessageBroker	0	0	N/A	N/A	N/A	N/A	6,650,000	0
Zenject SignalBus	600,000	600,000	N/A	N/A	N/A	N/A	3,200,000	600,000
Unity Atoms	0	0	0	N/A	N/A	N/A	0	N/A
ScriptableObject channel	0	0	0	N/A	N/A	N/A	0	0
UnityEvent	0	0	0	N/A	N/A	N/A	2,960,000	0
C# event	0	0	0	N/A	N/A	N/A	0	0
Unity SendMessage	200,000	200,000	200,000	N/A	N/A	N/A	N/A	N/A

github-actions · 2026-07-12T20:40:40Z

DxMessaging perf deltas vs master baseline (Unity 6000.3.16f1, Standalone IL2CPP)

Regenerated by CI (run logs). Changed beyond tolerance: yes. Regression gate: passed.

Reading the Delta column: the sign is normalized so + ALWAYS means better and - ALWAYS means worse, for every metric -- a + percentage is higher throughput (emits/sec) OR lower latency (wall clock ms). This Standalone IL2CPP player strips the profiler, so it cannot measure GC allocations or bytes; those columns are omitted from this comment entirely (not shown as n/a). Measured allocation counts and their regression coverage come from the in-editor Mono leg and the allocation test suite. When a leg DOES measure them, allocation/byte changes render as N fewer (better) / N more (worse).

Scenario	Baseline	Current	Delta
Empty Bus Dispatch	48.82 M emits/sec	49.46 M emits/sec	+1.32%
Untargeted Flood (One Handler)	40.02 M emits/sec	37.62 M emits/sec	-5.99%
Untargeted Flood (Two Handlers, One Priority)	38.09 M emits/sec	36.80 M emits/sec	-3.39%
Untargeted Flood (Three Handlers, One Priority)	36.29 M emits/sec	33.61 M emits/sec	-7.39%
Untargeted Flood (Four Handlers, One Priority)	34.21 M emits/sec	30.62 M emits/sec	-10.50%
Untargeted Flood (Four Handlers, Four Priorities)	34.89 M emits/sec	31.15 M emits/sec	-10.72%
Untargeted Flood (Sixteen Handlers, One Priority)	19.56 M emits/sec	18.20 M emits/sec	-6.94%
Untargeted Flood (One Inactive Handler)	40.05 M emits/sec	39.01 M emits/sec	-2.58%
Untargeted First Dispatch (Cold, Distinct Types)	0.198 ms	0.197 ms	+0.51%
Targeted Flood (No Matching Target)	15.81 M emits/sec	15.77 M emits/sec	-0.27%
Targeted Flood (One Listener)	12.03 M emits/sec	12.02 M emits/sec	-0.08%
Targeted Flood (Sixteen Listeners)	9.08 M emits/sec	8.62 M emits/sec	-4.98%
Targeted First Dispatch (Cold, Distinct Types)	0.200 ms	0.196 ms	+2.00%
Broadcast Flood (One Handler)	22.72 M emits/sec	21.50 M emits/sec	-5.34%
Broadcast First Dispatch (Cold, Distinct Types)	0.194 ms	0.188 ms	+3.09%
Interceptor Heavy (Four Interceptors)	4.09 M emits/sec	4.08 M emits/sec	-0.27%
Post-Processing Heavy (Four Post-Processors)	14.81 M emits/sec	14.54 M emits/sec	-1.87%
Message Bus Construction (1000)	17.334 ms	29.923 ms	-72.63%
Registration Token Construction (1000, Prebuilt Handler + Bus)	0.062 ms	0.060 ms	+3.23%
Registration Flood (1000 Types, Cold Bus)	558.092 ms	598.342 ms	-7.21%
Registration Flood (1000 Types, Warm JIT)	4.308 ms	4.413 ms	-2.44%
Untargeted Registration (Marginal, 1000 Same-Type)	0.418 ms	0.414 ms	+0.96%
Targeted Registration (Marginal, 1000 Same-Type)	0.454 ms	0.461 ms	-1.54%
Broadcast Registration (Marginal, 1000 Same-Type)	0.462 ms	0.460 ms	+0.43%
Deregistration Flood (1000 Types, Cold)	2.091 ms	1.974 ms	+5.60%
Deregistration Flood (1000 Types, Warm JIT)	1.501 ms	1.231 ms	+17.99%
Global -> 1 subscriber	35.70 M emits/sec	35.80 M emits/sec	+0.28%
Global -> 16 subscribers	18.81 M emits/sec	18.15 M emits/sec	-3.51%
Keyed/targeted -> 1 of many	11.87 M emits/sec	11.19 M emits/sec	-5.80%
Priority-ordered dispatch	31.16 M emits/sec	30.75 M emits/sec	-1.31%
Filtered/intercepted dispatch	9.32 M emits/sec	9.02 M emits/sec	-3.24%
Post-processing dispatch	16.16 M emits/sec	15.64 M emits/sec	-3.21%
Subscribe/unsubscribe churn	0.94 M emits/sec	0.91 M emits/sec	-2.97%
Struct message (no boxing)	35.27 M emits/sec	36.18 M emits/sec	+2.56%

Copilot · 2026-07-12T20:41:11Z

Copilot review

Please review exact head: the lifecycle contract now explicitly rejects any missing acquire/work/cleanup/release step before checking order.

Confirmed in 12dc1b1: positions.every((position) => position >= 0) now gates the order assertion, so any missing lifecycle step name (acquire/work/cleanup/release) fails the test explicitly rather than silently passing with a -1 in the sorted positions array. All 246 tests pass.

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 12dc1b1. Configure here.}

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

Eli Pinkerton (wallstop) · 2026-07-12T20:50:04Z

Copilot review

Please review exact head 957e109: cross-workflow missing-job diagnostics now include file and job id.

Eli Pinkerton (wallstop) · 2026-07-12T20:50:06Z

Cursor (@cursor) review

Please review exact head 957e109: cross-workflow missing-job diagnostics now include file and job id.

Copilot · 2026-07-12T20:51:26Z

Copilot review

Please review exact head 957e109: cross-workflow missing-job diagnostics now include file and job id.

Confirmed in 957e109: getJobBlock now takes an optional sourceName (defaults to "ci.yml" for backward compat) and the assertion message is ${sourceName}:${jobId} job must exist, so any renamed workflow/job produces a scoped diagnostic like unity-tests.yml:run-tests job must exist rather than the generic ci.yml must define.... All 246 tests pass.

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 957e109. Configure here.}

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated no new comments.

Eli Pinkerton (wallstop) added 4 commits July 11, 2026 21:58

Use build-lock App credentials and runner identity

057577e

Document build-lock App workflow requirements

b75a585

fix(ci): prove Unity cleanup before lock release

000dbef

Copilot AI review requested due to automatic review settings July 12, 2026 20:13

Copilot started reviewing on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:14 View session

cursor Bot reviewed Jul 12, 2026

View reviewed changes

Comment thread .github/workflows/unity-gameci-experiment.yml

Comment thread .github/workflows/unity-tests.yml

Copilot AI reviewed Jul 12, 2026

View reviewed changes

Copilot started work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:16 View session

fix(ci): keep Unity cleanup timeout non-masking

2f8866c

Copilot AI review requested due to automatic review settings July 12, 2026 20:19

Eli Pinkerton (wallstop) removed the request for review from Copilot July 12, 2026 20:19

Copilot finished work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:21

docs(ci): align GameCI cleanup proof contract

b2e9793

Copilot AI review requested due to automatic review settings July 12, 2026 20:28

Copilot started work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:28 View session

Copilot started reviewing on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:28 View session

cursor Bot reviewed Jul 12, 2026

View reviewed changes

Copilot finished work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:30

Copilot AI reviewed Jul 12, 2026

View reviewed changes

Comment thread scripts/__tests__/ci-aggregate-workflow.test.js

test(ci): require every lifecycle step

12dc1b1

Copilot AI review requested due to automatic review settings July 12, 2026 20:40

Copilot started work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:40 View session

Copilot started reviewing on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:40 View session

cursor Bot reviewed Jul 12, 2026

View reviewed changes

Copilot finished work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:41

Copilot AI reviewed Jul 12, 2026

View reviewed changes

Comment thread scripts/__tests__/ci-aggregate-workflow.test.js Outdated

test(ci): scope lifecycle job diagnostics

957e109

Copilot AI review requested due to automatic review settings July 12, 2026 20:50

Copilot started work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:50 View session

Copilot started reviewing on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:50 View session

Copilot finished work on behalf of Eli Pinkerton (wallstop) July 12, 2026 20:51

cursor Bot reviewed Jul 12, 2026

View reviewed changes

Copilot AI reviewed Jul 12, 2026

View reviewed changes

Eli Pinkerton (wallstop) enabled auto-merge (squash) July 12, 2026 20:59

Eli Pinkerton (wallstop) merged commit c5fe595 into master Jul 13, 2026
46 of 54 checks passed

Eli Pinkerton (wallstop) deleted the dev/wallstop/resource-lifecycle branch July 13, 2026 01:51

Uh oh!

Conversation

Eli Pinkerton (wallstop) commented Jul 12, 2026 • edited by cursor Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Verification

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI commented Jul 12, 2026

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

cursor Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI commented Jul 12, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

github-actions Bot commented Jul 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Dispatch throughput (Unity 6000.3.16f1)

Dispatch throughput - Standalone (IL2CPP)

Dispatch throughput - PlayMode (Mono)

Library comparison - throughput (Standalone (IL2CPP))

Library comparison - GC allocations per 10k ops (PlayMode (Mono))

Library comparison - GC allocated bytes per 10k ops (PlayMode (Mono))

Uh oh!

github-actions Bot commented Jul 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

DxMessaging perf deltas vs master baseline (Unity 6000.3.16f1, Standalone IL2CPP)

Uh oh!

Copilot AI commented Jul 12, 2026

Uh oh!

cursor Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

Eli Pinkerton (wallstop) commented Jul 12, 2026

Uh oh!

Copilot AI commented Jul 12, 2026

Uh oh!

cursor Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Eli Pinkerton (wallstop) commented Jul 12, 2026 •

edited by cursor Bot

Loading

github-actions Bot commented Jul 12, 2026 •

edited

Loading

github-actions Bot commented Jul 12, 2026 •

edited

Loading