Skip to content

test: retain combined performance interaction regressions - #512

Merged
raiseCatError merged 1 commit into
devfrom
test/20261027-performance-interactions
Oct 27, 2026
Merged

raiseCatError merged 1 commit into
devfrom
test/20261027-performance-interactions

Conversation

@raiseCatError

Copy link
Copy Markdown
Owner

The individually verified performance changes also need regression coverage for their interactions. Import the previously reviewed performanceIntegration.test.ts unchanged from 3173058d: output bursts interleaved with wrapped Unicode editing, scrolling and resizing; last-painted pointer/copy targets and exact clipboard payload; interrupt/stop timer cancellation and command/output order.

This PR changes tests and focused documentation only. Runtime, dependency, script, CI and test content matches the validated combined snapshot bd1d7750; no runtime feature, dependency, assertion relaxation or timeout change. The report records the four approved implementation merges and their exact passing dev push CI runs; the QA guide keeps Ghostty/Zed physical checks NOT RUN. It retains the measured +6.7% history typing p95 and +2.3% deletion p95 differences without causal attribution.

Validation at exact head 48d2530d0ca3c394693b06cd567bd5568f83f1c8: build and focused regression suite pass; full canonical npm run verify passes with two workers, short private T7 fixtures and explicit truecolor/xterm capabilities. Existing skips and Arcade TODO remain unchanged. The new test file enters canonical discovery/sharding automatically. No user clipboard or existing service session was modified.

Implementation PRs were merged sequentially into dev as #511, #506, #508, #510, each after source preflight and followed by its exact comprehensive push CI. After this PR's exact-head checks pass, the approved workflow permits merging it and requires final dev push CI including Fedora/timing. Final merged-source benchmarks and issue handoffs follow; no master/release/tag/deploy/Homebrew action is authorized.

Refs #505
Refs #507
Refs #509

Copy link
Copy Markdown
Owner Author

Integration stopped under the user's real-failed-gate rule. Exact-head CI 40504558737 did not pass: Node22 compatibility was cancelled after exceeding the existing 5m0s job limit. GitHub's failure annotation states that exact timeout. Shell setup ran 12:32:48–12:37:25 UTC (~4m37s); npm ci/build succeeded, then the test step ran 12:37:33–12:37:54 before cancellation. All six macOS/Ubuntu shards and Quality passed. Aggregate CI correctly failed because Node22 was not successful.

Head remains 48d2530d0ca3c394693b06cd567bd5568f83f1c8; PR remains open and unmerged. Fresh full local canonical verification passed at this exact head. No automatic retry, test weakening, timeout change, force-push or further code push was performed after the failed gate.

The four implementation merges are complete and each had green exact dev push CI: #511 fa3091d705d31255081172665801450a0d4c8890 / 40477837690; #506 fe40fd52fb68f0db81b3d35011d15e10e49119c4 / 40481488888; #508 acc61dad2f8ccea1d48e9b6ad7fa718ec6a431de / 40482503313; #510/current dev 317f2e80d5b1ce81d775657c7e5922acb75c94ee / 40483855963. Those push runs include Fedora and timing jobs.

Final test-PR merge/push verification and fresh final-dev benchmarks remain pending. Physical Ghostty/Zed QA stays NOT RUN; existing service sessions remain untouched. Prior measured +6.7% history-typing p95 and +2.3% deletion p95 differences remain reported without causal attribution. Cancellation logs/annotations and the integration checkpoint are retained at /Volumes/T7Tmp/ai-dev/nmsh-merge-20261027/.

@raiseCatError

Copy link
Copy Markdown
Owner Author

Final integration gate update: #512 merged as c690de055daeb7b33ed8b5e302ef01133b3e55b4 after the single authorized unchanged PR rerun passed, including Node22 and aggregate verification. Approved head remained 48d2530d0ca3c394693b06cd567bd5568f83f1c8; merge tree/parents matched fresh local simulation.

The resulting exact dev push CI40512364215 FAILED macOS2/3: existing hostProfiles.test.ts testcase “reattaching across profiles updates input modes while keeping the real shell environment” timed out waiting for the completed journal for echo SIZE-$(stty size) at line116 / liveFrontend.ts:229. Zero cancellations; process leak check passed. The test/helper are unchanged from originaldev. Root cause is unconfirmed; this is a real failed gate, separate from the earlier Node22 setup-budget timeout. No automatic rerun, fix push or rollback followed it.

All other five platform shards, Node22, Fedora, Quality and both timing jobs passed. Fresh merged-source local fullscreen/interaction tests also passed, but do not override the failed gate. Prior full canonical verification passed on the identical PR512 tree.

Final merged-source benchmarks are NOT RUN because the required push gate failed first. Earlier matched measurements remain in the performance report, including +6.7% history-typing p95 / +2.3% deletion p95 fixed-versus-pre-fix differences without causal attribution. No new final performance result is claimed.

Physical Ghostty/Zed QA remains NOT RUN; keep this work open / Needs Human Test. Use the merged physical QA guide. Existing user services/sessions remain alive. Next step: investigate the preserved macOS cross-profile attach/journal failure before resuming final validation. Evidence and regression sources remain under /Volumes/T7Tmp/ai-dev/nmsh-merge-20261027; no master/release/tag/deployment/Homebrew action occurred.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant