Add a benchmark lane (`vitest.bench.config.ts`, `pnpm run test:bench`,
gate mode `ci-bench`) and a required `node 24 / benchmarks` CI job that
runs it alone. Benchmarks synthesize their input in-process from fixed
parameters and fail on documented budgets:
- `open-generation.bench.ts`: a 200-turn released-v0 log with 500 text
and 125 reasoning deltas per reply (127,400 events, ~2.8 MB) encoded
through the frozen v0 codec; the migrating first `open()` must finish
within 2,000 ms in a child process capped at 128 MB of old space, and a
fresh process must open the published current generation within 500 ms.
- `conversation-fold.bench.client.ts`: 200 replies whose compact streams
hold 2,000 text + 500 reasoning deltas each, folded through every Chat
Definition by the real assembler; the fold must finish within 150 ms
and stay within 3x the fold of the same window with 100 deltas per
reply.
On this commit both gates fail: the migration exhausts the 128 MB heap
(4.8 s and 696 MB peak RSS without the cap; the pre-stack decode of the
same bytes took 34 ms and 168 MB) and the fold scales 11x with the delta
count. The stacked fixes bring both paths to O(records).
Resolve the 08-08 note conflict (keep the zero-build rewrite, drop the
SQLite busy-journal sentence removed with the SQLite backend) and drop the
now-removed session-persistence-sqlite built-package suite from the
lib-consuming self-skip inventory in the ci.yml comment, the
ci-workflow.spec.ts comment, and the partitioned-coverage note; the
inventory is now image-loadable, transform-corpus, and client-bundle.
client-bundle.client.spec.ts reads packages/client/ui-trajectory/lib/client.js
and skips all three cases when the bundle is absent, so it is a fourth
lib-consuming suite in the instrumented corpus. List it beside built-package
in the ci.yml comment, the ci-workflow.spec.ts comment, and the
partitioned-coverage note (both languages).
Strengthen the ci-workflow.spec.ts guard to match any 'pnpm run build'
spelling (corepack prefix, multi-line run blocks) instead of one exact
string, and complete the lib-consuming self-skip inventory with the
webworker-runtime transform-corpus import sweep alongside the packer
image assertions and the built-package check. Update the ci.yml comment
and the partitioned-coverage note (both languages) to match, and drop the
stale 'post-build' phase wording and the native-Windows build-wait
rationale from the coverage-exempt comment.
The windows-coverage job built the workspace before running the same
ci-coverage gates as Linux, but the instrumented corpus resolves
workspace imports to src through the tsconfig paths map and never
consumes lib/; the two lib-consuming suites (webworker-packer
image-loadable, session-persistence-sqlite built-package) self-skip on
unbuilt checkouts, exactly how the Linux lane already runs them. Remove
the build step so both lanes behave identically, and pin the zero-build
invariant in ci-workflow.spec.ts (red before this change, green after).
Agent Notes updated in place: corrected the build-wait rationale and the
wrong attribution of the packer assertions to the instrumented suite.
pnpm hardlinks node_modules files to the store on the same volume, and
TypeScript's native realpath resolves those links back to store paths
(F:/.pnpm-store/v11/files/...), producing TS6231 during tsc -b and vite
resolution. ReFS block cloning (package-import-method=clone) gives each
file an independent path while sharing physical blocks, avoiding the
leak without the copy cost. Clone mode needs the @reflink/reflink native
module, which the system corepack pnpm carries but pnpm/action-setup's
dest build omits, so installs run through corepack pnpm.
The install steps branch on the workspace filesystem: clone only on
ReFS, plain install on hosted NTFS (which rejects copy-on-write). The
serial-windows store points at F:\.pnpm-store to share the ReFS volume.
Agent Note 2026-08-30-windows-refs-store-block-clone-install records the
rationale; ci-workflow.spec asserts the branch.
Replace Vitest's hash-based --shard with a coordinator-side
longest-processing-time assignment. The coordinator collects the
instrumented inventory from a vitest list run (dropping the exempt
heavy suites that list does not exclude), reads per-file durations from
the Vitest results cache, and seeds heavy subprocess-bound suites into
different partitions. A weight-aware test fails when assignment ignores
recorded weights, verified by injecting a file-count-only rule.
Windows coverage measured partition spread of 442s (275-717s) under
hash sharding; a simulation with the same file durations and the new
assignment balances partitions to within 21s, cutting the critical
partition to roughly half.
Stacked on #3115: keep its windows-* setup-pnpm-js-<run_id>-<run_attempt>-<job>
destination, and extend the same isolation to non-Windows jobs in ci.yml
and ci-master.yml with setup-pnpm-<run_id>-<run_attempt>. This prevents
sequential self-hosted Windows jobs from tripping over a stale locked
pnpm.exe/reflink native module.
The per-case 15-30s budgets on the Windows native and coverage lanes fire
before oxlint, workflow-worker-thread, and other subprocess-spawning cases
finish under the loaded self-hosted pool; the failures rotate across cases
as load shifts, so per-case widening only moved the flake. Raise the lane
defaults (DSH_COVERAGE_TEST_TIMEOUT_MS and the native --testTimeout) to 90s,
align the oxlint and workflow-worker-thread case budgets, and keep the
built-bin SPAWN_TIMEOUT_MS at 60s under a 90s outer budget.
The windows-* jobs now install pnpm under a run/attempt/job-suffixed
destination, so the pre-install step that cleared the old fixed
setup-pnpm-js path no longer touches the actual destination and its
comment claims stale state. The suffix already gives every job a fresh
directory, so remove the four cleanup steps.
The windows-* jobs keep a separate standalone pnpm executable under
runner.temp/setup-pnpm-js. A previous job on the same self-hosted runner
can leave a locked @reflink native module there, so the next job's
pnpm/action-setup fails with EPERM during unlink before any test runs.
Suffix the destination with run_id, run_attempt, and job so every job
gets a fresh directory even when sequential jobs land on the same
runner; apply the same to the python SDK exe build. Update the pnpm
setup isolation note to record the Windows-specific destination.
Other PRs are blocked by Windows ACP half-close tests timing out. Keep the
coverage job running for signal, but remove it from all-checks-passed.needs
until the Windows skip fix is validated.
These tests spawn real ACP child subprocesses. On contended self-hosted
Windows runners the default 30s budget times out. Instead of raising the
global coverage timeout, give this file 2x the configured default
(DSH_COVERAGE_TEST_TIMEOUT_MS) so it follows future default changes.
After the 4-partition split, other PRs' windows coverage now fails on
process-bound subagent-acp tests timing out at 30s under self-hosted
concurrency. Give the coverage lane the same 60s per-test budget that the
earlier failover profile used.
Keep the 4-partition coverage profile, split the monolithic windows-native
job into smaller required jobs (build, coverage, native-tests) plus a
non-blocking observational job. Update ci-workflow.spec for the new topology.
Under high self-hosted concurrency, 8 partitions per Windows native job
triggered vitest fork worker startup timeouts. This branch lowers Windows
coverage to the same 4 partitions Linux uses, trading some single-job
coverage wall time for lower process-creation pressure.
Add node24-win-x64 to the required pull-request and public-release matrices on a native windows-2025 runner, and publish the same win_amd64 artifact from the GitLab tag pipeline. GitHub uses Git Bash for the shared release script while selecting the Windows venv's Scripts/python.exe explicitly; the Linux and macOS legs retain their existing commands and native checks.
Run the complete installed-wheel keyless suite and the trusted two-turn DeepSeek smoke on Windows exactly as on the existing targets. Make the minimal blackbox choose persistent PowerShell on Windows, keep advanced and restart snapshots platform-stable by disabling both one-shot shell variants, locate the generated dsh.exe console command, and validate text lines without assuming POSIX newlines.
Workflow tests pin the four-target matrix, Windows runner and wheel tag, cross-platform venv selection, GitLab publication dependency, and full blackbox invocation. The existing POSIX minimal snapshot changes only its platform-neutral prompt wording; Windows owns a separate model-visible snapshot.
* test(python): exercise installed wheels as black boxes
Add an installed-wheel mode that refuses source/editable imports, repository working directories, mismatched SDK/runtime versions, unpinned runtime dependencies, and executables outside the installed runtime distribution. The mode resolves the wheel-owned executable itself, so callers cannot accidentally prove an explicit checkout artifact.
Add a real-API scenario that drives two tool-using turns through the public synchronous SDK, verifies the file bytes outside the agent, checks completed turn/tool events and persistence, and projects provider failures without retaining credential-bearing error text. The existing deterministic scenario set remains the keyless behavior oracle.
Refs #2952.
* ci(python): require installed-wheel checks on every release target
Move the complete deterministic runtime scenarios behind construction and clean installation of the SDK and matching runtime wheels. Each native leg runs outside the checkout with source-resolution environment variables removed; Linux manylinux smokes assert the same installed provenance.
Expand the required pull-request call from Linux x64 to Linux x64, Linux arm64, and macOS arm64. Trusted heads receive only DEEPSEEK_API_KEY_EXTERNAL for a fail-loud live two-turn smoke on each carrier, while fork and Dependabot heads retain the full keyless path without exposing secrets.
Pin the reusable secret declaration, matrix call, aggregate dependency, untrusted-head condition, and live/keyless commands in the workflow contract test.
Refs #2952.
* docs(testing): make installed wheels the Python CI authority
Record the clean-wheel provenance boundary, complete keyless scenario set, trusted real-API contract, secret handling, and three-target required topology in a new implemented testing decision. Update the SEA distribution and portable-CI authorities plus the Python contributor reference to describe the same current state.
Archive the fully superseded Linux-x64-only decision after consolidating its rationale and alternatives into the new owner. Preserve its bilingual triplet as a sealed historical snapshot and redirect every active current-state reference.
Refs #2952.
Address the second review pass on PR #2768:
- Restore the serial-linux-selfhosted drill comment in ci-master.yml (it was
lost in the split boundary, leaving serial-linux-selfhosted as the only
comment-less job); adapted to the current-state fact that this workflow never
listens to pull_request while keeping the no-cache-steps reason, the
DSH_CI_FAILOVER_LINUX switch guidance, and push-triggered semantics.
- Update 2026-07-29-pnpm-setup-runner-isolation (en/zh + i18n): the regression
test and setup steps now span ci.yml and ci-master.yml.
- failover-runbook line 19: the push-reachable classification now names
ci-master.yml for the drills/wine cache and ci.yml for the pull-request jobs,
dropping the stale pull-request-gated phrasing.
- all-checks-passed comment in ci.yml notes that needs cannot reach across
workflow files (ci-master job are deliberately outside this PR verdict).
Verification: ci-workflow.spec.ts 13/13, both workflows YAML-parse with correct
concurrency, 3 translation pairs consistent, note-format 575.
Per review on PR #2768:
- Restore a concurrency block in ci.yml (cancel-in-progress: true) so a fresh
PR push cancels the superseded run; GitHub has no default PR auto-cancel,
so removing it would stack a second full 9-job run on every push.
Update the ci-workflow.spec assertion accordingly.
- Fix comment misplacements from the split boundary: ci.yml's all-checks-passed
now carries its own branch-protection comment (the drill comment that was
orphaned above it is gone), and ci-master.yml no longer ends with an orphaned
all-checks-passed comment.
- Restore the DeepSeek e2e workflow describe-block in ci-workflow.spec.ts that
the rewrite had silently deleted (e2e.yml is unchanged).
- serial-windows comment no longer claims non-blocking-for-PR/absent-from-needs
since ci-master never listens to pull_request.
- DSH_TELEMETRY_DISABLED safe-use comment restored in ci-master.yml env; split
rationale (mi gray segments in PR check) documented in .github/AGENTS.md.
Verification: ci-workflow.spec.ts 13/13, both workflows YAML-parse with correct
concurrency, git diff --check clean.
Split the single ci.yml into two workflows so the PR check panel stops listing
master-only obs jobs that skip (gray) and block the aggregate green-check:
- ci.yml is now pull_request-only, holding only the 9 PR jobs (node-24,
node-24-coverage, node-24-consumers, node-compat, python-sdk, python-runtime,
windows, windows-native, all-checks-passed). It drops the workflow-level
concurrency block since master carve-outs no longer apply.
- ci-master.yml (new) carries the six master/dispatch jobs (wine-apt-cache,
serial-linux-selfhosted, serial-macos, serial-windows, and both runner
benchmarks) with the push-exempt cancel-in-progress block and suite input.
It does not listen to pull_request, so its jobs never appear in PR checks.
ci.yml keeps the wine apt cache seed/restore (repo-scoped by key), so moving the
producer to ci-master.yml does not break the PR windows job's restore.
Update ci-workflow.spec.ts to assert the two-workflow split: ci-master owns the
drills/benchmarks and the push-exemption concurrency, ci.yml is PR-only without
concurrency. Update the serial-reference, portable-required, failover-runbook,
and pnpm-caching notes (en/zh + i18n) and .github/AGENTS.md to reflect ci-master.
Verification: scripts/ci-workflow.spec.ts 12/12, both workflows YAML-parse,
verify-translation-pairing consistent, verify-agent-note-format passes.
Address the review findings from the fresh pass on PR #2744:
- 2026-07-21 note L31: the enabled serial references run on the self-hosted
vm-backup/dsh-win-ci pools and the only remaining disabled hosted serial is
serial-macos (macos-latest); removed the stale hosted ubuntu-latest/windows-2025
serial framing and the outdated 'when enabled, serial / windows' clause.
- Drop remaining change-narrative from both notes: L19 (serial / linux, macos)
and the 2026-07-26 caching note L16/L34 now state only current facts without
deletion dates/PR numbers; the no-producer fact has one home (L16). zh.ms'
'直至其过期为逐出' corrected to '直至其被逐出'. Bilingual hashes re-recorded.
- ci.yml TODO notes that re-enabling serial-macos does not restore a Linux
hosted-cache producer and records the seeder-vs-remove decision direction.
- The Playwright restore's failover-skip comment is now self-contained (the
VM's persistent browser cache is warm) instead of pointing at the coveragelane
rationale, which is pnpm-store-specific.
Verification: scripts/ci-workflow.spec.ts passes (12/12), YAML re-parses,
both translation pairs consistent, git diff --check clean.
Apply review feedback on the serial-linux removal (PR #2744):
ci.yml:
- Rewrite the new comments as current-state statements, not change
narrative (dsh-prose-standard): the TODO names serial-macos as the one
remaining disabled hosted serial job; serial-macos's intro and the
self-hosted standby's frozen-archive note no longer narrate the deletion.
- The self-hosted standby's frozen-archive comment states its own reason
(full history to resolve DSH_ARCHIVE_BASE_REF against github.event.before)
instead of referenceing a now-nonexistent hosted serial reference.
- Move the hosted-cache comment above the pnpm restore so it covers both
restore-keys fallback steps, and describe the real consequence (matches the
archived entry until evict, then cold) instead of the false
'cold on a lockfile change'. Restore the per-step failover-skip note.
Agent Note 2026-07-26-pnpm-action-setup-for-symmetric-ci-caching:
- Update the restore-only bullet and the consequences closing line (it
described serial-linux as the active master-push producer of the pnpm store
cache) to state that no master job produces these hosted keys since the
producer was removed; the Problem and Alternatives sections are historical
context and are left unchanged. Re-record the bilingual pair hashes.
Verification: scripts/ci-workflow.spec.ts passes (12/12), YAML re-parses,
both translation pairs consistent, git diff --check clean.
serial-linux (hosted ubuntu-latest) has been `if: false` since 2026-07-30 and
never runs. Remove the dead job block and retire the dangling references:
- TODO(hosted-serial-ci) narrowed to the single remaining disabled hosted
serial job (serial-macos); the hosted linux definition is gone.
- The cache producer comment no longer claims serial-linux refreshes the
hosted pnpm/Playwright caches; there is currently no active master producer
for them, so restores are cold on a lockfile change.
- The self-hosted standby's frozen-archive comment no longer cross-references
the deleted job.
- serial-macos gains its own intro comment since the shared 'hosted reference
jobs below are disabled' lede was removed.
No runner allocation, required gate, or all-checks-passed.needs reference this
job; the aggregate is unchanged.
The windows wine-blocking job installs Wine with apt-get over the local
.deb archive, but apt re-downloads the full 100+ MB closure from the
Ubuntu mirror anyway. A degraded runner network stalled that transfer
past the job's 15-minute budget and cancelled the check. Install the
restored archive directly with dpkg (no repository access) and keep the
apt network install as the fallback when the archive cannot satisfy the
closure.
Tianyi Cui