kevin/deepseek-harness
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

ci: collapse cold setup waves

Browse Source
This commit is contained in:
Tianyi Cui
2026-07-22 15:44:11 +08:00
parent cc22492ddd
commit 0f57eafc36
3 changed files with 71 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/ci.yml
+45 -48
View File
@@ -319,7 +319,7 @@ jobs:
run: ${{ matrix.command }}
# One large runner pays hosted setup once, then the repository scheduler
# overlaps the complete unsharded primary gate inventory inside that box.
# overlaps the unsharded primary inventory except the production site build.
node-24:
if: github.event_name != 'workflow_dispatch' || inputs.suite == 'optimized-larger-runners'
runs-on: dsh-ubuntu-24-04-96core
@@ -383,6 +383,38 @@ jobs:
- name: Run complete unsharded primary Node CI concurrently
run: pnpm run check:ci:large-runner
# Keep only the longest independent Linux gate on a second coarse-grained
# runner so cold install variance does not push the primary box over a minute.
node-24-site:
if: github.event_name != 'workflow_dispatch' || inputs.suite == 'optimized-larger-runners'
runs-on: dsh-ubuntu-24-04-16core
name: node 24 / production site
steps:
- uses: actions/checkout@v6
- uses: actions/setup-node@v6
with:
node-version: ${{ env.PRIMARY_NODE_VERSION }}
- name: Enable corepack and resolve pnpm store path
id: pnpm-store
run: |
corepack enable
echo "path=$(pnpm store path --silent)" >> "$GITHUB_OUTPUT"
- uses: actions/cache@v4
with:
path: ${{ steps.pnpm-store.outputs.path }}
key: ${{ runner.os }}-node-${{ env.PRIMARY_NODE_VERSION }}-pnpm-${{ hashFiles('pnpm-lock.yaml') }}
restore-keys: |
${{ runner.os }}-node-${{ env.PRIMARY_NODE_VERSION }}-pnpm-
- name: Install (immutable)
run: pnpm install --frozen-lockfile
- name: Build documentation site (production SPA)
run: pnpm run docs:build
node-compat:
if: github.event_name != 'workflow_dispatch' || inputs.suite == 'optimized-larger-runners'
runs-on: ubuntu-latest
@@ -583,49 +615,13 @@ jobs:
shell: pwsh
run: ${{ matrix.command }}
# The two blocking Windows build surfaces share one setup and run concurrently.
windows-build:
# One Windows box shares setup across the required build/site checks and the
# complete observational portability inventory. run-gates reports failures
# from observational gates without allowing them to fail the required job.
windows:
if: github.event_name != 'workflow_dispatch' || inputs.suite == 'optimized-larger-runners'
runs-on: dsh-windows-2025-16core
name: windows / blocking builds
steps:
- uses: actions/checkout@v6
- uses: actions/setup-node@v6
with:
node-version: ${{ env.PRIMARY_NODE_VERSION }}
- name: Enable corepack and resolve pnpm store path
id: pnpm-store
shell: pwsh
run: |
corepack enable
"path=$(pnpm store path --silent)" >> $env:GITHUB_OUTPUT
- uses: actions/cache@v4
with:
path: ${{ steps.pnpm-store.outputs.path }}
key: ${{ runner.os }}-node-${{ env.PRIMARY_NODE_VERSION }}-pnpm-${{ hashFiles('pnpm-lock.yaml') }}
restore-keys: |
${{ runner.os }}-node-${{ env.PRIMARY_NODE_VERSION }}-pnpm-
- name: Install (immutable)
shell: pwsh
run: pnpm install --frozen-lockfile
- name: Run blocking Windows builds concurrently
shell: pwsh
env:
DSH_GATE_CONCURRENCY: '2'
run: pnpm run check:ci:windows-blocking
# Wider Windows portability gates remain observational, but one larger box
# now runs their complete unsharded inventory with in-runner concurrency.
windows-gates:
if: github.event_name != 'workflow_dispatch' || inputs.suite == 'optimized-larger-runners'
continue-on-error: true
runs-on: dsh-windows-2025-32core
name: windows node 24 / observational
name: windows node 24 / complete
env:
# Keep ESLint itself single-threaded: 16 ESLint workers took 174 seconds on
# this image. The outer scheduler still overlaps lint with the other gates.
@@ -670,9 +666,9 @@ jobs:
shell: pwsh
run: pnpm install --frozen-lockfile
- name: Run observational Windows gates concurrently
- name: Run blocking and observational Windows gates concurrently
shell: pwsh
run: pnpm run check:ci:windows-observational
run: pnpm run check:ci:windows-complete
# Manual reference runs deliberately avoid the optimized jobs above.
# Each host executes the complete, unsharded primary Node aggregate with one
@@ -1011,7 +1007,7 @@ jobs:
DSH_GATE_CONCURRENCY: ${{ matrix.workers }}
DSH_PUBLINT_CONCURRENCY: ${{ matrix.workers }}
DSH_SNAPSHOT_MAX_CONCURRENCY: ${{ matrix.workers }}
run: pnpm run check:ci:large-runner
run: pnpm run check:ci
- name: Run blocking Windows builds concurrently
if: matrix.platform == 'windows'
@@ -1023,8 +1019,9 @@ jobs:
# Single stable required check for branch protection: require "all checks
# passed" instead of enumerating matrix legs whose names change as lanes and
# node versions evolve. Every blocking job in THIS workflow must be listed in
# `needs`; explicitly observational jobs such as windows-gates stay out
# (`needs` cannot reach across workflow files; e2e.yml stays its own check).
# `needs`; observational Windows gates share the required Windows job but are
# marked non-blocking inside run-gates. (`needs` cannot reach across workflow
# files; e2e.yml stays its own check.)
# `if: always()` is load-bearing: without it a failed dependency
# would SKIP this job, and GitHub counts a skipped required check as passing
# — so this job always runs and fails on any non-success result, including
@@ -1032,7 +1029,7 @@ jobs:
all-checks-passed:
name: all checks passed
runs-on: ubuntu-latest
needs: [node-24, node-compat, python-sdk, windows-build]
needs: [node-24, node-24-site, node-compat, python-sdk, windows]
if: always() && (github.event_name != 'workflow_dispatch' || inputs.suite == 'optimized-larger-runners')
steps:
- name: Fail if any needed job did not succeed