[CI] Add cross-job fast-fail health check (Layer 3) (#21341)

.github/actions/check-stage-health/action.yml

@@ -0,0 +1,42 @@
+name: Check Stage Health
+description: Fail fast if any job in the current workflow run has already failed. Auto-skips for scheduled runs.
+
+inputs:
+  github-token:
+    description: 'GitHub token for API calls'
+    required: false
+    default: ${{ github.token }}
+
+runs:
+  using: composite
+  steps:
+    - name: Check stage health
+      uses: actions/github-script@v7
+      with:
+        github-token: ${{ inputs.github-token }}
+        script: |
+          // Skip for scheduled runs — they should collect all failures, not fast-fail
+          if (context.eventName === 'schedule') {
+            core.info('Skipping health check for scheduled run');
+            return;
+          }
+
+          const jobs = await github.paginate(github.rest.actions.listJobsForWorkflowRun, {
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            run_id: context.runId,
+            per_page: 100,
+          });
+          // Find jobs that failed from a real error, not from fast-fail cascade
+          const rootCauseFailures = jobs.filter(j => {
+            if (j.status !== 'completed' || j.conclusion !== 'failure') return false;
+            // If the failing step is the health check, it's a cascade — skip it
+            const failedStep = (j.steps || []).find(s => s.conclusion === 'failure');
+            if (failedStep && (failedStep.name.includes('check-stage-health') || failedStep.name.includes('Check stage health'))) {
+              return false;
+            }
+            return true;
+          });
+          if (rootCauseFailures.length > 0) {
+            core.setFailed(`Fast-fail: skipping — root cause job(s): ${rootCauseFailures.map(j => j.name).join(', ')}`);
+          }

.github/workflows/pr-test-jit-kernel.yml

@@ -44,6 +44,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -71,6 +74,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}

.github/workflows/pr-test-multimodal-gen.yml

@@ -62,6 +62,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -116,6 +119,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -167,6 +173,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}

.github/workflows/pr-test-sgl-kernel.yml

@@ -34,6 +34,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -69,6 +72,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -104,6 +110,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -151,6 +160,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}

.github/workflows/pr-test.yml

@@ -561,6 +561,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -614,6 +617,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -668,6 +674,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -729,6 +738,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -783,6 +795,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -839,6 +854,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -923,6 +941,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -977,6 +998,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -1047,6 +1071,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -1097,6 +1124,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -1157,6 +1187,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}
@@ -1223,6 +1256,9 @@ jobs:
         with:
           ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
 
+      - uses: ./.github/actions/check-stage-health
+        id: stage-health
+
       - uses: ./.github/actions/check-maintenance
         with:
           github-token: ${{ github.token }}