[CICD] Merge multiple chip pipelines into a single one

.github/configs/metax.yml

@@ -1,24 +1,24 @@
-# MetaX C500 Hardware Configuration
-# This file defines CI/CD settings for MetaX C500 GPU (MACA) testing
+# MetaX C550 Hardware Configuration
+# This file defines CI/CD settings for MetaX C550 GPU (MACA) testing
 # Test configurations are defined in tests/test_utils/config/platforms/metax.yaml
 
 hardware_name: metax
-display_name: "MetaX C500 Tests"
+display_name: "MetaX C550 Tests"
 
-# Docker images for MetaX C500 (MACA-based)
+# Docker images for MetaX C550 (MACA-based)
 ci_image: harbor.baai.ac.cn/flagscale/megatron-lm-with-te:202603231839
 ci_train_image: harbor.baai.ac.cn/flagscale/megatron-lm-with-te:202603231839
 ci_inference_image: harbor.baai.ac.cn/flagscale/megatron-lm-with-te:202603231839
 
-# Runner labels for MetaX C500 hardware
+# Runner labels for MetaX C550 hardware
 runner_labels: ["flagscale-metax-c550-gpu2-8c-256g"]
 
 # Container volumes (hardware-specific paths)
 container_volumes:
-  - /mnt/op/cicd/baai_datasets:/home/gitlab-runner/data
-  - /mnt/op/cicd/baai_tokenizers:/home/gitlab-runner/tokenizers
+  - /mnt/airs-business/cicd/baai_datasets:/home/gitlab-runner/data
+  - /mnt/airs-business/cicd/baai_tokenizers:/home/gitlab-runner/tokenizers
 
-# Container options for MetaX C500 (MACA runtime)
+# Container options for MetaX C550 (MACA runtime)
 container_options: >-
   --hostname=flagscale-ci
   --ipc=host

.github/workflows/all_tests.yml

@@ -0,0 +1,129 @@
+name: all_tests
+
+on:
+  push:
+    branches: ["main"]
+    paths-ignore:
+      - 'docs/**'
+      - 'docker/cuda/**'
+      - 'docker/build.sh'
+      - 'tools/install/**'
+      - 'requirements/**'
+      - '.github/workflows/build_image_cuda.yml'
+      - '.github/workflows/push_image_harbor.yml'
+  pull_request:
+    branches: ["main"]
+    paths-ignore:
+      - 'docs/**'
+      - 'docker/cuda/**'
+      - 'docker/build.sh'
+      - 'tools/install/**'
+      - 'requirements/**'
+      - '.github/workflows/build_image_cuda.yml'
+      - '.github/workflows/push_image_harbor.yml'
+  workflow_dispatch:
+    inputs:
+      platform:
+        description: Platform test suite to run
+        required: true
+        default: all
+        type: choice
+        options:
+          - all
+          - cuda
+          - ascend
+          - metax
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-${{ github.actor }}
+  cancel-in-progress: true
+
+jobs:
+  # ---------------------------------------------------------------------------
+  # Guard: skip CUDA tests when PR contains docker-related changes, because
+  # build_image_cuda.yml will handle build + test for those PRs.
+  # paths-ignore alone cannot guarantee mutual exclusivity when a PR touches
+  # both docker files and non-docker files, so we need this job-level check.
+  # ---------------------------------------------------------------------------
+  check_docker_changes:
+    runs-on: ubuntu-latest
+    if: github.event_name == 'pull_request'
+    outputs:
+      has_docker_changes: ${{ steps.check.outputs.has_docker_changes }}
+    steps:
+      - name: Check for docker-related file changes
+        id: check
+        env:
+          GH_TOKEN: ${{ github.token }}
+        run: |
+          CHANGED=$(gh api repos/${{ github.repository }}/pulls/${{ github.event.pull_request.number }}/files --paginate --jq '.[].filename')
+          if echo "$CHANGED" | grep -qE '^(docker/cuda/|docker/build\.sh$|tools/install/|requirements/|\.github/workflows/build_image_cuda\.yml$)'; then
+            echo "has_docker_changes=true" >> $GITHUB_OUTPUT
+          else
+            echo "has_docker_changes=false" >> $GITHUB_OUTPUT
+          fi
+
+  cuda_tests:
+    name: CUDA tests
+    needs: check_docker_changes
+    if: >-
+      (github.event_name != 'workflow_dispatch' || inputs.platform == 'all' || inputs.platform == 'cuda') &&
+      (github.event_name == 'push' || github.event_name == 'workflow_dispatch' || needs.check_docker_changes.outputs.has_docker_changes != 'true')
+    uses: ./.github/workflows/all_tests_common.yml
+    with:
+      platform: cuda
+
+  # Disable util the new devices ready
+  # ascend_tests:
+  #   name: Ascend tests
+  #   if: github.event_name != 'workflow_dispatch' || inputs.platform == 'all' || inputs.platform == 'ascend'
+  #   uses: ./.github/workflows/all_tests_common.yml
+  #   with:
+  #     platform: ascend
+
+  metax_tests:
+    name: MetaX tests
+    if: github.event_name != 'workflow_dispatch' || inputs.platform == 'all' || inputs.platform == 'metax'
+    uses: ./.github/workflows/all_tests_common.yml
+    with:
+      platform: metax
+
+  all_tests:
+    name: Verify all platform tests
+    needs:
+      - check_docker_changes
+      # - cuda_tests
+      # - ascend_tests
+      - metax_tests
+    runs-on: ubuntu-latest
+    if: always()
+    steps:
+      - name: Verify workflow status
+        run: |
+          failed=false
+
+          # if [ "${{ needs.check_docker_changes.outputs.has_docker_changes }}" = "true" ]; then
+          #   echo "⏭️ CUDA tests skipped - docker changes detected, handled by build_image_cuda workflow"
+          # elif [ "${{ needs.cuda_tests.result }}" != "success" ] && \
+          #      [ "${{ needs.cuda_tests.result }}" != "skipped" ]; then
+          #   echo "❌ CUDA tests failed"
+          #   failed=true
+          # fi
+
+          # if [ "${{ needs.ascend_tests.result }}" != "success" ] && \
+          #    [ "${{ needs.ascend_tests.result }}" != "skipped" ]; then
+          #   echo "❌ Ascend tests failed"
+          #   failed=true
+          # fi
+
+          if [ "${{ needs.metax_tests.result }}" != "success" ] && \
+             [ "${{ needs.metax_tests.result }}" != "skipped" ]; then
+            echo "❌ MetaX C550 tests failed"
+            failed=true
+          fi
+
+          if [ "$failed" = "true" ]; then
+            exit 1
+          fi
+
+          echo "✅ All requested platform tests passed!"

.github/workflows/all_tests_common.yml

@@ -55,11 +55,13 @@ jobs:
       env_name_serve: ${{ steps.config.outputs.env_name_serve }}
       env_name_rl: ${{ steps.config.outputs.env_name_rl }}
     steps:
-      - name: Checkout code
+      - name: Checkout code (attempt 1)
+        id: checkout_attempt_1
         uses: actions/checkout@v4
+        continue-on-error: true
         with:
-          repository: ${{ github.event.pull_request.head.repo.full_name }}
-          ref: ${{ github.event.pull_request.head.ref }}
+          repository: ${{ github.event.pull_request.head.repo.full_name || github.repository }}
+          ref: ${{ github.event.pull_request.head.ref || github.ref }}
           ssh-strict: true
           ssh-user: git
           persist-credentials: false
@@ -71,28 +73,42 @@ jobs:
           submodules: false
           set-safe-directory: true
 
-      - name: Cache source code
-        id: cache-source
-        uses: actions/cache@v4
+      - name: Checkout code (attempt 2)
+        if: steps.checkout_attempt_1.outcome == 'failure'
+        id: checkout_attempt_2
+        uses: actions/checkout@v4
+        continue-on-error: true
         with:
-          path: .
-          key: source-code-${{ github.sha }}
-
-      - name: Create source code tarball for distribution
-        run: |
-          tar -czf /tmp/flagscale-source.tar.gz \
-            --exclude='.git' \
-            --exclude='*.pyc' \
-            --exclude='__pycache__' \
-            .
+          repository: ${{ github.event.pull_request.head.repo.full_name || github.repository }}
+          ref: ${{ github.event.pull_request.head.ref || github.ref }}
+          ssh-strict: true
+          ssh-user: git
+          persist-credentials: false
+          clean: true
+          sparse-checkout-cone-mode: true
+          fetch-tags: false
+          show-progress: true
+          lfs: false
+          submodules: false
+          set-safe-directory: true
 
-      - name: Upload source code artifact
-        uses: actions/upload-artifact@v4
+      - name: Checkout code (attempt 3)
+        if: steps.checkout_attempt_2.outcome == 'failure'
+        id: checkout_attempt_3
+        uses: actions/checkout@v4
         with:
-          name: flagscale-source-${{ github.sha }}
-          path: /tmp/flagscale-source.tar.gz
-          retention-days: 1
-          compression-level: 0
+          repository: ${{ github.event.pull_request.head.repo.full_name || github.repository }}
+          ref: ${{ github.event.pull_request.head.ref || github.ref }}
+          ssh-strict: true
+          ssh-user: git
+          persist-credentials: false
+          clean: true
+          sparse-checkout-cone-mode: true
+          fetch-tags: false
+          show-progress: true
+          lfs: false
+          submodules: false
+          set-safe-directory: true
 
       - name: Load platform configuration
         id: config
@@ -149,7 +165,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -172,7 +187,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -191,7 +205,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -210,7 +223,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -229,7 +241,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_inference }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -248,7 +259,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_serve }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -267,7 +277,6 @@ jobs:
   #     runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
   #     container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
   #     container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-  #     source_artifact: flagscale-source-${{ github.sha }}
   #     pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
   #     env_name: ${{ needs.checkout_and_config.outputs.env_name_rl }}
   #     env_path: ${{ needs.checkout_and_config.outputs.env_path }}
@@ -286,7 +295,6 @@ jobs:
       runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
       container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
       container_options: ${{ needs.checkout_and_config.outputs.container_options }}
-      source_artifact: flagscale-source-${{ github.sha }}
       pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
       env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
       env_path: ${{ needs.checkout_and_config.outputs.env_path }}