[CICD] Add MetaX image build workflow

.github/configs/metax.yml

@@ -1,24 +1,27 @@
-# 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: ["flagscale-metax-c550-gpu2-8c-256g"]
+# Directory to store image tar files (shared between build and test workflows)
+tar_dir: /mnt/op/cicd/image_tar
+
+# 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
@@ -41,14 +44,14 @@ container_options: >-
 # =============================================================================
 # Package Manager Configuration
 # =============================================================================
-pkg_mgr: "conda"
+pkg_mgr: 'conda'
 
 # Environment path (conda installation path)
-env_path: "/opt/conda"
+env_path: '/opt/conda'
 
 # Conda environment names
 env_names:
-  train: "base"
-  hetero_train: ""
-  inference: ""
-  rl: ""
+  train: 'base'
+  hetero_train: ''
+  inference: ''
+  rl: ''

.github/workflows/all_tests_common.yml

@@ -58,8 +58,8 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v4
         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

.github/workflows/all_tests_metax.yml

@@ -6,20 +6,24 @@ on:
     paths-ignore:
       - 'docs/**'
       - 'docker/cuda/**'
+      - 'docker/metax/**'
       - 'docker/build.sh'
       - 'tools/install/**'
       - 'requirements/**'
       - '.github/workflows/build_image_cuda.yml'
+      - '.github/workflows/build_image_metax.yml'
       - '.github/workflows/push_image_harbor.yml'
   pull_request:
     branches: ["main"]
     paths-ignore:
       - 'docs/**'
       - 'docker/cuda/**'
+      - 'docker/metax/**'
       - 'docker/build.sh'
       - 'tools/install/**'
       - 'requirements/**'
       - '.github/workflows/build_image_cuda.yml'
+      - '.github/workflows/build_image_metax.yml'
       - '.github/workflows/push_image_harbor.yml'
 
 concurrency:

.github/workflows/build_image_cuda.yml

@@ -49,7 +49,7 @@ on:
         required: false
         type: string
         description: Override runs_on. Falls back to platform config if not set.
-        default: '["self-hosted", "Linux", "X64", "nvidia-0", "gpus-8"]'
+        default: '["flagscale-nvidia-a100-gpu2-32c-128g"]'
       container_volumes:
         required: false
         type: string
@@ -151,7 +151,7 @@ jobs:
   build:
     name: Build ${{ matrix.task }}
     needs: prepare
-    runs-on: [self-hosted, Linux, X64, nvidia-0, gpus-8]
+    runs-on: ${{ fromJson(inputs.runs_on) || '["flagscale-nvidia-a100-gpu2-32c-128g"]' }}
     env:
       HTTP_PROXY: ${{ vars.HTTP_PROXY }}
       HTTPS_PROXY: ${{ vars.HTTPS_PROXY }}
@@ -316,7 +316,7 @@ jobs:
   load_images:
     name: Load and push images
     needs: ['build', 'summary']
-    runs-on: [self-hosted, Linux, X64, nvidia-0, gpus-8]
+    runs-on: ${{ fromJson(inputs.runs_on) || '["flagscale-nvidia-a100-gpu2-32c-128g"]' }}
     steps:
       - name: Load train image from tar and push
         run: |
@@ -354,8 +354,7 @@ jobs:
       ci_train_image: ${{ needs.build.outputs.train_tag }}
       ci_inference_image: ${{ needs.build.outputs.inference_tag }}
       runs_on: >-
-        ${{ inputs.runs_on ||
-        '["self-hosted", "Linux", "X64", "nvidia-0", "gpus-8"]' }}
+        ${{ fromJson(inputs.runs_on) || '["flagscale-nvidia-a100-gpu2-32c-128g"]' }}
       container_volumes: >-
         ${{ inputs.container_volumes ||
         '["/home/flagscale_cicd/docker/docker_build/docker_data:/home/gitlab-runner/data",