.gitlab-ci.yml

workflow:
  rules:
    - if: $CI_PROJECT_NAMESPACE != "ADLR"
      when: never
    - if: $CI_COMMIT_BRANCH =~ /ci-/ && $CI_PIPELINE_SOURCE != "schedule"
      when: never
    - if: $CI_PIPELINE_SOURCE == "schedule"
      auto_cancel:
        on_new_commit: none
    - if: $CI_PIPELINE_SOURCE == "web"
    - if: $CI_COMMIT_REF_PROTECTED == "true"
      variables:
        FUNCTIONAL_TEST: "no"
    - if: $CI_MERGE_REQUEST_LABELS =~ /Run tests/ && $CI_MERGE_REQUEST_TARGET_BRANCH_SHA != ""
      variables:
        UNIT_TEST_REPEAT: 5
        UNIT_TEST_TIMEOUT: 50
        FUNCTIONAL_TEST: "yes"
        FUNCTIONAL_TEST_SCOPE: mr
        FUNCTIONAL_TEST_CLUSTER_A100: ""
        FUNCTIONAL_TEST_CLUSTER_H100: ""
    - if: $CI_MERGE_REQUEST_LABELS =~ /Run nightly/ && $CI_MERGE_REQUEST_TARGET_BRANCH_SHA != ""
      variables:
        UNIT_TEST_REPEAT: 5
        UNIT_TEST_TIMEOUT: 50
        FUNCTIONAL_TEST: "yes"
        FUNCTIONAL_TEST_SCOPE: nightly
        FUNCTIONAL_TEST_CLUSTER_A100: ""
        FUNCTIONAL_TEST_CLUSTER_H100: ""
    - if: $CI_MERGE_REQUEST_LABELS =~ /Run weekly/ && $CI_MERGE_REQUEST_TARGET_BRANCH_SHA != ""
      variables:
        UNIT_TEST_REPEAT: 5
        UNIT_TEST_TIMEOUT: 50
        FUNCTIONAL_TEST: "yes"
        FUNCTIONAL_TEST_SCOPE: weekly
        FUNCTIONAL_TEST_CLUSTER_A100: ""
        FUNCTIONAL_TEST_CLUSTER_H100: ""
    - if: $CI_PIPELINE_SOURCE == "merge_request_event" && $CI_MERGE_REQUEST_TARGET_BRANCH_SHA != ""
      variables:
        FUNCTIONAL_TEST: "no"
    - when: never
  auto_cancel:
    on_new_commit: interruptible

stages:
  - test 
  - functional_tests
  - convergence_tests
  - publish

default:
  interruptible: true

variables:
  FUNCTIONAL_TEST: 
    value: "yes"
    options:
      - "yes"
      - "no"
    description: To run the funtional test suite
  FUNCTIONAL_TEST_SCOPE:
    value: "mr"
    options:
      - "mr"
      - "nightly"
      - "weekly"
      - "pre-release"
      - "release"
    description: "Testsuite to run (only for FUNCTIONAL_TEST=yes)"
  FUNCTIONAL_TEST_CLUSTER_A100:
    value: "dgxa100_dracooci"
    options:
      - "dgxa100_dracooci"
      - "dgxa100_dracooci-ord"
    description: 'Cluster for A100 workloads'
  FUNCTIONAL_TEST_CLUSTER_H100:
    value: "dgxh100_eos"
    options:
      - "dgxh100_coreweave"
      - "dgxh100_eos"
    description: 'Cluster for H100 workloads'
  FUNCTIONAL_TEST_NAME:
    description: "Name of functional test run (only for pre-release and release)"
  PUBLISH: 
    value: "no"
    options: 
      - "yes"
      - "no"
    description: Build and publish a wheel to PyPi
  PUBLISH_SCOPE:
    value: "code-freeze"
    options:
      - "code-freeze"
      - "release"
    description: Type of publish (freeze or final release)

  # CI wide variables
  CI_MCORE_IMAGE: ${GITLAB_ENDPOINT}:5005/adlr/megatron-lm/mcore_ci
  CI_MCORE_DEV_IMAGE: ${GITLAB_ENDPOINT}:5005/adlr/megatron-lm/mcore_ci_dev
  CI_NEMO_IMAGE: ${GITLAB_ENDPOINT}:5005/adlr/megatron-lm/nemo_ci
  LINTING_IMAGE: ${GITLAB_ENDPOINT}:5005/adlr/megatron-lm/mcore_linting
  UNIT_TEST_TIMEOUT: 15
  UNIT_TEST_REPEAT: 1

include:
  - .gitlab/stages/00.pre.yml
  - .gitlab/stages/01.tests.yml
  - .gitlab/stages/02.functional-tests.yml
  - .gitlab/stages/03.publish.yml