Added DevOps PR and Nightly Build logic (#58007)

Summary: This PR adds Azure DevOps support for running custom PyTorch unit tests on PyTorch PR and Nightly builds. PR Builds on Azure DevOps: - Ensures that the wheel artifacts for a given PR build is ready - Once the wheels are ready, PyTorch custom tests are run on torch installation from build wheels Nightly Builds on Azure DevOps: - Cues 4 builds {Win,Linux}*{cpu, CUDA} to run PyTorch custom unit tests on nightly PyTorch builds. Pull Request resolved: https://github.com/pytorch/pytorch/pull/58007 Reviewed By: seemethere, mruberry Differential Revision: D28342428 Pulled By: malfet fbshipit-source-id: a454accf69163f9ba77845eeb54831ef91437981
2025-12-06 12:20:52 +01:00 · 2021-05-12 12:21:12 -07:00 · 2021-05-12 12:21:12 -07:00 · 53bc6f79f3
commit 53bc6f79f3
parent 7156168f71
7 changed files with 249 additions and 1 deletions
--- a/.azure_pipelines/job_templates/pytorch-template-unix.yml
+++ b/.azure_pipelines/job_templates/pytorch-template-unix.yml
@ -0,0 +1,51 @@
+# PyTorch build steps template with Unix images Azure DevOps Instances
+#
+# This build depends on 5 parameters set as an environment variables in the pipeline:
+#   - AZURE_DEVOPS_CLI_PAT: Secret var for authenticating to Azure DevOps
+#   - AZURE_STORAGE_KEY: Secret var for authenticating to Azure Storage
+#   - _TS_CLONE_P, _TS_P, _TS_SM_P: Secret vars for specific unit tests
+
+parameters:
+  name: ''
+  pool: ''
+  container_endpoint: ''
+  customMatrixes: ''
+
+jobs:
+- job: ${{parameters.name}}
+  timeoutInMinutes: 600
+  strategy:
+    matrix:
+      ${{ insert }}: ${{parameters.customMatrixes}}
+  pool:
+    name: ${{ parameters.pool}}
+  variables:
+    DECODE_PERCENTS: false
+
+  steps:
+  # Don't checkout repo contents to save time and CPU compute. Environment variables
+  # related to checkout branch such as $(BUILD_SOURCEBRANCH) are still available.
+  - checkout: none
+
+  # Delete pytorch_tests repo from previous builds if exists
+  - bash: rm -rf pytorch_tests/
+    displayName: Delete pytorch_tests repo from previous builds if exists
+
+  # Clone PyTorch Tests repository
+  - bash: |
+      B64_PAT=$(printf "%s"":$_ADOTOKEN" | base64)
+      git -c http.extraHeader="Authorization: Basic ${B64_PAT}" clone $(AZURE_DEVOPS_PYTORCH_TESTS_REPO_URL)
+      cd pytorch_tests
+      git checkout $(PYTORCH_TESTS_CHECKOUT_BRANCH)
+    env:
+      _ADOTOKEN: $(AZURE_DEVOPS_CLI_PAT)
+    displayName: Clone PyTorch Tests repo
+
+  # Run PyTorch Unit Tests
+  - bash: bash $(Build.SourcesDirectory)/pytorch_tests/scripts/linux/run.sh
+    env:
+      _AZURE_STORAGE_KEY: $(AZURE_STORAGE_KEY)
+      _TS_CLONE_P: $(TS_CLONE_PASSWORD)
+      _TS_P: $(TS_PAT)
+      _TS_SM_P: $(TS_SM_PAT)
+    displayName: Run PyTorch Unit Tests
--- a/.azure_pipelines/job_templates/pytorch-template-win.yml
+++ b/.azure_pipelines/job_templates/pytorch-template-win.yml
@ -0,0 +1,49 @@
+# PyTorch build steps template with Windows images Azure DevOps Instances
+#
+# This build depends on 5 parameters set as an environment variables in the pipeline:
+#   - AZURE_DEVOPS_CLI_PAT: Secret var for authenticating to Azure DevOps
+#   - AZURE_STORAGE_KEY: Secret var for authenticating to Azure Storage
+#   - _TS_CLONE_P, _TS_P, _TS_SM_P: Secret vars for specific unit tests
+
+parameters:
+  name: ''
+  pool: ''
+  customMatrixes: ''
+
+jobs:
+- job: ${{parameters.name}}
+  timeoutInMinutes: 600
+  strategy:
+    matrix:
+      ${{ insert }}: ${{parameters.customMatrixes}}
+  pool:
+    name: ${{ parameters.pool}}
+
+  steps:
+  # Don't checkout repo contents to save time and CPU compute. Environment variables
+  # related to checkout branch such as $(BUILD_SOURCEBRANCH) are still available.
+  - checkout: none
+
+  # Delete pytorch_tests repo from previous builds if exists
+  - script: if exist "pytorch_tests/" rmdir "pytorch_tests/" /q /s
+    displayName: Delete pytorch_tests repo from previous builds if exists
+
+  # Clone PyTorch Tests repository
+  - powershell: |
+      $env:B64Pat = [Convert]::ToBase64String([System.Text.Encoding]::UTF8.GetBytes(":$env:_ADOTOKEN"))
+      git -c http.extraHeader="Authorization: Basic $env:B64Pat" clone $env:AZURE_DEVOPS_pytorch_tests_REPO_URL
+      cd pytorch_tests
+      git checkout $(PYTORCH_TESTS_CHECKOUT_BRANCH)
+    env:
+      _ADOTOKEN: $(AZURE_DEVOPS_CLI_PAT)
+    displayName: Clone PyTorch Tests repo
+
+  # Run PyTorch Unit Tests
+  - script: call $(Build.SourcesDirectory)\pytorch_tests\scripts\windows\run.bat
+    env:
+      _ADOTOKEN: $(AZURE_DEVOPS_CLI_PAT)
+      _AZURE_STORAGE_KEY: $(AZURE_STORAGE_KEY)
+      _TS_CLONE_P: $(TS_CLONE_PASSWORD)
+      _TS_P: $(TS_PAT)
+      _TS_SM_P: $(TS_SM_PAT)
+    displayName: Run PyTorch Unit Tests
--- a/.azure_pipelines/job_templates/wheel-wait-job-template.yml
+++ b/.azure_pipelines/job_templates/wheel-wait-job-template.yml
@ -0,0 +1,14 @@
+# Main logic to initiate wait for PR artifact to be ready
+
+steps:
+- task: InvokeRESTAPI@1
+  displayName: 'Wait for job success and wheel ready'
+  timeoutInMinutes: 60
+  inputs:
+    connectionType: 'connectedServiceName'
+    serviceConnection: circleciconn
+    method: 'POST'
+    headers: '{"Content-Type":"application/json", "BranchName":"$(TARGET_BRANCH_TO_CHECK_PR)", "JobName":"$(TARGET_CIRCLECI_PR)", "PlanUrl":"$(System.CollectionUri)", "ProjectId":"$(System.TeamProjectId)", "HubName":"$(System.HostType)", "PlanId":"$(System.PlanId)", "JobId":"$(System.JobId)", "TimelineId":"$(System.TimelineId)", "TaskInstanceId":"$(System.TaskInstanceId)", "AuthToken":"$(System.AccessToken)"}'
+    body: ''
+    urlSuffix: 'api/JobStatus'
+    waitForCompletion: true
--- a/.azure_pipelines/job_templates/wheel-wait-template.yml
+++ b/.azure_pipelines/job_templates/wheel-wait-template.yml
@ -0,0 +1,49 @@
+# Initiate 5 agentless-server waiting jobs to check on the
+# status of PR artifact builds, for a maximum wait time of
+# 5 * 60 min =300 minutes. These jobs will pass immediately
+# once targeted CircleCI build is ready.
+
+jobs:
+- job: checkjob1
+  pool: server
+  timeoutInMinutes: 60
+  continueOnError: true
+
+  steps:
+  - template: wheel-wait-job-template.yml
+
+- job: checkjob2
+  pool: server
+  timeoutInMinutes: 60
+  dependsOn: checkjob1
+  continueOnError: true
+
+  steps:
+  - template: wheel-wait-job-template.yml
+
+- job: checkjob3
+  pool: server
+  timeoutInMinutes: 60
+  dependsOn: checkjob2
+  continueOnError: true
+
+  steps:
+  - template: wheel-wait-job-template.yml
+
+- job: checkjob4
+  pool: server
+  timeoutInMinutes: 60
+  dependsOn: checkjob3
+  continueOnError: true
+
+  steps:
+  - template: wheel-wait-job-template.yml
+
+- job: checkjob5
+  pool: server
+  timeoutInMinutes: 60
+  dependsOn: checkjob4
+  continueOnError: true
+
+  steps:
+  - template: wheel-wait-job-template.yml
--- a/.azure_pipelines/nightly-pytorch-tests-pipeline.yml
+++ b/.azure_pipelines/nightly-pytorch-tests-pipeline.yml
@ -0,0 +1,50 @@
+# PyTorch Nightly PyTorch Tests Builds Pipeline on Azure DevOps
+#
+# This pipeline runs custom PyTorch unit-tests on nightly
+# PyTorch wheels.
+
+stages:
+- stage: 'NightlyCustomTests'
+  displayName: 'Run custom unit tests on PyTorch wheels'
+  jobs:
+  - template: job_templates/pytorch-template-unix.yml
+    parameters:
+      name: ubuntu_1804_CPU_docker
+      pool: $(BUILD_POOL_LIN_1)
+      customMatrixes:
+        Nightly_Custom_Tests:
+          _DOCKER_IMAGE: $(DOCKER_IMAGE_LIN_1)
+          _PYTHON_VERSION: $(PYTHON_VERSION_LIN_1)
+          _CUDA_BUILD_VERSION: $(CUDA_BUILD_VERSION_LIN_1)
+          _RUN_TESTS: $(RUN_TESTS_LIN)
+
+  - template: job_templates/pytorch-template-unix.yml
+    parameters:
+      name: ubuntu_1804_GPU_docker
+      pool: $(BUILD_POOL_LIN_2)
+      customMatrixes:
+        Nightly_Custom_Tests:
+          _DOCKER_IMAGE: $(DOCKER_IMAGE_LIN_2)
+          _PYTHON_VERSION: $(PYTHON_VERSION_LIN_2)
+          _CUDA_BUILD_VERSION: $(CUDA_BUILD_VERSION_LIN_2)
+          _RUN_TESTS: $(RUN_TESTS_LIN)
+
+  - template: job_templates/pytorch-template-win.yml
+    parameters:
+      name: windows_2019_CPU
+      pool: $(BUILD_POOL_WIN_1)
+      customMatrixes:
+        Nightly_Custom_Tests:
+          _PYTHON_VERSION: $(PYTHON_VERSION_WIN_1)
+          _CUDA_BUILD_VERSION: $(CUDA_BUILD_VERSION_WIN_1)
+          _RUN_TESTS: $(RUN_TESTS_WIN)
+
+  - template: job_templates/pytorch-template-win.yml
+    parameters:
+      name: windows_2019_GPU
+      pool: $(BUILD_POOL_WIN_2)
+      customMatrixes:
+        Nightly_Custom_Tests:
+          _PYTHON_VERSION: $(PYTHON_VERSION_WIN_2)
+          _CUDA_BUILD_VERSION: $(CUDA_BUILD_VERSION_WIN_2)
+          _RUN_TESTS: $(RUN_TESTS_WIN)
--- a/.azure_pipelines/pytorch-tests-pipeline.yml
+++ b/.azure_pipelines/pytorch-tests-pipeline.yml
@ -0,0 +1,30 @@
+# PyTorch PR PyTorch Tests Builds Pipeline on Azure DevOps
+#
+# This pipeline:
+#   1) ensures that CircleCI builds for a given PR
+#      have finished, and that its artifacts are
+#      ready for download
+#   2) runs custom PyTorch unit-tests on PyTorch
+#      wheels generated during PR builds.
+
+stages:
+- stage: 'EnsureArtifactsReady'
+  displayName: 'Ensure PyTorch PR Artifacts are ready'
+  jobs:
+  - template: job_templates/wheel-wait-template.yml
+
+- stage: 'PRCustomTests'
+  displayName: 'Run custom unit tests on PyTorch wheels'
+  jobs:
+  - template: job_templates/pytorch-template-unix.yml
+    parameters:
+      name: ubuntu_1804_GPU_docker
+      pool: $(BUILD_POOL_PR)
+      customMatrixes:
+        PR_Custom_Tests:
+          _PYTHON_VERSION: $(PYTHON_VERSION_PR)
+          _CUDA_BUILD_VERSION: $(CUDA_BUILD_VERSION_PR)
+          _TARGET_CIRCLECI_BUILD: $(TARGET_CIRCLECI_PR)
+          _TARGET_BRANCH_TO_CHECK: $(TARGET_BRANCH_TO_CHECK_PR)
+          _DOCKER_IMAGE: $(DOCKER_IMAGE_PR)
+          _RUN_TESTS: $(RUN_TESTS_PR)
--- a/docker/pytorch/ubuntu_cpu_gpu/Dockerfile
+++ b/docker/pytorch/ubuntu_cpu_gpu/Dockerfile
@ -49,8 +49,10 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
        cmake \
        curl \
        git \
+        git-lfs \
        libjpeg-dev \
        libpng-dev \
+        openmpi-bin \
        wget && \
    rm -rf /var/lib/apt/lists/*
 RUN /usr/sbin/update-ccache-symlinks
@ -86,7 +88,10 @@ ARG TORCH_CUDA_ARCH_LIST_VAR
 RUN if [ -z "$TORCH_CUDA_ARCH_LIST_VAR" ] ; then \
    echo "Continuing CPU build ..."; \
  else \
-    echo "Setting CUDA env vars ..."; \
+    echo "Setting CUDA env vars and installing openmpi ..."; \
+    # Set MPI links to avoid libmpi_cxx.so.1 not found error
+    ln -s /usr/lib/x86_64-linux-gnu/libmpi_cxx.so.20 /usr/lib/x86_64-linux-gnu/libmpi_cxx.so.1; \
+    ln -s /usr/lib/x86_64-linux-gnu/libmpi.so.20.10.1 /usr/lib/x86_64-linux-gnu/libmpi.so.12; \
  fi
 # If the build argument TORCH_CUDA_ARCH_LIST_VAR is given, container will be
 # set for GPU/CUDA build, else for CPU build.