.github/workflows/benchmark_execution.yml

# Copyright 2022 The IREE Authors
#
# Licensed under the Apache License v2.0 with LLVM Exceptions.
# See https://llvm.org/LICENSE.txt for license information.
# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
#
# Workflow for execution-benchmark-related jobs. It is designed to be called
# from the main workflow ci.yml. The concurrency of this workflow is controlled
# by the caller's job.

name: Benchmark execution

on:
  workflow_call:
    inputs:
      runner-group:
        required: true
        type: string
      runner-env:
        required: true
        type: string
      e2e-test-artifacts-dir:
        required: true
        type: string
      e2e-test-artifacts-gcs-artifact-dir:
        required: true
        type: string
      benchmark-tools-gcs-artifact-dir:
        required: true
        type: string
    outputs:
      benchmark-results-dir:
        description: |
          Local path that stores all benchmark results.
          Empty if no benchmark runs.
        value: ${{ jobs.run_benchmarks.outputs.benchmark-results-dir }}
      benchmark-results-gcs-artifact-dir:
        description: |
          GCS path that stores all benchmark results.
          Empty if no benchmark runs.
        value: ${{ jobs.run_benchmarks.outputs.benchmark-results-gcs-artifact-dir }}

env:
  # This duplicates the variable from ci.yml. The variable needs to be in env
  # instead of the outputs of setup because it contains the run attempt and we
  # want that to be the current attempt, not whatever attempt the setup step
  # last ran in. It therefore can't be passed in via inputs because the env
  # context isn't available there.
  GCS_DIR: gs://iree-github-actions-${{ github.event_name == 'pull_request' && 'presubmit' || 'postsubmit' }}-artifacts/${{ github.run_id }}/${{ github.run_attempt }}
  BENCHMARK_CONFIG: ${{ inputs.e2e-test-artifacts-dir }}/execution-benchmark-config.json
  BENCHMARK_CONFIG_GCS_ARTIFACT: ${{ inputs.e2e-test-artifacts-gcs-artifact-dir }}/execution-benchmark-config.json

jobs:
  generate_matrix:
    runs-on:
      - self-hosted  # must come first
      - runner-group=${{ inputs.runner-group }}
      - environment=${{ inputs.runner-env }}
      - cpu
      - os-family=Linux
    outputs:
      benchmark-matrix: ${{ steps.generate.outputs.benchmark-matrix }}
    steps:
      - name: "Checking out repository"
        uses: actions/checkout@8f4b7f84864484a7bf31766abe9204da3cbe65b3 # v3.5.0
      - name: "Checking out runtime submodules"
        run: ./build_tools/scripts/git/update_runtime_submodules.sh
      - name: "Generating benchmark matrix"
        id: generate
        run: |
          gcloud storage cp "${BENCHMARK_CONFIG_GCS_ARTIFACT}" "${BENCHMARK_CONFIG}"
          echo benchmark-matrix=$(jq \
              'to_entries | map({"device_name": .key, "host_environment": .value.host_environment})' \
              "${BENCHMARK_CONFIG}") \
            >> "${GITHUB_OUTPUT}"

  run_benchmarks:
    needs: [generate_matrix]
    if: needs.generate_matrix.outputs.benchmark-matrix != '[]'
    strategy:
      # Matrix is dynamically generated by the job generate_matrix. So we only
      # runs the benchmarks specified in ${BENCHMARK_CONFIG}.
      # All tasks in matrix are seen as a single job in Github CI and the job
      # can output a single set of values.
      matrix:
        benchmark: ${{ fromJSON(needs.generate_matrix.outputs.benchmark-matrix) }}
    runs-on:
      - self-hosted  # must come first
      - runner-group=${{ inputs.runner-group }}
      - environment=${{ inputs.runner-env }}
      - machine-type=${{ matrix.benchmark.device_name }}
    env:
      DEVICE_NAME: ${{ matrix.benchmark.device_name }}
      PLATFORM_ARCH: ${{ matrix.benchmark.host_environment.platform }}-${{ matrix.benchmark.host_environment.architecture }}
      E2E_TEST_ARTIFACTS_GCS_ARTIFACT_DIR: ${{ inputs.e2e-test-artifacts-gcs-artifact-dir }}
      E2E_TEST_ARTIFACTS_DIR: ${{ inputs.e2e-test-artifacts-dir }}
      BENCHMARK_RESULTS_DIR: benchmark-results
    outputs:
      benchmark-results-dir: ${{ env.BENCHMARK_RESULTS_DIR }}
      # Ideally this should be defined in env, so it can be used in the upload
      # step. But Github CI doesn't allow us to access env.GCS in env.
      benchmark-results-gcs-artifact-dir: ${{ env.GCS_DIR }}/${{ env.BENCHMARK_RESULTS_DIR }}
    steps:
      - name: "Checking out repository"
        uses: actions/checkout@8f4b7f84864484a7bf31766abe9204da3cbe65b3 # v3.5.0
      - name: "Checking out runtime submodules"
        run: ./build_tools/scripts/git/update_runtime_submodules.sh
      - name: "Downloading benchmark tools"
        id: download-tools
        env:
          # See `build_benchmark_tools` step in ci.yml for the name format of
          # benchmark tools artifacts.
          BENCHMARK_TOOLS_ARCHIVE: ${{ env.PLATFORM_ARCH }}-benchmark-tools.tar
          BENCHMARK_TOOLS_GCS_ARTIFACT: ${{ inputs.benchmark-tools-gcs-artifact-dir }}/${{ env.PLATFORM_ARCH }}-benchmark-tools.tar
        run: |
          gcloud storage cp "${BENCHMARK_TOOLS_GCS_ARTIFACT}" "${BENCHMARK_TOOLS_ARCHIVE}"
          echo "benchmark-tools-archive=${BENCHMARK_TOOLS_ARCHIVE}" >> "${GITHUB_OUTPUT}"
      - name: "Downloading benchmark assets"
        id: download-assets
        run: |
          gcloud storage cp "${BENCHMARK_CONFIG_GCS_ARTIFACT}" "${BENCHMARK_CONFIG}"
          mkdir -p "${E2E_TEST_ARTIFACTS_DIR}"
          jq -r \
            --arg DEVICE_NAME "${DEVICE_NAME}" \
            --arg GCS_ARTIFACT_DIR "${E2E_TEST_ARTIFACTS_GCS_ARTIFACT_DIR}" \
            '.[$DEVICE_NAME] | .module_dir_paths | map("\($GCS_ARTIFACT_DIR)/\(.)") | join("\n")' \
            "${BENCHMARK_CONFIG}" | \
            gcloud storage cp -r --read-paths-from-stdin "${E2E_TEST_ARTIFACTS_DIR}"
          echo "benchmark-config=${BENCHMARK_CONFIG}" >> "${GITHUB_OUTPUT}"
      - name: "Unpacking benchmark tools"
        id: unpack-tools
        env:
          BENCHMARK_TOOLS_ARCHIVE: ${{ steps.download-tools.outputs.benchmark-tools-archive }}
          # See `build_benchmark_tools` step in ci.yml for the name format of
          # benchmark tools directory.
          BENCHMARK_TOOLS_DIR: ${{ env.PLATFORM_ARCH }}-benchmark-tools-dir
        run: |
          tar -xf ${BENCHMARK_TOOLS_ARCHIVE}
          echo "normal-benchmark-tools-dir=${BENCHMARK_TOOLS_DIR}/build/tools" >> "${GITHUB_OUTPUT}"
          echo "traced-benchmark-tools-dir=${BENCHMARK_TOOLS_DIR}/build-traced/tools" >> "${GITHUB_OUTPUT}"
          echo "tracy-capture-tool=${BENCHMARK_TOOLS_DIR}/build-traced/tracy-capture" >> "${GITHUB_OUTPUT}"
      - name: "Running benchmarks"
        env:
          IREE_EXECUTION_BENCHMARK_CONFIG: ${{ steps.download-assets.outputs.benchmark-config }}
          IREE_DOCKER_WRAPPER: ./build_tools/github_actions/docker_run.sh
          IREE_NORMAL_BENCHMARK_TOOLS_DIR: ${{ steps.unpack-tools.outputs.normal-benchmark-tools-dir }}
          IREE_TRACED_BENCHMARK_TOOLS_DIR: ${{ steps.unpack-tools.outputs.traced-benchmark-tools-dir }}
          IREE_TRACY_CAPTURE_TOOL: ${{ steps.unpack-tools.outputs.tracy-capture-tool }}
          IREE_TARGET_DEVICE_NAME: ${{ env.DEVICE_NAME }}
          IREE_E2E_TEST_ARTIFACTS_DIR: ${{ env.E2E_TEST_ARTIFACTS_DIR }}
          IREE_BENCHMARK_RESULTS: ${{ env.BENCHMARK_RESULTS_DIR }}/benchmark-results-${{ matrix.benchmark.device_name }}.json
          IREE_BENCHMARK_TRACES: ${{ env.BENCHMARK_RESULTS_DIR }}/benchmark-traces-${{ matrix.benchmark.device_name }}.tar.gz
        run: |
          mkdir -p ${BENCHMARK_RESULTS_DIR}
          ./build_tools/benchmarks/run_benchmarks.sh
      - name: "Uploading benchmark results"
        run: gcloud storage cp -r "${BENCHMARK_RESULTS_DIR}" "${GCS_DIR}/"