.github/workflows/run_iree_dispatch_profiler.yml

# Copyright 2023 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 running IREE Dispatch Profiler
# (https://github.com/openxla/iree/tree/main/experimental/dispatch_profiler).
#
# IREE Dispatch Profiler validates functional correctness and performs profiling
# on matmul, conv, and other ops on IREE CUDA and other back-ends.
#
# It uploads results to this a publicly accessible GCS bucket:
# https://storage.googleapis.com/dispatch-profiler-artifacts

name: dispatch_profiler

on:
  schedule:
    # Run daily at 2:00 PM (14:00) every weekday (Monday to Friday)
    - cron: "0 14 * * 1-5"
  workflow_dispatch:

concurrency:
  # A PR number if a pull request and otherwise the commit hash. This cancels
  # queued and in-progress runs for the same PR (presubmit) or commit
  # (postsubmit). The workflow name is prepended to avoid conflicts between
  # different workflows.
  group: ${{ github.workflow }}-${{ github.event.number || github.sha }}
  cancel-in-progress: true

jobs:
  setup:
    uses: ./.github/workflows/setup.yml

  build_all:
    needs: setup
    if: contains(fromJson(needs.setup.outputs.enabled-jobs), 'build_all')
    uses: ./.github/workflows/build_all.yml
    with:
      runner-group: ${{ needs.setup.outputs.runner-group }}
      runner-env: ${{ needs.setup.outputs.runner-env }}
      write-caches: ${{ needs.setup.outputs.write-caches }}

  profile_cuda:
    needs: [setup, build_all]
    if: contains(fromJson(needs.setup.outputs.enabled-jobs), 'profile_cuda')
    runs-on:
      - self-hosted # must come first
      - runner-group=${{ needs.setup.outputs.runner-group }}
      - environment=${{ needs.setup.outputs.runner-env }}
      - machine-type=a2-highgpu-1g
    env:
      RESULTS_DIR: dispatch-profiler-results-cuda
      GCS_UPLOAD_PARENT_DIR: "gs://dispatch-profiler-artifacts/cuda"
      GCS_UPLOAD_DIR_NAME: ${{ needs.setup.outputs.artifact-upload-dir }}
      BUILD_DIR: ${{ needs.build_all.outputs.build-dir }}
      BUILD_DIR_ARCHIVE: ${{ needs.build_all.outputs.build-dir-archive }}
      BUILD_DIR_GCS_ARTIFACT: ${{ needs.build_all.outputs.build-dir-gcs-artifact }}
      IREE_SHA: ${{ github.sha }}
    steps:
      - name: "Checking out repository"
        uses: actions/checkout@8f4b7f84864484a7bf31766abe9204da3cbe65b3 # v3.5.0
      - name: "Downloading build dir archive"
        run: gcloud storage cp "${BUILD_DIR_GCS_ARTIFACT}" "${BUILD_DIR_ARCHIVE}"
      - name: "Extracting install from build dir archive"
        run: tar -vxf "${BUILD_DIR_ARCHIVE}" "${BUILD_DIR}/install"
      - name: "Running IREE Dispatch Profiler on CUDA"
        run: |
          mkdir "${RESULTS_DIR}"
          ./build_tools/github_actions/docker_run.sh \
            --gpus all \
            gcr.io/iree-oss/nvidia-bleeding-edge@sha256:522491c028ec3b4070f23910c70c8162fd9612e11d9cf062a13444df7e88ab70 \
            ./experimental/dispatch_profiler/profile_all.sh "${BUILD_DIR}/install/bin" \
            "${RESULTS_DIR}"
      - name: "Uploading results"
        run: |
          GCS_ARTIFACT_DIR="$(date +'%Y-%m-%d').sha_${IREE_SHA}.timestamp_$(date +'%s')"
          gcloud storage cp "${RESULTS_DIR}/**" "${GCS_UPLOAD_PARENT_DIR}/${GCS_ARTIFACT_DIR}/"