inductor-periodic #1156
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: inductor-periodic | |
| on: | |
| push: | |
| tags: | |
| - ciflow/inductor-periodic/* | |
| workflow_dispatch: | |
| schedule: | |
| # Run every 4 hours during the week and every 12 hours on the weekend | |
| - cron: 45 0,4,8,12,16,20 * * 1-5 | |
| - cron: 45 4,12 * * 0,6 | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref_name }}-${{ github.ref_type == 'branch' && github.sha }}-${{ github.event_name == 'workflow_dispatch' }} | |
| cancel-in-progress: true | |
| permissions: read-all | |
| jobs: | |
| get-default-label-prefix: | |
| name: get-default-label-prefix | |
| uses: pytorch/pytorch/.github/workflows/_runner-determinator.yml@main | |
| if: ${{ (github.event_name != 'schedule' || github.repository == 'pytorch/pytorch') && github.repository_owner == 'pytorch' }} | |
| with: | |
| triggering_actor: ${{ github.triggering_actor }} | |
| issue_owner: ${{ github.event.pull_request.user.login || github.event.issue.user.login }} | |
| curr_branch: ${{ github.head_ref || github.ref_name }} | |
| curr_ref_type: ${{ github.ref_type }} | |
| linux-focal-cuda12_4-py3_10-gcc9-periodic-dynamo-benchmarks-build: | |
| name: cuda12.4-py3.10-gcc9-sm86-periodic-dynamo-benchmarks | |
| uses: ./.github/workflows/_linux-build.yml | |
| needs: get-default-label-prefix | |
| with: | |
| runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}" | |
| build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm86 | |
| docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks | |
| cuda-arch-list: '8.6' | |
| test-matrix: | | |
| { include: [ | |
| { config: "dynamo_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamo_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamo_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamo_eager_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamo_eager_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_eager_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_eager_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_aot_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_aot_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_aot_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_aot_eager_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_aot_eager_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| ]} | |
| secrets: inherit | |
| linux-focal-cuda12_4-py3_10-gcc9-periodic-dynamo-benchmarks-test: | |
| name: cuda12.4-py3.10-gcc9-sm86-periodic-dynamo-benchmarks | |
| uses: ./.github/workflows/_linux-test.yml | |
| needs: linux-focal-cuda12_4-py3_10-gcc9-periodic-dynamo-benchmarks-build | |
| with: | |
| build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm86 | |
| docker-image: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-periodic-dynamo-benchmarks-build.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-periodic-dynamo-benchmarks-build.outputs.test-matrix }} | |
| secrets: inherit | |
| linux-focal-rocm6_3-py3_10-periodic-dynamo-benchmarks-build: | |
| if: github.repository_owner == 'pytorch' | |
| name: rocm6_3-py3_10-periodic-dynamo-benchmarks | |
| uses: ./.github/workflows/_linux-build.yml | |
| with: | |
| build-environment: linux-focal-rocm6_3-py3_10 | |
| docker-image-name: pytorch-linux-focal-rocm-n-py3 | |
| sync-tag: rocm-build | |
| test-matrix: | | |
| { include: [ | |
| { config: "dynamo_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamo_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamo_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamo_eager_timm", shard: 1, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamo_eager_timm", shard: 2, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "aot_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "aot_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "aot_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "aot_eager_timm", shard: 1, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "aot_eager_timm", shard: 2, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamic_aot_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamic_aot_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamic_aot_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamic_aot_eager_timm", shard: 1, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| { config: "dynamic_aot_eager_timm", shard: 2, num_shards: 2, runner: "linux.rocm.gpu.mi300.2" }, | |
| ]} | |
| secrets: inherit | |
| linux-focal-rocm6_3-py3_10-periodic-dynamo-benchmarks-test: | |
| permissions: | |
| id-token: write | |
| contents: read | |
| name: rocm6_3-py3_10-periodic-dynamo-benchmarks | |
| uses: ./.github/workflows/_rocm-test.yml | |
| needs: linux-focal-rocm6_3-py3_10-periodic-dynamo-benchmarks-build | |
| with: | |
| build-environment: linux-focal-rocm6_3-py3_10 | |
| docker-image: ${{ needs.linux-focal-rocm6_3-py3_10-periodic-dynamo-benchmarks-build.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-focal-rocm6_3-py3_10-periodic-dynamo-benchmarks-build.outputs.test-matrix }} | |
| secrets: inherit | |
| linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp: | |
| name: cuda12.4-py3.10-gcc9-sm80 | |
| uses: ./.github/workflows/_linux-build.yml | |
| needs: | |
| - get-default-label-prefix | |
| with: | |
| runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}" | |
| build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm80 | |
| docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks | |
| cuda-arch-list: '8.0' | |
| test-matrix: | | |
| { include: [ | |
| { config: "inductor_torchbench_smoketest_perf", shard: 1, num_shards: 1, runner: "linux.aws.a100" }, | |
| ]} | |
| secrets: inherit | |
| linux-focal-cuda12_4-py3_10-gcc9-inductor-test-gcp: | |
| name: cuda12.4-py3.10-gcc9-sm80 | |
| uses: ./.github/workflows/_linux-test.yml | |
| needs: linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp | |
| with: | |
| build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm80 | |
| docker-image: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp.outputs.test-matrix }} | |
| # disable monitor in smoke perf tests for more investigation | |
| disable-monitor: true | |
| secrets: inherit | |
| linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build: | |
| name: linux-jammy-cpu-py3.9-gcc11-periodic-dynamo-benchmarks | |
| uses: ./.github/workflows/_linux-build.yml | |
| needs: get-default-label-prefix | |
| with: | |
| build-environment: linux-jammy-py3.9-gcc11-build | |
| docker-image-name: pytorch-linux-jammy-py3.9-gcc11-inductor-benchmarks | |
| runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}" | |
| test-matrix: | | |
| { include: [ | |
| { config: "cpu_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_inductor_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_inductor_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_freezing_avx2_huggingface", shard: 1, num_shards: 1, runner: "linux.10xlarge.avx2" }, | |
| { config: "cpu_inductor_freezing_avx2_torchbench", shard: 1, num_shards: 2, runner: "linux.10xlarge.avx2" }, | |
| { config: "cpu_inductor_freezing_avx2_torchbench", shard: 2, num_shards: 2, runner: "linux.10xlarge.avx2" }, | |
| { config: "cpu_inductor_freezing_avx2_timm", shard: 1, num_shards: 2, runner: "linux.10xlarge.avx2" }, | |
| { config: "cpu_inductor_freezing_avx2_timm", shard: 2, num_shards: 2, runner: "linux.10xlarge.avx2" }, | |
| ]} | |
| secrets: inherit | |
| linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-test: | |
| name: linux-jammy-cpu-py3.9-gcc11-periodic-dynamo-benchmarks | |
| uses: ./.github/workflows/_linux-test.yml | |
| needs: linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build | |
| with: | |
| build-environment: linux-jammy-py3.9-gcc11-build | |
| docker-image: ${{ needs.linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build.outputs.test-matrix }} | |
| secrets: inherit | |
| linux-focal-cuda12_4-py3_10-gcc9-inductor-build: | |
| name: cuda12.4-py3.10-gcc9-sm86 | |
| uses: ./.github/workflows/_linux-build.yml | |
| needs: get-default-label-prefix | |
| with: | |
| build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm86 | |
| docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks | |
| cuda-arch-list: '8.6' | |
| runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}" | |
| sync-tag: linux-focal-cuda12_4-py3_10-gcc9-inductor-build | |
| test-matrix: | | |
| { include: [ | |
| { config: "dynamic_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_inductor_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_inductor_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "dynamic_inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_inductor_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_inductor_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "aot_inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| ]} | |
| secrets: inherit | |
| linux-focal-cuda12_4-py3_10-gcc9-inductor-test: | |
| name: cuda12.4-py3.10-gcc9-sm86 | |
| uses: ./.github/workflows/_linux-test.yml | |
| needs: linux-focal-cuda12_4-py3_10-gcc9-inductor-build | |
| with: | |
| build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm86 | |
| docker-image: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build.outputs.test-matrix }} | |
| secrets: inherit | |
| linux-jammy-cpu-py3_9-gcc11-inductor-build: | |
| name: linux-jammy-cpu-py3.9-gcc11-inductor | |
| uses: ./.github/workflows/_linux-build.yml | |
| needs: get-default-label-prefix | |
| with: | |
| build-environment: linux-jammy-py3.9-gcc11-build | |
| docker-image-name: pytorch-linux-jammy-py3.9-gcc11-inductor-benchmarks | |
| runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}" | |
| sync-tag: linux-jammy-cpu-py3_9-gcc11-inductor-build | |
| test-matrix: | | |
| { include: [ | |
| { config: "cpu_inductor_freezing_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_freezing_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_freezing_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_amp_freezing_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_amp_freezing_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_amp_freezing_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_amp_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_inductor_amp_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_freezing_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_freezing_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_freezing_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_amp_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "cpu_aot_inductor_amp_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_aot_inductor_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_aot_inductor_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_aot_inductor_amp_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| { config: "dynamic_cpu_aot_inductor_amp_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" }, | |
| ]} | |
| secrets: inherit | |
| linux-jammy-cpu-py3_9-gcc11-inductor-test: | |
| name: linux-jammy-cpu-py3.9-gcc11-inductor | |
| uses: ./.github/workflows/_linux-test.yml | |
| needs: linux-jammy-cpu-py3_9-gcc11-inductor-build | |
| with: | |
| build-environment: linux-jammy-py3.9-gcc11-build | |
| docker-image: ${{ needs.linux-jammy-cpu-py3_9-gcc11-inductor-build.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-jammy-cpu-py3_9-gcc11-inductor-build.outputs.test-matrix }} | |
| secrets: inherit |