inductor-cu124 #578
This file contains 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-cu124 | |
| on: | |
| push: | |
| tags: | |
| - ciflow/inductor-cu124/* | |
| 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 | |
| - cron: 29 8 * * * # about 1:29am PDT, for mem leak check and rerun disabled tests | |
| 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: | |
| linux-focal-cuda12_4-py3_10-gcc9-inductor-build: | |
| # Should be synced with the one in inductor.yml, but this doesn't run inductor_timm | |
| name: cuda12.4-py3.10-gcc9-sm86 | |
| uses: ./.github/workflows/_linux-build.yml | |
| with: | |
| sync-tag: linux-focal-cuda12_4-py3_10-gcc9-inductor-build | |
| 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: "inductor", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "inductor", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "inductor_distributed", shard: 1, num_shards: 1, runner: "linux.g5.12xlarge.nvidia.gpu" }, | |
| { config: "inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { 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" }, | |
| { config: "inductor_cpp_wrapper_abi_compatible", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| ]} | |
| secrets: | |
| HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }} | |
| 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: | |
| sync-tag: linux-focal-cuda12_4-py3_10-gcc9-inductor-test | |
| 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: | |
| HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }} | |
| linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp: | |
| name: cuda12.4-py3.10-gcc9-sm80 | |
| uses: ./.github/workflows/_linux-build.yml | |
| with: | |
| 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.gcp.a100" }, | |
| ]} | |
| secrets: | |
| HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }} | |
| 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 }} | |
| use-gha: anything-non-empty-to-use-gha | |
| secrets: | |
| HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }} | |
| linux-focal-cuda12_4-py3_12-gcc9-inductor-build: | |
| name: cuda12.4-py3.12-gcc9-sm86 | |
| uses: ./.github/workflows/_linux-build.yml | |
| with: | |
| build-environment: linux-focal-cuda12.4-py3.12-gcc9-sm86 | |
| docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3.12-gcc9-inductor-benchmarks | |
| cuda-arch-list: '8.6' | |
| test-matrix: | | |
| { include: [ | |
| { config: "inductor", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| { config: "inductor", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" }, | |
| ]} | |
| linux-focal-cuda12_4-py3_12-gcc9-inductor-test: | |
| name: cuda12.4-py3.12-gcc9-sm86 | |
| uses: ./.github/workflows/_linux-test.yml | |
| needs: linux-focal-cuda12_4-py3_12-gcc9-inductor-build | |
| with: | |
| build-environment: linux-focal-cuda12.4-py3.12-gcc9-sm86 | |
| docker-image: ${{ needs.linux-focal-cuda12_4-py3_12-gcc9-inductor-build.outputs.docker-image }} | |
| test-matrix: ${{ needs.linux-focal-cuda12_4-py3_12-gcc9-inductor-build.outputs.test-matrix }} |