mirror of
https://github.com/saymrwulf/pytorch.git
synced 2026-05-14 20:57:59 +00:00
d4871750d9
This PR * makes changes to the workflow files and scripts so we can run CI workflows on the MI300 runners * skips and fixes several tests, failed on MI300, observed in https://github.com/pytorch/pytorch/pull/140989 Skipped due to unsupported Float8_e4m3fn data type on MI300 (need to update test code to use datatypes supported by MI300): - distributed.tensor.parallel.test_micro_pipeline_tp.py::MicroPipelineTPTest::test_fuse_all_gather_scaled_matmul_A_dims_\*_gather_dim_\* (24 tests across inductor/distributed configs) - distributed.tensor.parallel.test_micro_pipeline_tp.py::test_fuse_scaled_matmul_reduce_scatter_A_dims_\*_scatter_dim_\* (12 tests across inductor/distributed configs)) - inductor.test_loop_ordering::LoopOrderingTest::test_fp8_cast_and_t - inductor.test_loop_ordering::LoopOrderingTest::test_fp8_pattern_2 Skipped due to AssertionError on MI300: - inductor.test_mkldnn_pattern_matcher.py::test_qconv2d_int8_mixed_bf16 - distributed._tools.test_sac_ilp::TestSACILP::test_sac_ilp_case1 Skipped: - test_cuda.py::TestCudaMallocAsync::test_clock_speed - test_cuda.py::TestCudaMallocAsync::test_power_draw - test_torch.py::TestTorchDeviceTypeCUDA::test_deterministic_cumsum_cuda Skipped flaky tests on MI300: - distributed.test_c10d_gloo.py::ProcessGroupGlooTest::test_gather_stress_cuda - inductor.test_cpu_repro::CPUReproTests::test_lstm_packed_unbatched_False* (256 tests) Fixed: - test_matmul_cuda.py::TestFP8MatmulCudaCUDA::test_float8_basics_cuda Features: - inductor/test_fp8.py - declare a new function to convert FP8 datatypes to ROCm supported FP8 datatypes. It keeps test names for CUDA and ROCm and allows to enable Inductor FP8 tests on CPU Pull Request resolved: https://github.com/pytorch/pytorch/pull/143673 Approved by: https://github.com/jeffdaily, https://github.com/malfet, https://github.com/pruthvistony Co-authored-by: saienduri <saimanas.enduri@amd.com> Co-authored-by: Jithun Nair <jithun.nair@amd.com> Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com> |
||
|---|---|---|
| .. | ||
| _bazel-build-test.yml | ||
| _binary-build-linux.yml | ||
| _binary-test-linux.yml | ||
| _binary-upload.yml | ||
| _docs.yml | ||
| _linux-build.yml | ||
| _linux-test.yml | ||
| _mac-build.yml | ||
| _mac-test-mps.yml | ||
| _mac-test.yml | ||
| _rocm-test.yml | ||
| _runner-determinator.yml | ||
| _win-build.yml | ||
| _win-test.yml | ||
| _xpu-test.yml | ||
| assigntome-docathon.yml | ||
| auto_request_review.yml | ||
| build-almalinux-images.yml | ||
| build-libtorch-images.yml | ||
| build-magma-linux.yml | ||
| build-magma-windows.yml | ||
| build-manywheel-images-s390x.yml | ||
| build-manywheel-images.yml | ||
| build-triton-wheel.yml | ||
| check-labels.yml | ||
| check_mergeability_ghstack.yml | ||
| cherry-pick.yml | ||
| close-nonexistent-disable-issues.yml | ||
| create_release.yml | ||
| delete_old_branches.yml | ||
| docathon-sync-label.yml | ||
| docker-builds.yml | ||
| docker-release.yml | ||
| generated-linux-aarch64-binary-manywheel-nightly.yml | ||
| generated-linux-binary-libtorch-cxx11-abi-main.yml | ||
| generated-linux-binary-libtorch-cxx11-abi-nightly.yml | ||
| generated-linux-binary-libtorch-pre-cxx11-main.yml | ||
| generated-linux-binary-libtorch-pre-cxx11-nightly.yml | ||
| generated-linux-binary-manywheel-main.yml | ||
| generated-linux-binary-manywheel-nightly.yml | ||
| generated-linux-s390x-binary-manywheel-nightly.yml | ||
| generated-macos-arm64-binary-libtorch-cxx11-abi-nightly.yml | ||
| generated-macos-arm64-binary-wheel-nightly.yml | ||
| generated-windows-binary-libtorch-debug-main.yml | ||
| generated-windows-binary-libtorch-debug-nightly.yml | ||
| generated-windows-binary-libtorch-release-main.yml | ||
| generated-windows-binary-libtorch-release-nightly.yml | ||
| generated-windows-binary-wheel-nightly.yml | ||
| inductor-micro-benchmark-x86.yml | ||
| inductor-micro-benchmark.yml | ||
| inductor-perf-compare.yml | ||
| inductor-perf-test-nightly-aarch64.yml | ||
| inductor-perf-test-nightly-macos.yml | ||
| inductor-perf-test-nightly-x86.yml | ||
| inductor-perf-test-nightly.yml | ||
| inductor-periodic.yml | ||
| inductor-rocm.yml | ||
| inductor-unittest.yml | ||
| inductor.yml | ||
| lint-autoformat.yml | ||
| lint-bc.yml | ||
| lint.yml | ||
| linux-aarch64.yml | ||
| llm_td_retrieval.yml | ||
| mac-mps.yml | ||
| nightly-s3-uploads.yml | ||
| nightly.yml | ||
| nitpicker.yml | ||
| periodic.yml | ||
| pull.yml | ||
| revert.yml | ||
| rocm.yml | ||
| runner-determinator-validator.yml | ||
| runner_determinator_script_sync.yaml | ||
| s390.yml | ||
| scorecards.yml | ||
| slow.yml | ||
| stale.yml | ||
| target-determination-indexer.yml | ||
| target_determination.yml | ||
| test-check-binary.yml | ||
| torchbench.yml | ||
| trunk.yml | ||
| trymerge.yml | ||
| tryrebase.yml | ||
| unstable-periodic.yml | ||
| unstable.yml | ||
| update-viablestrict.yml | ||
| update_pytorch_labels.yml | ||
| upload-test-stats-while-running.yml | ||
| upload-test-stats.yml | ||
| upload-torch-dynamo-perf-stats.yml | ||
| upload_test_stats_intermediate.yml | ||
| weekly.yml | ||
| xpu.yml | ||