OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
b5c4f46bb9
pytorch/test
History
Han Qi b5c4f46bb9 Add functions to setup PrivateUse1 as a python backend device. (#157859) 
Fixes #156052 and #156444.

This PR setup the privateuseone key in Python to be used as a python backend for pytorch.
Meaning that, after calling `setup_privateuseone_for_python_backend('npy')`, one can use a subclass to with that device to hold arbitrary python data as "device data" and use `torch.library` to register ops that takes that Tensor.

Changes done in this PR:

1. Register an vanilla Device Guard: I extended NoOpDeviceGuard to have allow device index of 0 and to not raise errors when event related functions are accessed. If I don't do those, when calling backward I would get errors. (CPU backend uses NoOpDeviceGuard just fine, although there seems to be special treatment of CPU in the autograd engine.
2. Tensor subclass allows not having `__torch_dispatch__` if the device is not CUDA or CPU. The comment of the check suggests it was to avoid segfault when calling into ops that expects a storage. Here we have a different device so will not call into those ops.
3. python function that invokes the other incantations to setup the privateusekey backend.

This took inspiration of https://github.com/bdhirsh/pytorch_open_registration_example and https://github.com/tinygrad/tinygrad/blob/master/extra/torch_backend/wrapped_tensor.cpp; great thanks to @bdhirsh and @geohot.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/157859
Approved by: https://github.com/albanD
2025-10-01 21:32:59 +00:00
..
ao/sparsity [testing] Add test owner labels for some ao sparse tests (#163203) 2025-09-18 16:08:13 +00:00
autograd
backends/xeon
benchmark_utils [BE][3/6] fix typos in test/ (#157637) 2025-07-17 12:08:33 +00:00
compiled_autograd_skips [ca] skip on some PYTORCH_TEST_WITH_DYNAMO=1 autograd tests (#156374) 2025-06-21 18:33:38 +00:00
cpp Migrate DeviceType to torch/headeronly (#163999) 2025-09-30 23:13:27 +00:00
cpp_api_parity [BE][PYFMT] migrate PYFMT for test/[a-h]*/ to ruff format (#144555) 2025-06-24 04:53:54 +00:00
cpp_extensions [OpenReg] Add AMP Integration guide for accelerators (#162050) 2025-09-30 12:27:11 +00:00
custom_backend
custom_operator Using the latest torch.library.register_fake API instead of torch.library.impl_abstract (#158839) 2025-07-25 02:37:30 +00:00
distributed [Replicate][Pipeline Parallelism] integration of new replicate function with pipeline parallelism (#164031) 2025-10-01 18:01:16 +00:00
distributions [Reland][2/N]Port several test files under test/distributed to Intel GPU (#159473) 2025-09-17 06:42:27 +00:00
dynamo Use dataclass features in two classes (#164221) 2025-10-01 03:20:39 +00:00
dynamo_expected_failures Add a RECORD_FUNCTION for Python fallback so it shows in profile (#160573) 2025-10-01 14:10:44 +00:00
dynamo_skips Add functions to setup PrivateUse1 as a python backend device. (#157859) 2025-10-01 21:32:59 +00:00
error_messages
expect [fx] Allow customization of submod name in split graph (#164035) 2025-10-01 16:26:14 +00:00
export [export] Preserve nn_module_stack for aliased nn modules (#164311) 2025-10-01 18:04:33 +00:00
forward_backward_compatibility Fused RMSNorm implementation (#153666) 2025-07-22 22:25:44 +00:00
functorch [1/N] Apply UP035 rule in tests (#163947) 2025-09-29 01:42:01 +00:00
fx [doc]: Small typos (#162982) 2025-09-16 17:42:19 +00:00
higher_order_ops [hop] support local_map + SAC (#163322) 2025-09-24 04:57:40 +00:00
inductor Skip windows unittest in fbcode (#164363) 2025-10-01 20:18:19 +00:00
inductor_expected_failures [dynamo] Graph break on on user-defined class in compiled region (#161670) 2025-09-10 04:39:20 +00:00
inductor_skips
jit [Code Clean] Remove deadcodes about Python3.9 [7/N] (#163646) 2025-09-24 07:30:50 +00:00
jit_hooks
lazy [BE][PYFMT] migrate PYFMT for test/[i-z]*/ to ruff format (#144556) 2025-07-29 03:26:09 +00:00
mobile [BE][PYFMT] migrate PYFMT for test/[i-z]*/ to ruff format (#144556) 2025-07-29 03:26:09 +00:00
nn Skip test_conv3d_cudnn_broken on ROCM (#164138) 2025-09-29 16:56:51 +00:00
onnx [1/N] Apply UP035 rule in tests (#163947) 2025-09-29 01:42:01 +00:00
optim [optim] prevent unintended aliasing in lr_scheduler; update type annotations/docs (#163120) 2025-09-25 06:58:58 +00:00
package [Torch Package] Make get names of OrderedImporters support fallback to importers (#155743) 2025-08-06 02:26:10 +00:00
profiler Add a RECORD_FUNCTION for Python fallback so it shows in profile (#160573) 2025-10-01 14:10:44 +00:00
quantization half support for fused_moving_avg_obs_fake_quant() op (#164175) 2025-09-30 19:35:17 +00:00
scripts [BE][PYFMT] migrate PYFMT for test/[i-z]*/ to ruff format (#144556) 2025-07-29 03:26:09 +00:00
strobelight/examples
test_img
torch_np [BE][PYFMT] migrate PYFMT for test/[i-z]*/ to ruff format (#144556) 2025-07-29 03:26:09 +00:00
typing Revert "Add __init__.pyi to torch/linalg (#160750)" 2025-09-02 16:53:55 +00:00
xpu [WOQ] Add XPU kernel for _weight_int8pack_mm (#160938) 2025-09-19 07:37:14 +00:00
_test_bazel.py
allowlist_for_publicAPI.json Clean up 'torch.onnx' entries from public API allowlist (#162850) 2025-09-15 16:14:43 +00:00
bench_mps_ops.py [BE] Remove macos-13 guard from bench_mps_ops (#159732) 2025-08-03 20:53:58 +00:00
conftest.py [CI][run_test] Fix rerun logic for failing at exit (#155853) 2025-06-17 17:51:40 +00:00
create_dummy_torchscript_model.py
HowToWriteTestsUsingFileCheck.md
linear.py
load_torchscript_model.py
minioptest_failures_dict.json
mkl_verbose.py
mkldnn_verbose.py
pytest_shard_custom.py
run_doctests.sh
run_test.py Add functions to setup PrivateUse1 as a python backend device. (#157859) 2025-10-01 21:32:59 +00:00
simulate_nccl_errors.py
slow_tests.json Update slow tests (#163493) 2025-09-29 11:58:17 +00:00
test_accelerator.py Add UT for torch.accelerator memory-related API (#155200) 2025-08-08 17:41:22 +00:00
test_ao_sparsity.py [testing] Add test owner labels for some ao sparse tests (#163203) 2025-09-18 16:08:13 +00:00
test_appending_byte_serializer.py
test_autocast.py
test_autograd_fallback.py Fix TestAutogradFallback flaky tests under Dynamo: migrate to lib._destroy() (#159443) 2025-07-30 19:30:55 +00:00
test_autograd.py Clear custom autograd Function ctx.to_save earlier (#161171) 2025-09-02 03:26:31 +00:00
test_autoload.py
test_binary_ufuncs.py [TEST][CUDA] Use proper dtype in test_cuda_tensor_pow_scalar_tensor_cuda (#163070) 2025-09-16 18:28:50 +00:00
test_bundled_images.py
test_bundled_inputs.py
test_ci_sanity_check_fail.py
test_comparison_utils.py
test_compile_benchmark_util.py
test_complex.py
test_content_store.py [ROCm] Enabling several UTs (#161715) 2025-09-09 15:49:21 +00:00
test_cpp_api_parity.py Revert "Stop parsing command line arguments every time common_utils is imported. (#156703)" 2025-08-04 20:37:39 +00:00
test_cpp_extensions_aot.py Replace setup.py install with pip install (#156711) 2025-09-29 15:15:10 +00:00
test_cpp_extensions_jit.py [MPSHooks] Release pending command encoder (#164093) 2025-09-29 17:50:12 +00:00
test_cpp_extensions_mtia_backend.py
test_cpp_extensions_stream_and_event.py
test_cuda_expandable_segments.py
test_cuda_multigpu.py [ROCm] Enabling several UTs (#161715) 2025-09-09 15:49:21 +00:00
test_cuda_nvml_based_avail.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_cuda_primary_ctx.py [ROCm][CI] skip TestCudaPrimaryCtx.test_set_device_0 (#163693) 2025-09-23 22:15:10 +00:00
test_cuda_sanitizer.py
test_cuda_trace.py
test_cuda.py [CUDA][Expandable Segments] Follow-up cleanups for even more expandable segments tests (#163297) 2025-09-30 15:39:14 +00:00
test_custom_ops.py Add utility to get computed kernel in torch.library (#158393) 2025-08-13 21:00:59 +00:00
test_dataloader.py Skip test_ind_worker_queue on Windows and macOS (flaky) (#162555) 2025-09-10 07:05:14 +00:00
test_datapipe.py [BC breaking] Remove deprecated imports for torch.utils.data.datapipes.iter.grouping (#163438) 2025-09-23 05:02:06 +00:00
test_decomp.py write conv1d decomposition (#163080) 2025-09-17 19:22:38 +00:00
test_determination.py
test_dispatch.py Typo fixes for "overridden" in comments and function names (#155944) 2025-06-14 03:37:38 +00:00
test_dlpack.py Upgrades dlpack to v1.1 to include fp8/fp4 (#162195) 2025-09-17 16:39:11 +00:00
test_dynamic_shapes.py [dynamic shapes] unbacked-safe slicing (#161414) 2025-09-30 01:15:19 +00:00
test_expanded_weights.py Revert "Stop parsing command line arguments every time common_utils is imported. (#156703)" 2025-08-04 20:37:39 +00:00
test_extension_utils.py
test_fake_tensor.py [FakeTensor] Supplement the relevant logic for converting conv1d to conv2d in meta_conv (#160408) 2025-09-26 15:45:02 +00:00
test_file_check.py
test_flop_counter.py [ROCm] Transformer/SDPA unit test parity (#163745) 2025-09-25 17:14:19 +00:00
test_foreach.py Fix requires_cuda to requires_cuda_and_triton (#160222) 2025-08-10 07:05:52 +00:00
test_function_schema.py
test_functional_autograd_benchmark.py
test_functional_optim.py
test_functionalization_of_rng_ops.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_functionalization.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_futures.py
test_fx_experimental.py [fx] fix split_module with symint (#160093) 2025-08-13 05:50:15 +00:00
test_fx_passes.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_fx_reinplace_pass.py [dynamic shapes] avoid unnecessary slices (#157528) 2025-07-10 06:34:46 +00:00
test_fx.py [Code Clean] Remove deadcodes about Python3.9 [8/N] (#163728) 2025-09-25 05:12:46 +00:00
test_hop_infra.py
test_hub.py
test_import_stats.py
test_indexing.py Revert "Fix index_add for int64 input + zerodim index (#161511)" 2025-08-27 15:38:11 +00:00
test_itt.py
test_jit_autocast.py Revert "Stop parsing command line arguments every time common_utils is imported. (#156703)" 2025-08-04 20:37:39 +00:00
test_jit_disabled.py
test_jit_fuser_legacy.py Revert "Stop parsing command line arguments every time common_utils is imported. (#156703)" 2025-08-04 20:37:39 +00:00
test_jit_fuser_te.py Remove tensorexpr tests (#158928) 2025-08-09 02:21:22 +00:00
test_jit_fuser.py Revert "Stop parsing command line arguments every time common_utils is imported. (#156703)" 2025-08-04 20:37:39 +00:00
test_jit_legacy.py Revert "Stop parsing command line arguments every time common_utils is imported. (#156703)" 2025-08-04 20:37:39 +00:00
test_jit_llga_fuser.py
test_jit_profiling.py
test_jit_simple.py
test_jit_string.py
test_jit.py [Code Clean] Remove deadcodes about Python3.9 [8/N] (#163728) 2025-09-25 05:12:46 +00:00
test_jiterator.py
test_kernel_launch_checks.py [testing] Add test owner labels for some cuda? tests (#163296) 2025-09-26 18:26:56 +00:00
test_legacy_vmap.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_license.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_linalg.py Adding check for square matrix for input tensor in matrix_exp backwar… (#163357) 2025-10-01 03:12:30 +00:00
test_logging.py
test_masked.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_maskedtensor.py Fix MaskedTensor to device ignored mask (#151205) 2025-07-21 21:44:49 +00:00
test_matmul_cuda.py Fix non-TMA loads in grouped MM Triton kernel (#163895) 2025-10-01 12:21:13 +00:00
test_meta.py [Inductor] support mixed dtype in the native_layer_norm_backward meta function (#159830) 2025-09-17 20:29:12 +00:00
test_metal.py
test_mkl_verbose.py
test_mkldnn_fusion.py
test_mkldnn_verbose.py
test_mkldnn.py Enable TF32 as fp32 internal precision for matmul/linear/conv (#157520) 2025-07-17 08:57:34 +00:00
test_mobile_optimizer.py
test_model_exports_to_core_aten.py Replace export_for_training with export (#162396) 2025-09-10 14:19:34 +00:00
test_module_tracker.py
test_modules.py [ROCm][CI] unskip some test_memory_format tests (#162766) 2025-09-12 15:02:40 +00:00
test_monitor.py
test_mps.py [SDPA] [MPS] Fixes regression in 2.8.0 for scaled_dot_product_attention using mps (#163598) 2025-09-29 16:09:46 +00:00
test_multiprocessing_spawn.py Test multiprocessing spawn timing fix (#160672) 2025-08-15 00:11:55 +00:00
test_multiprocessing.py Revert "[CI] Use setup-python from for Mac tests (#155698)" 2025-06-12 14:42:32 +00:00
test_namedtensor.py Fix tensor creation with empty names crash (#163957) 2025-09-26 17:41:00 +00:00
test_namedtuple_return_api.py
test_native_functions.py
test_native_mha.py
test_nestedtensor.py Add NestedTensor dispatch for _is_any_true/_is_all_true (#162096) 2025-09-22 20:22:44 +00:00
test_nn.py better error handling for rrelu when lower or upper range is infinite (#160965) 2025-09-30 05:01:32 +00:00
test_nnapi.py
test_numa_binding.py Revert "[RELAND] Always build USE_DISTRIBUTED (#160449) and Make distributed modules importable even when backend not built (#159889) (#162594)" 2025-09-25 13:47:46 +00:00
test_numba_integration.py [testing] Add test owner labels for some cuda? tests (#163296) 2025-09-26 18:26:56 +00:00
test_numpy_interop.py Throw invalid_argument instead of RuntimeError when parameters exceed… (#158267) 2025-07-25 23:49:46 +00:00
test_opaque_obj.py [opaque_obj] Add set_payload + docs (#163276) 2025-09-22 20:02:29 +00:00
test_openmp.py
test_ops_fwd_gradients.py
test_ops_gradients.py
test_ops_jit.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_ops.py [1/N]Enable some tests in test_ops.TestCommon on Intel GPU (#159944) 2025-09-29 09:08:04 +00:00
test_optim.py Fix LBFGS wolfe max iteration (#161488) 2025-09-16 12:07:50 +00:00
test_out_dtype_op.py
test_overrides.py Revert "[Reland] Use std::string_view in torchgen (#158625)" 2025-09-13 07:52:50 +00:00
test_package.py
test_per_overload_api.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_prims.py [Bugfix] Match eager stride semantics for cloned tensors with preserve_format in compile (#163017) 2025-09-19 19:41:33 +00:00
test_privateuseone_python_backend.py Add functions to setup PrivateUse1 as a python backend device. (#157859) 2025-10-01 21:32:59 +00:00
test_proxy_tensor.py [dynamic shapes] unbacked-safe slicing (#161414) 2025-09-30 01:15:19 +00:00
test_pruning_op.py
test_public_bindings.py [1/N] Apply UP035 rule in tests (#163947) 2025-09-29 01:42:01 +00:00
test_python_dispatch.py Remove guard_size_oblivious from default contiguity python check, and add aten.sym_is_contiguous. [attempt2] (#160869) 2025-09-08 22:59:13 +00:00
test_pytree.py [BE][pytree] cleanup parameterized pytree tests (#160842) 2025-09-05 20:15:29 +00:00
test_quantization.py Remove pytorch quant docs since we are moving to torchao (#157766) 2025-07-11 03:21:47 +00:00
test_reductions.py PyTorch histc fix for values with large magnitudes (#163506) 2025-09-25 20:55:25 +00:00
test_rename_privateuse1_to_existing_device.py [Device] Add support for PrivateUse1 device type in parse_type function (#157609) 2025-07-17 01:27:44 +00:00
test_scaled_matmul_cuda.py [ROCm] fix carveout feature (#164303) 2025-10-01 19:25:41 +00:00
test_scatter_gather_ops.py fix deterministic scatter_add path for multi-d tensors (#162866) 2025-09-15 06:50:00 +00:00
test_schema_check.py [inductor] slow test some Windows UTs. (#160267) 2025-08-10 18:35:42 +00:00
test_segment_reductions.py [ROCm] Enabling several UTs (#161715) 2025-09-09 15:49:21 +00:00
test_serialization.py added class or module info for functions blocked by weight-only load (#159935) 2025-08-12 20:52:25 +00:00
test_set_default_mobile_cpu_allocator.py
test_shape_ops.py
test_show_pickle.py
test_sort_and_select.py Add dtype checks in meta dispatch for various ordering ops (#159556) 2025-08-14 17:06:27 +00:00
test_sparse_csr.py [Fix] Adding missing f prefixes to formatted strings [4/N] (#164068) 2025-09-29 04:07:07 +00:00
test_sparse_semi_structured.py Remove test conditions for CUDA<12 (#163495) 2025-09-23 07:52:00 +00:00
test_sparse.py [MPS] [Sparse] unique_dim and sparse broadcast (#163694) 2025-09-26 23:03:13 +00:00
test_spectral_ops.py
test_stateless.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_static_runtime.py
test_subclass.py
test_sympy_utils.py [1/N] Apply UP035 rule in tests (#163947) 2025-09-29 01:42:01 +00:00
test_tensor_creation_ops.py Reland #161649, vectorize stored in cat for all dtypes (#162440) 2025-09-18 13:50:44 +00:00
test_tensorboard.py Fix protobuf test comparison by parsing proto instead of raw strings (#162644) 2025-09-12 16:26:54 +00:00
test_tensorexpr_pybind.py
test_tensorexpr.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_testing.py [1/N] Apply UP035 rule in tests (#163947) 2025-09-29 01:42:01 +00:00
test_throughput_benchmark.py
test_torch.py [BE] Move indexing tests to test_indexing (#160994) 2025-08-21 00:42:55 +00:00
test_transformers.py [cuDNN][SDPA] Disable dropout for cuDNN SDPA on 9.11 - 9.13 (#163903) 2025-09-26 23:50:09 +00:00
test_type_hints.py
test_type_info.py
test_type_promotion.py [BE] Raise ValueError from torch.cat meta func (#158249) 2025-07-20 23:49:18 +00:00
test_typing.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_unary_ufuncs.py [BUG] Fix nonzero_static crash on CUDA when the input is a empty tensor (#162578) 2025-09-15 05:44:15 +00:00
test_utils_config_module.py
test_utils_filelock.py
test_utils.py [BE] Remove bottleneck (#163210) 2025-09-18 12:08:13 +00:00
test_view_ops.py unify broadcast_shapes functions and avoid duplicates (#160251) 2025-08-16 00:54:32 +00:00
test_vulkan.py
test_weak.py [BE][2/6] fix typos in test/ (test/test_*.py) (#157636) 2025-07-09 11:02:23 +00:00
test_xnnpack_integration.py
test_xpu.py Add a new API torch.xpu.can_device_access_peer for Intel GPU (#162705) 2025-09-16 18:00:22 +00:00