mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
978b572652
This PR is a follow-up of RFC https://github.com/pytorch/pytorch/issues/115545. In this PR, we intend to provide a registration API dedicated to eager-through-torch.compile. The major workflow of this API will be as follows. - Load cache - Check cache according to the input tensors - Cache Hit: Run the cached kernel directly - Cache Miss: Run the AOTI to produce kernel and run the produced kernel. If AOTI fails to produce the kernel, invoke the python fallback function. Currently, this PR always fallback to python kernel now and cache mechanism will be implemented in another PR - https://github.com/pytorch/pytorch/pull/116368 Differential Revision: [D57164385](https://our.internmc.facebook.com/intern/diff/D57164385) Pull Request resolved: https://github.com/pytorch/pytorch/pull/121387 Approved by: https://github.com/desertfire, https://github.com/jansel, https://github.com/zou3519, https://github.com/jgong5 |
||
|---|---|---|
| .. | ||
| byte_order.cpp | ||
| byte_order.h | ||
| cpp_stacktraces.cpp | ||
| cpp_stacktraces.h | ||
| cuda_enabled.h | ||
| device_lazy_init.cpp | ||
| device_lazy_init.h | ||
| disable_torch_function.cpp | ||
| disable_torch_function.h | ||
| init.cpp | ||
| init.h | ||
| invalid_arguments.cpp | ||
| invalid_arguments.h | ||
| nested.cpp | ||
| nested.h | ||
| numpy_stub.h | ||
| object_ptr.cpp | ||
| object_ptr.h | ||
| out_types.cpp | ||
| out_types.h | ||
| pybind.cpp | ||
| pybind.h | ||
| pycfunction_helpers.h | ||
| pyobject_preservation.cpp | ||
| pyobject_preservation.h | ||
| python_arg_parser.cpp | ||
| python_arg_parser.h | ||
| python_compat.h | ||
| python_dispatch.cpp | ||
| python_dispatch.h | ||
| python_numbers.h | ||
| python_raii.h | ||
| python_scalars.h | ||
| python_strings.h | ||
| python_stub.h | ||
| python_symnode.cpp | ||
| python_symnode.h | ||
| python_torch_function_mode.h | ||
| python_tuples.h | ||
| pythoncapi_compat.h | ||
| schema_info.cpp | ||
| schema_info.h | ||
| six.h | ||
| structseq.cpp | ||
| structseq.h | ||
| tensor_apply.cpp | ||
| tensor_apply.h | ||
| tensor_dtypes.cpp | ||
| tensor_dtypes.h | ||
| tensor_flatten.cpp | ||
| tensor_flatten.h | ||
| tensor_layouts.cpp | ||
| tensor_layouts.h | ||
| tensor_list.cpp | ||
| tensor_list.h | ||
| tensor_memoryformats.cpp | ||
| tensor_memoryformats.h | ||
| tensor_new.cpp | ||
| tensor_new.h | ||
| tensor_numpy.cpp | ||
| tensor_numpy.h | ||
| tensor_qschemes.cpp | ||
| tensor_qschemes.h | ||
| tensor_types.cpp | ||
| tensor_types.h | ||
| throughput_benchmark-inl.h | ||
| throughput_benchmark.cpp | ||
| throughput_benchmark.h | ||
| torch_dispatch_mode.h | ||
| variadic.cpp | ||
| variadic.h | ||
| verbose.cpp | ||
| verbose.h | ||