Wei-Sheng Chin 9c9b05bc4f Expose functions used in custom backend in torch_python dll (#148213) ...

Fixes #148208. There are solutions for exposing symbols implicitly from inline functions (i.e., inline function A calls non-inline function B in foo.h. Code includes foo.h has to see the symbol B in DLL).

Solution 1: tag the entire struct where the inline functions are defined as member functions with TORCH_PYTHON_API --- this PR does this for python_arg_parser.h. An alternative solution exists but will slow down dispatching a lot --- drop inline keyword and move implementation to .cc file.

Solution 2: tag individual functions with TORCH_PYTHON_API. This PR does this for python_tensor.h.

Related discussion about hiding torch_python symbols: https://github.com/pytorch/pytorch/pull/142214

Pull Request resolved: https://github.com/pytorch/pytorch/pull/148213
Approved by: https://github.com/malfet

2025-03-07 02:34:37 +00:00

..

python_tensor.cpp

[BE][clang-format] make macro PyObject_HEAD_INIT(type) and PyVarObject_HEAD_INIT(type, size) have its own line (#136949)

2024-10-02 18:39:22 +00:00

python_tensor.h

Expose functions used in custom backend in torch_python dll (#148213)

2025-03-07 02:34:37 +00:00