frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-21 05:34:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

Respect TorchDispatchMode for shallow_copy_and_detach (#83372)

Browse Source 
I noticed I was missing tensor creations with modes when I tried
to delete proxy tensor.  This was the cause.

Hypothetically, all PyInterpreter calls could get this treatment.
But I think it only matters for detach; the rest do not return
Tensors and most modes will not be interested in them.

Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/83372
Approved by: https://github.com/zou3519
This commit is contained in:
Edward Z. Yang
2022-08-15 20:03:12 -07:00
committed by PyTorch MergeBot
parent 1665715cb0
commit a3907ca92d
4 changed files with 42 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
torch/csrc/autograd/python_variable.cpp
View File

@ -2144,7 +2144,8 @@ py::object torchDispatchFromTensorImpl(
c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::
unsafe_reclaim_from_nonowning(const_cast<c10::TensorImpl*>(self)));
auto self_p = py::reinterpret_steal<py::object>(THPVariable_Wrap(self_t));
TORCH_INTERNAL_ASSERT(isPythonTensor(self_t));
// NB: this may not be a python tensor if you got here from a mode!
// TORCH_INTERNAL_ASSERT(isPythonTensor(self_t));
append_overloaded_tensor(&overloaded_args, self_p.ptr());
auto args = py::reinterpret_steal<py::object>(PyTuple_New(1));
PyTuple_SET_ITEM(args.ptr(), 0, self_p.release().ptr());