Skip to content

Fix rare hang in DeepSpeed Async I/O wait by releasing the Python GIL #7727

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sfc-gh-truwase merged 1 commit into from
Dec 18, 2025
+2 −1
Merged

Fix rare hang in DeepSpeed Async I/O wait by releasing the Python GIL #7727

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion csrc/aio/py_lib/py_ds_aio.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -124,5 +124,6 @@ PYBIND11_MODULE(TORCH_EXTENSION_NAME, m)

.def("wait",
&deepspeed_aio_handle_t::wait,
"Wait for (ongoing) asynchronous operations to complete");
"Wait for (ongoing) asynchronous operations to complete",
py::call_guard<py::gil_scoped_release>());
Copy link
Copy Markdown
Collaborator

@sfc-gh-truwase sfc-gh-truwase Dec 17, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there any BC concerns with python versions?

Copy link
Copy Markdown
Collaborator Author

@xylian86 xylian86 Dec 18, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

py::call_guard<py::gil_scoped_release>() requires pybind11 ≥ 2.2.0 (added in v2.2.0 (2017-08-31) per the changelog), So this should be fine—this is an old requirement relative to current pybind11 releases. And there’s no Python-version BC concern.

}