BUG: Importing pandas interferes with mutexes in unrelated modules.

[alexcthomas]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

#include <iostream>
#include <pybind11/pybind11.h>
#include <mutex>

namespace py = pybind11;

void create_mutex() {
    py::gil_scoped_release release;
    std::cout << "Creating mutex\n";
    std::mutex local_mutex;
    std::cout << "Mutex created\n";
    {
        std::cout << "Attempting to lock mutex\n";
        std::lock_guard<std::mutex> lock(local_mutex);
        std::cout << "Mutex locked and unlocked\n";
    }
    std::cout << "Exiting create_mutex\n";
}

PYBIND11_MODULE(TestModule, m) {
    m.doc() = "Test Module";
    m.def("create_mutex", &create_mutex, "Create mutex");
}

Issue Description

Creating a python module with the above code, then running it in python:

import pandas as pd
import TestModule as tm
tm.create_mutex()

crashes without error message after outputting:

Creating mutex
Mutex created
Attempting to lock mutex

The issue occurs whether the gil is released or not. (using pybind11::gil_scoped_acquire acquire;)

Expected Behavior

Should output this without crashing.

Creating mutex
Mutex created
Attempting to lock mutex
Mutex locked and unlocked
Exiting create_mutex

Removing the pandas import from the python allows it to do this.

Installed Versions

INSTALLED VERSIONS

commit : d9cdd2e
python : 3.12.5.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.19045
machine : AMD64
processor : AMD64 Family 23 Model 49 Stepping 0, AuthenticAMD
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : English_United Kingdom.1252

pandas : 2.2.2
numpy : 2.1.0
pytz : 2024.1
dateutil : 2.9.0.post0
setuptools : None
pip : 24.1.2
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : None
pandas_datareader : None
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : None
bottleneck : None
dataframe-api-compat : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : None
numba : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : 17.0.0
pyreadstat : None
python-calamine : None
pyxlsb : None
s3fs : None
scipy : None
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
zstandard : None
tzdata : 2024.1
qtpy : None
pyqt5 : None
pybind11: : 2.13.5

[rhshadrach]

Thanks for the report, further investigations and PRs to fix are welcome!

[alexcthomas]

On further investigation seems to be a pyarrow issue not a pandas issue, so I'm closing the ticket.

I've put together a repo to make replicating this easier https://github.com/alexcthomas/pandas-br59603