[issue30703] test_multiprocessing_forkserver hangs on the master branch

[STINNER Victor]

STINNER Victor added the comment:

If I understood corretly, the problem is that the Python C signal handler is not reentrant because it calls Py_AddPendingCall() which uses a lock and a list.

Before, the signal handler queued a new call to checksignals_witharg() (which just calls PyErr_CheckSignals()) for each received signal. Now, we only queue a single call to checksignals_witharg().

To prevent reentrency issues, can't we hardcoded a call to PyErr_CheckSignals() in ceval.c when SIGNAL_PENDING_CALLS() is called?

Py_AddPendingCall() feature is rarely used, it's mostly used for processing signals, no? Calling PyErr_CheckSignals() when no signal was received is cheap, so it shouldn't hurt.

----------

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue30703>
_______________________________________