gh-102512: Fix threading after os.fork() called from a foreign thread #102517
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
If os.fork() is called from a DummyThread (a thread started not by the threading module, in which somebody called threading.current_thread()), it becomes main thread. Threading shutdown logic relies on the main thread having tstate_lock, so initialize it for the DummyThread too. Fixes python#102512
|
More radical change #113261 was applied instead. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
threading._shutdown() relies on _main_thread having _tstate_lock not None (there is assert for that). When fork is called from a foreign thread (aka DummyThread), it gets promoted to main thread, but remains very simplistic DummyThread. Especially, nobody initializes its _tstate_lock. threading._after_fork() handles the case of current thread not being in _active dict at all (by creating new MainThread object), but it doesn't handle the case of having DummyThread there already.
Fix this by initializing
_tstate_lockin a DummyThread too.See linked issue for more details, including reproducer.
The PR includes also regression test
os.fork()called from DummyThread confuses threading shutdown logic #102512