-
-
Notifications
You must be signed in to change notification settings - Fork 30.4k
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.
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
gh-117657: Fix race involving GC and heap initialization #119923
Conversation
The `_PyThreadState_Bind()` function is called before the first `PyEval_AcquireThread()` so it's not synchronized with the stop the world GC. We had a race where `gc_visit_heaps()` might visit a thread's heap while it's being initialized. Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before `tstate_mimalloc_bind()` is called). The race was reproducible by running: `python Lib/test/test_importlib/partial/pool_in_threads.py`.
I guess we could have also used the presence of |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM!
Thanks @colesbury for the PR 🌮🎉.. I'm working now to backport this PR to: 3.13. |
…nGH-119923) The `_PyThreadState_Bind()` function is called before the first `PyEval_AcquireThread()` so it's not synchronized with the stop the world GC. We had a race where `gc_visit_heaps()` might visit a thread's heap while it's being initialized. Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before `tstate_mimalloc_bind()` is called). The race was reproducible by running: `python Lib/test/test_importlib/partial/pool_in_threads.py`. (cherry picked from commit e69d068) Co-authored-by: Sam Gross <colesbury@gmail.com>
GH-120038 is a backport of this pull request to the 3.13 branch. |
…19923) (#120038) The `_PyThreadState_Bind()` function is called before the first `PyEval_AcquireThread()` so it's not synchronized with the stop the world GC. We had a race where `gc_visit_heaps()` might visit a thread's heap while it's being initialized. Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before `tstate_mimalloc_bind()` is called). The race was reproducible by running: `python Lib/test/test_importlib/partial/pool_in_threads.py`. (cherry picked from commit e69d068) Co-authored-by: Sam Gross <colesbury@gmail.com>
…n#119923) The `_PyThreadState_Bind()` function is called before the first `PyEval_AcquireThread()` so it's not synchronized with the stop the world GC. We had a race where `gc_visit_heaps()` might visit a thread's heap while it's being initialized. Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before `tstate_mimalloc_bind()` is called). The race was reproducible by running: `python Lib/test/test_importlib/partial/pool_in_threads.py`.
…n#119923) The `_PyThreadState_Bind()` function is called before the first `PyEval_AcquireThread()` so it's not synchronized with the stop the world GC. We had a race where `gc_visit_heaps()` might visit a thread's heap while it's being initialized. Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before `tstate_mimalloc_bind()` is called). The race was reproducible by running: `python Lib/test/test_importlib/partial/pool_in_threads.py`.
…n#119923) The `_PyThreadState_Bind()` function is called before the first `PyEval_AcquireThread()` so it's not synchronized with the stop the world GC. We had a race where `gc_visit_heaps()` might visit a thread's heap while it's being initialized. Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before `tstate_mimalloc_bind()` is called). The race was reproducible by running: `python Lib/test/test_importlib/partial/pool_in_threads.py`.
The
_PyThreadState_Bind()
function is called beforePyEval_AcquireThread()
so it's not synchronized with the stop the world GC. We had a race wheregc_visit_heaps()
might visit a thread's heap while it's being initialized.Use a simple atomic int to avoid visiting heaps for threads that are not yet fully initialized (i.e., before
tstate_mimalloc_bind()
is called).The race was reproducible by running:
python Lib/test/test_importlib/partial/pool_in_threads.py
.