GeezerVison is showing github.com — gh-153176: Fix destroy_interpreter() test helper clearing a non-current thread state by zangjiucheng · Pull Request #153307 · python/cpython
Conversation
zangjiucheng
commented
Jul 8, 2026
•
edited
Loading
Uh oh!
There was an error while loading. Please reload this page.
Summary
Fixes #153176.
On free-threaded debug builds, _testinternalcapi.destroy_interpreter(id, basic=True)
aborted the process with:
Assertion `tstate == (_PyThreadStateImpl *)_PyThreadState_GET()' failed.
The root cause is in the test helper, not in obmalloc. destroy_interpreter()
calls PyThreadState_Clear(t1) / PyThreadState_Delete(t1) while t2 is the
current thread state. Clearing a non-current thread state runs
_PyThreadState_ClearMimallocHeaps(t1), and on a debug build
mi_heap_collect_ex() treats MI_ABANDON as >= MI_FORCE, so it calls
_mi_abandoned_reclaim_all() and reclaims abandoned pages back into t1's
heaps. That reaches _PyMem_mi_page_reclaimed(), where the reclaimed page
belongs to t1 but the current thread state is t2, so the assertion fires.
The assertion is correct — reclaim should only happen into the current thread's
own heap — so this keeps it and fixes the helper instead: t1 is left for
Py_EndInterpreter() to clean up, matching the rule that PyThreadState_Clear()
must be called on the current thread. Thanks @kumaraditya303 for pinpointing the
root cause.
Test plan
-
./python -m test test_free_threading.test_interpreters -v
- With the assertion in place, reproduced the abort on a free-threaded
debug build before the fix; confirmed it no longer reproduces after.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
|
I also think we can just relax this assertion. Thanks for your fix!
|
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
…ent thread state
_testinternalcapi.destroy_interpreter(basic=True) called PyThreadState_Clear()
and PyThreadState_Delete() on t1 while t2 was the current thread state. On a
free-threaded debug build this reclaimed mimalloc pages into a heap not owned
by the current thread, tripping the assertion in _PyMem_mi_page_reclaimed().
Leave t1 for Py_EndInterpreter() to clean up. Add a regression test.
Co-authored-by: Kumar Aditya <[email protected]>
Co-authored-by: Xiaowei Lu <[email protected]>
|
The PR title and description needs to be updated.
|
zangjiucheng reacted with thumbs up emoji
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
zangjiucheng
changed the title
gh-153176: Fix crash in _PyMem_mi_page_reclaimed on free-threaded debug builds
gh-153176: Fix destroy_interpreter() test helper clearing a non-current thread state
Jul 23, 2026
Uh oh!
There was an error while loading. Please reload this page.
⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️
Hi! The buildbot s390x Fedora Stable LTO + PGO 3.x (tier-3) has failed when building commit 6e98393.
What do you need to do:
- Don't panic.
- Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
- Go to the page of the buildbot that failed (https://buildbot.python.org/#/builders/1627/builds/3097) and take a look at the build logs.
- Check if the failure is related to this commit (6e98393) or if it is a false positive.
- If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.
You can take a look at the buildbot page here:
https://buildbot.python.org/#/builders/1627/builds/3097
Summary of the results of the build (if available):
Click to see traceback logs
Note: switching to '6e983938834a8b6c7043369b3c5de2b5d2084ea4'.
You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.
If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:
git switch -c <new-branch-name>
Or undo this operation with:
git switch -
Turn off this advice by setting config variable advice.detachedHead to false
HEAD is now at 6e983938834a8 gh-153176: Fix destroy_interpreter() test helper clearing a non-current thread state (#153307)
Switched to and reset branch 'main'
find: ‘build’: No such file or directory
find: ‘build’: No such file or directory
find: ‘build’: No such file or directory
find: ‘build’: No such file or directory
make[2]: [Makefile:3474: clean-retain-profile] Error 1 (ignored)
./Modules/socketmodule.c: In function ‘getsockaddrarg’:
./Modules/socketmodule.c:2646:9: warning: ‘strncpy’ specified bound 64 equals destination size [-Wstringop-truncation]
2646 | strncpy((char *)sa->salg_name, name, sizeof(sa->salg_name));
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/bin/ld.bfd: final link failed: No space left on device
collect2: error: ld returned 1 exit status
make[2]: *** [Makefile:1891: Programs/_freeze_module] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [Makefile:996: profile-gen-stamp] Error 2
make: *** [Makefile:1008: profile-run-stamp] Error 2
|
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️
Hi! The buildbot s390x Fedora Stable LTO 3.x (tier-3) has failed when building commit 6e98393.
What do you need to do:
- Don't panic.
- Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
- Go to the page of the buildbot that failed (https://buildbot.python.org/#/builders/1654/builds/3074) and take a look at the build logs.
- Check if the failure is related to this commit (6e98393) or if it is a false positive.
- If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.
You can take a look at the buildbot page here:
https://buildbot.python.org/#/builders/1654/builds/3074
Summary of the results of the build (if available):
Click to see traceback logs
Note: switching to '6e983938834a8b6c7043369b3c5de2b5d2084ea4'.
You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.
If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:
git switch -c <new-branch-name>
Or undo this operation with:
git switch -
Turn off this advice by setting config variable advice.detachedHead to false
HEAD is now at 6e983938834a8 gh-153176: Fix destroy_interpreter() test helper clearing a non-current thread state (#153307)
Switched to and reset branch 'main'
Assembler messages:
Fatal error: can't create Modules/resource.o: No space left on device
Assembler messages:
Fatal error: can't create Modules/_multiprocessing/posixshmem.o: No space left on device
make: *** [Makefile:3781: Modules/resource.o] Error 1
make: *** Waiting for unfinished jobs....
make: *** [Makefile:3791: Modules/_multiprocessing/posixshmem.o] Error 1
/tmp/ccTKtqke.s: Assembler messages:
/tmp/ccTKtqke.s: Fatal error: can't write 332 bytes to section .text of Modules/expat/xmlparse.o: 'No space left on device'
Assembler messages:
Fatal error: can't create Modules/termios.o: No space left on device
make: *** [Makefile:3789: Modules/termios.o] Error 1
/tmp/cc8EHqEG.s: Assembler messages:
Fatal error: can't create Modules/syslogmodule.o: No space left on device
Assembler messages:
/tmp/cc8EHqEG.s: Fatal error: can't write 3951 bytes to section .gnu.lto_packed_name_dawg.83.a1122b6197080a2a of Modules/unicodedata.o: 'No space left on device'
make: *** [Makefile:3787: Modules/syslogmodule.o] Error 1
/tmp/ccTKtqke.s: Fatal error: Modules/expat/xmlparse.o: No such file or directory
make: *** [Makefile:1614: Modules/expat/xmlparse.o] Error 1
/tmp/cc8EHqEG.s: Fatal error: Modules/unicodedata.o: No such file or directory
make: *** [Makefile:3771: Modules/unicodedata.o] Error 1
/tmp/ccYg48Cu.s: Assembler messages:
/tmp/ccYg48Cu.s: Fatal error: can't write 3952 bytes to section .gnu.debuglto_.debug_info of Modules/selectmodule.o: 'No space left on device'
/tmp/ccvInZys.s: Assembler messages:
/tmp/ccvInZys.s: Fatal error: Modules/mmapmodule.o: No space left on device
make: *** [Makefile:3777: Modules/mmapmodule.o] Error 1
make: *** [Makefile:3783: Modules/selectmodule.o] Error 1
/tmp/cc8yXini.s: Assembler messages:
/tmp/cc8yXini.s: Fatal error: can't write 3951 bytes to section .gnu.lto_.decls.eddbfbeabee625fb of Modules/socketmodule.o: 'No space left on device'
/tmp/cc8yXini.s: Fatal error: Modules/socketmodule.o: No such file or directory
make: *** [Makefile:3785: Modules/socketmodule.o] Error 1
find: ‘build’: No such file or directory
find: ‘build’: No such file or directory
find: ‘build’: No such file or directory
find: ‘build’: No such file or directory
make: [Makefile:3474: clean-retain-profile] Error 1 (ignored)
|
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
Reviewers
colesbury
Awaiting requested review from colesbury
Uh oh!
There was an error while loading. Please reload this page.
Assignees
No one assigned
Milestone
No milestone
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.
|