Skip to content

Conversation

@nascheme
Copy link
Member

@naschemenascheme commented Jul 11, 2025

I don't think it makes sense to make these functions thread-safe. They are already unsafe when used in the default build. And, given that they can operate on any sequence object, trying to lock the sequence object doesn't make sense.

I added a unit test because I believe these functions should not crash or produce TSAN warnings when they are used on a sequence being mutated in another thread.


📚 Documentation preview 📚: https://cpython-previews--136555.org.readthedocs.build/

@naschemenascheme added docs Documentation in the Doc dir topic-free-threading labels Jul 11, 2025
@bedevere-appbedevere-appbot added the tests Tests in the Lib/test dir label Jul 11, 2025
Copy link
Member

@picnixzpicnixz left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Possible suggestions:

  • Use a .. note:: if you think users should know about it.
  • Use a .. warning:: if you think users should be careful about it.
  • Do not use either of them if you think it's just good that users know about it (a note or a warning is quite visible in the sense that it creates a box where the entire content is inside that box).
  • Use a subsection about thread-safetiness. Or use a small warning box just saying that the functions are not thread-safe, and add a link to a section at the bottom of the page that contains more detailed explanations.

cc @hugovk

@nascheme
Copy link
MemberAuthor

Using "note" seems appropriate to me in this case. In the long term, I suggest it makes sense to have a thread-safety (or concurrency-safety) section for any module that has unusual rules (non-atomic functions, shared state). In the "warnings" doc, I added a "Concurrent safety of Context Managers" section to it. That's a case where the rules are quite complex and a dedicated section makes sense.

@naschemenascheme marked this pull request as ready for review July 14, 2025 18:50
@rhettingerrhettinger removed their request for review July 15, 2025 03:17
@@ -0,0 +1,57 @@
importunittest
fromtest.supportimportimport_helper, threading_helper
fromthreadingimportThread, Barrier
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
from threading import Thread, Barrier

this would fix lint CI

@naschemenascheme enabled auto-merge (squash) July 30, 2025 02:21
@naschemenascheme merged commit 5236b02 into python:mainJul 30, 2025
41 checks passed
@github-project-automationgithub-project-automationbot moved this from Todo to Done in Docs PRsJul 30, 2025
@miss-islington-app
Copy link

Thanks @nascheme for the PR 🌮🎉.. I'm working now to backport this PR to: 3.13, 3.14.
🐍🍒⛏🤖

miss-islington pushed a commit to miss-islington/cpython that referenced this pull request Jul 30, 2025
(cherry picked from commit 5236b02) Co-authored-by: Neil Schemenauer <nas-github@arctrix.com>
miss-islington pushed a commit to miss-islington/cpython that referenced this pull request Jul 30, 2025
(cherry picked from commit 5236b02) Co-authored-by: Neil Schemenauer <nas-github@arctrix.com>
@bedevere-app
Copy link

GH-137221 is a backport of this pull request to the 3.14 branch.

@bedevere-appbedevere-appbot removed the needs backport to 3.14 bugs and security fixes label Jul 30, 2025
@bedevere-app
Copy link

GH-137222 is a backport of this pull request to the 3.13 branch.

@bedevere-appbedevere-appbot removed the needs backport to 3.13 bugs and security fixes label Jul 30, 2025
@bedevere-bot
Copy link

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot s390x Fedora Stable LTO + PGO 3.x (tier-3) has failed when building commit 5236b02.

What do you need to do:

  1. Don't panic.
  2. Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
  3. Go to the page of the buildbot that failed (https://buildbot.python.org/#/builders/1627/builds/893) and take a look at the build logs.
  4. Check if the failure is related to this commit (5236b02) or if it is a false positive.
  5. 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/893

Summary of the results of the build (if available):

Click to see traceback logs
remote: Enumerating objects: 9, done. remote: Counting objects: 14% (1/7) remote: Counting objects: 28% (2/7) remote: Counting objects: 42% (3/7) remote: Counting objects: 57% (4/7) remote: Counting objects: 71% (5/7) remote: Counting objects: 85% (6/7) remote: Counting objects: 100% (7/7) remote: Counting objects: 100% (7/7), done. remote: Compressing objects: 50% (1/2) remote: Compressing objects: 100% (2/2) remote: Compressing objects: 100% (2/2), done. remote: Total 9 (delta 5), reused 5 (delta 5), pack-reused 2 (from 1)  From https://github.com/python/cpython * branch main -> FETCH_HEAD Note: switching to '5236b0281b91a874b14cf15f3fdef9b7beffb22f'. 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 5236b0281b9 GH-116738: document thread-safety of bisect (GH-136555) Switched to and reset branch 'main' find: ‘build’: No such file or directoryfind: ‘build’: No such file or directoryfind: ‘build’: No such file or directoryfind: ‘build’: No such file or directorymake[2]: [Makefile:3398: clean-retain-profile] Error 1 (ignored)Python/ceval.c: In function ‘_PyEvalFramePushAndInit_Ex’: Python/ceval.c:1916:38: warning: ‘stack_array’ may be used uninitialized [-Wmaybe-uninitialized] 1916 | _PyInterpreterFrame *new_frame = _PyEvalFramePushAndInit( |^~~~~~~~~~~~~~~~~~~~~~~~ 1917 | tstate, func, locals, |~~~~~~~~~~~~~~~~~~~~~ 1918 | newargs, nargs, kwnames, previous |~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1919 | ); |~ Python/ceval.c:1839:1: note: by argument 4 of type ‘const union _PyStackRef *’ to ‘_PyEvalFramePushAndInit’ declared here 1839 | _PyEvalFramePushAndInit(PyThreadState *tstate, _PyStackRef func, |^~~~~~~~~~~~~~~~~~~~~~~ Python/ceval.c:1885:17: note: ‘stack_array’ declared here 1885 | _PyStackRef stack_array[8]; |^~~~~~~~~~~ test_denial_of_service_prevented_int_to_str (test.test_int.IntStrDigitLimitsTests.test_denial_of_service_prevented_int_to_str) Regression test: ensure we fail before performing O(N**2) work. ... FAIL test_denial_of_service_prevented_str_to_int (test.test_int.IntStrDigitLimitsTests.test_denial_of_service_prevented_str_to_int) Regression test: ensure we fail before performing O(N**2) work. ... ok test_disabled_limit (test.test_int.IntStrDigitLimitsTests.test_disabled_limit) ... ok test_int_from_other_bases (test.test_int.IntStrDigitLimitsTests.test_int_from_other_bases) ... ok test_int_max_str_digits_is_per_interpreter (test.test_int.IntStrDigitLimitsTests.test_int_max_str_digits_is_per_interpreter) ... ok test_max_str_digits (test.test_int.IntStrDigitLimitsTests.test_max_str_digits) ... ok test_max_str_digits_edge_cases (test.test_int.IntStrDigitLimitsTests.test_max_str_digits_edge_cases) Ignore the +/- sign and space padding. ... ok test_power_of_two_bases_unlimited (test.test_int.IntStrDigitLimitsTests.test_power_of_two_bases_unlimited) The limit does not apply to power of 2 bases. ... ok test_sign_not_counted (test.test_int.IntStrDigitLimitsTests.test_sign_not_counted) ... ok test_underscores_ignored (test.test_int.IntStrDigitLimitsTests.test_underscores_ignored) ... ok test_denial_of_service_prevented_int_to_str (test.test_int.IntSubclassStrDigitLimitsTests.test_denial_of_service_prevented_int_to_str) Regression test: ensure we fail before performing O(N**2) work. ... ok test_denial_of_service_prevented_str_to_int (test.test_int.IntSubclassStrDigitLimitsTests.test_denial_of_service_prevented_str_to_int) Regression test: ensure we fail before performing O(N**2) work. ... ok test_disabled_limit (test.test_int.IntSubclassStrDigitLimitsTests.test_disabled_limit) ... ok test_int_from_other_bases (test.test_int.IntSubclassStrDigitLimitsTests.test_int_from_other_bases) ... ok test_int_max_str_digits_is_per_interpreter (test.test_int.IntSubclassStrDigitLimitsTests.test_int_max_str_digits_is_per_interpreter) ... ok test_max_str_digits (test.test_int.IntSubclassStrDigitLimitsTests.test_max_str_digits) ... ok test_max_str_digits_edge_cases (test.test_int.IntSubclassStrDigitLimitsTests.test_max_str_digits_edge_cases) Ignore the +/- sign and space padding. ... ok test_power_of_two_bases_unlimited (test.test_int.IntSubclassStrDigitLimitsTests.test_power_of_two_bases_unlimited) The limit does not apply to power of 2 bases. ... ok test_sign_not_counted (test.test_int.IntSubclassStrDigitLimitsTests.test_sign_not_counted) ... ok test_underscores_ignored (test.test_int.IntSubclassStrDigitLimitsTests.test_underscores_ignored) ... ok test_basic (test.test_int.IntTestCases.test_basic) ... ok test_error_message (test.test_int.IntTestCases.test_error_message) ... ok test_int_base_bad_types (test.test_int.IntTestCases.test_int_base_bad_types) Not integer types are not valid bases; issue16772. ... ok test_int_base_indexable (test.test_int.IntTestCases.test_int_base_indexable) ... ok test_int_base_limits (test.test_int.IntTestCases.test_int_base_limits) Testing the supported limits of the int() base parameter. ... ok test_int_memoryview (test.test_int.IntTestCases.test_int_memoryview) ... ok test_int_returns_int_subclass (test.test_int.IntTestCases.test_int_returns_int_subclass) ... ok test_int_subclass_with_index (test.test_int.IntTestCases.test_int_subclass_with_index) ... ok test_int_subclass_with_int (test.test_int.IntTestCases.test_int_subclass_with_int) ... ok test_intconversion (test.test_int.IntTestCases.test_intconversion) ... ok test_invalid_signs (test.test_int.IntTestCases.test_invalid_signs) ... ok test_issue31619 (test.test_int.IntTestCases.test_issue31619) ... ok test_keyword_args (test.test_int.IntTestCases.test_keyword_args) ... ok test_no_args (test.test_int.IntTestCases.test_no_args) ... ok test_non_numeric_input_types (test.test_int.IntTestCases.test_non_numeric_input_types) ... ok test_round_with_none_arg_direct_call (test.test_int.IntTestCases.test_round_with_none_arg_direct_call) ... ok test_small_ints (test.test_int.IntTestCases.test_small_ints) ... ok test_string_float (test.test_int.IntTestCases.test_string_float) ... ok test_underscores (test.test_int.IntTestCases.test_underscores) ... ok test_unicode (test.test_int.IntTestCases.test_unicode) ... ok test_pylong_compute_powers (test.test_int.PyLongModuleTests.test_pylong_compute_powers) ... ok test_pylong_int_divmod (test.test_int.PyLongModuleTests.test_pylong_int_divmod) ... ok test_pylong_int_to_decimal (test.test_int.PyLongModuleTests.test_pylong_int_to_decimal) ... ok test_pylong_int_to_decimal_2 (test.test_int.PyLongModuleTests.test_pylong_int_to_decimal_2) ... skipped "resource 'cpu' is not enabled" test_pylong_misbehavior_error_path_from_str (test.test_int.PyLongModuleTests.test_pylong_misbehavior_error_path_from_str) ... ok test_pylong_misbehavior_error_path_to_str (test.test_int.PyLongModuleTests.test_pylong_misbehavior_error_path_to_str) ... ok test_pylong_roundtrip (test.test_int.PyLongModuleTests.test_pylong_roundtrip) ... ok test_pylong_roundtrip_huge (test.test_int.PyLongModuleTests.test_pylong_roundtrip_huge) ... skipped "resource 'cpu' is not enabled" test_pylong_str_to_int (test.test_int.PyLongModuleTests.test_pylong_str_to_int) ... ok test_whitebox_dec_str_to_int_inner_failsafe (test.test_int.PyLongModuleTests.test_whitebox_dec_str_to_int_inner_failsafe) ... skipped "resource 'cpu' is not enabled" test_whitebox_dec_str_to_int_inner_monster (test.test_int.PyLongModuleTests.test_whitebox_dec_str_to_int_inner_monster) ... ok ====================================================================== FAIL: test_denial_of_service_prevented_int_to_str (test.test_int.IntStrDigitLimitsTests.test_denial_of_service_prevented_int_to_str) Regression test: ensure we fail before performing O(N**2) work. ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/buildbot/buildarea/3.x.cstratak-fedora-stable-s390x.lto-pgo/build/Lib/test/test_int.py", line 611, in test_denial_of_service_prevented_int_to_strself.assertLessEqual(sw_fail_huge.seconds, sw_convert.seconds/2) ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^AssertionError: 0.3910390669479966 not less than or equal to 0.01879003643989563 ---------------------------------------------------------------------- Ran 51 tests in 1.646s FAILED (failures=1, skipped=3) test test_int failed make: *** [Makefile:1032: profile-run-stamp] Error 2

kumaraditya303 added a commit that referenced this pull request Jul 30, 2025
* GH-116738: document thread-safety of bisect (GH-136555) (cherry picked from commit 5236b02) Co-authored-by: Neil Schemenauer <nas-github@arctrix.com> Co-authored-by: Kumar Aditya <kumaraditya@python.org>
hugovk pushed a commit that referenced this pull request Jul 30, 2025
Co-authored-by: Neil Schemenauer <nas-github@arctrix.com>
Agent-Hellboy pushed a commit to Agent-Hellboy/cpython that referenced this pull request Aug 19, 2025
kumaraditya303 pushed a commit to miss-islington/cpython that referenced this pull request Sep 9, 2025
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Labels

docsDocumentation in the Doc dirskip newstestsTests in the Lib/test dirtopic-free-threading

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

4 participants

@nascheme@bedevere-bot@picnixz@kumaraditya303