Description

sergey-miryanov

opened on Aug 14, 2026

Issue body actions

Bug report

Bug description:

If an external tool sampling GC statistics does not pause the target process (and we designed get_gc_stats not to pause), it can read torn data under certain circumstances.

There are two distinct problems:

1. Weakly ordered platforms: The stores of ts_start and ts_stop may be reordered by the CPU/optimizer. We need to add memory barriers to mitigate this.

2. Large memcpy window: We are copying a sufficiently large memory region via memcpy. If two or more GC cycles occur during this copy, we can end up with inconsistent data spanning multiple GC states.

@maurycy has an idea to use a sequence counter for consistency checks — I think he should definitely give it a try.

cc @pablogsal

CPython versions tested on:

CPython main branch

Operating systems tested on:

No response

Linked PRs

• gh-155811: Add a seqcount to gc_stats to prevent torn reads #155828
• [3.15] gh-155811: Add a seqcount to gc_stats to prevent torn reads (GH-155828) #158829

Reacted by Maurycy Pawłowski-Wieroński

Activity

1. sergey-miryanov

   added

   type-bugAn unexpected behavior, bug, or errorAn unexpected behavior, bug, or error

   3.15bugs and security fixesbugs and security fixes

   3.16new features, bugs and security fixesnew features, bugs and security fixes

   interpreter-core(Objects, Python, Grammar, and Parser dirs)(Objects, Python, Grammar, and Parser dirs)

   extension-modulesC modules in the Modules dirC modules in the Modules dir

   on Aug 14, 2026
2. sergey-miryanov commented on Aug 14, 2026

   

   sergey-miryanov

   on Aug 14, 2026

   ContributorAuthor

   More actions

   I don't think we should retry more than once if the GC stats read is inconsistent. Complex retry logic should be handled by the external tools.

   Reacted by Maurycy Pawłowski-Wieroński
3. pablogsal

   added 2 commits that reference this issue on Oct 5, 2026

   pythongh-155811: Retry an inconsistent GC snapshot once

   577305e

   gh-155811: Add a seqcount to `gc_stats` to prevent torn reads (#155828)

   5fecd44
4. maurycy commented on Oct 5, 2026

   

   maurycy

   on Oct 5, 2026

   Contributor

   More actions

   @sergey-miryanov All good now? :)
5. sergey-miryanov commented on Oct 5, 2026

   

   sergey-miryanov

   on Oct 5, 2026

   ContributorAuthor

   More actions

   Since this is a bug fix, I think it should be backported.

   I don't think we'll get this into 3.15.0, so I'm perfectly fine with waiting for 3.15.1.

   cc @hugovk
6. hugovk commented on Oct 5, 2026

   

   hugovk

   on Oct 5, 2026

   Member

   More actions

   Yep, this can wait for .1

   • Core team: very little will be merged into the 3.15 branch between now and 3.15.0 final. Docs improvements always welcome.

   https://discuss.python.org/t/python-3-15-following-tradition-lets-have-a-surprise-rc3/109313

   Reacted by Maurycy Pawłowski-Wieroński and Sergey Miryanov
7. pablogsal

   added a commit that references this issue on Oct 5, 2026

   [3.15] gh-155811: Add a seqcount to `gc_stats` to prevent torn reads (G…

   0281240
8. sergey-miryanov

   closed this as completedon Oct 5, 2026
9. pablogsal

   added a commit that references this issue on Oct 5, 2026

   Revert "[3.15] pythongh-155811: Add a seqcount to `gc_stats` to preve…

   693eccc
10. Kronuz

    added 6 commits that reference this issue on Oct 5, 2026

    pythongh-146527, pythongh-155811: Backport out-of-process GC statisti…

    a9a6d3a

    pythongh-146527, pythongh-155811: Backport out-of-process GC statisti…

    5f652ff

    pythongh-146527, pythongh-155811: Backport out-of-process GC statisti…

    70b3c2d

    pythongh-146527, pythongh-155811: Backport out-of-process GC statisti…

    3453253

    pythongh-146527, pythongh-155811: Backport out-of-process GC statisti…

    c8d47d4

    pythongh-146527, pythongh-155811: Backport out-of-process GC statisti…

    7ecdc72
11. sergey-miryanov

    reopened this on Oct 7, 2026
12. hugovk

    closed this as completedon Oct 11, 2026
13. sergey-miryanov commented on Oct 11, 2026

    

    sergey-miryanov

    on Oct 11, 2026

    ContributorAuthor

    More actions

    @hugovk Backport was reverted in c49b3dc, so I reopened it. Do you mind I reopen it again? :)
14. pablogsal

    added a commit that references this issue on Oct 11, 2026

    [3.15] Restore deferred backports for 3.15.1, including GH-158874 (#1…

    4ceb3fa
15. hugovk commented on Oct 11, 2026

    

    hugovk

    on Oct 11, 2026

    Member

    More actions

    Re-added in #158882?

    Reacted by Maurycy Pawłowski-Wieroński and Sergey Miryanov

    Reacted by Sergey Miryanov and Maurycy Pawłowski-Wieroński
16. sergey-miryanov commented on Oct 11, 2026

    

    sergey-miryanov

    on Oct 11, 2026

    ContributorAuthor

    More actions

    Yes, I missed that. Thanks!

    Reacted by Hugo van Kemenade

Issue actions