Skip to content

test_mmap are leaked #104698

Closed
Closed
test_mmap are leaked#104698
Labels
type-bugAn unexpected behavior, bug, or error

Description

@Eclips4
Eclips4
opened on May 20, 2023

Tried on current main branch.

 ./python -m test -R 3:3 test_mmap
Running Debug|x64 interpreter...
0:00:00 Run tests sequentially
0:00:00 [1/1] test_mmap
beginning 6 repetitions
123456
......
test_mmap leaked [13, 13, 13] references, sum=39
test_mmap failed (reference leak)

== Tests result: FAILURE ==

1 test failed:
    test_mmap

Total duration: 1.3 sec
Tests result: FAILURE

OS: Windows 10 & WSL Ubuntu 20.04

Linked PRs

Activity

  1. Eclips4
    added
    type-bugAn unexpected behavior, bug, or error
    on May 20, 2023

  2. Eclips4 commented on May 20, 2023

    @Eclips4
    Eclips4
    on May 20, 2023
    MemberAuthor

    PR which introduce it: #103990

  3. AlexWaygood commented on May 20, 2023

    @AlexWaygood
    AlexWaygood
    on May 20, 2023
    Member

    PR which introduce it: #103990

    Cc. @Agent-Hellboy / @JelleZijlstra / @sunmy2019

  4. chgnrdv commented on May 20, 2023

    @chgnrdv
    chgnrdv
    on May 20, 2023
    Contributor

    It seems like Py_buffer in mmap_gfind, mmap_write_method and mmap_ass_subscript should be released if CHECK_VALID fails.

  5. Eclips4 commented on May 20, 2023

    @Eclips4
    Eclips4
    on May 20, 2023
    MemberAuthor

    It seems like Py_buffer in mmap_gfind, mmap_write_method and mmap_ass_subscript should be released if CHECK_VALID fails.

    Yeah, you're right. I'll soon send a PR.

  6. sunmy2019 commented on May 20, 2023

    @sunmy2019
    sunmy2019
    on May 20, 2023
    Member

    It seems like Py_buffer in mmap_gfind, mmap_write_method and mmap_ass_subscript should be released if CHECK_VALID fails.

    Yeah, I forgot about it.

    I'll soon send a PR.

    Thanks!

  7. Eclips4
    added a commit that references this issue on May 20, 2023

    pythongh-104698: Fix reference leak in mmapmodule.c

    f973151

  8. JelleZijlstra commented on May 20, 2023

    @JelleZijlstra
    JelleZijlstra
    on May 20, 2023
    Member

    Sorry for this, should have run the refleak buildbots before merging. Thanks @chgnrdv for the diagnosis and @Eclips4 for the PR.

  9. pablogsal
    added a commit that references this issue on May 21, 2023

    gh-104698: Fix reference leak in mmapmodule.c (#104700)

    99b6418
  10. miss-islington
    added a commit that references this issue on May 21, 2023

    pythongh-104698: Fix reference leak in mmapmodule.c (pythonGH-104700)

    914dbdc
  11. JelleZijlstra
    closed this as completedon May 21, 2023
  12. JelleZijlstra
    added a commit that references this issue on May 21, 2023

    [3.11] gh-104698: Fix reference leak in mmapmodule.c (GH-104700) (#10…

    b2e0201

  13. sunmy2019 commented on May 21, 2023

    @sunmy2019
    sunmy2019
    on May 21, 2023
    Member

    Sorry for this, should have run the refleak buildbots before merging.

    Can we make it mandatory?

  14. Eclips4 commented on May 21, 2023

    @Eclips4
    Eclips4
    on May 21, 2023
    MemberAuthor

    Sorry for this, should have run the refleak buildbots before merging.

    Can we make it mandatory?

    I think, it's a good idea. Also, running refleak buildbots make sense in PR's which changes/adding C code. So, there need a check.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    type-bugAn unexpected behavior, bug, or error

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions