[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Eric Snow added the comment: Thanks, Alexey. I've merged the PR. As to finding and fixing leaks, I'd welcome any help you have to offer. :) Feel free to open an separate issue. One issue covering all the ones you find should be enough. If you find any non-trivial leaks or other issues

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

miss-islington added the comment: New changeset 36433221f06d649dbd7e13f5fec948be8ffb90af by Miss Islington (bot) (Alexey Izbyshev) in branch 'master': bpo-35984: _xxsubinterpreters: Fix memory leak in _channel_send() (GH-11845)

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Alexey Izbyshev added the comment: Thank you for your introduction about _xxsubinterpreters, Eric. This particular leak is easy: it's right in _channel_send(). I've submitted a PR. I've also done a quick scan of neighboring code, and it seems there are other leaks as well, e.g.: *

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Change by Alexey Izbyshev : -- keywords: +patch pull_requests: +11876 stage: test needed -> patch review ___ Python tracker ___ ___

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Eric Snow added the comment: FYI, the _xxsubinterpreters module serves several purposes. First, it allows us to more thoroughly test the subinterpreter functionality of CPython (doing so via _testembed and _testcapi is messy), effectively in test__xxsubinterpreters. Second, it is the

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Eric Snow added the comment: Alexey, even though the refleak does not appear to be related to your earlier PR, you are welcome to keep working on this issue. :) If you do then please add me as a reviewer on whatever PR you make. Also, I'd be glad to answer any questions you have about the

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

STINNER Victor added the comment: > I'll look into it later today. An obvious guess is that my test simply > exposed an existing leak because the exception code path wasn't tested before > AFAIK, but I need to check it. Right. I don't think that your change introduced a regression.

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Alexey Izbyshev added the comment: I'll look into it later today. An obvious guess is that my test simply exposed an existing leak because the exception code path wasn't tested before AFAIK, but I need to check it. -- assignee: -> izbyshev ___

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

STINNER Victor added the comment: I'm mentoring Alexey Izbyshev. @Alexey: Do you want to work on this issue? -- ___ Python tracker ___

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

STINNER Victor added the comment: $ ./python -m test.bisect_cmd -R 3:3 test__xxsubinterpreters found: test.test__xxsubinterpreters.ShareableTypeTests.test_non_shareable_int -- ___ Python tracker

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

Change by STINNER Victor : -- nosy: +izbyshev ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

STINNER Victor added the comment: > Bisecting shows 16f842da3c862d76a1177bd8ef9579703c24fa5a is the first bad > commit. This was introduced in PR11822. That's issue bpo-35972. -- ___ Python tracker

[issue35984] test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=1

New submission from Pablo Galindo Salgado : The reafleak buildbots have detected memory block leaks in test__xxsubinterpreters: Ran 112 tests in 4.721s OK (skipped=5) . test__xxsubinterpreters leaked [3, 4, 3] memory blocks, sum=10 1 test failed again: test__xxsubinterpreters == Tests