[issue7825] test_threadsignals leaks references

Florent Xicluna added the comment: Now it's fixed for 2.7 with r83187. (thanks for the hint, Antoine) No issue on 3.x, AFAICT. -- resolution: -> fixed stage: patch review -> committed/rejected status: open -> closed ___ Python tracker

[issue7825] test_threadsignals leaks references

Antoine Pitrou added the comment: I haven't seen test_threadsignals leaking, either on 3.1 or 3.2. Was it a transient problem? Note that if you want for test-specific threads to end, you can use the reap_threads() utility from the test.support module. -- nosy: +pitrou ___

[issue7825] test_threadsignals leaks references

Florent Xicluna added the comment: I did not commit the patch because I'm not sure it is the right approach. I need someone else to comment on this. -- ___ Python tracker ___ ___

[issue7825] test_threadsignals leaks references

Mark Lawrence added the comment: The patch is small and easily applied, but no point me trying it as I've only a Windows box. Also note that my earlier comment was incorrect as I misread the original post. -- ___ Python tracker

[issue7825] test_threadsignals leaks references

Mark Lawrence added the comment: Has the underlying code been fixed or could we still use the patch to test_threadsignals? -- nosy: +BreamoreBoy ___ Python tracker ___ _

[issue7825] test_threadsignals leaks references

New submission from STINNER Victor : test_threadsignals leaks references. I'm unable to understand where. Maybe somewhere around Py_AddPendingCall()? - $ ./python Lib/test/regrtest.py -R 3:2: test_threadsignals test_threadsignals beginning 5 repetitions 12345 . test_threadsignals lea