Repository navigation
test_type_comments leaks references and memory blocks #80062
Description
| BPO | 35881 |
|---|---|
| Nosy | @gvanrossum, @vstinner, @ambv, @pablogsal, @tirkarthi |
| Superseder |
Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.
Show more details
GitHub fields:
assignee = None
closed_at = <Date 2019-02-01.14:09:03.830>
created_at = <Date 2019-02-01.14:02:45.794>
labels = ['interpreter-core', '3.8', 'tests']
title = 'test_type_comments leaks references and memory blocks'
updated_at = <Date 2019-02-01.15:14:41.462>
user = 'https://github.com/pablogsal'bugs.python.org fields:
activity = <Date 2019-02-01.15:14:41.462>
actor = 'vstinner'
assignee = 'none'
closed = True
closed_date = <Date 2019-02-01.14:09:03.830>
closer = 'vstinner'
components = ['Interpreter Core', 'Tests']
creation = <Date 2019-02-01.14:02:45.794>
creator = 'pablogsal'
dependencies = []
files = []
hgrepos = []
issue_num = 35881
keywords = []
message_count = 5.0
messages = ['334676', '334679', '334680', '334681', '334686']
nosy_count = 5.0
nosy_names = ['gvanrossum', 'vstinner', 'lukasz.langa', 'pablogsal', 'xtreak']
pr_nums = []
priority = 'normal'
resolution = 'duplicate'
stage = 'resolved'
status = 'closed'
superseder = '35879'
type = None
url = 'https://bugs.python.org/issue35881'
versions = ['Python 3.8']
Activity
pablogsal commentedon Feb 1, 2019
The refleak buildbots have detected that test_type_comments is leaking references. Example failure:
https://buildbot.python.org/all/#/builders/1/builds/490/steps/4/logs/stdio
test_type_comments leaked [37, 37, 37] references, sum=111
test_type_comments leaked [11, 12, 11] memory blocks, sum=34
1 test failed again:
test_type_comments
Bisecting shows that PR11645 is the one that introduced the refleaks:
https://github.com/python/cpython/pull/11645/files#
After some hours of debugging, I have identified the (possible) cause of the majority of the refleaks.
The type_comments created in ast.c with new_type_comment() never reaches 0 reference counts. This **seems** to be because the cleanup for the stmt_ty elements where the type_comments are included never call DECREF on these elements when destroyed unless I am missing something fundamental.
tirkarthi commentedon Feb 1, 2019
This looks like the same as https://bugs.python.org/issue35879
vstinner commentedon Feb 1, 2019
Yeah, as usual Pablo is way too slow: he posted the same bug 1h30 after me.
pablogsal commentedon Feb 1, 2019
:(
vstinner commentedon Feb 1, 2019
I'm kidding :-) Sorry for not reviewing emails from buildbot-status.