New submission from Alexander Mohr <[email protected]>:
while investigating https://github.com/boto/botocore/issues/1464 I used
tracemalloc (like I've done before in 3.5.2) to try to figure out where the
leak was. To my surprise tracemalloc listed stacks that didn't make any sense.
Strangest example is the top result when running the attached script against
python 3.6.5 in the following manner: PYTHONMALLOC=malloc /valgrind/bin/python3
/tmp/test.py head_object
The top hit is listed as:
21 memory blocks: 4.7 KiB
File "/tmp/test.py", line 28
raise
File "/tmp/test.py", line 47
test(s3_client)
File "/tmp/test.py", line 65
main()
how is it that the "raise" is a leak? It doesn't make any sense to me
specially given that no contexts are used in that call. Further that line is
never hit because the exception is not re-thrown.
Further a bunch of regular expression allocs don't make any sense either given
that I've cleared the cache before doing snapshots.
if someone could shed some light on why this is/isn't a bug that would be
great. It seems to me that the callstacks are not related at all to the leak.
----------
components: Library (Lib)
files: tracemalloc_test.py
messages: 317002
nosy: thehesiod
priority: normal
severity: normal
status: open
title: strange tracemalloc results
versions: Python 3.6
Added file: https://bugs.python.org/file47600/tracemalloc_test.py
_______________________________________
Python tracker <[email protected]>
<https://bugs.python.org/issue33565>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
