Memory leak in Python target fixed #246
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
edwardalee
approved these changes
Jun 28, 2023
lhstrh
reviewed
Jun 28, 2023
petervdonovan
approved these changes
Jun 29, 2023
There was a problem hiding this comment.
Other than the lingering comments, this looks good to me. I like that it hooks the reference count into our destructor mechanism -- this seems like the right way to get our memory management system to interact properly with CPython's. As before, I'm really grateful that I didn't have to do this because this looks really hard.
petervdonovan
requested changes
Jun 29, 2023
There was a problem hiding this comment.
We should not merge this yet because there are test failures; see the discussion at lf-lang/lingua-franca#1873
Co-authored-by: Edward A. Lee <[email protected]>
It was incorporated to assist with debugging.
Co-authored-by: Edward A. Lee <[email protected]>
Co-authored-by: Edward A. Lee <[email protected]>
Co-authored-by: Marten Lohstroh <[email protected]>
Co-authored-by: Marten Lohstroh <[email protected]>
The function is called by the deserializer during federated execution.
This commit fixes segfault for unfederated execution, but leads to memory leak during federated execution.
petervdonovan
force-pushed
the
memory_leak_fix
branch
2 times, most recently
from
df0379b to
d28a9f5
Compare
petervdonovan
approved these changes
Jul 11, 2023
lhstrh
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This pull request addresses a memory leak issue observed in the current implementation. As shown in the following screenshot, it is noticeable that with each iteration, the size of Python objects constructed using C extensions continues to grow.
This problem could be attributed to several key issues, which are as follows:
PyObject Reference Count: The reference counts for the majority of
PyObjectsare not being appropriately managed.Port Assignment: Rather than using the
_LF_SETmacro to assign values to thegeneric_port_instance_structin the current implementation, the more suitable method would be to uself_set_tokenand define a destructor for the token to ensure proper garbage collection.Function Definition: The existing
_lf_free_token_valuefunction is incorrect and fails to invoke the destructor.Comments have been added to the code within the commits to provide additional context and insight into the changes made. Other related changes have also been made in the code generator repository (LINK).
Please feel free to comment if further modifications are required. Similar issues persist during the federated multiport setting. To address this, a separate PR will be submitted later.