sollhui opened a new pull request, #63967: URL: https://github.com/apache/doris/pull/63967
### What problem does this PR solve? PackedFileWriter buffers data for files smaller than small_file_threshold_bytes before deciding whether to pack them into a packed file or switch to direct write. The buffered data is stored in a std::string. After the buffered data is flushed to the inner writer or submitted to PackedFileManager, the old code only called clear(), which resets size but keeps capacity. When segment file writers are still retained by upper-level rowset structures after close, this retained capacity can keep a large amount of memory alive and show up under PackedFileWriter::appendv in memory profiling: <img width="800" height="1180" alt="image" src="https://github.com/user-attachments/assets/7e0e2c40-c35b-4bfc-b45b-aeed31c29771"; /> This change reserves the final append size before buffering to reduce repeated std::string growth, and releases the buffer capacity after the data has been flushed or submitted. ### Release note None ### Check List (For Author) - Test <!-- At least one of them must be included. --> - [ ] Regression test - [ ] Unit Test - [ ] Manual test (add detailed scripts or steps below) - [ ] No need to test or manual test. Explain why: - [ ] This is a refactor/code format and no logic has been changed. - [ ] Previous test can cover this change. - [ ] No code files have been changed. - [ ] Other reason <!-- Add your reason? --> - Behavior changed: - [ ] No. - [ ] Yes. <!-- Explain the behavior change --> - Does this need documentation? - [ ] No. - [ ] Yes. <!-- Add document PR link here. eg: https://github.com/apache/doris-website/pull/1214 --> ### Check List (For Reviewer who merge this PR) - [ ] Confirm the release note - [ ] Confirm test cases - [ ] Confirm document - [ ] Add branch pick label <!-- Add branch pick label that this PR should merge into --> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
