pnowojski commented on pull request #18173: URL: https://github.com/apache/flink/pull/18173#issuecomment-1002588560
> Maybe we can pass the local pool instance or some abstract interface to NetworkBufferPool when calling any NetworkBufferPool method, then check if the corresponding lock is locked before acquiring the NetworkBufferPool lock, Do you mean introducing one more lock, above `NetworkBufferPool`? Why would that be necessary? Wouldn't it be enough to pass some locking "object" or "abstraction" to `LocalBufferPool` (when it's being created in the `NetworkBufferPool`, and make sure that the `LocalBufferPool` locks on the object before calling `NetworkBufferPool`. And then inside `NetworkBufferPool` only assert/check state that the lock has been acquired? > maybe in a different PR I'm not sure what would be a benefit of having this fix in a separate PR? It's not a release blocker, so I think we can fix it properly in this PR/issue. -- 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: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
