This repository has been archived by the owner on Sep 26, 2019. It is now read-only.
Fix deadlock scenario in AsyncOperationProcessor and re-enable WorldStateDownloaderTest #1126
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
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
PR description
AsyncOperationProcessor could wind up deadlocked if it is processing the last requests in a download which complete while its output pipe is full. Instead of blocking waiting to be able to output the completed results, it loops back round to see if it can start more tasks but winds up blocked on the
pipe.get()call. That call will never return because the pipeline is just waiting for the final requests to complete and will then close but we're now stuck.Since the aim of each pipeline stage is to keep its output buffer full, we now block until the outputs can be added, avoiding the deadlock. This potentially means some requests which could have been started are delayed but if the downstream stages are the bottleneck that won't matter anyway.