ChatThrottleLib: Allow direct sending while in a blocked state #18
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
With the changes to support the concept of a blocked queue, there's a slight latency defect possible when CTL is in a state where its priorities consist solely of blocked queues. The direct submission checks in the Send functions eagerly return false in the case where the 'bQueueing' boolean is set, and force their data to be queued up for submission ~potentially up to 0.08s later. The problem is that this boolean is only ever cleared in the case where CTL has no data whatsoever left in any queue. This means that if CTL has only blocked (and as such, totally idle) queues the boolean remains set, and we force messages that could be perfectly sendable right away to instead be queued up unnecessarily. This attempts to resolve the issue as follows; - Firstly, we now clear the 'bQueueing' flag in the OnUpdate function if we enter a state where there's no data that can be *immediately* sent. This means that the flag is unset if we only consist of blocked queues. - The OnUpdate function restores the 'bQueueing' flag when transitioning back from a fully-blocked state to a sendable state. - The direct send checks have been split out to a new function and in addition to checking the state of 'bQueueing', now also check if a named queue exists within a certain priority bracket. If it does, then we do not directly send the message. The new check is required to deal with cases where a caller attempts to send a message on a specific queue while 'bQueueing' is false - as we don't want to directly send a message ahead of everything else in the queue that it's supposed to belong to.
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.
With the changes to support the concept of a blocked queue, there's a slight latency defect possible when CTL is in a state where its priorities consist solely of blocked queues.
The direct submission checks in the Send functions eagerly return false in the case where the 'bQueueing' boolean is set, and force their data to be queued up for submission ~potentially up to 0.08s later.
The problem is that this boolean is only ever cleared in the case where CTL has no data whatsoever left in any queue, including blocked ones. This means that if CTL only has blocked queues the boolean remains set, we force messages that could be perfectly sendable right away on other unblocked prefixes to be queued up unnecessarily.
This attempts to resolve the issue as follows;
The new check is required to deal with cases where a caller attempts to send a message on a blocked queue while 'bQueueing' is false - as we don't want to directly send a message ahead of everything else in the queue that it's supposed to belong to.