mpsc: completely split bounded and unbounded and optimize unbounded receiver #1337
Conversation
This file contains hidden or 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
The major downside of code sharing is performance overhead (I've noticed it in profiler) in an unbounded queue: `clone`/`drop` of `UnboundedSender` perform unnecessary atomic increment/decrement of `num_senders` field The downside of this patch is the increase of LOC because of a copy-paste of `next_message` function. `next_message` functions could be merged if `unpark_one()` and `dec_num_messages()` calls could be swapped. And I don't see why `dec_num_messages()` cannot be called before `unpark_one()`. However the original comment by Carl Lerche says that the order is important, so I decided to keep it as is. rust-lang@193b53a
Mechanically inline `UnboundedInner` into `BoundedInner`
Instead have a local non-atomic `num_received` variable and considers queue exhausted when the number of sent messages is equal to the number of received messages. Before: ``` test unbounded_uncontended ... bench: 111,980 ns/iter (+/- 8,636) ``` After: ``` test unbounded_uncontended ... bench: 103,043 ns/iter (+/- 7,359) ``` It's hard (or impossible, I tried to and failed) to do the same optimization for the bounded queue, because bounded sender needs to atomically track message queue size. Thus previous commit completely splits bounded and unbounded queue implementations.
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Outdated suggestions cannot be applied.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.
On top of #1326 two commits:
The first finally splits bounded and unbounded queues.
The second replaces atomic decrement of
num_messageswith a local increment ofUnboundedReceiver::num_sentfield.Bench before:
After:
(Edit: numbers after revert of atomic waker change.)
It's hard (or impossible, I tried to and failed) to do the same
optimization for the bounded queue because bounded sender needs
to atomically track message queue size. Thus the previous commit
completely splits bounded and unbounded queue implementations.