test: benchmark MPMC on Tokio and replace Pollster - #300
Merged
Conversation
tisonkun
enabled auto-merge (squash)
September 11, 2026 06:50
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
MPMC comparisons currently measure dedicated OS threads driving async operations through Pollster. Add Tokio task benchmarks for asyncband, async-channel, and Flume, covering bounded/unbounded queues, current-thread/four-worker runtimes, and four producer/consumer topologies. Share the fixtures between the asyncband-only and ecosystem suites.
Replace Pollster in integration tests and channel benchmarks with the existing
asyncband::blocking::FutureExtAPI, and remove the workspace dependency.Design Notes
Keep the dedicated-thread workload as
blocking_threads: it measures synchronous callers bridging into async channel APIs, including polling and parking overhead. The newtokio_tasksworkload runs every producer and consumer in a spawned task; consumers compete until disconnection and drain all messages.Regression coverage checks completion, received counts/checksums, capacity-one backpressure, uneven consumer counts, and last-sender drop. Related to #298.
Validation
cargo x lint,cargo x check,cargo x test,cargo +1.86.0 x test,cargo x miri, andcargo x bench --no-run.