fix(divan): support threaded walltime benchmarks - #189
Draft
not-matthias wants to merge 2 commits into
Draft
Conversation
Exercise Divan native threads with four workers in the walltime benchmark suite. Keep the simulation compatibility path single-threaded because its benchmark options do not expose the native threads setting.
Serialize the process-wide instrument-hooks transport across all Rust FFI handles. Aggregate worker timestamps into one benchmark marker envelope per Divan sampling round so benchmark bodies remain parallel without corrupting FIFO requests.
Hooray! CodSpeed harness just leveled up!The base and head of this comparison were measured with different runner settings, so their benchmark values are not directly comparable. What changed between base and head:
Re-run the base with the same settings to get a valid performance comparison. Comparing 
|
Unable to generate the flame graphsThe performance report has correctly been generated, but there was an internal error while generating the flame graphs for this run. We're working on fixing the issue. Feel free to contact us on Discord or at support@codspeed.io if the issue persists. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Prevent threaded Divan walltime benchmarks from concurrently writing to the process-wide instrument-hooks FIFOs.
The Rust FFI boundary now serializes every transport operation across all handles. Divan workers record monotonic bounds locally, then the coordinator emits one marker envelope for the complete parallel sampling round. Benchmark bodies remain parallel.
A native four-worker benchmark keeps this path covered by the existing walltime integration matrix. The simulation compatibility path remains single-threaded because it does not expose Divan's native thread option.