Replace stack pool mutex with lock-free queue #4196
Merged
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Outdated suggestions cannot be applied.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.
Alternate implementation of #3922 using a lock-free queue instead of thread locals. I benchmarked this by creating 100 threads and running a simple WASM function in a loop 30,000 times in each thread. This PR reduces the total time by ~75% on my 10-core CPU, and does not create meaningful impact on single-threaded performance. Also, since the queue is shared among threads, there is no risk of high memory consumption in highly concurrent scenarios.