AI is changing the world. Don't stay behind. Clear summaries, community insight, delivered without the noise. Subscribe to never miss a beat.

Bun has an open PR adding shared-memory threads to JavaScriptCore

Shared-memory threads for JavaScriptCore (experimental, not working yet) by Jarred-Sumner · Pull Request #249 · oven-sh/WebKit

github.com

June 20, 2026

293 min read

🔥🔥🔥🔥🔥

54/100

Summary

Shared-memory threads for JavaScriptCore allow functions to run on separate threads while sharing the same heap and objects, eliminating the need for structured cloning or message passing. This experimental feature is currently not functional.

Key Takeaways

Shared-memory threads for JavaScriptCore allow functions to run on another thread without message passing or cloning, sharing the same heap and objects directly.
The new threading model enables parallel execution without a global lock, passing all thread test suites, but is still in an experimental phase with additional cleanup and testing required.
The API allows for easy thread creation and management, enabling closures to access shared variables directly, improving performance and simplifying code compared to traditional worker models.
A shared cache can be implemented using a single Map, allowing for efficient memoization across threads without the need for separate caches or complex inter-thread communication.

Read original article

Community Sentiment

Mixed

Positives

The potential for true parallelism in JavaScript engines could revolutionize performance, making applications more efficient and responsive.
The excitement around implementing shared-global multithreading indicates a significant advancement in JavaScript capabilities, which could lead to more powerful web applications.

Concerns

Concerns about the trustworthiness of the codebase and the oversight of a single individual raise significant doubts about the project's stability and reliability.
The perception that AI tools are not effective at multi-threading code highlights a critical limitation in current AI capabilities, which could hinder development efforts.
The PR's description is criticized for being filled with LLMisms, suggesting a lack of clarity and readability that could impede understanding and adoption.