IdeaWell

Project ideas from Hacker News discussions.

Asahi Linux on M3

Original Article

Hacker News Discussion

📝 Discussion Summary (Click to expand)

Theme 1 – Apple’s silicon is macOS‑centric and lacks open documentation
- “Their Silicon was written to be OS agnostic, Apple’s was not.” – BloodyIron
- “Apple silicon has features specifically designed for macOS or software running on macOS. e.g.: Rosetta 2.” – bayindirh
- “Apple doesn’t … post documents to explain how to use their hardware. … Apple consumes it, so Apple always has a customer.” – MBCook

Theme 2 – Reverse‑engineering Apple’s custom chips is a massive, impressive undertaking
- “Reverse-engineering Apple's custom silicon is basically the modern equivalent of repairing a spaceship while it's actively launching. Incredible work by the Asahi team!” – Krish1577
- “The market for people who want to install Linux on a Mac laptop is very tiny… not worth the huge effort investment needed to document an entire platform for public consumption.” – Aurornis
- “Intel and AMD post documents… Apple doesn’t. So it’s not an equivalent effort by any stretch.” – MBCook

Theme 3 – Apple’s business incentives discourage official Linux support (walled‑garden lock‑in)
- “They want control on the walled garden… If they gave you a way to install linux easily then you could use it far past their intended lifespan and wouldnt keep you locked into the apple ecosystem.” – alt227
- “Because then they'll have to compete with Linux, which opens them up to someone else making a cheaper computer that does the same stuff.” – ajross
- “Apple could contribute as well… Why are you demanding a user does that for free instead of the big corp that made the platform?” – izacus

🚀 Project Ideas

Generating project ideas…

Asahi Spec Hub

Summary

  • Centralized, community‑maintained repository of Apple Silicon hardware specifications (registers, memory ordering, controller interfaces) gathered from reverse‑engineering efforts.
  • Provides searchable docs, diff‑tracking between chip generations, and exportable C header files for driver developers.

Details

Key Value
Target Audience Linux kernel developers, reverse engineers, Asahi contributors
Core Feature Collaborative spec database with versioned diffs and auto‑generated header exports
Tech Stack React/TypeScript frontend, Node.js/Express backend, PostgreSQL, Git‑LFS for binary blobs
Difficulty Medium
Monetization Hobby

Notes

  • HN users lament the lack of docs: “Whatever1: Why Apple does not at least provide the specs to the community? So much wasted effort when they can just share a pdf.” – a spec hub directly addresses this pain point.
  • Lowering the barrier to entry will encourage more contributors and spark discussion about open hardware documentation on Apple Silicon.

Thunderbolt Shim for Apple Silicon

Summary

  • Open‑source kernel module that adds Thunderbolt 3/4 support on Apple Silicon Macs by reusing Intel’s publicly available Thunderbolt controller specs and mapping them to the undocumented Apple registers discovered by the community.
  • Enables plug‑and‑play Thunderbolt peripherals (displays, docks, external GPUs) under Linux without waiting for a full vendor driver.

Details

Key Value
Target Audience Linux laptop users needing Thunderbolt peripherals (external displays, docks, GPUs) on Apple Silicon
Core Feature Kernel driver that translates Intel Thunderbolt register maps to Apple Silicon equivalents, providing hot‑plug support
Tech Stack C (Linux kernel), Makefile, optional eBPF for register tracing; relies on community‑supplied register map from Asahi Spec Hub
Difficulty High
Monetization Hobby

Notes

  • Commenters note the feasibility: “bigyabai: Intel and AMD both support Thunderbolt on Linux, it's not impossible per‑se.” – a shim leverages existing specs to close the gap.
  • Delivers immediate usability gains for docks and external GPUs, likely to generate lively discussion about open driver strategies.

Apple Silicon GPU Compute Runtime (ASGCR)

Summary

  • Userspace library that exposes Apple GPU’s compute cores via a Vulkan‑compatible compute API, allowing ML frameworks (llama.cpp, PyTorch) to run models directly on M‑series GPUs under Linux.
  • Bridges the gap until a full graphics driver is ready, delivering near‑native performance for AI workloads.

Details

Key Value
Target Audience ML engineers, researchers, developers wanting to run LLMs or other compute workloads on Apple Silicon Macs under Linux
Core Feature Userspace compute runtime that reverse‑engineers GPU command buffers and exposes Vulkan‑style compute queues
Tech Stack Rust (core library), Vulkan SDK, libdrm, Python bindings; utilizes existing AGX reverse‑engineering tooling (e.g., Asahi’s agxcommand decoder)
Difficulty Medium‑High
Monetization Hobby

Notes

  • Users express desire for GPU compute: “xrd: Is this someone that only wants to run Linux and wants to use their GPU on Linux? …” and “brundolf: … M1 and M2 have very functional GPU drivers.” – ASGCR targets the missing compute piece for M3 and later.
  • Enables AI workloads on Linux, likely to spark discussion about cross‑platform ML performance and motivate further GPU driver efforts.

Read Later