🌗 RPython GC 的分配速度有多快？
➤ 深入分析 RPython 垃圾回收器的分配效率
✤ https://pypy.org/posts/2025/06/rpython-gc-allocation-speed.html
本文探討了 RPython 垃圾回收器 (GC) 的分配速度。作者透過一個簡單的基準測試程式，測量了在 64 位元架構上分配包含單一整數欄位的物件的效能。測試結果顯示，RPython GC 在沒有初始化欄位的情況下，可以以高達 34.35 GB/s 的速度分配物件，即使初始化欄位，也能達到 29.7 GB/s。作者將 RPython GC 的效能與 Boehm GC 進行比較，並使用 perf 工具分析了快取命中率、指令週期和分支等指標。研究發現，RPython GC 的分配速度非常快，且垃圾回收佔總執行時間的比例相對較小。
+ 這篇文章對於瞭解 RPython 垃圾回收器的效能非常有幫助，提供了詳細的測試數據和分析結果。
+ 我一直很好奇 PyPy 的效能，這篇文章讓我對 RPython
#程式設計 #垃圾回收 #PyPy #RPython #效能分析

🚀🚀 "Tech genius discovers the astonishing speed of #RPython GC allocation... by writing *another* benchmark program. Because, clearly, the world was waiting with bated breath for this groundbreaking revelation instead of solving actual problems. 🙄"
https://pypy.org/posts/2025/06/rpython-gc-allocation-speed.html #TechGenius #Benchmarking #GCAllocation #GroundbreakingRevelation #ProblemSolving #HackerNews #ngated

How fast can the RPython GC allocate?

https://pypy.org/posts/2025/06/rpython-gc-allocation-speed.html

#HackerNews #RPython #GC #Allocation #RPython #Performance #Memory #Management #Hacker #News

I finally think I've figured out a basic #Nix heap using the #Haskell/#GHC technique of flagging every heap object to detect infinite loops. Cache invalidation really is hard... Fortunately, Nix is pure, so I don't have to invalidate any heap objects directly. #RPython makes it easy to hack, although it's still a bit of a slog to debug segfaults.

I think I might have to actually look at #CppNix now, or at least reread Eelco's thesis. I need to understand the differences between evaluation and instantiation.

I apparently was overly emotional last night and started an #RPython version of #Nix. I have about half a parser so far.

If I keep working on this, the goals will be to have a smaller footprint in terms of code and CPU (but not necessarily RAM) than CppNix. This sounds quite achievable because the RPython toolkit comes with a JIT generator: insert interpreter, receive JIT.

Contributions will be welcome once everything is available to the public. Wait a day or two, please.

Germane to the current #NixOS / #nixpkgs drama, I think.

What would be a good and punny name for an #RPython implementation of a #Nix evaluator?

New blog post: Which Interpreters are Faster, AST or Bytecode?

It gives a high-level overview over our OOPLSA paper, which investigates interpreters on top of metacompilation systems.
Will be presented next week at @splashcon.

https://stefan-marr.de/2023/10/ast-vs-bytecode-interpreters/

#graalvm #truffle #rpython

Apparently not a joke. A lot of people seem to be quite enthused about this.

#Pypy, a Python runtime in Python, tried this ages ago. Their #RPython cross-compilation toolchain was extremely useful, built up to target #WASM as a back-end. Really helped push Web Assembly forward.

Conclusion: it's not worth it. It's almost never worth it.

Instead: transpile to ES6 for native runtime, zero download overhead, and zero startup overhead. JavaScripthon is an excellent one I've even contributed to.