{
  "id": 14335801,
  "title": "Torvalds Let Claude Code Write 8 Kernel Patches: 3x Reads on One Laptop, Not Merged",
  "url": "https://urgent.news/2026/10/14/torvalds-let-claude-code-write-8-kernel-patches-3x-reads-on-one",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-14T01:01:26.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/gabrielanhaia/torvalds-let-claude-code-write-8-kernel-patches-3x-reads-on-one-laptop-not-merged-21la"
  },
  "original_language": "en",
  "account": "Linus Torvalds pushed a set of eight kernel patches to his own Linux tree, which rework how the page cache serves small reads. The patches were written by Claude Code, an AI coding agent from Anthropic, under Torvalds' direction. The patches improved the performance of reading small files on a laptop with a 3x increase in reads per second. However, the patches are not merged into the main kernel. The patches work by using RCU (Read-Copy-Update) to avoid unnecessary memory updates when multiple cores read the same file. The patches also introduce a safety valve that forces an expedited grace period for CPU cores with more than 1,024 pages queued. The patches have been tested on a 4-core, 8-thread laptop, and the results show significant improvements in performance, but the patches are not applicable to larger machines.",
  "summary": "Book: AI Agents Pocket Guide Also by me: Thinking in Go (2-book series) — Complete Guide to Go Programming + Hexagonal Architecture in Go My project: Hermes IDE | GitHub — an IDE for developers who ship with Claude Code and other AI coding tools Me: xgabriel.com | GitHub The branch name gives it away: noref-filemap-read-with-llm . Linus Torvalds pushed it to his own Linux tree , and Phoronix…",
  "key_points": [
    "Claude Code wrote eight kernel patches to improve page cache performance.",
    "Patches increase reads per second by 3x on a 4-core laptop.",
    "Patches not merged into main kernel due to limited applicability."
  ],
  "editors_take": "Using AI coding agents to optimize kernel performance demonstrates how automated tools can achieve significant hardware-specific gains while remaining limited by their inability to scale to larger systems.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}