<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><title>Linus Torvalds Working On A Hefty Linux Kernel Improvement With Claude Code — Live Feed</title><link>https://www.live-feeds.com/feed/linus-torvalds-working-on-a-hefty-linux-kernel-improvement-with-claude-code</link><atom:link xmlns:atom="http://www.w3.org/2005/Atom" href="https://www.live-feeds.com/feed/linus-torvalds-working-on-a-hefty-linux-kernel-improvement-with-claude-code/rss.xml" rel="self" type="application/rss+xml"/><description>Continuously updated, source-cited coverage.</description>
<item><title>Linus Torvalds Uses Claude Code for Experimental Linux Kernel Optimization</title><link>https://www.live-feeds.com/feed/linus-torvalds-working-on-a-hefty-linux-kernel-improvement-with-claude-code</link><guid isPermaLink="false">https://www.live-feeds.com/feed/linus-torvalds-working-on-a-hefty-linux-kernel-improvement-with-claude-code#u130217</guid><pubDate>Sun, 18 Oct 2026 02:36:29 +0000</pubDate><description>Linus Torvalds published an experimental eight-patch branch, noref-filemap-read-with-llm, developed with Claude Code to optimize page-cache reads. The patches aim to reduce folio-reference contention and improve memory management. In a benchmark using an eight-thread, 16-byte-read setup, Torvalds reported throughput increasing from approximately 10 million to 30-33 million reads per second. While Torvalds stated he uses AI for tasks he is bad at, the code remains experimental and has not been merged into the mainline Linux kernel following the discovery of a race condition by reviewers.Why it </description></item>
</channel></rss>