<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Cpu on Shibo Chen</title><link>https://shibo-chen.github.io/tags/cpu/</link><description>Recent content in Cpu on Shibo Chen</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 28 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://shibo-chen.github.io/tags/cpu/index.xml" rel="self" type="application/rss+xml"/><item><title>General Purpose FTW</title><link>https://shibo-chen.github.io/blog/general-purpose-ftw/</link><pubDate>Fri, 28 Aug 2026 00:00:00 +0000</pubDate><guid>https://shibo-chen.github.io/blog/general-purpose-ftw/</guid><description>You can&amp;rsquo;t guess the next model, computing was never only matmul, and the expensive parts have to stay busy. Computing has never concentrated on one workload — the thing that needs justifying isn&amp;rsquo;t generality, it&amp;rsquo;s specialization.</description></item></channel></rss>