<?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>parsing on Sparse Notes</title>
    <link>https://sparsenotes.com/tags/parsing/</link>
    <description>Recent content in parsing on Sparse Notes</description>
    <image>
      <url>https://sparsenotes.com/images/og-default.png</url>
      <link>https://sparsenotes.com/images/og-default.png</link>
    </image>
    <generator>Hugo -- gohugo.io</generator>
    <lastBuildDate>Sat, 17 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://sparsenotes.com/tags/parsing/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Accelerating the Zig Tokenizer with AVX-512 (Niles Salter)</title>
      <link>https://sparsenotes.com/posts/2026/10/accelerating-the-zig-tokenizer-avx512-niles-salter/</link>
      <pubDate>Sat, 17 Oct 2026 00:00:00 +0000</pubDate>
      
      <guid>https://sparsenotes.com/posts/2026/10/accelerating-the-zig-tokenizer-avx512-niles-salter/</guid>
      <description>Niles Salter shows how bit-parallel tricks (subtract, count-trailing-zeros, vector compaction, ternary logic) turn a branchy byte-at-a-time tokenizer into a branchless one, taking Zig source tokenization from about 0.5 GB/s to about 2 GB/s on one core.</description>
    </item>
    
  </channel>
</rss>
