<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Coroutines on Pi Stack</title>
    <link>https://www.pistack.xyz/tags/coroutines/</link>
    <description>Recent content in Coroutines on Pi Stack</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Sun, 21 Jun 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://www.pistack.xyz/tags/coroutines/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Self-Hosted Coroutine Libraries: libaco vs greenlet vs libco vs boost::context — Asymmetric Coroutines for High-Concurrency C/C&#43;&#43; Systems</title>
      <link>https://www.pistack.xyz/posts/2026-06-21-coroutine-libraries-libaco-greenlet-libco-boost-context/</link>
      <pubDate>Sun, 21 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://www.pistack.xyz/posts/2026-06-21-coroutine-libraries-libaco-greenlet-libco-boost-context/</guid>
      <description>&lt;h2 id=&#34;why-coroutines-matter-for-high-concurrency-systems&#34;&gt;Why Coroutines Matter for High-Concurrency Systems&lt;/h2&gt;&#xA;&lt;p&gt;Traditional multi-threaded programming with one OS thread per task hits a scalability wall: each thread consumes 8-16 MB of stack space and incurs kernel scheduling overhead. At 10,000 concurrent connections, thread stacks alone consume 80-160 MB of memory. At 100,000 connections, the overhead becomes prohibitive. Coroutines solve this by multiplexing thousands of logical tasks onto a small number of OS threads — each coroutine uses only kilobytes of state rather than megabytes.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
