
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Chaos and Order</title>
      <link>https://www.youngju.dev/blog</link>
      <description>천천히 올바르게. AI Researcher &amp; DevOps Engineer Youngju&#39;s blog. GPU/CUDA, LLM, MLOps, Kubernetes AI workloads, and data engineering — plus mindset essays on confidence, routines, health, and sport psychology.</description>
      <language>ko</language>
      <managingEditor>fjvbn2003@gmail.com (Youngju Kim)</managingEditor>
      <webMaster>fjvbn2003@gmail.com (Youngju Kim)</webMaster>
      <lastBuildDate>Sun, 09 Aug 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://www.youngju.dev/tags/chroot/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget.en</guid>
    <title>It Was Not the CPU That Was Slow but the Syscall Path — Lessons From Turning a Phone Into a Server</title>
    <link>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget.en</link>
    <description>Through the case of turning a CMF Phone 1 into a personal infrastructure server, this post lays out where the real cost of a Linux compatibility layer actually lands. After failing to flash postmarketOS, the author kept stock Android and used Termux as the host control plane; ordinary web services ran fine under PRoot, but the Chromium-based workload alone could not reach the CPU that was sitting idle, because of the userspace translation layer. The fix was not code but a change of execution path — rooting the phone and entering a real chroot so that native syscalls are used. Along with the general principle that follows, we also cover why a compatibility layer must never be mistaken for an isolation boundary.</description>
    <pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>linux</category><category>termux</category><category>chroot</category><category>proot</category><category>self-hosting</category><category>namespace</category><category>android</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget.ja</guid>
    <title>遅かったのはCPUではなくシステムコールの経路だった — スマホをサーバーにして学ぶこと</title>
    <link>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget.ja</link>
    <description>CMF Phone 1を個人インフラのサーバーにした事例を通じて、Linux互換レイヤの実際のコストがどこに付くのかを整理します。著者はpostmarketOSのフラッシュに失敗したあとストックのAndroidを維持し、Termuxをホストの制御プレーンとしました。一般的なウェブサービスはPRootで十分でしたが、Chromiumベースのワークロードだけはユーザースペースの変換レイヤのせいで余っているCPUに到達できませんでした。解決はコードではなく実行経路を変えること、つまりroot化して本物のchrootへ入りネイティブなシステムコールを使うことでした。ここから出てくる一般原則とともに、互換レイヤを隔離境界と取り違えてはいけない理由も扱います。</description>
    <pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>linux</category><category>termux</category><category>chroot</category><category>proot</category><category>self-hosting</category><category>namespace</category><category>android</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget</guid>
    <title>느린 건 CPU가 아니라 시스템 콜 경로였다 — 폰을 서버로 만들며 배우는 것</title>
    <link>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget</link>
    <description>CMF Phone 1을 개인 인프라 서버로 만든 사례를 통해 리눅스 호환 계층의 실제 비용이 어디에 붙는지 정리합니다. 저자는 postmarketOS 플래싱에 실패한 뒤 스톡 안드로이드를 유지하고 Termux를 호스트 제어 평면으로 삼았으며, 일반 웹 서비스는 PRoot로 충분했지만 Chromium 기반 워크로드만은 유저스페이스 변환 계층 때문에 남는 CPU에 도달하지 못했습니다. 해결은 코드가 아니라 실행 경로를 바꾸는 것, 즉 루팅 후 실제 chroot로 진입해 네이티브 시스템 콜을 쓰는 것이었습니다. 여기서 나오는 일반 원칙과 함께, 호환 계층을 격리 경계로 착각하면 안 되는 이유도 다룹니다.</description>
    <pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>linux</category><category>termux</category><category>chroot</category><category>proot</category><category>self-hosting</category><category>namespace</category><category>android</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget.zh</guid>
    <title>慢的不是 CPU，是系统调用路径 —— 把手机做成服务器学到的东西</title>
    <link>https://www.youngju.dev/blog/linux/2026-08-09-syscall-path-is-the-performance-budget.zh</link>
    <description>本文借一个把 CMF Phone 1 变成个人基础设施服务器的案例，梳理 Linux 兼容层的真实代价究竟落在哪里。作者刷 postmarketOS 失败之后保留了原厂 Android，把 Termux 当作宿主控制平面；一般的 Web 服务用 PRoot 就够了，唯独基于 Chromium 的负载因为用户态转换层而够不着那些闲着的 CPU。解法不是改代码，而是改执行路径 —— root 之后进入真正的 chroot，用上原生系统调用。文中在给出由此得到的一般原则的同时，也讨论了为什么绝不能把兼容层误当成隔离边界。</description>
    <pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>linux</category><category>termux</category><category>chroot</category><category>proot</category><category>self-hosting</category><category>namespace</category><category>android</category>
  </item>

    </channel>
  </rss>
