<?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>Embedded on Christian Grölings Blog</title>
    <link>https://www.christiangroeling.de/categories/embedded/</link>
    <description>Recent content in Embedded on Christian Grölings Blog</description>
    <generator>Hugo -- 0.163.3</generator>
    <language>en</language>
    <lastBuildDate>Fri, 17 Jul 2026 06:00:10 +0200</lastBuildDate>
    <atom:link href="https://www.christiangroeling.de/categories/embedded/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Freestanding is not a flag: building a bare-metal benchmark harness for Cortex-M4F</title>
      <link>https://www.christiangroeling.de/posts/2026-07-17-freestanding-is-not-a-flag/</link>
      <pubDate>Fri, 17 Jul 2026 06:00:10 +0200</pubDate>
      <guid>https://www.christiangroeling.de/posts/2026-07-17-freestanding-is-not-a-flag/</guid>
      <description>&lt;p&gt;I wanted to know two things about a routine on a Cortex-M4F: how many cycles it costs, and how
many bytes of flash it adds. Both answers are only worth having if nothing else in the image
contributes to either figure.&lt;/p&gt;
&lt;p&gt;That became &lt;a href=&#34;https://github.com/chgroeling/nucleo-bench&#34;&gt;nucleo-bench&lt;/a&gt;: a bare-metal benchmarking harness for the STM32F446RE. You hand it an
algorithm, it runs the thing a few million times with the core&amp;rsquo;s cycle counter running, and it
tells you what that cost. No operating system, no vendor hardware abstraction layer, no vendor
startup code.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
