<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Llvm22 on ./Code</title><link>https://blog.ouankou.com/tags/llvm22/</link><description>Recent content in Llvm22 on ./Code</description><generator>Hugo</generator><language>en-US</language><copyright>© Anjia Wang</copyright><lastBuildDate>Mon, 04 May 2026 20:11:12 -0700</lastBuildDate><atom:link href="https://blog.ouankou.com/tags/llvm22/index.xml" rel="self" type="application/rss+xml"/><item><title>What Changed When REX Was Re-Evaluated Against LLVM 22</title><link>https://blog.ouankou.com/2026/05/01/what-changed-when-rex-was-re-evaluated-against-llvm-22/</link><pubDate>Fri, 01 May 2026 00:00:00 +0000</pubDate><guid>https://blog.ouankou.com/2026/05/01/what-changed-when-rex-was-re-evaluated-against-llvm-22/</guid><description>&lt;p&gt;The previous post closed the LLVM 21 performance story. That mattered, but it was not the end of the engineering question.&lt;/p&gt;
&lt;p&gt;The next question was more uncomfortable:&lt;/p&gt;
&lt;div class="highlight"&gt;&lt;div style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;-webkit-text-size-adjust:none;"&gt;
&lt;table style="border-spacing:0;padding:0;margin:0;border:0;"&gt;&lt;tr&gt;&lt;td style="vertical-align:top;padding:0;margin:0;border:0;"&gt;
&lt;pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;-webkit-text-size-adjust:none;"&gt;&lt;code&gt;&lt;span style="white-space:pre;-webkit-user-select:none;user-select:none;margin-right:0.4em;padding:0 0.4em 0 0.4em;color:#7f7f7f"&gt;1
&lt;/span&gt;&lt;span style="white-space:pre;-webkit-user-select:none;user-select:none;margin-right:0.4em;padding:0 0.4em 0 0.4em;color:#7f7f7f"&gt;2
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;&lt;/td&gt;
&lt;td style="vertical-align:top;padding:0;margin:0;border:0;;width:100%"&gt;
&lt;pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;-webkit-text-size-adjust:none;"&gt;&lt;code class="language-text" data-lang="text"&gt;&lt;span style="display:flex;"&gt;&lt;span&gt;If the OpenMP offloading runtime changes under us,
&lt;/span&gt;&lt;/span&gt;&lt;span style="display:flex;"&gt;&lt;span&gt;do the REX fixes still hold?
&lt;/span&gt;&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/table&gt;
&lt;/div&gt;
&lt;/div&gt;&lt;p&gt;That is the question this post answers.&lt;/p&gt;
&lt;p&gt;This is not another optimization post. No new benchmark-specific rewrite was added for LLVM 22. The point was to test whether the current REX design remained valid when the underlying OpenMP offloading stack moved from LLVM 21 to a source-built LLVM 22 toolchain.&lt;/p&gt;</description></item></channel></rss>