<?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>Branch Prediction on 梦兽编程</title><link>https://rexai.top/en/tags/branch-prediction/</link><description>Recent content in Branch Prediction on 梦兽编程</description><generator>Hugo -- 0.163.3</generator><language>en</language><copyright>梦兽编程</copyright><lastBuildDate>Mon, 10 Aug 2026 06:30:00 +0800</lastBuildDate><atom:link href="https://rexai.top/en/tags/branch-prediction/index.xml" rel="self" type="application/rss+xml"/><item><title>One if Made This Rust Filter 4x Slower: The CPU Was Guessing for You</title><link>https://rexai.top/en/languages/rust/2026-08-10-rust-branchless-filter-if/</link><pubDate>Mon, 10 Aug 2026 06:30:00 +0800</pubDate><guid>https://rexai.top/en/languages/rust/2026-08-10-rust-branchless-filter-if/</guid><description>Filtering 1M floats: why is keeping half the slowest case? I reproduced the HN-top branch-prediction experiment on Windows (i5-12400F, rustc 1.97.1): a branchless rewrite cut the worst case from 5.1ms to 1.5ms (3.4x faster) — and learned when the trick backfires</description></item></channel></rss>