<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Yestino - The Signal · AArch64 FP16</title><link>https://yestino.com/zh-CN/entities/aarch64-fp16-06d32a</link><description>与 AArch64 FP16 相关的全部事件</description><language>zh-CN</language><atom:link href="https://yestino.com/zh-CN/entities/aarch64-fp16-06d32a/feed.xml" rel="self" type="application/rss+xml"/><item><title>trunk/377ffb934bd31a1538eb67a5ee1bc5cba10a4660: AArch64: FP16 matmul (#198498)</title><link>https://yestino.com/zh-CN/events/trunk-377ffb934bd31a1538eb67a5ee1bc5cba10a4660-aarch64-fp16-5900ce</link><guid isPermaLink="true">https://yestino.com/zh-CN/events/trunk-377ffb934bd31a1538eb67a5ee1bc5cba10a4660-aarch64-fp16-5900ce</guid><pubDate>Fri, 25 Sep 2026 08:55:01 GMT</pubDate><description>Enable dispatch to optimized oneDNN kernel for non-contigous shapes for FP16 MM on AArch64 with this: oneDNN picks up gemm:acl impl as compared to ref impl Pul…
来源：PyTorch Releases</description></item></channel></rss>