<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Yestino - The Signal · CUDA TMA</title><link>https://yestino.com/zh-CN/entities/cuda-tma-a9984e</link><description>与 CUDA TMA 相关的全部事件</description><language>zh-CN</language><atom:link href="https://yestino.com/zh-CN/entities/cuda-tma-a9984e/feed.xml" rel="self" type="application/rss+xml"/><item><title>trunk/4a81d5787ea55c5a1cb73afec4cd30a171f8cb5b: [inductor] Gate remaining CUDA TMA tests on CUDA devices (#197141)</title><link>https://yestino.com/zh-CN/events/trunk-4a81d5787ea55c5a1cb73afec4cd30a171f8cb5b-inductor-gate-54868a</link><guid isPermaLink="true">https://yestino.com/zh-CN/events/trunk-4a81d5787ea55c5a1cb73afec4cd30a171f8cb5b-inductor-gate-54868a</guid><pubDate>Thu, 17 Sep 2026 00:29:34 GMT</pubDate><description>Problem When Triton&apos;s CPU backend is installed, generic host TMA checks can enable CUDA TMA tests on unsupported GPUs such as A100 and MI300.
来源：PyTorch Releases</description></item></channel></rss>