【Hacker News搬运】AMD 7900XTX的文档

hackernews

Title: Documentation for the AMD 7900XTX

AMD 7900XTX的文档

Text:

Url: https://github.com/geohot/7900xtx

该页面是GitHub上geohot的7900xtx项目的页面。项目标题为"GitHub - geohot/7900xtx"，作者是geohot。页面提供了多种功能，包括自动化任何工作流程的动作、托管和管理包的包、查找和修复漏洞的安全性、即时开发环境的Codespaces、通过AI编写更好代码的Copilot、管理代码更改的代码审查、计划和跟踪工作的Issues、在代码之外的协作的Discussions等。此外，还有用于探索所有功能、文档、GitHub技能、博客等导航链接。页面还提供了针对企业、团队、初创企业和教育的解决方案链接。资源部分包括学习路径、白皮书、电子书、网络研讨会、客户故事和合作伙伴等。GitHub Sponsors是资助开源开发者的项目，The ReadME Project是GitHub社区文章项目。此外，还有按主题、趋势和收藏分类的仓库。页面底部提供了登录和注册链接。

Post by: mpereira

Comments:

roenxi: So for reference, this geohot is George Hotz who has a company Tiny Corp [0] in the space. Among a long list of things, he had a moment in the spotlight recently for a long rant [1] where he "gave up on AMD" because their drivers sucked more than he expected. The situation is reasonably complex - AMD have some programmers working on ROCm who seemed to be operating at standard fare when that is really not what AMD needs right now (I personally, suspect there is/was a PM in a key position who didn't "get it", although I am not sure what it is either). As far as I know it got the attention of Lisa Su.I'm cheering him on, even though his complaints were a little melodramatic. My experience is the driver technically supports everything I could possibly want. The problem is if I spend an evening trying to do anything with OpenCL or ROCm the kernel hard-locks and I go to bed early. If the problem inside is what it looks like from the outside (repeating myself, a key manager somewhere just doesn't get the space) they really need some pressure from grumpy customers like George to realign their software development process.That context might be related to this particular case. I see a suspicious folder called "crash" in this repo.[0] <a href="https://tinygrad.org/" rel="nofollow">https://tinygrad.org/</a>[1] <a href="https://www.youtube.com/watch?v=Mr0rWJhv9jU" rel="nofollow">https://www.youtube.com/watch?v=Mr0rWJhv9jU</a> - as I recall he ran the demo suite and it crashed.

roenxi: 因此，作为参考，这个地理热点是乔治·霍茨，他在太空中有一家公司Tiny Corp[0]。在一长串事情中，他最近有一段时间在聚光灯下长篇大论[1]，他“；放弃AMD”；因为他们的司机比他预想的要糟糕。情况相当复杂——AMD有一些程序员在ROCm上工作，他们似乎在以标准的价格运行，而这实际上不是AMD现在所需要的（我个人怀疑，有一位关键职位的PM没有得到它，尽管我也不确定它是什么）。据我所知，它引起了苏的注意；我为他加油，尽管他的抱怨有点夸张。我的经验是，司机在技术上支持我可能想要的一切。问题是，如果我花了一个晚上的时间试图用OpenCL或ROCm做任何事情，内核就会硬锁，我会很早就上床睡觉。如果内部的问题是从外部看是什么样子的（重复我的话，某个地方的关键经理就是没有得到空间），他们真的需要来自像George这样脾气暴躁的客户的一些压力来重新调整他们的软件开发过程 这种情况可能与这一特殊情况有关。我看到一个名为“；崩溃”；在这次回购中 [0]<a href=“https://；&#x2F；tinygrad.org&#x2F”rel=“nofollow”>https://&#x2F；tinygrad.org&#x2F</a> [1]<a href=“https://；&#x2F；www.youtube.com&#x2F！watch？v=Mr0rWJhv9jU”rel=“nofollow”>https://&#x2F；www.youtube.com&#x2F；看v=Mr0rWJhv9jU</a>-我记得他运行了演示套件，但它崩溃了。

glimshe: I used to be an AMD diehard fan. Bought their processors and GPUs for close to 20 years until I simply couldn't take it anymore - 5 years ago I started buying NVIDIA GPUs with Intel CPUs and honestly never looked back. Sure I pay more, but it's worth paying more for reliability and software support. Anything else and I'm losing money out of ideology to give to a business that doesn't have its act together.

glimshe: 我曾经是AMD的铁杆粉丝。买了他们的处理器和GPU将近20年，直到我不能；别再这么想了——5年前，我开始购买带有英特尔CPU的NVIDIA GPU，老实说，我从未回头。当然我付的更多，但它；值得为可靠性和软件支持付出更多的代价。任何其他东西和我；I’我出于意识形态而赔钱给一家没有；Don't have its act together。

snvzz: This is quite crude on what it is.e.g. how does it differ from gpuopen[0] ISA documentation released by AMD themselves?0. <a href="https://gpuopen.com/amd-isa-documentation/" rel="nofollow">https://gpuopen.com/amd-isa-documentation/</a>

snvzz: 这是相当粗糙的。例如，它与AMD自己发布的gpuopen[0]ISA文档有何不同 0<a href=“https://；&#x2F；gpuopen.com&#x2F，amd isa documentation&#x2F”rel=“nofollow”>https://&#x2F；gpuopen.com&#x2F；amd isa文档</一

tekno45: Can anyone explain what this level of documentation allows? Could someone make a CUDA compatible tool with this?

tekno45: 有人能解释一下这种级别的文档允许什么吗？有人能用这个制作一个与CUDA兼容的工具吗？

corey_moncure: We're talking about the company that couldn't fix a mouse cursor corruption bug affecting all of their GPU lines for what, 15, 20 years?

corey_moncure: 我们；我们谈论的是一家不能；t修复影响所有GPU线的鼠标光标损坏错误，持续15、20年？