Tiny Corp Puts Their AMD-Powered Compute Boxes "On Hold"

Written by Michael Larabel in Hardware on 20 March 2024 at 06:54 AM EDT. 57 Comments
Tiny Corp has been frustrated before with AMD / ROCm and planned to drop AMD graphics cards in their planned compute boxes over it only to go back to AMD GPUs with their open-source driver stack later. It's now happened again following frustrations over firmware binaries. After recently lobbying AMD to at least open-source some relevant pieces of their firmware and at ~70% confidence over their plans, Tiny Corp announced on Tuesday they are dropping AMD GPUs again from their compute plans.

Tiny Corp is now said to be dropping AMD GPUs (or putting it "on hold") from their compute boxes and instead will try getting Intel Arc Graphics working for their compute needs or potentially need to resort to using NVIDIA GeForce graphics cards, which would be rather ironic given their walled garden and lack of any open-source firmware or user-space. Tiny Corp tweeted:
"The AMD tinybox is on hold until we can build and run the relevant firmware on our GPUs.

The driver is still very unstable, and when it crashes or hangs we have no way of debugging it. We have no way of dumping the state of a GPU. Apparently it isn't just the MES causing these issues, it's also the Command Processor (CP).

After seeing how open @tenstorrent is, it's hard to deal with this. With tenstorrent, I feel confident that if there's an issue, I can debug and fix it. With AMD, I don't.

We are exploring Intel, working on adding Level Zero support to tinygrad. We also added a $400 bounty for XMX support. We are also (sadly) exploring a 6x4090 box. At least we know the software is good there.

We will revisit AMD once we have an open and reproducible build process for the driver and firmware. We are willing to dive really deep into hardware to make it amazing. But without access, we can't."

It was also added:
"I have spoken with AMD on multiple occasions, we have gotten through to top people, and they have been quite nice to us. I believe they want to be more open, and obviously they don't want their driver to have bugs.

Unfortunately, this access and responses prolonged this decision, part of me wishes they just said it's a consumer card, you get what you pay for and we could have switched earlier. We probably tried too hard to make it work.

We have an amazing team at tinygrad. Someday, we are going to make our own chips, and I figure if we can make our own chips, we better be able to make the 7900XTX software great. But we can't if we don't have access. The firmware is complex, undocumented, closed source, and signed, all struggles we wouldn't have with our own hardware.

If and when the firmware is open and installable, if we aren't too far along with a different chip, we are down to put resources into writing fuzzers and rewriting whatever needs to be rewritten. The 7900XTX hardware seems great, but we aren't going to put resources into fixing a black box."

Back on the matter of their lobbying for the MES firmware sources, it was responded:
"We thought about it more. While the announcement is movement in the right direction (you'll hopefully see it soon), it's actually still unusable for us in any practical way.

Also, they released this changelog with a prerelease version of ROCm 6.0.3. At first, it's cool, finally some insight into what these things are.

But that cool morphs into horrifying when you read this document more carefully. There's so much complexity hidden deep inside this GPU that isn't surfaced in the driver or the user space. It reveals how much of a black box this thing really is."

So at least for now, the AMD graphics are "on hold" at Tiny Corp.
Related News
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week