In case someone from AMD is reading this: guys, you need to fix stability issues on Linux, or your Epyc is DOA. People are having some serious lock-up trouble with Ryzen, even with the latest AGESA updates and kernels. This is the only issue preventing me from recommending to purchase threadripper workstations at work. We do need that pcie bandwidth for GPU, but we absolutely can't tolerate instability.
I sure hope so, but given the epic fail of Ryzen, I'm going to wait for others to beta test it for me, and/or for some sort of explanation and fix from AMD.
Well then you likely won't get the information you need as quickly as you need it.
Buy one, test it in your environment, file bug reports if you come across any. You don't need to replace your entire fleet in one go; Plan a phased rollout, starting with one workstation running the most aggressive of your workloads.
I had a lot of issues: obviously Ubuntu 17.04 doesn't boot, but Fedora 26 crashed a lot seemingly no matter what I changed. Windows 10 also crashed fairly often. I was about say that it wasn't as bad as F26, but given the automatic restart on crash that I kept seeing, it might have been as bad.
Currently using Arch/4.12-zen/nvidia with Gigabyte Gaming K5/Ryzen 7 1700X/GTX1060 and since kicking nouveau out, I have had no crashes at all. I am hopeful that this means my Ryzen woes are over.
AMD has confirmed that they've been able to replicate the Linux bug and that it only happens on Ryzen, ThreadRipper and Epyc do not suffer from that issue.
They also have also inadvertently confirmed that their Linux testing efforts are lacking. Which just blows my mind TBH, because they're also in the server market, dominated completely by Linux.
