See this thread: https://community.amd.com/message/2815893
Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
11–20 of 70 posts
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#12Not the first bug reported for Ryzen. Wasn't there a couple of others too, one with linux locking up and another triggered by the ocaml compiler emitting opcodes refering AH/BH/CH/DH registers in a tight loop? Edit: Sorry, the ocaml bug was Intel Skylake. It's interesting how so many new CPUs have breaking bugs. Feels like it's been quiet since the original pentium F00F bug and then all of a sudden everyone's new CPU…
The OCaml thing was Skylake. But at least that one was fixed back in May.
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#13Does this affect only BSDs then? I have a Ryzen machine I bought and put into a Jenkins cluster. It does builds on a bunch of VMs occasionally and pegs the CPUs. I have had no issue so far.
It affects Linux as well.
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#14Not the first bug reported for Ryzen. Wasn't there a couple of others too, one with linux locking up and another triggered by the ocaml compiler emitting opcodes refering AH/BH/CH/DH registers in a tight loop? Edit: Sorry, the ocaml bug was Intel Skylake. It's interesting how so many new CPUs have breaking bugs. Feels like it's been quiet since the original pentium F00F bug and then all of a sudden everyone's new CPU…
The ocaml compiler found a bug in Intel's Skylake/Kaby Lake. If they found one for Ryzen too, I haven't heard about it.
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#15Its like we have all collectively forgotten that the first release of a hardware or software project is just an expensive beta. I don't buy 1st gen microarches because i'm not interested in paying top dollar to be a tester. What I find odd, is that apparently both intel and AMD have also forgotten this, as Intel seems to be moving toward making their enterprise customers the beta testers, while AMD seems so desperate for marketshare as to have released zen as a volume product before releasing it as a high end one. Meaning that if they have to do a recall, they are both losing their entire margin, as well as having to replace a large number of devices.
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#16> When a cpu-thread stalls in this manner it appears to stall INSIDE the microcode for IRETQ. It doesn't make it to the return pc, and the cpu thread cannot take any IPIs or other hardware interrupts while in this state. So maybe fixable with a microcode update?
Most cpu errata are. (Most of the rest are ignorable).
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#17Not the first bug reported for Ryzen. Wasn't there a couple of others too, one with linux locking up and another triggered by the ocaml compiler emitting opcodes refering AH/BH/CH/DH registers in a tight loop? Edit: Sorry, the ocaml bug was Intel Skylake. It's interesting how so many new CPUs have breaking bugs. Feels like it's been quiet since the original pentium F00F bug and then all of a sudden everyone's new CPU…
The ocaml compiler found a bug in Intel's Skylake/Kaby Lake. If they found one for Ryzen too, I haven't heard about it. https://lists.debian.org/debian-devel/2017/06/msg00308.html
You are right, of course.
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#18Uh oh, this might be it for AMD this year
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#19Potentially related to this issue over 5 years ago? https://it.slashdot.org/story/12/03/06/0136243/amd-confirms-...
Re: Ryzen CPU HyperThreads break if 100% busy and interrupted to top of Memory
#20I bought a custom-built Rzyen-based PC around Easter, and I have experienced some issues with it; I am not sure, though, were to put the blame (CPU, motherboard/firmware, operating system (which is openSUSE Tumbleweed)).
Under heavy load, the machine has performed most gracefully. However, the machine does freeze (almost) completely when left idle for a while (usually > 1 hour). When it happens, it sometimes still responds to pings, but nothing more; if I try to ssh into it from my laptop, I do not even receive a TCP ACK.
Unfortunately, I guess, there is no Kernel Panic, so no memory dump I could inspect or send to somebody who actually knows how to make sense of it.
On the upside, I have gotten into the habit of putting the machine into standby when I leave it alone for more than a couple of minutes, and I was pleasantly surprised that Suspend-to-Disk is a very acceptable option with an M.2 SSD. ;-P
Asus (who built my mainboard) releases firmware updates (which include microcode updates) on a fairly regular basis, so I hope this problem will be fixed eventually. I knew there was a risk of something like this happening when I got this machine, and overall, I am not disappointed. Otherwise, I am _very_ happy with the machine.