Overclock.net banner

Ryzen compilation segfaults

2K views 11 replies 6 participants last post by  SteelBox 
#1 ·
I saw this on reddit.

Will a regular user ever encounter that error or it is only on linux?
 
#2 ·
If your CPU is newer than July (check IHS date) it won't be affected.

Also with the newer Linux kernels (4.11 , 4.12+) + AGESA 1.0.0.6 people don't seem to have as many issues

If it were a Windows issue as well you'd hear about it a lot faster...
 
#3 ·
Quote:
Originally Posted by AlphaC View Post

If your CPU is newer than July (check IHS date) it won't be affected.

Also with the newer Linux kernels (4.11 , 4.12+) + AGESA 1.0.0.6 people don't seem to have as many issues

If it were a Windows issue as well you'd hear about it a lot faster...
It is older than July, I checked date, on my R5 1600 UA number is 1708. I have dual boot: windows and Linux. I usually use my PC for gaming, movies, music, web design, audio and video editing. That error in Linux is only in compilation, that means programing?
 
#4 ·
Quote:
Originally Posted by SteelBox View Post

It is older than July, I checked date, on my R5 1600 UA number is 1708. I have dual boot: windows and Linux. I usually use my PC for gaming, movies, music, web design, audio and video editing. That error in Linux is only in compilation, that means programing?
Yes.

Not even video editing triggers any issues
 
#6 ·
I have a day 1 CPU and have had no issues im aware of. Its largely a Linux compiling but with GCC. I also have linux whom which I compile my cod eon but have not had it yet still.

I am curious if anyone knows why this happens, as I see OpCode errors is it missing an instruction set to read the first 4 bits or what?
 
#7 ·
I now saw this on reddit:

"Former AMD employee and CPU engineer here.

I would be concerned under Windows. The CPU doesn't know it's running Linux, doesn't know it's running gcc. Any software with the wrong stream of instructions (or data? or pattern of jumps? or pattern of faults? etc) could potentially trigger this issue.

Without some clarification from AMD on the issue, I would assume it's rare but also possible to provoke on any OS given the right load. It might be impossible to fix in software or firmware.

This is a serious bug. To compare against some classics: the Pentium F00F bug wouldn't manifest itself with any normal software. The famous Pentium floating point bug really was innocuous and Intel really got unlucky in the PR department on that one.

The saving graces for AMD are that this bug was caught early enough, does not affect server or high end parts (like the terrible amd Barcelona cache coherency bug that led to a big recall) and is rare enough that a widespread recall isn't necessary.

If I owned an affected ryzen, I would probably RMA it. I keep computers several years and don't want to wonder about the CPU every time something crashes for years."

https://www.reddit.com/r/Amd/comments/6wckrj/should_windows_users_be_worried_about_the/

So it is serious?
 
#9 ·
We don't know...
This problem has been observed only during heavy compilation under Linux oder some BSD variants....so far.

When you think about RMAing your processor, first make shure yours is realy affected by the problem. You can use the script kill-ryzen.sh to reproduce the segfault. It start to compile the gcc source on all available threads. If your processor is affected by the problem, the compilers workers will start to segfault after some time (from a several minutes to a few hours) one after the other. If your processor is not affected the script will run out of memory after sevearl hours depending on the number of threads and your amount of RAM and all or nearly all worker stop at the same time.
Of course you have to check this at stock settings.

My old 1800x was affected by this bug. kill-ryzen took from a few minutes up to two hours to show segfaults. I RMAed the processor and got my replacement chip on Monday. No segfaults since then.

Hope this helps.
 
#10 ·
Yes it is serious, as most probably this is hardware problem that can't be fixed with AGESA and AMD engineers doesn't know how to fix it at the moment, unfortunately mine 1700
mad.gif
also has the problem, so I'm also thinking of RMA if they can't come up with something better next several weeks. Seems like this problem is affecting many CPU and is more spread than initially though.
The RMA process is quite inconvenient for me but it is what it is ....
thumb.gif
 
#11 ·
What will happen if it happens in gaming (or something else) on Windows, will computer freeze or what?

For now my CPU is working very well, stable. If I am affected with this problem I would RMA after couple of months, after the Zen+ release. I bought my Ryzen on Amazon, would it be problem on Amazon for swithing my Ryzen to Zen+, would Amazon policies allow that or will I get the same Ryzen that I sent to RMA?
 
#12 ·
Some users says that crashing in Ashes of singularity can be connected to this problem:

"Both my machines have the Linux SEGV issue (R7-1700X, R7-1700) additionally - both machines can't run the Ashes of the Singulariy Escalation built-in Benchmark in either DX12 or VULKAN mode without crashing to the desktop. Disabling SMT seems to make the benchmark work but I believe the two are related (Linux SegV / AOTS crashing) - note, nothing overclocked on the system, just BIOS defaults.
I should have the R7-1700X back from AMD this upcoming week, and hopefully the R7-1700 a few days later. If it fixes the CTD's with AOTS Benchmark (DX12 / Vulkan) with only thing that was changed in the systems were the CPU's - then my theory is that Windows Games are also affected.
I'm not the only one that ran into the problem - couple others have reported it as well. If you have a Ryzen CPU, try running the AOTS:Singularity Benchmark a couple of times in DX12 or Vulkan mode only - if it CTDs, try turning off SMT."

https://www.reddit.com/r/Amd/comments/6w4d3e/should_windows_users_care_about_the_ryzen/

Also some mention that Shadow of mordor because high CPU usage crashes because of that segfault problem....
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top