Ryzen linux stability concerns

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Patrick

Patrick

Administrator
Staff member
Dec 21, 2010
12,513
5,804
113
Just to give you an idea with gcc and Ryzen (we used 5.x, 6.x and 7.x) we were doing 100x make -j$(nproc) runs at 8min 30s+ per run on Ryzen with Ubuntu 14.04 / 16.04 w/ 4.10 kernels and never had an issue on the 7 or so systems running. That is >3000 kernel compiles worth.

Not saying it is not an issue, but not something we saw in our testing as causing a problem.

Now, seg faults due to SMT and/or old kernels we ran into non-stop during the early days. Most of that is fixed now.

The EPYC systems are working well.

Also - there is a Ryzen 7 1700 system in our hosting cluster that has been up since April without issue.
 
  • Like
Reactions: gigatexal, Stux and eva2000
Deslok

Deslok

Well-Known Member
Jul 15, 2015
1,122
125
63
34
deslok.dyndns.org
Patrick said:
Just to give you an idea with gcc and Ryzen (we used 5.x, 6.x and 7.x) we were doing 100x make -j$(nproc) runs at 8min 30s+ per run on Ryzen with Ubuntu 14.04 / 16.04 w/ 4.10 kernels and never had an issue on the 7 or so systems running. That is >3000 kernel compiles worth.

Not saying it is not an issue, but not something we saw in our testing as causing a problem.

Now, seg faults due to SMT and/or old kernels we ran into non-stop during the early days. Most of that is fixed now.

The EPYC systems are working well.

Also - there is a Ryzen 7 1700 system in our hosting cluster that has been up since April without issue.
Click to expand...
That's what I was curious about, it seems tied to only that workload, and fixed on epyc/threadripper(likely a stepping update)
 
You must log in or register to reply here.
Top Bottom