https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=219399
--- Comment #18 from Nils Beyer <n...@renzel.net> ---
SF,
sorry, but what five values do you mean exactly? I'm currently not at work so I
cannot test anything for the rest of the weekend.
But, because you are experiencing improvments by playing around with voltages,
it's probably more a voltage than a timing problem.
One thing that wonders me most is that the CPU keeps very, very cool at 54°C
during compiling stuff with 14 threads. We're talking about an octa-core CPU
with 3.0 GHz. Even OCing to 3.4GHz, the temperature stays at 58°C.
For my taste this temperature is too low, and this suggests that the Ryzen CPU
is permanently undervoltaged.
If you feel boldly, you can try the following:
- reset all settings to the default values in your BIOS (also SoC voltage)
- set the CPU VCore voltage staticly to 1.412V
and see if that helps in your situation. This has been suggested here:
https://community.amd.com/message/2805287?tstart=0#2805287
(last sentence)
Because I'm home, I cannot test that at the moment.
I know that this has not much to do with FreeBSD at all, but for some reason
this behaviour is triggered using it. We have another Ryzen-based sever running
Win2008R2, and it doesn't show that behaviour at all...
BTW: my poudriere build system performed a hard reset (instead of a black
screen crash) recently after 8.5 hours - luckily; because so I can still access
it now...
--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
