Comment 320 for bug 1690085

Revision history for this message
In , sineways (sineways-linux-kernel-bugs) wrote :

Ryzen 3 1200 Quad-Core (Made: 17/28) with an ASRock AB350 Pro4 running BIOS 3.20 (AGESA 1.0.0.6b)

System gets hardlocks (anywhere from 1-3, 10-30 days) that cause even SysRq to not work. The chip is in a server that varies between idle and active, but the hang seems to happen regardless of CPU usage state.

Only once did I get a "BUG: soft lockup". And in that situation I was able to REISUB sysrq my way to a clean reboot.

I've been following this bug for a while, but I'm not sure of what a proper solution might be (if any) for those that have found one at this point.

Below is some possibly helpful information:
Kernel: 4.14.0-0.bpo.3-amd64 #1 SMP Debian 4.14.13-1~bpo9+1 (2018-01-14) x86_64 GNU/Linux
Grub: rcu_nocbs=0-3

-
python zenstates.py --list
P0 - Enabled - FID = 7C - DID = 8 - VID = 3A - Ratio = 31.00 - vCore = 1.18750
P1 - Enabled - FID = 8C - DID = A - VID = 50 - Ratio = 28.00 - vCore = 1.05000
P2 - Enabled - FID = 7C - DID = 10 - VID = 6A - Ratio = 15.50 - vCore = 0.88750
P3 - Disabled
P4 - Disabled
P5 - Disabled
P6 - Disabled
P7 - Disabled
C6 State - Package - Enabled
C6 State - Core - Enabled

At a loss for where to go now. I did not enable CONFIG_RCU_NOCB_CPU=y as I thought that disabled C6 states, and that wasn't desirable, but perhaps I'm mistaken.

Appreciate any guidance
Doug