Comment 705 for bug 1690085

I have to kind of wonder, doesn't AMD like, want to sell these chips (in Epyc) to server farms? Ones that will run Linux? Or do Epyc/TR just not have this problem? My experience with this bug has been a bit bizarre, especially since AMD's GPUs have really great Linux support these days.

My report:
- Ryzen 1600
- CORSAIR Vengeance LPX 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 3000 (PC4 24000)
- Gigabyte AB-350 Gaming 3
- RX 460 Sapphire 2GB
- TP-Link TL-WN722N USB Wifi -- This is its own bag of worms
- misc: SATA SSDs, HDDs
I've had this crash problem ever since I built this PC, HOWEVER for a very long time, the LTS kernel 4.14 didn't have the crashing/hanging/random reboot. Usually, if I installed a new kernel, that kernel would be stable for a week or so (including lots of suspend/unsuspend) and then develop the crashing problem. Meaning I mainly just stayed on 4.14. But today was a special day, because 4.14 developed the problem as well now!
IDK if this has been discussed further up already, but I'd like to highlight two of those points:
- freshly installed kernels don't have the problem at first? what could cause this?
- 4.14 didn't have it either? Again, what did they change to break it?
Moreover, the crashing is WORSE THAN EVER now. Like hourly.

So I came back to this thread and I'm trying disabling Global C-States, as my older motherboard firmware doesn't have the other options that have been mentioned ITT. I'll wait and see if that solves it, and if not, I'll try the python script, and then maybe updating my motherboard so I can access more settings.