Comment 505 for bug 1690085

Sorry guys, typical Current Idle doesn't fix it for me.

I can't believe this is not yet nailed down. I kinda lost believe in the AMD comeback.

I do have this issue.

My setup is a 2700X on a Gigabyte B450 AORUS PRO.

I did hunt this freeze issue down to the behaviour decribed here: System is suddenly stuck during normal office use. Everything stays where is is. No restart, whoops, panic, log entries of any type.

Only reset-button helps.

I did try to diable Global C-State Control in UEFI - didn't help.
After finding this thread today I set Power Supply Idle Control to "Typical Current Idle" and set global C-State back to Auto (since I thought that doesn't help). Sadly, my system froze after 3h.

That is with the first and stock UEFI version F1 (AGESA 1.0.0.4).
I've just seen that there is an updated UEFI version F2. I've just updated and set global c-states to disabled AND Power Supply Idle Control to "Typical Current Idle".

"sudo ./zenstates.py -l" is now:
C6 State - Package - Disabled
C6 State - Core - Disabled

I'll come back with the results.

This makes the whole AMD 1xxx and 2xxx seem unstable.

Questions:
- Why does AMD not care? Is this maybe not NOT a general issue and only affecting a small percentage of users?

- If Windows doesn't suffer from that issue - can we find out why and do a "if amd then powersave like windows"?
Or is it just the lack of information/support by vendor so that we don't know what exactly the issue is, what is causing it and how to work around it?