Comment 428 for bug 1690085

(In reply to Jonathan from comment #342)
> Anyone care to see if this has an effect on this bug?

Better save your time, on my Ryzen 2500U; Kernel 4.15 (Ubuntu's), nor 4.16 (Fedora's) nor 4.17rc6+7 (Mainline) had an effect on this.

Ubuntu & mainline kernels needed "pci=noacpi" or "acpi=noirq" to even boot.
Fedora's just needed "nomodeset", if you wanted to boot with XFCE, and nothing at all with Gnome.

Doing a backup with fsarchiver
```fsarchiver savefs -o -j8 -A -a -v -Z3 $BackupfileDestination $Partition```
__always lead to a freeze within 1 - 5 runs__). Fsarchiver causes a lot of up- and downclocking (and boosting above baseclock), while it produces only partial cpu-load).
Full idle never caused a freeze neither did full load (handbrake, kill-ryzen.sh).

On kernel 4.13 the runs of fsarchiver never caused a freeze (max cpu-clock 2ghz).
Also, the setting of scaling_governor to powersave on all other tested kernels prevented that freezes (cpu clock was locked to 1,6ghz in that case).

Things I (unsuccessful) tried to get rid of the freezes:
*blacklisting amdgpu
*kernel @ CONFIG_RCU_NOCB_CPU and rcu_nocbs=0-7
*rmmod ath9k (rmmod ath10k_pci ath10k_core)
*30.5.2018 Ubuntu's microcode update for amd64
*disabled ASLR
*new mesa (18.0rc5 > 18.1.1. > 18.2 dev)
*c6-disabled via service
via grub:
*acpi=strict
*libata.noacpi
*amdgpu.dpm=0
*.dc=0
*.audio=0
*.bapm=0
*noibrs noibpb nopti
*libata.force=noncq
*pcie_aspm=off & the others
*amd_iommu=off & the other options (including iommu + options)
*pci=nocrs
*nomodeset
*all acpi= parameters
*processor.max_cstate=5