After resume from suspend, AMD integrated GPU temps not available

[DjoleDzele]

Manjaro kernel 5.13 and 5.12, after installing Disable-c6, after resuming from suspend and when running command "sensors" it shows:

amdgpu-pci-0600
Adapter: PCI adapter
vddgfx: N/A
vddnb: N/A
edge: N/A
power1: N/A

but when i start system normally i can see all these temperatures correctly instead of N/A, what could be the issue?

[jfredrickson]

The only thing this package does is tell systemd to run zenstates.py --c6-disable upon boot or after suspend/hibernate. That script comes from ZenStates-Linux. So I'd like to first figure out if the issue is with ZenStates-Linux or if there's really something wonky going on by having systemd run it.

1. Disable this systemd unit: sudo systemctl disable disable-c6.service
2. Reboot
3. Check sensors output, is it correct?
4. Run sudo zenstates.py --c6-disable
5. Check sensors output, is it correct?
6. Suspend then resume
7. Check sensors output, is it correct?
8. Run sudo zenstates.py --c6-disable
9. Check sensors output, is it correct?

If sensors is giving you issues at any of the 4 steps above, it is likely an issue with ZenStates-Linux, not this systemd unit.

[DjoleDzele]

3- GOOD
5- GOOD
7- N/A
9- N/A

The reason why i started using this script is that on kernels 5.12 and 5.13 i was unable at all to power up laptop from suspend - sleep, which works correctly on kernel 5.10 without any setting or tweaks, this script seems to make it work good on these newer ones as well, but no idea what happens with amd gpu temp, its not important issue i guess but thought to mention it. also if it matters its Asus Tuf A15, AMD Ryzen 7 5800h, and BIOS is so basic there are almost no options to change i looked there as well no power options at all