[amdgpu/Navi] [powerplay] Failed to send message errors

Submitted by Shmerl @shmerl

Assigned to Default DRI bug account

Link to original bug (#111555)

Description

I get periodic errors like this in dmesg, which coincides with intermittent system stalls:

[Wed Sep 4 00:43:43 2019] amdgpu: [powerplay] Failed to send message 0x12, response 0xfffffffb param 0x6
[Wed Sep 4 00:43:43 2019] amdgpu: [powerplay] Failed to export SMU metrics table!
[Wed Sep 4 00:44:53 2019] amdgpu: [powerplay] Failed to send message 0xe, response 0xfffffffb, param 0x80
[Wed Sep 4 00:44:53 2019] amdgpu: [powerplay] Failed to send message 0xf, response 0xfffffffb, param 0xa90000
[Wed Sep 4 00:45:30 2019] amdgpu: [powerplay] Failed to send message 0x12, response 0xfffffffb, param 0x6
[Wed Sep 4 00:45:35 2019] amdgpu: [powerplay] Failed to send message 0x12, response 0xfffffffb param 0x6
[Wed Sep 4 00:45:35 2019] amdgpu: [powerplay] Failed to export SMU metrics table!

I'm running kernel 5.3-rc7
GPU: Sapphire Pulse RX 5700XT (Navi 10) with firmware from https://people.freedesktop.org/~agd5f/radeon_ucode/navi10/
Distro: Debian testing / KDE.

I noticed, that it starts happening often when I'm using ksysguard, which queries lm-sensors for amdgpu temperature and fan speed.