Comment 722 for bug 1690085

(In reply to raulvior.bcn from comment #635)
ASUSTeK COMPUTER INC. CROSSHAIR VI HERO 7403 08/20/2019
AMD Ryzen 7 1800X Eight-Core Processor
16410MB
2560x1440 pixels
Radeon RX 580 Series (POLARIS10, DRM 3.27.0, 5.0.0-25-generic, LLVM 8.0.0)
Linux 5.0.0-25-generic (x86_64) #26-Ubuntu SMP Thu Aug 1 12:04:58 UTC 2019
GNU C library / (Ubuntu GLIBC 2.29-0ubuntu2) 2.29
Ubuntu 19.04
BOOT_IMAGE=/vmlinuz-5.0.0-25-generic root=UUID=XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX ro quiet splash vt.handoff=1

With the latest BIOS (which provides AGESA 1.0.0.3ABB), Power Supply Idle option did work. This option increases core voltage. Knowing that increasing core voltage stabilized the system for more than 48 hours, I suspected the "performance" governor could be problematic. Because even though it made the CPU to operate at higher frequencies, the voltage was not increased. Values of 0.35, 0.5 or 0.85V were still showing up despite the minimum 3 GHz frequency instead of 2 GHz.

I have disabled "power supply idle" UEFI option and started to use the "ondemand" governor, which has a lower minimum frequency of 2 GHz.

The system continues to be stable. This is the longest time I have had the system without reboots.

Linux reports the following bugs firmware bugs:
[ 0.065489] ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored
[ 0.586638] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.586749] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.586830] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.586920] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.586983] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587042] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587110] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587163] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587235] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587309] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587378] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587448] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587522] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587590] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587646] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)
[ 0.587699] [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0)

Which effectively turns off MWAIT. cpupower idle-info output:

CPUidle driver: acpi_idle
CPUidle governor: menu
analyzing CPU 0:

Number of idle states: 3
Available idle states: POLL C1 C2
POLL:
Flags/Description: CPUIDLE CORE POLL IDLE
Latency: 0
Usage: 42208
Duration: 98245
C1:
Flags/Description: ACPI HLT
Latency: 1
Usage: 2147693
Duration: 605512928
C2:
Flags/Description: ACPI IOPORT 0x414
Latency: 400
Usage: 4338245
Duration: 71750833391

For my problem, which might not be the same as for the rest of users here, not having MWAIT did not stop the system to reboot randomly. And despite the kernel not detecting MWAIT, the processor still boosts up to 4.1 GHz as intended. C6 States are enabled. The CPU might enter them after entering C1 with HLT. Zenstates.py output:

P0 - Enabled - FID = 90 - DID = 8 - VID = 20 - Ratio = 36.00 - vCore = 1.35000
P1 - Enabled - FID = 80 - DID = 8 - VID = 2C - Ratio = 32.00 - vCore = 1.27500
P2 - Enabled - FID = 84 - DID = C - VID = 68 - Ratio = 22.00 - vCore = 0.90000
P3 - Disabled
P4 - Disabled
P5 - Disabled
P6 - Disabled
P7 - Disabled
C6 State - Package - Enabled
C6 State - Core - Enabled

I'll report in the future whether if I have had another reboot. I'm nonitoring the system with the Vitals GNOME extension (https://extensions.gnome.org/extension/1460/vitals/). The kernel module asus-wmi-sensors provide the actual data (https://github.com/electrified/asus-wmi-sensors).