I am running configuration from $SUBJECT (configured with prime-select nvidia).
Everything seems to be working fine-ish, but once I start executing any GPU heavy process (for example glmark2) it goes from:
nvidia-smi -q -d PERFORMANCE
==============NVSMI LOG==============
Timestamp : Mon Feb 10 10:56:45 2020
Driver Version : 440.59
CUDA Version : 10.2
Attached GPUs : 1
GPU 00000000:01:00.0
Performance State : P0
Clocks Throttle Reasons
Idle : Not Active
Applications Clocks Setting : Not Active
SW Power Cap : Not Active
HW Slowdown : Not Active
HW Thermal Slowdown : Not Active
HW Power Brake Slowdown : Not Active
Sync Boost : Not Active
SW Thermal Slowdown : Not Active
Display Clock Setting : Not Active
to:
$ nvidia-smi -q -d PERFORMANCE
==============NVSMI LOG==============
Timestamp : Mon Feb 10 10:57:15 2020
Driver Version : 440.59
CUDA Version : 10.2
Attached GPUs : 1
GPU 00000000:01:00.0
Performance State : P0
Clocks Throttle Reasons
Idle : Active
Applications Clocks Setting : Not Active
SW Power Cap : Not Active
HW Slowdown : Not Active
HW Thermal Slowdown : Not Active
HW Power Brake Slowdown : Not Active
Sync Boost : Not Active
SW Thermal Slowdown : Not Active
Display Clock Setting : Not Active
There is not thermal throttling going on since temperatures are looking fine.
I have read on forums there that it might be caused due to HardDPMS, so I have tried to disable DPMS altogether, but it has no effect.
Section "Files"
ModulePath "/usr/lib/x86_64-linux-gnu/nvidia/xorg/"
ModulePath "/usr/lib/xorg/modules"
EndSection
Section "Extensions"
Option "DPMS" "Disable"
EndSection
Section "ServerLayout"
Identifier "layout"
Screen 0 "nvidia"
Inactive "intel"
EndSection
Section "Device"
Identifier "nvidia"
Driver "nvidia"
BusID "PCI:1:0:0"
EndSection
Section "Screen"
Identifier "nvidia"
Device "nvidia"
Option "AllowEmptyInitialConfiguration"
EndSection
Section "Device"
Identifier "intel"
Driver "modesetting"
EndSection
Section "Screen"
Identifier "intel"
Device "intel"
EndSection
Once throttling kicks in only logout or couple of AC disconnects help.
EDIT: this happens with trying to run with prime offload configuration as well.
I have tried to set HardDPMS to false in Device Section in xorg, based on this question [1] but it did not produce any difference.
EDIT2: I have sent my report over the mail to linux-bugs@nvidia.com