Enabling GPU Performance Counters

Graphics / Linux Linux
1

Hello,

I want to enable “GPU Performance Counters” to all users. I have read step by step NVIDIA Development Tools Solutions - ERR_NVGPUCTRPERM: Permission issue with Performance Counters | NVIDIA Developer, but it doesn’t work. My system runs a Rocky Linux 8.10 with kernel 4.18.0-553.5.1.el8_10.x86_64 and Nvidia driver 535.54.03.

After run “modinfo nvidia”, I have seen all parameters:

parm:           NvSwitchRegDwords:NvSwitch regkey (charp)
parm:           NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp)
parm:           NVreg_ResmanDebugLevel:int
parm:           NVreg_RmLogonRC:int
parm:           NVreg_ModifyDeviceFiles:int
parm:           NVreg_DeviceFileUID:int
parm:           NVreg_DeviceFileGID:int
parm:           NVreg_DeviceFileMode:int
parm:           NVreg_InitializeSystemMemoryAllocations:int
parm:           NVreg_UsePageAttributeTable:int
parm:           NVreg_EnablePCIeGen3:int
parm:           NVreg_EnableMSI:int
parm:           NVreg_TCEBypassMode:int
parm:           NVreg_EnableStreamMemOPs:int
parm:           NVreg_RestrictProfilingToAdminUsers:int
parm:           NVreg_PreserveVideoMemoryAllocations:int
parm:           NVreg_EnableS0ixPowerManagement:int
parm:           NVreg_S0ixPowerManagementVideoMemoryThreshold:int
parm:           NVreg_DynamicPowerManagement:int
parm:           NVreg_DynamicPowerManagementVideoMemoryThreshold:int
parm:           NVreg_EnableGpuFirmware:int
parm:           NVreg_EnableGpuFirmwareLogs:int
parm:           NVreg_OpenRmEnableUnsupportedGpus:int
parm:           NVreg_EnableUserNUMAManagement:int
parm:           NVreg_MemoryPoolSize:int
parm:           NVreg_KMallocHeapMaxSize:int
parm:           NVreg_VMallocHeapMaxSize:int
parm:           NVreg_IgnoreMMIOCheck:int
parm:           NVreg_NvLinkDisable:int
parm:           NVreg_EnablePCIERelaxedOrderingMode:int
parm:           NVreg_RegisterPCIDriver:int
parm:           NVreg_EnableResizableBar:int
parm:           NVreg_EnableDbgBreakpoint:int
parm:           NVreg_RegistryDwords:charp
parm:           NVreg_RegistryDwordsPerDevice:charp
parm:           NVreg_RmMsg:charp
parm:           NVreg_GpuBlacklist:charp
parm:           NVreg_TemporaryFilePath:charp
parm:           NVreg_ExcludedGpus:charp
parm:           NVreg_DmaRemapPeerMmio:int
parm:           NVreg_RmNvlinkBandwidth:charp
parm:           rm_firmware_active:charp

I have created a new file /etc/modprobe.d/nvidia.conf with and added this line:

options nvidia NVreg_RestrictProfilingToAdminUsers=0

However, when I reboot system, I have got some errors from “nvidia-persistenced” daemon and, after that, nvidia module has not been loaded, so I can’t work with Nvidia hardware.
I have done a second test, disabling previosly “nvidia-persistenced” daemon, but after reboot, result is bad, too…

I have check with “cat /proc/driver/nvidia/params” all Nvidia module parameters:

ResmanDebugLevel: 4294967295
RmLogonRC: 1
ModifyDeviceFiles: 1
DeviceFileUID: 0
DeviceFileGID: 0
DeviceFileMode: 438
InitializeSystemMemoryAllocations: 1
UsePageAttributeTable: 4294967295
EnableMSI: 1
EnablePCIeGen3: 0
MemoryPoolSize: 0
KMallocHeapMaxSize: 0
VMallocHeapMaxSize: 0
IgnoreMMIOCheck: 0
TCEBypassMode: 0
EnableStreamMemOPs: 0
EnableUserNUMAManagement: 1
NvLinkDisable: 0
RmProfilingAdminOnly: 1
PreserveVideoMemoryAllocations: 0
EnableS0ixPowerManagement: 0
S0ixPowerManagementVideoMemoryThreshold: 256
DynamicPowerManagement: 2
DynamicPowerManagementVideoMemoryThreshold: 200
RegisterPCIDriver: 1
EnablePCIERelaxedOrderingMode: 0
EnableResizableBar: 0
EnableGpuFirmware: 18
EnableGpuFirmwareLogs: 2
EnableDbgBreakpoint: 0
OpenRmEnableUnsupportedGpus: 0
DmaRemapPeerMmio: 1
RegistryDwords: “”
RegistryDwordsPerDevice: “”
RmMsg: “”
GpuBlacklist: “”
TemporaryFilePath: “”
ExcludedGpus: “”

There, parameters takes as name “RmProfilingAdminOnly”, not “NVreg_RestrictProfilingToAdminUsers”… Maybe could be it the problem?

Thanks.

2

Hi,

It seems I have got it to work with this file in /etc/modprobe.d:

options nvidia NVreg_RestrictProfilingToAdminUsers=0
options nvidia_drm NVreg_RestrictProfilingToAdminUsers=0
options nvidia_modeset NVreg_RestrictProfilingToAdminUsers=0
options nvidia_uvm NVreg_RestrictProfilingToAdminUsers=0

After reboot, parameters has taken value 0 and I have been able to run my test.

Thanks.

1 Like
3

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.