We just installed nhpc 24.3 and people are trying to use nsys on a grace node (ubuntu 22). Note the s/w was installed without privileges under our shared cluster filesystem. When we do
module load nvidia/nvhpc/24.3
Then, if I run,
nsys status -e
Output:
Timestamp counter supported: Yes
CPU Profiling Environment Check
Root privilege: disabled
Linux Kernel Paranoid Level = 1
Linux Distribution = Ubuntu
Linux Kernel Version = 6.2.0-1015-nvidia-64k: OK
Linux perf_event_open syscall available: OK
Sampling trigger event available: OK
Intel(c) Last Branch Record support: Not Available
Kernel module: Not Available
CPU Profiling Environment (process-tree): OK
CPU Profiling Environment (system-wide): Fail
See the product documentation at https://docs.nvidia.com/nsight-systems for more information,
including information on how to set the Linux Kernel Paranoid Level.
what to make of the kernel module message? Is it expected? Not clear what kernel means here: linux module or userland GPU kernel code?
We also installed the .deb nsight package but see the same output.
In order to get some of the information from linux perf that we use for CPU sampling we need the linux kernel paranoid level to be <=2, however you are, so that should be fine for most features. The line at the bottom is telling you what you need to do to reset the paranoid level.
Here is a quick chart of what CPU profiling capabilities are available at what paranoid level from the User Guide.