You can't install NVML anymore. NVML has been part of CUDA since CUDA version 8. See: https://developer.nvidia.com/gpu-deployment-kit
```
The GPU Deployment Kit is part of the CUDA Toolkit, in versions 8 and later, and is no longer available as a separate download.
```
The package is called: `cuda-nvml-dev-9-2.ppc64le`
The entry in the passwd table has to use the key `switch`, otherwise the sshusername and password is not detected.
See `man switches` page:
```
sshusername
The remote login user name. It can be for ssh or telnet. If it is for telnet,
set protocol to "telnet". If the sshusername is blank, the username, password
and protocol will be retrieved from the passwd table with "switch" as the key.
sshpassword
The remote login password. It can be for ssh or telnet. If it is for telnet, set
protocol to "telnet". If the sshusername is blank, the username, password and
protocol will be retrieved from the passwd table with "switch" as the key.
```
See `man passwd` page:
```
key
The type of component this user/pw is for. Valid values: blade (management module),
ipmi (BMC), system (nodes), omapi (DHCP), hmc, ivm, cec, frame, switch.
```
* To fix issue that lose activation success log
* update onie doc to add rspconfig sshcfg command (#5539)
* fix osarch missing from copycds (#5543)
* Modify release information for 2.14.3 release
* update goconserver quickstart document
* Enhancement for using site cache in plugin (#5535)
* site cache when run plugin does not work very well
- using cache from plugin when getNodesAttribs/getNodeAttribs (pass it into DB process from plugin process)
- Site cache is a whole hash, so to use cache when by the hash is there, instead of the specified key is there.
It is because that there might be no key defined in site table.
* with XCATBYPASS, to populate site hash before scan_plugins. Then only 1 query for site table to do whole things.
* cache site when init plugins on service nodes
* missing to comment the old codes query from xCAT DB process
* Change cuda installation document for POWER 9
* Update the example script for CUDA setup on POWER 9
* Change CUDA installation document, ask user to use postscripts instead of postbootscripts
* Tweak the nvidia_patch.conf
* Redo initrd image for both diskless and diskfull compute node
* Fix typo
