document nvidia software management

_toc.yml

@@ -39,7 +39,8 @@ parts:
       sections:
         - file: rhel8/installation
         - file: rhel8/packages
-        - file: rhel8/desktop
+        - file: rhel8/nvidia
         - file: rhel8/kerberos
+        - file: rhel8/desktop
         - file: rhel8/vendor_documentation
 

rhel8/index.md

@@ -22,8 +22,6 @@ A lot works already out of the box, but no guarantee can be given unless your sp
 
 ## Major Known Issues
 
-- sssd_kcm switches Kerberos credential cache midsession [PSILINUX-120](https://jira.psi.ch/browse/PSILINUX-120)
-
 ## Major Missing Features
 
 - not all interesting 3rd-party packages available yet  [PSILINUX-113](https://jira.psi.ch/browse/PSILINUX-113)
@@ -107,6 +105,7 @@ which is IMHO OK to not allow a normal user to do changes there.
 
 * [Installation](installation)
 * [Package Management](packages)
+* [CUDA and Nvidia Drivers](nvidia)
 * [Kerberos](kerberos)
 * [Desktop](desktop)
 * [Vendor Documentation](vendor_documentation)

rhel8/nvidia.md

@@ -0,0 +1,47 @@
+# CUDA and Proprietary Nvidia GPU drivers on RHEL 8
+
+Managing Nvidia software comes with its own set of challenges.
+For the most common cases are covered by our Puppet configuration.
+Those are discussed in the first chapter, more details you find more below.
+
+## Hiera Configuration
+Changes in Hiera are forwared by Puppet to the node, but **not applied**.
+They are applied on **reboot**.
+Alternatively you might execute `/opt/pli/libexec/ensure-nvidia-software` in a safe moment (no process using CUDA and the desktop will be restarted).
+
+### I need CUDA
+
+Set in Hiera `nvidia::cuda::install_software: true` and it will automatically install the suitable Nvidia drivers and newest possible CUDA version.
+
+To enable `nvidia_persistenced` you additionally need to set `nvidia::cuda::nvidia_persistenced::enable: true`.
+
+### I need a specific CUDA version
+
+Then you can additionally set `nvidia::cuda::version` to the desired version.
+The version must be fully specified (all three numbers, with X.Y.0 for the GA version).
+
+Note that newer CUDA versions do not support older drivers, for details see Table 3 in the [CUDA Release Notes](https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html).
+
+### I just need the Nvidia drivers
+Nothing needs to be done, they are installed by default when Nvidia GPUs or accelerators are found.
+
+### I do not want the Nvidia drivers
+
+Set in Hiera `nvidia::driver::enable: false`. Note this will be ignored if CUDA is enabled (see above).
+
+Note they do not get automatically removed when already installed. That you would need to do by hand.
+
+### I need the Nvidia drivers from a specific driver branch
+
+The driver branch can be selected in Hiera with `nvidia::driver::branch`. It will then use the latest driver version of that branch. Note that only production branches are available in the PSI package repository.
+
+### I need a Nvidia driver of a given version
+
+This is not recommended, still it is possible to do so by setting the exact driver version (X.Y.Z, excluding the package iteration number) in Hiera with `nvidia::driver::version`.
+
+If the driver version is too old, it will install an older kernel version and you will need a second reboot to activate it.
+
+## Versioning Mess
+
+## Manual Operation
+