Add nvidia runtime class

This commit is contained in:
2025-08-05 01:25:46 +02:00
parent 0eb95a74b3
commit 0b355971a2
3 changed files with 43 additions and 0 deletions

5
apps/misc/nvidia.yaml Normal file
View File

@@ -0,0 +1,5 @@
apiVersion: node.k8s.io/v1
kind: RuntimeClass
metadata:
name: nvidia
handler: nvidia

26
apps/nvidia.yaml Normal file
View File

@@ -0,0 +1,26 @@
apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: nvidia-device-plugin
namespace: argocd
spec:
project: default
destination:
server: https://kubernetes.default.svc
namespace: nvidia
syncPolicy:
automated:
prune: true
selfHeal: true
syncOptions:
- CreateNamespace=true
managedNamespaceMetadata:
labels:
pod-security.kubernetes.io/enforce: privileged
source:
repoURL: https://nvidia.github.io/k8s-device-plugin
targetRevision: 0.13.0
chart: nvidia-device-plugin
helm:
valuesObject:
runtimeClassName: nvidia

View File

@@ -20,6 +20,16 @@ patches:
kernel:
modules:
- name: zfs
- |-
machine:
kernel:
modules:
- name: nvidia
- name: nvidia_uvm
- name: nvidia_drm
- name: nvidia_modeset
sysctls:
net.core.bpf_jit_harden: 1
nodes:
- hostname: kadan
ipAddress: kube.sdg.moe
@@ -34,3 +44,5 @@ nodes:
systemExtensions:
officialExtensions:
- siderolabs/zfs
- siderolabs/nonfree-kmod-nvidia-lts
- siderolabs/nvidia-container-toolkit-lts