ROCm / k8s-device-plugin

Kubernetes (k8s) device plugin to enable registration of AMD GPU to a container cluster

ROCm/k8s-device-plugin Issues

[Documentation]: Introduction link broken
Closed a month ago
[Issue]: namespace configuration in Helm chart
Closed a month ago
question about privileged mode for labeller
Updated 2 months ago4
[Issue]: need a way to overwrite node_selector labels
Updated 2 months ago
Permissions for the /dev/{kfd,dri/renderXXXX} devices in containers
Updated 2 months ago1
[Documentation]: clarification on kernel/driver installation
Closed 2 months ago1
[Documentation]: MI200 support
Closed 2 months ago1
[Feature]: Support detection, allocation and resetting of GPU partitions in CDNA cards
Updated 3 months ago
Does this device plugin support kubevirt about GPU?
Closed 3 years ago1
Is AMD Radeon Vega 8 supported?
Closed 5 months ago18
[Issue]: benchmark example is broken
Closed 4 months ago1
[Issue]: ArtifactHUB helm install points to non-existant chart repo
Closed 5 months ago1
[Issue]: Helm Chart unavailable
Closed 5 months ago2
Runtime Error with AMD GPU Helm Chart Installation in Kubernetes
Updated 5 months ago
404 link in README.md
Closed 5 months ago
GPU isolation options
Updated 5 months ago
Libraries/binaries mounted in the container (analogous to NVIDIA_DRIVER_CAPABILITIES)
Updated 5 months ago
Registration failing on K3s
Closed a year ago12
EKS Support
Updated a year ago
Node detected and labelled but has no capacity?
Closed a year ago8
Is it possible to restrict vram/cu per pod?
Closed 4 years ago2
add podLabels and podAnnotations to labeller's and device plugin's pods
Updated a year ago
device plugin daemonset tries to load itself on non-amd64 nodes
Closed a year ago1
Dead chart repo. Please fix
Closed a year ago1
Issues in helm chart
Updated 2 years ago1
Helm Incompitable Version with AWS EKS Cluster
Closed 2 years ago
Prometheus metrics of gpu resources
Closed 2 years ago1
Talos Vega 11 Fail to glob topology properties
Closed 2 years ago3
Not Working On Ubuntu 20.04 Kernel 5.8 HWE
Closed 3 years ago3
would all pods using the same GPU are assigned to the same NUMA?
Closed 3 years ago1
Create helm chart repository
Closed 3 years ago1
wrong
Closed 3 years ago
Is amdgpu pro required?
Closed 3 years ago1
Can't find any GPU
Closed 3 years ago3
[Q] Pod can specify GPU device-id?
Closed 4 years ago1
Deprecated k8s apiversion in yaml
Closed 4 years ago1
New release compatible with k8s 1.10?
Closed 6 years ago1