NVIDIA / gpu-operator / 35103690707

44%

main: 44%

push

github

feat: add daemonset affinity and nodeSelector to ClusterPolicy

Expose spec.daemonsets.affinity and spec.daemonsets.nodeSelector on
ClusterPolicy and wire them through Helm. User nodeSelector labels are
merged additively with existing operand nodeSelector entries, including
protected keys such as nvidia.com/gpu.deploy and nvidia.com/mps.capable.
Cluster policy affinity is merged with existing daemonset affinity using a
cartesian product on NodeSelectorTerms so constraints stay additive.

Inline the full corev1.Affinity OpenAPI schema on daemonsets (matching
NVIDIADriver nodeAffinity), regenerate ClusterPolicy and GPUCluster CRDs,
and sync helm chart CRDs via make sync-crds.

Signed-off-by: Aaron <amark@g.jct.ac.il>

44 of 106 new or added lines in 2 files covered. (41.51%)

5754 of 13073 relevant lines covered (44.01%)

0.44 hits per line