Skip to content

Add GDRCOPY_ENABLED=true to device-plugin when GDRCopy is enabled #1749

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cdesiniotis merged 1 commit into from
Oct 3, 2025
Merged

Add GDRCOPY_ENABLED=true to device-plugin when GDRCopy is enabled #1749

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions controllers/object_controls.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,6 +133,8 @@ const (
GDSEnabledEnvName = "GDS_ENABLED"
// MOFEDEnabledEnvName is the env name to enable MOFED devices injection with device-plugin
MOFEDEnabledEnvName = "MOFED_ENABLED"
// GDRCopyEnabledEnvName is the envvar that enables injection of the GDRCopy device node with the device-plugin
GDRCopyEnabledEnvName = "GDRCOPY_ENABLED"
// ServiceMonitorCRDName is the name of the CRD defining the ServiceMonitor kind
ServiceMonitorCRDName = "servicemonitors.monitoring.coreos.com"
// DefaultToolkitInstallDir is the default toolkit installation directory on the host
Expand Down Expand Up @@ -1500,6 +1502,10 @@ func TransformDevicePlugin(obj *appsv1.DaemonSet, config *gpuv1.ClusterPolicySpe
setContainerEnv(mainContainer, MOFEDEnabledEnvName, "true")
}

if config.GDRCopy != nil && config.GDRCopy.IsEnabled() {
setContainerEnv(mainContainer, GDRCopyEnabledEnvName, "true")
}

// apply plugin configuration through ConfigMap if one is provided
err = handleDevicePluginConfig(obj, config)
if err != nil {
Expand Down
38 changes: 38 additions & 0 deletions controllers/transforms_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -871,6 +871,44 @@ func TestTransformDevicePlugin(t *testing.T) {
},
}).WithContainer(corev1.Container{Name: "dummy"}).WithPullSecret("pull-secret").WithRuntimeClassName("nvidia"),
},
{
description: "transform device plugin, gds and gdrcopy enabled",
ds: NewDaemonset().
WithContainer(corev1.Container{Name: "nvidia-device-plugin"}),
cpSpec: &gpuv1.ClusterPolicySpec{
DevicePlugin: gpuv1.DevicePluginSpec{
Repository: "nvcr.io/nvidia/cloud-native",
Image: "nvidia-device-plugin",
Version: "v1.0.0",
ImagePullPolicy: "IfNotPresent",
},
Toolkit: gpuv1.ToolkitSpec{
Enabled: newBoolPtr(true),
InstallDir: "/path/to/install",
},
GDRCopy: &gpuv1.GDRCopySpec{
Enabled: newBoolPtr(true),
},
GPUDirectStorage: &gpuv1.GPUDirectStorageSpec{
Enabled: newBoolPtr(true),
},
},
expectedDs: NewDaemonset().WithContainer(corev1.Container{
Name: "nvidia-device-plugin",
Image: "nvcr.io/nvidia/cloud-native/nvidia-device-plugin:v1.0.0",
ImagePullPolicy: corev1.PullIfNotPresent,
Env: []corev1.EnvVar{
{Name: GDSEnabledEnvName, Value: "true"},
{Name: MOFEDEnabledEnvName, Value: "true"},
{Name: GDRCopyEnabledEnvName, Value: "true"},
{Name: "NVIDIA_MIG_MONITOR_DEVICES", Value: "all"},
{Name: CDIEnabledEnvName, Value: "true"},
{Name: DeviceListStrategyEnvName, Value: "cdi-annotations,cdi-cri"},
{Name: CDIAnnotationPrefixEnvName, Value: "cdi.k8s.io/"},
{Name: NvidiaCDIHookPathEnvName, Value: "/path/to/install/toolkit/nvidia-cdi-hook"},
},
}).WithRuntimeClassName("nvidia"),
},
}

for _, tc := range testCases {
Expand Down