Skip to content

add nvidia-modeset device if present #122

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

Open
danisla wants to merge 1 commit into
base: master
Choose a base branch
from
Open

add nvidia-modeset device if present #122

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: 5 additions & 1 deletion pkg/gpu/nvidia/alpha_plugin_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,13 +98,16 @@ func TestNvidiaGPUManagerAlphaAPI(t *testing.T) {
testNvidiaCtlDevice := path.Join(testDevDir, nvidiaCtlDevice)
testNvidiaUVMDevice := path.Join(testDevDir, nvidiaUVMDevice)
testNvidiaUVMToolsDevice := path.Join(testDevDir, nvidiaUVMToolsDevice)
testNvidiaModesetDevice := path.Join(testDevDir, nvidiaModesetDevice)
os.Create(testNvidiaCtlDevice)
os.Create(testNvidiaUVMDevice)
os.Create(testNvidiaUVMToolsDevice)
testGpuManager.defaultDevices = []string{testNvidiaCtlDevice, testNvidiaUVMDevice, testNvidiaUVMToolsDevice}
os.Create(testNvidiaModesetDevice)
testGpuManager.defaultDevices = []string{testNvidiaCtlDevice, testNvidiaUVMDevice, testNvidiaUVMToolsDevice, testNvidiaModesetDevice}
defer os.Remove(testNvidiaCtlDevice)
defer os.Remove(testNvidiaUVMDevice)
defer os.Remove(testNvidiaUVMToolsDevice)
defer os.Remove(testNvidiaModesetDevice)

gpu1 := path.Join(testDevDir, "nvidia1")
gpu2 := path.Join(testDevDir, "nvidia2")
Expand Down Expand Up @@ -179,6 +182,7 @@ func TestNvidiaGPUManagerAlphaAPI(t *testing.T) {
as.Contains(retDevices, testNvidiaCtlDevice)
as.Contains(retDevices, testNvidiaUVMDevice)
as.Contains(retDevices, testNvidiaUVMToolsDevice)
as.Contains(retDevices, testNvidiaModesetDevice)
resp, err = client.Allocate(context.Background(), &pluginapi.AllocateRequest{
DevicesIDs: []string{"nvidia1", "nvidia3"},
})
Expand Down
6 changes: 5 additions & 1 deletion pkg/gpu/nvidia/beta_plugin_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,13 +45,16 @@ func TestNvidiaGPUManagerBetaAPI(t *testing.T) {
testNvidiaCtlDevice := path.Join(testDevDir, nvidiaCtlDevice)
testNvidiaUVMDevice := path.Join(testDevDir, nvidiaUVMDevice)
testNvidiaUVMToolsDevice := path.Join(testDevDir, nvidiaUVMToolsDevice)
testNvidiaModesetDevice := path.Join(testDevDir, nvidiaModesetDevice)
os.Create(testNvidiaCtlDevice)
os.Create(testNvidiaUVMDevice)
os.Create(testNvidiaUVMToolsDevice)
testGpuManager.defaultDevices = []string{testNvidiaCtlDevice, testNvidiaUVMDevice, testNvidiaUVMToolsDevice}
os.Create(testNvidiaModesetDevice)
testGpuManager.defaultDevices = []string{testNvidiaCtlDevice, testNvidiaUVMDevice, testNvidiaUVMToolsDevice, testNvidiaModesetDevice}
defer os.Remove(testNvidiaCtlDevice)
defer os.Remove(testNvidiaUVMDevice)
defer os.Remove(testNvidiaUVMToolsDevice)
defer os.Remove(testNvidiaModesetDevice)

gpu1 := path.Join(testDevDir, "nvidia1")
gpu2 := path.Join(testDevDir, "nvidia2")
Expand Down Expand Up @@ -123,6 +126,7 @@ func TestNvidiaGPUManagerBetaAPI(t *testing.T) {
as.Contains(retDevices, testNvidiaCtlDevice)
as.Contains(retDevices, testNvidiaUVMDevice)
as.Contains(retDevices, testNvidiaUVMToolsDevice)
as.Contains(retDevices, testNvidiaModesetDevice)
resp, err = client.Allocate(context.Background(), &pluginapi.AllocateRequest{
ContainerRequests: []*pluginapi.ContainerAllocateRequest{
{DevicesIDs: []string{"nvidia1", "nvidia3"}}}})
Expand Down
12 changes: 10 additions & 2 deletions pkg/gpu/nvidia/manager.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,8 +34,11 @@ const (
// If the driver installed correctly, these two devices will be there.
nvidiaCtlDevice = "nvidiactl"
nvidiaUVMDevice = "nvidia-uvm"
// Optional device.
nvidiaUVMToolsDevice = "nvidia-uvm-tools"

// Optional devices.
nvidiaUVMToolsDevice = "nvidia-uvm-tools"
nvidiaModesetDevice = "nvidia-modeset"

nvidiaDeviceRE = `^nvidia[0-9]*$`
gpuCheckInterval = 10 * time.Second
pluginSocketCheckInterval = 1 * time.Second
Expand Down Expand Up @@ -149,6 +152,11 @@ func (ngm *nvidiaGPUManager) Start() error {

ngm.defaultDevices = []string{nvidiaCtlDevicePath, nvidiaUVMDevicePath}

nvidiaModesetDevicePath := path.Join(ngm.devDirectory, nvidiaModesetDevice)
if _, err := os.Stat(nvidiaModesetDevicePath); err == nil {
ngm.defaultDevices = append(ngm.defaultDevices, nvidiaModesetDevicePath)
}

nvidiaUVMToolsDevicePath := path.Join(ngm.devDirectory, nvidiaUVMToolsDevice)
if _, err := os.Stat(nvidiaUVMToolsDevicePath); err == nil {
ngm.defaultDevices = append(ngm.defaultDevices, nvidiaUVMToolsDevicePath)
Expand Down
6 changes: 5 additions & 1 deletion pkg/gpu/nvidia/multiple_versions_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,13 +46,16 @@ func TestNvidiaGPUManagerMultuipleAPIs(t *testing.T) {
testNvidiaCtlDevice := path.Join(testDevDir, nvidiaCtlDevice)
testNvidiaUVMDevice := path.Join(testDevDir, nvidiaUVMDevice)
testNvidiaUVMToolsDevice := path.Join(testDevDir, nvidiaUVMToolsDevice)
testNvidiaModesetDevice := path.Join(testDevDir, nvidiaModesetDevice)
os.Create(testNvidiaCtlDevice)
os.Create(testNvidiaUVMDevice)
os.Create(testNvidiaUVMToolsDevice)
testGpuManager.defaultDevices = []string{testNvidiaCtlDevice, testNvidiaUVMDevice, testNvidiaUVMToolsDevice}
os.Create(testNvidiaModesetDevice)
testGpuManager.defaultDevices = []string{testNvidiaCtlDevice, testNvidiaUVMDevice, testNvidiaUVMToolsDevice, testNvidiaModesetDevice}
defer os.Remove(testNvidiaCtlDevice)
defer os.Remove(testNvidiaUVMDevice)
defer os.Remove(testNvidiaUVMToolsDevice)
defer os.Remove(testNvidiaModesetDevice)

gpu1 := path.Join(testDevDir, "nvidia1")
gpu2 := path.Join(testDevDir, "nvidia2")
Expand Down Expand Up @@ -125,6 +128,7 @@ func TestNvidiaGPUManagerMultuipleAPIs(t *testing.T) {
as.Contains(retDevices, testNvidiaCtlDevice)
as.Contains(retDevices, testNvidiaUVMDevice)
as.Contains(retDevices, testNvidiaUVMToolsDevice)
as.Contains(retDevices, testNvidiaModesetDevice)
resp, err = clientBeta.Allocate(context.Background(), &pluginbeta.AllocateRequest{
ContainerRequests: []*pluginbeta.ContainerAllocateRequest{
{DevicesIDs: []string{"nvidia1", "nvidia3"}}}})
Expand Down