nvidia

package

v0.0.0-...-f849f17 Latest Latest Go to latest Published: Mar 8, 2023 License: Apache-2.0 Imports: 29 Imported by: 0

Details

Valid go.mod file

The Go module system was introduced in Go 1.11 and is the official dependency management solution for Go.
Redistributable license

Redistributable licenses place minimal restrictions on how software can be used, modified, and redistributed.
Tagged version

Modules with tagged versions give importers more predictable builds.
Stable version

When a project reaches major version v1 it is considered stable.
Learn more about best practices

Repository

github.com/xpucube/xpu-device-plugin

Links

Open Source Insights

Documentation ¶

Index ¶

Constants
func NewSharedGPUManager(enableMPS, healthCheck bool, bp ShareUnit) *sharedGPUManager
func StackTrace(all bool) string
type NvidiaDevicePlugin
- func NewNvidiaDevicePlugin(mps, healthCheck bool) (*NvidiaDevicePlugin, error)
type ShareUnit

Constants ¶

View Source

const (
	OptimisticLockErrorMsg = "the object has been modified; please apply your changes to the latest version and try again"

	EnvResourceIndex           = "OPENXPU_XPU_SHARES_INDEX"
	EnvResourceByPod           = "OPENXPU_XPU_SHARES_POD"
	EnvResourceByContainer     = "OPENXPU_XPU_SHARES"
	EnvResourceByDev           = "OPENXPU_XPU_SHARES_TOTAL"
	EnvAssignedFlag            = "OPENXPU_XPU_SHARES_ALLOCATED"
	EnvResourceAssumeTime      = "OPENXPU_XPU_SHARES_FILTER_STAMP"
	EnvResourceAssignTime      = "OPENXPU_XPU_SHARES_ALLOCATED_STAMP"
	EnvNodeLabelForDisableCGPU = "xpu.disable.isolation"

	GiBPrefix    = ShareUnit("GiB")
	MiBPrefix    = ShareUnit("MiB")
	XPUShareUnit = 1000
)

Variables ¶

This section is empty.

Functions ¶

func NewSharedGPUManager ¶

func NewSharedGPUManager(enableMPS, healthCheck bool, bp ShareUnit) *sharedGPUManager

func StackTrace ¶

func StackTrace(all bool) string

Types ¶

type NvidiaDevicePlugin ¶

type NvidiaDevicePlugin struct {
	sync.RWMutex
	// contains filtered or unexported fields
}

NvidiaDevicePlugin implements the Kubernetes device plugin API

func NewNvidiaDevicePlugin ¶

func NewNvidiaDevicePlugin(mps, healthCheck bool) (*NvidiaDevicePlugin, error)

NewNvidiaDevicePlugin returns an initialized NvidiaDevicePlugin

func (*NvidiaDevicePlugin) Allocate ¶

func (m *NvidiaDevicePlugin) Allocate(ctx context.Context,
	reqs *pluginapi.AllocateRequest) (*pluginapi.AllocateResponse, error)

Allocate which return list of devices.

func (*NvidiaDevicePlugin) GetDeviceNameByIndex ¶

func (m *NvidiaDevicePlugin) GetDeviceNameByIndex(index uint) (name string, found bool)

func (*NvidiaDevicePlugin) GetDevicePluginOptions ¶

func (m *NvidiaDevicePlugin) GetDevicePluginOptions(context.Context, *pluginapi.Empty) (*pluginapi.DevicePluginOptions, error)

func (*NvidiaDevicePlugin) ListAndWatch ¶

func (m *NvidiaDevicePlugin) ListAndWatch(e *pluginapi.Empty, s pluginapi.DevicePlugin_ListAndWatchServer) error

ListAndWatch lists devices and update that list according to the health status

func (*NvidiaDevicePlugin) PreStartContainer ¶

func (m *NvidiaDevicePlugin) PreStartContainer(context.Context, *pluginapi.PreStartContainerRequest) (*pluginapi.PreStartContainerResponse, error)

func (*NvidiaDevicePlugin) Register ¶

func (m *NvidiaDevicePlugin) Register(kubeletEndpoint, resourceName string) error

Register registers the device plugin for the given resourceName with Kubelet.

func (*NvidiaDevicePlugin) Serve ¶

func (m *NvidiaDevicePlugin) Serve() error

Serve starts the gRPC server and register the device plugin to Kubelet

func (*NvidiaDevicePlugin) Start ¶

func (m *NvidiaDevicePlugin) Start() error

Start starts the gRPC server of the device plugin

func (*NvidiaDevicePlugin) Stop ¶

func (m *NvidiaDevicePlugin) Stop() error

Stop stops the gRPC server

type ShareUnit ¶

type ShareUnit string

ShareUnit describes GPU Share

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL