Skip to content
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

Add method IsFabricAttached to Device interface #44

Merged
merged 1 commit into from
Sep 28, 2024
Merged

Add method IsFabricAttached to Device interface #44

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
42 changes: 42 additions & 0 deletions pkg/nvlib/device/device.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@ type Device interface {
GetMigDevices() ([]MigDevice, error)
GetMigProfiles() ([]MigProfile, error)
GetPCIBusID() (string, error)
IsFabricAttached() (bool, error)
IsMigCapable() (bool, error)
IsMigEnabled() (bool, error)
VisitMigDevices(func(j int, m MigDevice) error) error
Expand Down Expand Up @@ -208,6 +209,47 @@ func (d *device) IsMigEnabled() (bool, error) {
return (mode == nvml.DEVICE_MIG_ENABLE), nil
}

// IsFabricAttached checks if a device is attached to a GPU fabric.
func (d *device) IsFabricAttached() (bool, error) {
if d.lib.hasSymbol("nvmlDeviceGetGpuFabricInfo") {
klueska marked this conversation as resolved.
Show resolved Hide resolved
info, ret := d.GetGpuFabricInfo()
if ret == nvml.ERROR_NOT_SUPPORTED {
return false, nil
}
klueska marked this conversation as resolved.
Show resolved Hide resolved
if ret != nvml.SUCCESS {
return false, fmt.Errorf("error getting GPU Fabric Info: %v", ret)
}
if info.State != nvml.GPU_FABRIC_STATE_COMPLETED {
return false, nil
}
if nvml.Return(info.Status) != nvml.SUCCESS {
return false, nil
}

return true, nil
}

if d.lib.hasSymbol("nvmlDeviceGetGpuFabricInfoV") {
info, ret := d.GetGpuFabricInfoV().V2()
if ret == nvml.ERROR_NOT_SUPPORTED {
return false, nil
}
if ret != nvml.SUCCESS {
return false, fmt.Errorf("error getting GPU Fabric Info: %v", ret)
}
if info.State != nvml.GPU_FABRIC_STATE_COMPLETED {
return false, nil
}
if nvml.Return(info.Status) != nvml.SUCCESS {
return false, nil
}

return true, nil
}

return false, nil
}

// VisitMigDevices walks a top-level device and invokes a callback function for each MIG device configured on it.
func (d *device) VisitMigDevices(visit func(int, MigDevice) error) error {
capable, err := d.IsMigCapable()
Expand Down