Skip to content

feat(gpu): start device topology registry in base plugin #1108

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
luomingmeng wants to merge 2 commits into
base: main
Choose a base branch
from
Open

feat(gpu): start device topology registry in base plugin #1108

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0fc4ff7
feat(gpu): start device topology registry in base plugin
luomingmeng Mar 26, 2026
a5c84a7
fix(device): handle nil return from PickLatestDeviceTopology
luomingmeng Mar 26, 2026
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
2 changes: 1 addition & 1 deletion pkg/agent/qrm-plugins/gpu/baseplugin/base.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,6 +98,7 @@ func NewBasePlugin(

// Run starts the asynchronous tasks of the plugin
func (p *BasePlugin) Run(stopCh <-chan struct{}) {
go p.DeviceTopologyRegistry.Run(stopCh)
go func() {
select {
case <-p.stateInitializedCh:
Expand All @@ -108,7 +109,6 @@ func (p *BasePlugin) Run(stopCh <-chan struct{}) {
return
}
}()
go p.DeviceTopologyRegistry.Run(stopCh)
}

// GetState may return a nil state because the state is only initialized when InitState is called.
Expand Down
2 changes: 2 additions & 0 deletions pkg/agent/qrm-plugins/gpu/staticpolicy/policy.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,6 +149,8 @@ func (p *StaticPolicy) Start() (err error) {
periodicalhandler.ReadyToStartHandlersByGroup(appqrm.QRMGPUPluginPeriodicalHandlerGroupName)
}, 5*time.Second, p.stopCh)

p.BasePlugin.Run(p.stopCh)

return nil
}

Expand Down
7 changes: 6 additions & 1 deletion pkg/util/machine/device.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,7 +224,12 @@ func (r *DeviceTopologyRegistry) GetLatestDeviceTopology(deviceNames []string) (
return nil, err
}

return PickLatestDeviceTopology(topologiesMap), nil
latestTopology := PickLatestDeviceTopology(topologiesMap)
if latestTopology == nil {
return nil, fmt.Errorf("no latest device topology")
}

return latestTopology, nil
}

// GetDeviceNUMAAffinity retrieves a map of a certain device A to the list of devices in device B that it has an affinity with.
Expand Down
Loading