serving support schedulerName？

[631068264]

What version of Knative?

I use kubeflow 1.6.1 . According to it use Knative 1.2.5

I want to use GPU share feature for predictor via kubeflow + volcano . But I get error

I apply this

apiVersion: "serving.kserve.io/v1beta1"
kind: "InferenceService"
metadata:
  name: "volcano"
spec:
  predictor:
    schedulerName: volcano
    containers:
      - name: kserve-container
        image: xxx:8080/library/model/firesmoke:v1
        env:
          - name: MODEL_NAME
            value: volcano
        command:
          - python
          - -m
          - fire_smoke
        resources:
          limits:
            volcano.sh/gpu-memory: 2048

error

Events:
  Type     Reason         Age                    From                Message
  ----     ------         ----                   ----                -------
  Warning  InternalError  3m4s (x17 over 8m32s)  v1beta1Controllers  fails to reconcile predictor: admission webhook "validation.webhook.serving.knative.dev" denied the request: validation failed: must not set the field(s): spec.template.spec.schedulerName

Expected Behavior

Actual Behavior

Steps to Reproduce the Problem

[maidol]

source code

...
	if cfg.Features.PodSpecSchedulerName != config.Disabled {
		out.SchedulerName = in.SchedulerName
	}
	if cfg.Features.PodSpecInitContainers != config.Disabled {
		out.InitContainers = in.InitContainers
	}
...

to enabled special schedulername feature

kubectl edit cm config-features -n knative-serving

apiVersion: v1
kind: ConfigMap
metadata:
  name: config-features
  namespace: knative-serving
data:
  kubernetes.podspec-schedulername: enabled # enabled special schedulername feature