fix synchronous pvc/pv bind error when volcano schedule pod with pvc

Signed-off-by: gj199575 <409237405@qq.com>

pkg/scheduler/cache/cache.go

@@ -138,7 +138,6 @@ type SchedulerCache struct {
 	vcInformerFactory vcinformer.SharedInformerFactory
 
 	BindFlowChannel chan *schedulingapi.TaskInfo
-	bindCache       []*schedulingapi.TaskInfo
 	batchNum        int
 
 	// A map from image name to its imageState.
@@ -768,24 +767,22 @@ func (sc *SchedulerCache) Evict(taskInfo *schedulingapi.TaskInfo, reason string)
 }
 
 // Bind binds task to the target host.
-func (sc *SchedulerCache) Bind(tasks []*schedulingapi.TaskInfo) error {
-	go func(taskArray []*schedulingapi.TaskInfo) {
-		tmp := time.Now()
-		errTasks, err := sc.Binder.Bind(sc.kubeClient, taskArray)
-		if err == nil {
-			klog.V(3).Infof("bind ok, latency %v", time.Since(tmp))
-			for _, task := range tasks {
-				sc.Recorder.Eventf(task.Pod, v1.EventTypeNormal, "Scheduled", "Successfully assigned %v/%v to %v",
-					task.Namespace, task.Name, task.NodeName)
-			}
-		} else {
-			for _, task := range errTasks {
-				klog.V(2).Infof("resyncTask task %s", task.Name)
-				sc.VolumeBinder.RevertVolumes(task, task.PodVolumes)
-				sc.resyncTask(task)
-			}
+func (sc *SchedulerCache) Bind(taskArray []*schedulingapi.TaskInfo) error {
+	tmp := time.Now()
+	errTasks, err := sc.Binder.Bind(sc.kubeClient, taskArray)
+	if err == nil {
+		klog.V(3).Infof("bind ok, latency %v", time.Since(tmp))
+		for _, task := range taskArray {
+			sc.Recorder.Eventf(task.Pod, v1.EventTypeNormal, "Scheduled", "Successfully assigned %v/%v to %v",
+				task.Namespace, task.Name, task.NodeName)
+		}
+	} else {
+		for _, task := range errTasks {
+			klog.V(2).Infof("resyncTask task %s", task.Name)
+			sc.VolumeBinder.RevertVolumes(task, task.PodVolumes)
+			sc.resyncTask(task)
 		}
-	}(tasks)
+	}
 	return nil
 }
 
@@ -988,16 +985,17 @@ func (sc *SchedulerCache) AddBindTask(taskInfo *schedulingapi.TaskInfo) error {
 }
 
 func (sc *SchedulerCache) processBindTask() {
+	var bindCache []*schedulingapi.TaskInfo = make([]*schedulingapi.TaskInfo, 10)
 	for {
 		select {
 		case taskInfo, ok := <-sc.BindFlowChannel:
 			if !ok {
 				return
 			}
 
-			sc.bindCache = append(sc.bindCache, taskInfo)
-			if len(sc.bindCache) == sc.batchNum {
-				sc.BindTask()
+			bindCache = append(bindCache, taskInfo)
+			if len(bindCache) == sc.batchNum {
+				go sc.BindTask(bindCache)
 			}
 		default:
 		}
@@ -1007,17 +1005,18 @@ func (sc *SchedulerCache) processBindTask() {
 		}
 	}
 
-	if len(sc.bindCache) == 0 {
+	if len(bindCache) == 0 {
 		return
 	}
 
-	sc.BindTask()
+	go sc.BindTask(bindCache)
 }
 
-func (sc *SchedulerCache) BindTask() {
-	klog.V(5).Infof("batch bind task count %d", len(sc.bindCache))
+func (sc *SchedulerCache) BindTask(bindCache []*schedulingapi.TaskInfo) {
+	klog.V(5).Infof("batch bind task count %d", len(bindCache))
+
 	successfulTasks := make([]*schedulingapi.TaskInfo, 0)
-	for _, task := range sc.bindCache {
+	for _, task := range bindCache {
 		if err := sc.VolumeBinder.BindVolumes(task, task.PodVolumes); err != nil {
 			klog.Errorf("task %s/%s bind Volumes failed: %#v", task.Namespace, task.Name, err)
 			sc.VolumeBinder.RevertVolumes(task, task.PodVolumes)
@@ -1038,8 +1037,6 @@ func (sc *SchedulerCache) BindTask() {
 	for _, task := range successfulTasks {
 		metrics.UpdateTaskScheduleDuration(metrics.Duration(task.Pod.CreationTimestamp.Time))
 	}
-
-	sc.bindCache = sc.bindCache[0:0]
 }
 
 // Snapshot returns the complete snapshot of the cluster from cache