Merge pull request #62464 from choury/fix-double-rlock-in-cpumanger

Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

avoid dobule RLock() in cpumanager

**What this PR does / why we need it**:

We met a deadlock when removing pod.
kubelet keeps logging:
```
Pod "xxxx" is terminated, but some containers are still running 
```
After debug, we found it stuck in `SetDefaultCPUSet` [here](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/cm/cpumanager/policy_static.go#L184)  while another goroutine are calling `GetCPUSetOrDefault` [here](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/cm/cpumanager/cpu_manager.go#L256).

According  golang/go#15418, It is not safe to double RLock a RWMutex.

**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:

none

**Special notes for your reviewer**:

**Release note**:

```release-note
removed unsafe double RLock in cpumanager
```

pkg/kubelet/cm/cpumanager/state/state_mem.go

@@ -56,9 +56,6 @@ func (s *stateMemory) GetDefaultCPUSet() cpuset.CPUSet {
 }
 
 func (s *stateMemory) GetCPUSetOrDefault(containerID string) cpuset.CPUSet {
-	s.RLock()
-	defer s.RUnlock()
-
 	if res, ok := s.GetCPUSet(containerID); ok {
 		return res
 	}