lock: Moving locking to handler layer. (#3381)

This is implemented so that the issues like in the
following flow don't affect the behavior of operation.

```
GetObjectInfo()
.... --> Time window for mutation (no lock held)
.... --> Time window for mutation (no lock held)
GetObject()
```

This happens when two simultaneous uploads are made
to the same object the object has returned wrong
info to the client.

Another classic example is "CopyObject" API itself
which reads from a source object and copies to
destination object.

Fixes #3370
Fixes #2912

cmd/xl-v1-bucket.go

@@ -36,10 +36,6 @@ func (xl xlObjects) MakeBucket(bucket string) error {
 		return traceError(BucketNameInvalid{Bucket: bucket})
 	}
 
-	bucketLock := nsMutex.NewNSLock(bucket, "")
-	bucketLock.Lock()
-	defer bucketLock.Unlock()
-
 	// Initialize sync waitgroup.
 	var wg = &sync.WaitGroup{}
 
@@ -153,10 +149,6 @@ func (xl xlObjects) GetBucketInfo(bucket string) (BucketInfo, error) {
 		return BucketInfo{}, BucketNameInvalid{Bucket: bucket}
 	}
 
-	bucketLock := nsMutex.NewNSLock(bucket, "")
-	bucketLock.RLock()
-	defer bucketLock.RUnlock()
-
 	bucketInfo, err := xl.getBucketInfo(bucket)
 	if err != nil {
 		return BucketInfo{}, toObjectErr(err, bucket)
@@ -227,10 +219,6 @@ func (xl xlObjects) DeleteBucket(bucket string) error {
 		return BucketNameInvalid{Bucket: bucket}
 	}
 
-	bucketLock := nsMutex.NewNSLock(bucket, "")
-	bucketLock.Lock()
-	defer bucketLock.Unlock()
-
 	// Collect if all disks report volume not found.
 	var wg = &sync.WaitGroup{}
 	var dErrs = make([]error, len(xl.storageDisks))