- To reduce chances of throttling, VM cache is set to nil when Azure node provisioning state is deleting ([#87635](https://github.com/kubernetes/kubernetes/pull/87635), [@feiskyer](https://github.com/feiskyer)) [SIG Cloud Provider]

- Ensure Job status updates are batched by 1s. This fixes an unlikely scenario when a sequence of immediately 
  completing pods could trigger a sequence of non-batched Job status updates. ([#118470](https://github.com/kubernetes/kubernetes/pull/118470), [@mimowo](https://github.com/mimowo)) [SIG Apps]

* kube-proxy: Add a lower-bound for conntrack (128k default) ([#33051](https://github.com/kubernetes/kubernetes/pull/33051), [@thockin](https://github.com/thockin))
* Use patched golang1.7.1 for cross-builds targeting darwin ([#33803](https://github.com/kubernetes/kubernetes/pull/33803), [@ixdy](https://github.com/ixdy))

- Pkg k8s.io/apiserver/pkg/storage/cacher, method (*Cacher) Wait(context.Context) error ([#125450](https://github.com/kubernetes/kubernetes/pull/125450), [@mauri870](https://github.com/mauri870)) [SIG API Machinery]

### Bug or Regression

- Kubelet config file will be backed up to `/etc/kubernetes/tmp/` folder with `kubeadm-kubelet-config` append with a random suffix as the filename ([#114695](https://github.com/kubernetes/kubernetes/pull/114695), [@chendave](https://github.com/chendave)) [SIG Cluster Lifecycle]

- Kube-scheduler now retries scheduling Pods rejected by the PodTopologySpread plugin when related Pods are created, deleted, updated,...

}

void NoOpDeallocator(void* data, size_t, void*) {}

TEST(CAPI, MalformedTensor) {
  // See https://github.com/tensorflow/tensorflow/issues/7394
  // num_dims = 0 implies a scalar, so should be backed by at least 4 bytes of
  // data.
  TF_Tensor* t =
      TF_NewTensor(TF_FLOAT, nullptr, 0, nullptr, 0, &NoOpDeallocator, nullptr);
  ASSERT_TRUE(t == nullptr);
}

TEST(CAPI, AllocateTensor) {

  fstype: ext3
```

([#56147](https://github.com/kubernetes/kubernetes/pull/56147), [@andyzhangx](https://github.com/andyzhangx))

### Scheduling

* Fixed a bug the in scheduler cache by using Pod UID as the cache key instead of namespace/name ([#61069](https://github.com/kubernetes/kubernetes/pull/61069), [@anfernee](https://github.com/anfernee))

     * character matches any number of characters in part of a name. If
     * the expression begins with one or more '?'s then exactly that
     * many characters will be matched whereas if it ends with '?'s
     * it will match that many characters <i>or less</i>.
     * <p>
     * Wildcard expressions will not filter workgroup names or server names.
     *
     * <blockquote><pre>

copy(em[k-len(prefix)-len(hashed):], prefix) copy(em[k-len(hashed):], hashed) return em, nil } // VerifyPKCS1v15 verifies an RSASSA-PKCS1-v1.5 signature. // // hash is the name of the hash function as returned by [crypto.Hash.String] // or the empty string to indicate that the message is signed directly. func VerifyPKCS1v15(pub *PublicKey, hash string, hashed []byte, sig []byte) error { fipsSelfTest() fips140.RecordApproved() checkApprovedHashNam(hash) return verifyPKCS1v15(pub, hash, hashed, sig) } func verifyPKCS1v15(pub...