If watching objects with a filter, when an update was made that no longer matched the filter a DELETE event was correctly sent. However, the object that was returned by that delete was not the (correct) version before the update, but instead, the newer version. That meant the new object was not matched by the filter. This was a regression from behavior between cached watches on the server side and uncached watches, and thus broke downstream API clients. ([#46223](https://github.com/kubernetes/ku...

### Bug or Regression

- Kubeadm: fixed bug where v1beta3's ClusterConfiguration.APIServer.TimeoutForControlPlane is not respected in newer versions of kubeadm where v1beta4 is the default. ([#133776](https://github.com/kubernetes/kubernetes/pull/133776), [@HirazawaUi](https://github.com/HirazawaUi)) [SIG Cluster Lifecycle]

### Other (Cleanup or Flake)

* gitRepo volumes in pods no longer require git 1.8.5 or newer, older git versions are supported too now. ([#62394](https://github.com/kubernetes/kubernetes/pull/62394), [@jsafrane](https://github.com/jsafrane))

- `apiserver_storage_objects` (a newer version of `etcd_object_counts`) is promoted and marked as stable. ([#100082](https://github.com/kubernetes/kubernetes/pull/100082), [@logicalhan](https://github.com/logicalhan))

### Uncategorized

  3. Single-stack clusters do not need to change, but may choose to use the more specific flags.  Users can use either the older `--node-cidr-mask-size` flag or one of the newer `--node-cidr-mask-size-ipv4` or `--node-cidr-mask-size-ipv6` flags to configure the per-node IP mask size, provided that the flag's IP family matches the cluster's IP family (--cluster-cidr). ([#104691](https://github.com/kubernetes/kubernetes/pull/104691),...

            AVX512_BF16, AMX, etc.
            ([Intel Cascade Lake](https://www.intel.com/content/www/us/en/products/platforms/details/cascade-lake.html)
            and newer CPUs.)
            *   [Example performance speedups.](https://medium.com/intel-analytics-software/leverage-intel-deep-learning-optimizations-in-tensorflow-129faa80ee07)

### Failing Test

- Allowed Azure Disk e2es to use newer topology labels if available from nodes. ([#117216](https://github.com/kubernetes/kubernetes/pull/117216), [@gnufied](https://github.com/gnufied))

* Fix public IP not found issues for VMSS nodes ([#80703](https://github.com/kubernetes/kubernetes/pull/80703), [@feiskyer](https://github.com/feiskyer))
* Fix a conformance test that was skipped when client had a newer micro version than server. ([#80598](https://github.com/kubernetes/kubernetes/pull/80598), [@smarterclayton](https://github.com/smarterclayton))

* Implemented `kubeadm upgrade plan` for checking whether you can upgrade your cluster to a newer version ([#48899](https://github.com/kubernetes/kubernetes/pull/48899), [@luxas](https://github.com/luxas))
    * Implemented `kubeadm upgrade apply` for upgrading your cluster from one version to an other

" ) // This file contains two implementations of AES-GCM. The first implementation // (useGHASH) uses the KMCTR instruction to encrypt using AES in counter mode // and the KIMD instruction for GHASH. The second implementation (useGCM) uses // the newer KMA instruction which performs both operations (but still requires // KIMD to hash large nonces). // Keep in sync with crypto/tls.hasAESGCMHardwareSup. var useGHASH = cpu.S390XHasAES && cpu.S390XHasAESCTR && cpu.S390XHasGHASH var useGCM = useGHASH...