- Extended resources backed by DRA feature allowed cluster operator to specify `extendedResourceName` in `DeviceClass`, and application operator to continue using extended resources in pod's requests to request for DRA devices matching the DeviceClass.
  

- First pod with required affinity terms can schedule only on nodes with matching topology keys. ([#91168](https://github.com/kubernetes/kubernetes/pull/91168), [@ahg-g](https://github.com/ahg-g)) [SIG Scheduling]

- Fixed recursive LIST from watch cache returning object matching key. ([#125584](https://github.com/kubernetes/kubernetes/pull/125584), [@serathius](https://github.com/serathius)) [SIG API Machinery and Testing]

- Fix azure disk naming matching issue due to case sensitive comparison ([#81720](https://github.com/kubernetes/kubernetes/pull/81720), [@andyzhangx](https://github.com/andyzhangx))

* kubectl cp now safely allows unpacking of symlinks that may point outside the destination directory ([#82384](https://github.com/kubernetes/kubernetes/pull/82384), [@tallclair](https://github.com/tallclair))
* fix azure disk naming matching issue due to case sensitive comparison ([#81720](https://github.com/kubernetes/kubernetes/pull/81720), [@andyzhangx](https://github.com/andyzhangx))

* Limit use of tags when calling EC2 API to prevent API throttling for very large clusters ([#76749](https://github.com/kubernetes/kubernetes/pull/76749), [@mcrute](https://github.com/mcrute))
* fix azure disk naming matching issue due to case sensitive comparison ([#81720](https://github.com/kubernetes/kubernetes/pull/81720), [@andyzhangx](https://github.com/andyzhangx))

- Kube-apiserver: The timeout used when making health check calls to etcd can now be configured with `--etcd-healthcheck-timeout`. The default timeout is 2 seconds, matching the previous behavior. ([#93244](https://github.com/kubernetes/kubernetes/pull/93244), [@Sh4d1](https://github.com/Sh4d1)) [SIG API Machinery]

- Added optional `--detect-local-mode` flag to kube-proxy. Valid values are "ClusterCIDR" (default matching previous behavior) and "NodeCIDR" ([#87748](https://github.com/kubernetes/kubernetes/pull/87748), [@satyasm](https://github.com/satyasm)) [SIG Cluster Lifecycle, Network and Scheduling]

- The errors in `k8s.io/apimachinery/pkg/api/meta` now support for the `stdlibs` `errors.Is` matching, including when wrapped. ([#111808](https://github.com/kubernetes/kubernetes/pull/111808), [@alvaroaleman](https://github.com/alvaroaleman))

* kubeadm now detects the Docker cgroup driver and starts the kubelet with the matching driver. This eliminates a common error experienced by new users in when the Docker cgroup driver is not the same as the one set for the kubelet due to different Linux distributions setting different cgroup drivers for Docker, making it hard to start...