Potential impacts:

  Previously, for IPv6, if the podSubnet had a mask lower than /112, kubeadm calculated a node-mask to be multiple of eight and splitting the available bits to maximise the number used for nodes. ([#95723](https://github.com/kubernetes/kubernetes/pull/95723), [@aojea](https://github.com/aojea)) [SIG Cluster Lifecycle]

    * action required: Cloud providers other than GCE might want to update their deployments or sample yaml files to reuse the role created via add-on.

### Other notable changes

- Failed pods associated with a job with `parallelism = 1` are recreated by the job controller honoring exponential backoff delay again. However, for jobs with `parallelism > 1`, pods might be created without exponential backoff delay. ([#115021](https://github.com/kubernetes/kubernetes/pull/115021), [@nikhita](https://github.com/nikhita)) [SIG Apps]

pkg syscall (netbsd-arm64-cgo), const SCM_CREDS = 4
pkg syscall (netbsd-arm64-cgo), const SCM_CREDS ideal-int
pkg syscall (netbsd-arm64-cgo), const SCM_RIGHTS = 1
pkg syscall (netbsd-arm64-cgo), const SCM_RIGHTS ideal-int
pkg syscall (netbsd-arm64-cgo), const SCM_TIMESTAMP = 8
pkg syscall (netbsd-arm64-cgo), const SCM_TIMESTAMP ideal-int
pkg syscall (netbsd-arm64-cgo), const S_IEXEC = 64

- Add possibility to configure controller manager to use IPv6 dual stack:
  use `--cluster-cidr="<cidr1>,<cidr2>"`.
  Notes:
  1. Only the first two CIDRs are used (soft limits for Alpha, might be lifted later on).
  2. Only the "RangeAllocator" (default) is allowed as a value for `--cidr-allocator-type`. Cloud allocators are not compatible with IPv6 dual stack

pkg syscall (darwin-arm64), const RUSAGE_SELF ideal-int
pkg syscall (darwin-arm64), const SCM_CREDS = 3
pkg syscall (darwin-arm64), const SCM_CREDS ideal-int
pkg syscall (darwin-arm64), const SCM_RIGHTS = 1
pkg syscall (darwin-arm64), const SCM_RIGHTS ideal-int
pkg syscall (darwin-arm64), const SCM_TIMESTAMP = 2
pkg syscall (darwin-arm64), const SCM_TIMESTAMP ideal-int
pkg syscall (darwin-arm64), const SCM_TIMESTAMP_MONOTONIC = 4

- Fixes issue [#68899](https://github.com/kubernetes/kubernetes/issues/68899) where pods might schedule on an unschedulable node. ([#68984](https://github.com/kubernetes/kubernetes/issues/68984), [@k82cn](https://github.com/k82cn))

### Not Very Notable (that is, non-user-facing)

* Fixes issue [[#68899](https://github.com/kubernetes/kubernetes/pull/68899)](https://github.com/kubernetes/kubernetes/issues/68899) where pods might schedule on an unschedulable node. ([#68984](https://github.com/kubernetes/kubernetes/pull/68984), [@k82cn](https://github.com/k82cn))

- Failed pods associated with a job with `parallelism = 1` are recreated by the job controller honoring exponential backoff delay again. However, for jobs with `parallelism > 1`, pods might be created without exponential backoff delay. ([#115022](https://github.com/kubernetes/kubernetes/pull/115022), [@nikhita](https://github.com/nikhita)) [SIG Apps]