go tool fix -r cftype <pkg>
	go tool fix -r jni <pkg>

It will replace nil with 0 in the appropriate places.

The EGLDisplay case was introduced in Go 1.12. Use the egl rewrite
to auto-update code from Go 1.11 and earlier:

	go tool fix -r egl <pkg>

The EGLConfig case was introduced in Go 1.15. Use the eglconf rewrite
to auto-update code from Go 1.14 and earlier:

	go tool fix -r eglconf <pkg>

- Fixed a data race in kubelet image manager that can cause static pod workers to silently stop working. ([#88915](https://github.com/kubernetes/kubernetes/pull/88915), [@roycaihw](https://github.com/roycaihw)) [SIG Node]

import org.jspecify.annotations.Nullable;

/**
 * Static utility methods pertaining to {@link InetAddress} instances.
 *
 * <p><b>Important note:</b> Unlike {@code InetAddress.getByName()}, the methods of this class never
 * cause DNS services to be accessed. For this reason, you should prefer these methods as much as
 * possible over their JDK equivalents whenever you are expecting to handle only IP address string

Nope ! C'est ça la morale de l'histoire.

La concurrence est différente du parallélisme. C'est mieux sur des scénarios **spécifiques** qui impliquent beaucoup d'attente. À cause de ça, c'est généralement bien meilleur que le parallélisme pour le développement d'applications web. Mais pas pour tout.

Donc pour équilibrer tout ça, imaginez l'histoire suivante :

  upgrading an existing pre-v1.6.4 cluster to Kubernetes v1.6.5 will cause an unintentional
  [overwrite of manual edits to GCP Health Checks](https://github.com/kubernetes/ingress/issues/842)
  managed by the GLBC Ingress Controller. This can cause the health checks to start failing,
  requiring you to reapply the manual edits.

        * Then upgrade nodes to Kubernetes v1.3.
            * As nodes are upgraded, they will automatically, by default, opt-in to attach/detach controller management, which will cause the controller to start managing attaches/detaches for volumes that get scheduled to those nodes.

### Bug or Regression

- Fix deprecated version for pod_scheduling_duration_seconds that caused the metric to be hidden by default in 1.29. ([#123042](https://github.com/kubernetes/kubernetes/pull/123042), [@alculquicondor](https://github.com/alculquicondor)) [SIG Instrumentation and Scheduling]

    if (closed) return
    closed = true

    if (!started) return // Nothing to shut down.
    val serverSocket = this.serverSocket ?: return // If this is null, start() must have failed.

    // Cause acceptConnections() to break out.
    serverSocket.closeQuietly()

    // Await shutdown.
    for (queue in taskRunner.activeQueues()) {
      if (!queue.idleLatch().await(5, TimeUnit.SECONDS)) {

- `ElasticIndexedJob` is graduated to GA. ([#125751](https://github.com/kubernetes/kubernetes/pull/125751), [@ahg-g](https://github.com/ahg-g)) [SIG Apps and Testing]
- `pause`: Added a `-v` flag to the Windows variant of the pause binary, which
  prints the version of pause and exits. The Linux pause binary already has this flag. ([#125067](https://github.com/kubernetes/kubernetes/pull/125067), [@neolit123](https://github.com/neolit123))

### Failing Test

      // a copy of the entries array. Likewise, localSize is the same as size except in that case.
      // It's possible to keep using this Builder after calling buildKeepingLast(), so we need to
      // ensure that its state is not corrupted by removing duplicates that should cause a later
      // buildOrThrow() to fail, or by changing the size.
      @Nullable Entry<K, V>[] localEntries;