and used to compile Fortran files in that package. All the LDFLAGS directives
in any package in the program are concatenated and used at link time. All the
pkg-config directives are concatenated and sent to pkg-config simultaneously
to add to each appropriate set of command-line flags.

When the cgo directives are parsed, any occurrence of the string ${SRCDIR}
will be replaced by the absolute path to the directory containing the source

* Fix for kube-proxy healthcheck handling an update that simultaneously removes one port and adds another. ([#44365](https://github.com/kubernetes/kubernetes/pull/44365), [@MrHohn](https://github.com/MrHohn))

* Fixes a performance issue ([#51899](https://github.com/kubernetes/kubernetes/pull/51899)) identified in large-scale clusters when deleting thousands of pods simultaneously across hundreds of nodes, by actively removing containers of deleted pods, rather than waiting for periodic garbage collection and batching resulting pod API deletion requests. ([#53233](https://github.com/kubernetes/kubernetes/pull/53233), [...

##### **Other**

- kube-apiserver: fixes error creating system priority classes when starting multiple apiservers simultaneously ([#67372](https://github.com/kubernetes/kubernetes/pull/67372), [@tanshanshan](https://github.com/tanshanshan))

    cache.put(key, "1");
    // simultaneous update for same key, some null, some non-null
    doParallelCacheOp(
        count,
        n -> {
          cache.asMap().compute(key, (k, v) -> n % 2 == 0 ? v + delimiter + n : null);
        });
    assertThat(cache.size()).isAtMost(1);
  }

  public void testCompute() {
    cache.put(key, "1");
    // simultaneous deletion
    doParallelCacheOp(

- Added to `MutableFeatureGate` the ability to override the default setting of feature gates, to allow default-enabling a feature on a component-by-component basis instead of for all affected components simultaneously. ([#122647](https://github.com/kubernetes/kubernetes/pull/122647), [@benluddy](https://github.com/benluddy))

## Dispatch

For synchronous calls, you bring your own thread and are responsible for managing how many simultaneous requests you make. Too many simultaneous connections wastes resources; too few harms latency.

Example: Limit a MinIO cluster to accept at max 1600 simultaneous S3 API requests across all nodes of the cluster.

```sh
export MINIO_API_REQUESTS_MAX=1600
export MINIO_ROOT_USER=your-access-key
export MINIO_ROOT_PASSWORD=your-secret-key

 * typically false on Windows.
 *
 * If this returns false we won't permit simultaneous reads and writes. When writes commit we need
 * to delete the previous snapshots, and that won't succeed if the file is open. (We do permit
 * multiple simultaneous reads.)
 *
 * @param file a file in the directory to check. This file shouldn't already exist!
 */