To reproduce TeamCity build use `-Pteamcity=true` flag. Local builds don't run proguard and have jar compression disabled by default.

**OPTIONAL:** Some artifacts, mainly Maven plugin ones, are built separately with Maven.
Refer to [libraries/ReadMe.md](libraries/ReadMe.md) for details.

To build Kotlin/Native, see
[kotlin-native/README.md](kotlin-native/README.md#building-from-source).

   * under the assumption that for long-running instances, observed
   * contention levels will recur, so the cells will eventually be
   * needed again; and for short-lived ones, it does not matter.
   */

  /**
   * Padded variant of AtomicLong supporting only raw accesses plus CAS. The value field is placed
   * between pads, hoping that the JVM doesn't reorder them.
   *

    for (Future<?> result : results) {
      if (!result.isCancelled()) {
        result.get(10, SECONDS);
      }
      // TODO(cpovirk): Verify that the cancelled futures are exactly ones that we expect.
    }

    assertThat(logHandler.getStoredLogRecords()).isEmpty();
  }

  public void testAvoidsStackOverflow_manySubmitted() throws Exception {

    for (Future<?> result : results) {
      if (!result.isCancelled()) {
        result.get(10, SECONDS);
      }
      // TODO(cpovirk): Verify that the cancelled futures are exactly ones that we expect.
    }

    assertThat(logHandler.getStoredLogRecords()).isEmpty();
  }

  public void testAvoidsStackOverflow_manySubmitted() throws Exception {

    /**
     * Shortcut for {@code getService(DependencyCollector.class).collect(...)}
     *
     * @param artifact artifact for which to get the dependencies, including transitive ones
     * @return root node of the dependency graph for the given artifact
     *
     * @see org.apache.maven.api.services.DependencyCollector#collect(Session, Artifact)

  // zero and not specified. Defaults to 1.
  // +optional
  optional int32 replicas = 1;

  // Label selector for pods. Existing ReplicaSets whose pods are
  // selected by this will be the ones affected by this deployment.
  // It must match the pod template's labels.
  optional k8s.io.apimachinery.pkg.apis.meta.v1.LabelSelector selector = 2;

  // Template describes the pods that will be created.

        // spec says that we may repeat the request without modifications. Modern browsers also
        // repeat the request (even non-idempotent ones.)
        if (!client.retryOnConnectionFailure) {
          // The application layer has directed us not to retry the request.
          return null
        }

        val requestBody = userResponse.request.body

`ModelServer` collates all computation graphs exposed to it (from multiple
`SavedModel`) and executes them in parallel on available executors. Running
TensorFlow in a multitenant design mixes the risks described above with the
inherent ones from multitenant configurations. The primary areas of concern are
tenant isolation, resource allocation, model sharing and hardware attacks.

### Tenant isolation

   * under the assumption that for long-running instances, observed
   * contention levels will recur, so the cells will eventually be
   * needed again; and for short-lived ones, it does not matter.
   */

  /**
   * Padded variant of AtomicLong supporting only raw accesses plus CAS. The value field is placed
   * between pads, hoping that the JVM doesn't reorder them.
   *

        excludedPackages.add("org.gradle.plugins.binaries.model");

        // Exclude classes that were moved in a different package but the deprecated ones are not removed yet
        excludedPackages.add("org.gradle.platform.base.test");
        return excludedPackages;
    }