// multiple streams and so se::Events must be defined and waited for. If
  // 'use_multiple_streams' is true, 'allocate_xla_tensors' must also be true
  // because we track inter-stream dependencies through events inside XlaTensor
  // objects.
  XlaComputationLaunchContext(xla::LocalClient* client,
                              se::DeviceMemoryAllocator* xla_allocator,

    return absl::OkStatus();
  }

  // We conservatively assume that functions will both read and write resource
  // variables.  In the future we may consider doing some form of
  // inter-procedural analysis.
  if (MayCallFunction(n, flib_def)) {
    *out_resource_op_kind = XlaResourceOpKind::kReadWrite;
  } else {
    *out_resource_op_kind = std::nullopt;
  }

  return absl::OkStatus();
}

Artifacts are typically designed to be used or consumed by users or other projects, or deployed to hosting systems.
In such cases, the artifact is a single file.
Directories are common in the case of inter-project dependencies to avoid the cost of producing the publishable artifact.

[[sub:terminology_capability]]
== Capability

A capability identifies a feature offered by one or multiple components.

		return false
	}
	for i := range a {
		if a[i] != b[i] {
			return false
		}
	}
	return true
}

// This is an unusual case.  Service has a number of inter-related fields and
// in order to avoid breaking clients we try really hard to infer what users
// mean when they change them.
//
// Services are effectively a discriminated union, where `type` is the

	// we are almost done and this node is not suitable for preemption. The only
	// condition that we could check is if the "pod" is failing to schedule due to
	// inter-pod affinity to one or more victims, but we have decided not to
	// support this case for performance reasons. Having affinity to lower
	// priority pods is not a recommended configuration anyway.

  C2_DOCKER_IP=$(docker inspect -f "{{ .NetworkSettings.Networks.kind.IPAddress }}" "${C2_NODE}")
  if [ "${POD_TO_POD_AND_SERVICE_CONNECTIVITY}" -eq 1 ]; then
    # Set up routing rules for inter-cluster direct pod to pod & service communication
    C1_POD_CIDR=$(KUBECONFIG="${C1_KUBECONFIG}" kubectl get node -ojsonpath='{.items[0].spec.podCIDR}')

+
A build scan will make it easier to visualize what's happening in the build.
For Gradle builds, you'll be able to see the project structure, the dependencies (regular and inter-project ones), what plugins are being used and the console output of the build.
+
Your build may fail at this point, but that's ok; the scan will still run.

				continue

			print("{0}: {1}".format(obj.type, dtype))

# TODO: print interface's methods and dynamic type's func pointers thereof.
#rsc: "to find the number of entries in the itab's Fn field look at
# itab.inter->numMethods
# i am sure i have the names wrong but look at the interface type
# and its method count"
# so Itype will start with a commontype which has kind = interface

#

    TfeControlType:$control
  );
}

def TfExecutor_EnterOp : TfExecutor_Op<"Enter",
    [ControlOperandsAfterAllData, HasParent<"GraphOp">,
     PredOpTrait<"data operand must be broadcastable to result",
                 TF_OpIsBroadcastableToRes<0, 0>>]>{
  let summary = [{
    The "tf_executor.Enter" operation forwards its input to Tensorflow while
    loop.
  }];

  let description = [{

[it acts as`pivot` for marking the loop termination condition](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/kernels/control_flow_ops.h#L115-L118).

### `tf_executor.Enter` Operation

[`tf_executor.Enter`](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/kernels/control_flow_ops.h##77-L79):
takes a single input and a`name` string attribute that identifies the execution