lastTs  Time
	idx     int       // next index into []batch
	dataOff int       // next index into batch.data
	ev      baseEvent // last read event
}

func (b *batchCursor) nextEvent(batches []batch, freq frequency) (ok bool, err error) {
	// Batches should generally always have at least one event,
	// but let's be defensive about that and accept empty batches.
	for b.idx < len(batches) && len(batches[b.idx].data) == b.dataOff {
		b.idx++

		r.cpuSamples = r.gen.cpuSamples

		// Reset frontier.
		for m, batches := range r.gen.batches {
			bc := &batchCursor{m: m}
			ok, err := bc.nextEvent(batches, r.gen.freq)
			if err != nil {
				return Event{}, err
			}
			if !ok {
				// Turns out there aren't actually any events in these batches.
				continue
			}
			r.frontier = heapInsert(r.frontier, bc)
		}

	trace   *Trace
	gen     uint64
	batches []*Batch
	strings map[string]uint64
	stacks  map[stack]uint64

	// Options applied when Trace.Generate is called.
	ignoreStringBatchSizeLimit bool
	ignoreStackBatchSizeLimit  bool
}

// Batch starts a new event batch in the trace data.
//
// This is convenience function for generating correct batches.
func (g *Generation) Batch(thread trace.ThreadID, time Time) *Batch {

import org.gradle.api.internal.artifacts.ivyservice.resolveengine.graph.DependencyGraphVisitor;

import java.util.LinkedHashSet;
import java.util.Set;

/**
 * A visitor which batches up all conflicts and reports them all at once at the end of
 * the resolution.
 */
public class FailOnVersionConflictGraphVisitor implements DependencyGraphVisitor {

// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

package runtime

// traceExpWriter is a wrapper around trace writer that produces traceEvExperimentalBatch
// batches. This means that the data written to the writer need not conform to the standard
// trace format.
type traceExpWriter struct {
	traceWriter
	exp traceExperiment
}

				expect(0, 0)
			},
		},
		{
			name: "Should send full updates in batches",
			test: func(updateCh chan *model.PushRequest, expect func(partial, full int32)) {
				updateCh <- &model.PushRequest{Full: true}
				updateCh <- &model.PushRequest{Full: true}
				expect(0, 1)
			},
		},
		{
			name: "Should send full updates in batches, partial updates immediately",

            thread.block()
            stream.received(bytes)
        }

        then:
        1 * dispatch.onOutput({ it instanceof ReadStdInEvent }) >> { instant.requested }
    }

    def "can read multiple batches of input from client"() {
        def dispatch = Mock(OutputEventListener)
        def stream = new StdInStream(dispatch)
        def text1 = "some text"
        def bytes1 = text1.bytes
        def text2 = "more"

// RUN: tf-tfrt-opt -split-input-file -tfrt-reconfig-batch-op="tfrt-min-num-batch-threads=2 tfrt-min-max-enqueued-batches=3" %s | FileCheck %s --dump-input=always

// -----

// The num_batch_threads is lowered bound to 2 from the original attribute of 1

// CHECK-LABEL: func private @batched_function
func.func private @batched_function(%arg0: tensor<1x3xf32>) -> tensor<1x3xf32> {
  %2 = "tf.Identity"(%arg0) : (tensor<1x3xf32>) -> tensor<1x3xf32>

  // The TFLite fused op does not support batching yet.
  // TODO(b/158709815): Add support for batches with padded NMS.
  auto boxes_type =
      mlir::dyn_cast<RankedTensorType>(func_.getFunctionType().getInput(0));
  if (boxes_type == nullptr || !boxes_type.hasRank() ||
      boxes_type.getRank() != 2) {
    return func_.emitWarning() << "TFLite does not support batched input for "
                                  "non_max_suppression_padded";
  }

	lastTime traceTime // when we wrote the last event
	pos      int       // next write offset in arr
	lenPos   int       // position of batch length value
}

// traceBuf is per-M tracing buffer.
//
// TODO(mknyszek): Rename traceBuf to traceBatch, since they map 1:1 with event batches.
type traceBuf struct {
	_ sys.NotInHeap
	traceBufHeader