}
        block->cond_var.SignalAll();
        return;
      case FetchState::FETCHING:
        block->cond_var.WaitWithTimeout(&block->mu, absl::Minutes(1));
        if (block->state == FetchState::FINISHED) {
          return TF_SetStatus(status, TF_OK, "");
        }
        // Re-loop in case of errors.
        break;
      case FetchState::FINISHED:
        return TF_SetStatus(status, TF_OK, "");
    }
  }

  switch (data_type) {
    case TF_DataType::TF_BOOL:
      dtype.code = DLDataTypeCode::kDLBool;
      break;
    case TF_DataType::TF_HALF:
    case TF_DataType::TF_FLOAT:
    case TF_DataType::TF_DOUBLE:
      dtype.code = DLDataTypeCode::kDLFloat;
      break;
    case TF_DataType::TF_INT8:
    case TF_DataType::TF_INT16:
    case TF_DataType::TF_INT32:
    case TF_DataType::TF_INT64:

    calls++;
    char c = (filename == "a") ? 'a' : (filename == "b") ? 'b' : 'x';
    if (offset > 0) {
      // The first block is lower case and all subsequent blocks are upper case.
      c = toupper(c);
    }
    memset(buffer, c, n);
    TF_SetStatus(status, TF_OK, "");
    return n;
  };
  // This cache has space for 4 blocks; we'll read from two files.

  switch (default_value.value_case()) {
    case tensorflow::AttrValue::kS: {
      const string& v = default_value.s();
      TFE_OpSetAttrString(op, attr_name, v.data(), v.size());
      break;
    }
    case tensorflow::AttrValue::kI:
      TFE_OpSetAttrInt(op, attr_name, static_cast<int64_t>(default_value.i()));
      break;
    case tensorflow::AttrValue::kF:

#include <functional>
#include <iostream>
#include <list>
#include <map>
#include <memory>
#include <string>
#include <vector>

#include "absl/base/thread_annotations.h"
#include "absl/synchronization/mutex.h"
#include "absl/synchronization/notification.h"
#include "tensorflow/c/env.h"
#include "tensorflow/c/logging.h"
#include "tensorflow/c/tf_status.h"

                                                size_t storage_size,
                                                TF_Status* status);

// Infer OpInfo::TensorProperties for graph nodes inputs/outputs.
//
// Typical use case, is to infer tensor properties from a graph, before doing
// optimization pass. Nodes modified during optimization pass have to be
// invalidated, to prevent further incorrect optimizations based on wrong shape

inline bool IsDtypeTrainable(DataType dtype) {
  switch (dtype) {
    case DT_HALF:
    case DT_BFLOAT16:
    case DT_FLOAT:
    case DT_DOUBLE:
    case DT_COMPLEX64:
    case DT_COMPLEX128:
    case DT_RESOURCE:
    case DT_VARIANT:
      return true;
    case DT_QINT8:
    case DT_QINT16:
    case DT_QINT32:
    case DT_QUINT8:
    case DT_QUINT16:
      return tensorflow::flags::Global()

-       case 9:								      \
-	 __u->__ui = __c * 0x01010101;					      \
-	 __u = __extension__ ((void *) __u + 4);			      \
-       case 5:								      \
-	 __u->__ui = __c * 0x01010101;					      \
-	 __u = __extension__ ((void *) __u + 4);			      \
-       case 1:								      \
-	 __u->__uc = (unsigned char) __c;				      \
-	 break;								      \
-									      \
-       case 16:								      \

#include "tensorflow/core/protobuf/config.pb.h"

using tensorflow::string;

namespace tensorflow {
namespace {

static bool HasSubstr(absl::string_view base, absl::string_view substr) {
  bool ok = absl::StrContains(base, substr);
  EXPECT_TRUE(ok) << base << ", expected substring " << substr;
  return ok;
}

TEST(CAPI, MonitoringCounter0) {
  TF_Status* status = TF_NewStatus();
  auto* counter =

// Creates a new TensorHandle from memory residing in the physical device
// device_name. Takes ownership of the memory, and will call deleter to release
// it after TF no longer needs it or in case of error.
//
// Custom devices must use TFE_NewCustomDeviceTensorHandle instead.
TF_CAPI_EXPORT extern TFE_TensorHandle* TFE_NewTensorHandleFromDeviceMemory(