}

  const tensorflow::OpDef_ArgDef& input_arg = op_def->input_arg()[input_index];

  if (input_arg.number_attr().empty()) {
    status->status = tensorflow::errors::NotFound(
        op_name, " does not have number_attr() defined.");
    return nullptr;
  }

  // The returned string is owned by OpRegistry, so liveness is not a concern.
  return input_arg.number_attr().c_str();
}

    std::unique_ptr<TF_Status, decltype(&TF_DeleteStatus)> await_status(
        TF_NewStatus(), TF_DeleteStatus);
    // Wait until all pending nodes have completed since they may have a
    // reference to default_cancellation_manager_. We ignore the status return
    // since we already have a bad status to propagate.
    TFE_ContextAsyncWait(context, await_status.get());
    // Reset the cancellation manager on a bad status. Otherwise we'll cancel

        // artificial restriction and require that when num_opers=-1, such
        // nodes must have a single output.
        if (node->num_outputs() != 1) {
          return InvalidArgument(
              "When `num_opers` is set to -1, nodes referenced in `inputs` "
              "must have a single output. Node ",
              node->name(), " has ", node->num_outputs(),

    variable->Assign(context, initial_value.get(), status.get());
    ASSERT_TRUE(TF_GetCode(status.get()) == TF_OK) << TF_Message(status.get());
  }

  // Read from the variable and verify that we have a parallel tensor.
  {
    TensorHandlePtr read = variable->Read(context, status.get());
    ASSERT_TRUE(TF_GetCode(status.get()) == TF_OK) << TF_Message(status.get());
    std::array<TensorHandlePtr, 2> components;

  TFE_TensorHandle* h1 = TestVariable(ctx, 2.0, task2_name);
  TFE_TensorHandle* h2 = TestVariable(ctx, 3.0, task0_name);

  // Add a sync point to make sure that variables have been initialized
  // before the function execution starts.
  TFE_ContextAsyncWait(ctx, status);
  EXPECT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);

  // Pack 3 variable handles into one TFE_TensorHandle.

                                            absl::string_view result,
                                            size_t read) {
  // Result isn't a null-terminated string so we have to wrap it inside a
  // `string_view`
  if (length == read && content_view.substr(offset, length) ==
                            absl::string_view(result).substr(0, read))
    return ::testing::AssertionSuccess();
  else

TEST(RamFileBlockCacheTest, ParallelReads) {
  // This fetcher won't respond until either `callers` threads are calling it
  // concurrently (at which point it will respond with success to all callers),
  // or 10 seconds have elapsed (at which point it will respond with an error).
  const int callers = 4;
  BlockingCounter counter(callers);
  auto fetcher = [&counter](const string& filename, size_t offset, size_t n,

    block->lru_iterator = lru_list_.begin();
  }

  // Check for inconsistent state. If there is a block later in the same file
  // in the cache, and our current block is not block size, this likely means
  // we have inconsistent state within the cache. Note: it's possible some
  // incomplete reads may still go undetected.
  if (block->data.size() < block_size_) {
    Key fmax = std::make_pair(key.first, std::numeric_limits<size_t>::max());

#include "tensorflow/core/platform/file_system_helper.h"
#include "tsl/platform/errors.h"

// TODO(b/139060984): After all filesystems are converted, all calls to
// methods from `FileSystem` will have to be replaced to calls to private
// methods here, as part of making this class a singleton and the only way to
// register/use filesystems.

namespace tensorflow {

using UniquePtrTo_TF_Status =

  server_def_0.mutable_default_session_config()->set_isolate_session_state(
      false);
  tensorflow::ServerDef server_def_1 =
      ReplaceTaskInServerDef(server_def_0, /*task_index=*/0);

  // These server defs have task index set to 0.
  string serialized_server_def_0 = server_def_0.SerializeAsString();
  string serialized_server_def_1 = server_def_1.SerializeAsString();

  // Create two worker tasks.