const std::vector<int64_t>& unneeded_gradients,
      gtl::ArraySlice<AbstractTensorHandle*> output_gradients,
      absl::Span<AbstractTensorHandle*> result) const override;

  // Builds a tensor filled with ones with the same shape and dtype as `t`.
  Status BuildOnesLike(const TapeTensor& t,
                       AbstractTensorHandle** result) const override;

  // Looks up the ID of a Gradient.

// while `op` is alive.
TF_CAPI_EXPORT extern const TFE_OpAttrs* TFE_OpGetAttrs(const TFE_Op* op);
// Add attributes in `attrs` to `op`.
//
// Does not overwrite or update existing attributes, but adds new ones.
TF_CAPI_EXPORT extern void TFE_OpAddAttrs(TFE_Op* op, const TFE_OpAttrs* attrs);

// Serialize `attrs` as a tensorflow::NameAttrList protocol buffer (into `buf`),
// containing the op name and a map of its attributes.

      ["grep", "-rlE", '(failures|errors)="[1-9]', sys.argv[1]]
  )
except subprocess.CalledProcessError as e:
  print("No failures found to log!")
  exit(0)

# For test cases, only show the ones that failed that have text (a log)
seen = collections.Counter()
runfiles_matcher = re.compile(r"(/.*\.runfiles/)")


for f in files.strip().splitlines():
  # Just ignore any failures, they're probably not important

//
// `text` corresponds to a text representation of an ApiDefs protocol message.
// (https://www.tensorflow.org/code/tensorflow/core/framework/api_def.proto).
//
// The provided ApiDefs will be merged with existing ones in the map, with
// precedence given to the newly added version in case of conflicts with
// previous calls to TF_ApiDefMapPut.
TF_CAPI_EXPORT extern void TF_ApiDefMapPut(TF_ApiDefMap* api_def_map,

          if (op_it->second.output_tensor_info[j].GetID() == id) {
            found = true;
            Gradient* ones_like = nullptr;
            TF_RETURN_IF_ERROR(vspace.BuildOnesLike(
                op_it->second.output_tensor_info[j], &ones_like));
            (*result)[id].push_back(ones_like);
            break;
          }
        }
        if (!found) {
          return errors::Internal(

* `tf.ones`, `tf.zeros`, `tf.fill`, `tf.ones_like`, `tf.zeros_like` now take an additional Layout argument that controls the output layout of their results.

    AbstractTensorHandlePtr Conj_X(temp_output);

    // Creates Ones
    name = "OnesLike_Log1p_Grad_X";
    TF_RETURN_IF_ERROR(OnesLike(ctx, Conj_X.get(), &temp_output, name.c_str()));

    AbstractTensorHandlePtr Ones_X(temp_output);

    name = "Add_Log1p_Grad_X";
    // Calculate 1 + Conj(X)
    TF_RETURN_IF_ERROR(
        AddV2(ctx, Ones_X.get(), Conj_X.get(), &temp_output, name.c_str()));

`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