### Model sharing

If the multitenant design allows sharing models, make sure that tenants and
users are aware of the security risks detailed here and that they are going to
be practically running code provided by other users. Currently there are no good
ways to detect malicious models/graphs/checkpoints, so the recommended way to
mitigate the risk in this scenario is to sandbox the model execution.

### Hardware attacks

                                               TF_Status* status,
                                               void* device_info);

  // Method to execute an operation.
  //
  // Arguments provide enough information to reconstruct the original `TFE_Op`,
  // or construct a transformed version, by inspecting the passed `op`.
  //
  // TFE_OpGetDevice(op) records the original placement of the operation. It may

    unique_tensor_ptr;

TF_Tensor* BoolTensor(int32_t v);

// Create a tensor with values of type TF_INT8 provided by `values`.
TF_Tensor* Int8Tensor(const int64_t* dims, int num_dims, const char* values);

// Create a tensor with values of type TF_INT32 provided by `values`.
TF_Tensor* Int32Tensor(const int64_t* dims, int num_dims,
                       const int32_t* values);

# limitations under the License.
#
# ==============================================================================
#
# Repacks libtensorflow tarballs into $DIR with provided $TARBALL_SUFFIX,
# and also repacks libtensorflow-src.jar into a standardized format.

# Helper function to copy a srcjar after moving any source files
# directly under the root to the "maven-style" src/main/java layout
#

// Implementation detail for the unified execution APIs for Eager and tracing
// backends (graph/MLIR).
//
// This defines a set of abstract classes that are intended to provide the
// functionality of the opaque C types exposed in the public APIs defined in the
// `c_api_unified_experimental.h` header.
// =============================================================================

// result of `gradient_checker` and `[*]_grad`. The tests here test the
// functionality of `gradient_checker` by comparing the result with expected
// manual user-provided gradients.
class GradientCheckerTest
    : public ::testing::TestWithParam<std::tuple<const char*, bool, bool>> {
 protected:
  void SetUp() override {
    TF_StatusPtr status(TF_NewStatus());

/// filesystem (e.g. GCS).
class RamFileBlockCache {
 public:
  /// The callback executed when a block is not found in the cache, and needs to
  /// be fetched from the backing filesystem. This callback is provided when the
  /// cache is constructed. It returns total bytes read ( -1 in case of errors
  /// ). The `status` should be `TF_OK` as long as the read from the remote

    which is a NumPy-compatible API for writing TF programs. This module
    provides class `ndarray`, which mimics the `ndarray` class in NumPy, and
    wraps an immutable `tf.Tensor` under the hood. A subset of NumPy functions
    (e.g. `numpy.add`) are provided. Their inter-operation with TF facilities is
    seamless in most cases. See

  size_t max_bytes = kDefaultMaxCacheSize;
  uint64_t max_staleness = kDefaultMaxStaleness;

  // Apply the overrides for the block size (MB), max bytes (MB), and max
  // staleness (seconds) if provided.
  const char* block_size_env = std::getenv(kBlockSize);
  if (block_size_env && absl::SimpleAtoi(block_size_env, &value)) {
    block_size = value * 1024 * 1024;
  }

    visibility = ["//visibility:public"],
)

# DO NOT ADD ANY NEW EXCEPTIONS TO THIS LIST!
# Instead, please use public APIs or public build rules TF provides.
# If you need functionality that is not exposed, we will work with you to expand our public APIs.
# TODO(b/173549186): Move Google-internal TF code out of learning/brain
package_group(
    name = "internal",