# These flags share the user's gcloud credentials with the container, so that bazel
# inside the container can authenticate. Note: TF's CI does not have any credential
# stored here.

TFCI_BAZEL_COMMON_ARGS="--repo_env=TF_PYTHON_VERSION=$TFCI_PYTHON_VERSION --config release_gpu_linux"
TFCI_BAZEL_TARGET_SELECTING_CONFIG_PREFIX=linux_cuda
TFCI_BUILD_PIP_PACKAGE_ARGS="--repo_env=WHEEL_NAME=tensorflow"
TFCI_DOCKER_ARGS="--gpus all"
TFCI_LIB_SUFFIX="-gpu-linux-x86_64"

# ==============================================================================

setup_file() {
    bazel version  # Start the bazel server

    # Fixes "fatal: detected dubious ownership in repository" for Docker.
    git config --system --add safe.directory '*'
    git config --system protocol.file.allow always

    # Note that you could generate a list of all the affected targets with e.g.:

# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
# Disable Docker

#   container and start fresh, run "docker rm -f tf". Removing the container
#   destroys some temporary bazel data and causes longer builds.
#
#   You will need the NVIDIA Container Toolkit for GPU testing:
#   https://github.com/NVIDIA/nvidia-container-toolkit
#
#   Note: if you interrupt a bazel command on docker (ctrl-c), you
#   will need to run `docker exec tf pkill bazel` to quit bazel.
#

    execution time and the sharding
    [could create an overhead on the test execution](https://github.com/bazelbuild/bazel/issues/2113#issuecomment-264054799).

2.  Using [Docker](https://www.docker.com) and TensorFlow's CI scripts.

    ```bash
    # Install Docker first, then this will build and run cpu tests
    tensorflow/tools/ci_build/ci_build.sh CPU bazel test //tensorflow/...
    ```

    See

# despite lacking Nvidia CUDA support.
TFCI_BUILD_PIP_PACKAGE_ARGS="--repo_env=WHEEL_NAME=tensorflow"
TFCI_DOCKER_ENABLE=1
TFCI_DOCKER_IMAGE=gcr.io/tensorflow-sigs/build-arm64:tf-2-16-multi-python
TFCI_DOCKER_PULL_ENABLE=1
TFCI_DOCKER_REBUILD_ARGS="--target=tf ci/official/containers/linux_arm64"
TFCI_INDEX_HTML_ENABLE=1
TFCI_LIB_SUFFIX="-cpu-linux-arm64"
TFCI_OUTPUT_DIR=build_output

# CuDNN: https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#ubuntu-network-installation
libcudnn8=8.9.6.50-1+cuda12.2
libcudnn8-dev=8.9.6.50-1+cuda12.2

# This can be removed once NVIDIA publishes a cuda-12.3.2 Docker image.
# For now it ensures that we install at least version 12.3.107 of PTXAS,
# since 12.3.103 has a bug.

# the shared library, only export the core TF API functions to avoid
# causing library conflicts (e.g., those reported in github issue 1924).
# On Linux, tell the linker (-Wl,<option>) to use a version script that
# excludes all but a subset of function names.
# On MacOS, the linker does not support version_script, but has an
# an "-exported_symbols_list" command.  -z defs disallows undefined
# symbols in object files.

build --action_env PYTHON_BIN_PATH="/usr/local/bin/python3"
build --python_path="/usr/local/bin/python3"

# Build TensorFlow v2
build --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1

# Use lld as the linker
build --linkopt="-fuse-ld=lld"
build --linkopt="-lm"
build --linkopt="-Wl,--undefined-version"

# Prevent double-compilation of some TF code, ref. b/183279666 (internal)