Unable to build LiteRT for Android

[yg-dickson]

bazel build -c opt --cxxopt=--std=c++17 --config=android_arm64
--fat_apk_cpu=x86,x86_64,arm64-v8a,armeabi-v7a
--define=android_dexmerger_tool=d8_dexmerger
--define=android_incremental_dexing_tool=d8_dexbuilder
//tensorflow/lite/java:tensorflow-lite
Starting local Bazel server and connecting to it...
INFO: Reading 'startup' options from /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc: --windows_enable_symlinks
INFO: Options provided by the client:
Inherited 'common' options: --isatty=1 --terminal_columns=123
INFO: Reading rc options for 'build' from /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc:
Inherited 'common' options: --experimental_repo_remote_exec
INFO: Reading rc options for 'build' from /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc:
'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --features=-force_no_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility --config=v2 --config=litert_prefixes
INFO: Found applicable config definition build:short_logs in file /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc: --output_filter=DONT_MATCH_ANYTHING
INFO: Found applicable config definition build:v2 in file /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1
INFO: Found applicable config definition build:litert_prefixes in file /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc: --define=litert_prefixes=external/org_tensorflow/
INFO: Found applicable config definition build:android_arm64 in file /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc: --config=android --cpu=arm64-v8a --fat_apk_cpu=arm64-v8a
INFO: Found applicable config definition build:android in file /Users/wang/IdeProjects/tflite-ai/client/LiteRT/.bazelrc: --crosstool_top=//external:android/crosstool --host_crosstool_top=@bazel_tools//tools/cpp:toolchain --dynamic_mode=off --define=xnn_enable_avxvnniint8=false --noenable_platform_specific_config --copt=-w --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --define=with_xla_support=false
DEBUG: /private/var/tmp/wang/c286a78069d91d2d2c206403e296a20d/external/local_tsl/third_party/py/python_repo.bzl:98:14:
HERMETIC_PYTHON_VERSION variable was not set correctly, using default version.
Python 3.11 will be used.
To select Python version, either set HERMETIC_PYTHON_VERSION env variable in
your shell:
export HERMETIC_PYTHON_VERSION=3.12
OR pass it as an argument to bazel command directly or inside your .bazelrc
file:
--repo_env=HERMETIC_PYTHON_VERSION=3.12
DEBUG: /private/var/tmp/wang/c286a78069d91d2d2c206403e296a20d/external/local_tsl/third_party/py/python_repo.bzl:109:10: Using hermetic Python 3.11
ERROR: Skipping '//tensorflow/lite/java:tensorflow-lite': no such package 'tensorflow/lite/java': BUILD file not found in any of the following directories. Add a BUILD file to a directory to mark it as a package.

• /Users/wang/IdeProjects/tflite-ai/client/LiteRT/tensorflow/lite/java
  WARNING: Target pattern parsing failed.
  ERROR: no such package 'tensorflow/lite/java': BUILD file not found in any of the following directories. Add a BUILD file to a directory to mark it as a package.
• /Users/wang/IdeProjects/tflite-ai/client/LiteRT/tensorflow/lite/java
  INFO: Elapsed time: 2.220s
  INFO: 0 processes.
  FAILED: Build did NOT complete successfully (0 packages loaded)

[yg-dickson]

Is there any tutorial for building LiteRT for Android?

[gabrielpopa]

try bazel build -c opt --cxxopt=--std=c++17 --config=android_arm64 //tflite/c:tensorflowlite_c

don't forget to git submodule update --init --recursive before.

you may need to clean it with bazel clean --expunge

[gabrielpopa]

also you can use bazel query //... to query all the targets that you may need to build

[yg-dickson]

My computer is Mac Studio Apple M1 Max
My operation steps are as follows
Step 1 git clone https://github.com/google-ai-edge/LiteRT.git
Step 2 git submodule update --init --recursive
Step 3 bazel build -c opt --cxxopt=--std=c++17 --config=android_arm64 //tflite/c:tensorflowlite_c

Still error
ERROR: /Users/wangchengyun/IdeProjects/sit/LiteRT/tflite/kernels/BUILD:869:11: no such package '@local_xla//xla/tsl/lib/random': BUILD file not found in directory 'xla/tsl/lib/random' of external repository @local_xla. Add a BUILD file to a directory to mark it as a package. and referenced by '//tflite/kernels:builtin_op_kernels' ERROR: Analysis of target '//tflite/c:tensorflowlite_c' failed; build aborted: INFO: Elapsed time: 36.480s INFO: 0 processes. FAILED: Build did NOT complete successfully (96 packages loaded, 1002 targets configured) currently loading: @org_tensorflow//tensorflow/compiler/mlir/lite Fetching repository @local_config_cc; Running xcode-locator Fetching repository @local_config_xcode; Building xcode-locator Fetching repository @llvm-project; starting Fetching repository @Ruy; starting Fetching /private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/ruy; Extracting 3286a34cc8de6149ac6844107dfdffac91531e72.zip Fetching /private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/cpuinfo; Extracting 3c8b1533ac03dd6531ab6e7b9245d488f13a82a5.zip ➜ LiteRT git:(main)

[gabrielpopa]

did you also hit ./configure before?

try to add:

local_repository(
    name = "local_xla",
    path = "third_party/tensorflow/third_party/xla",
)

into WORKSPACE file.

[yg-dickson]

did you also hit ./configure before?

try to add:

local_repository(
    name = "local_xla",
    path = "third_party/tensorflow/third_party/xla",
)

into WORKSPACE file.

Do I need to execute ./configure first to complete the configuration? I executed ./configure and still got an error.

[gaikwadrahul8]

Hi, @yg-dickson
I apologize for the delayed response and I believe you're following Build LiteRT for Android official documentation specifically Set up build environment without Docker section

• Ensure you have the Bazel version 6.5.0 compatible with TensorFlow
• Check that Android SDK and NDK paths are correct
• Please give it try with python version 3.11 or 3.10 if you've multiple python versions installed in your system

Please follow the below instructions and let us know is it working as expected or not ?

# Clone the repository
git clone https://github.com/google-ai-edge/LiteRT.git
cd LiteRT
./configure

# Initialize submodules
git submodule update --init --recursive

# Clean previous builds
bazel clean

# Try building with specific configurations
bazel build -c opt \
    --cxxopt=--std=c++17 \
    --config=android_arm64 \
    //tflite/c:tensorflowlite_c \
   --verbose_failures

Thank you for your cooperation and patience.

[yg-dickson]

I did follow this step.

(fastapi) ➜ sit git clone https://github.com/google-ai-edge/LiteRT.git
Cloning into 'LiteRT'...
remote: Enumerating objects: 8549, done.
remote: Counting objects: 100% (650/650), done.
remote: Compressing objects: 100% (202/202), done.
remote: Total 8549 (delta 489), reused 455 (delta 448), pack-reused 7899 (from 2)
Receiving objects: 100% (8549/8549), 49.10 MiB | 23.72 MiB/s, done.
Resolving deltas: 100% (4132/4132), done.
(fastapi) ➜ sit cd LiteRT
(fastapi) ➜ LiteRT git:(main) ./configure
You have bazel 6.5.0 installed.
Please specify the location of python. [Default is /opt/miniconda3/envs/fastapi/bin/python3]:

Found possible Python library paths:
/opt/miniconda3/envs/fastapi/lib/python3.11/site-packages
Please input the desired Python library path to use. Default is [/opt/miniconda3/envs/fastapi/lib/python3.11/site-packages]

Do you wish to build TensorFlow with ROCm support? [y/N]: n
No ROCm support will be enabled for TensorFlow.

Do you wish to build TensorFlow with CUDA support? [y/N]: n
No CUDA support will be enabled for TensorFlow.

Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -Wno-sign-compare]:

Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: y
Searching for NDK and SDK installations.

Please specify the home path of the Android NDK to use. [Default is /Users/wangchengyun/library/Android/Sdk/ndk-bundle]: /Users/wangchengyun/Library/Android/sdk/ndk/25.1.8937393

Please specify the (min) Android NDK API level to use. [Available levels: [16, 17, 18, 19, 21, 22, 23, 24, 26, 27, 28, 29, 30, 31, 32, 33]] [Default is 21]: 21

Please specify the home path of the Android SDK to use. [Default is /Users/wangchengyun/library/Android/Sdk]: /Users/wangchengyun/Library/Android/sdk

Please specify the Android SDK API level to use. [Available levels: ['24', '28', '29', '30', '31', '32', '33', '34', '35']] [Default is 35]: 30

Please specify an Android build tools version to use. [Available versions: ['30.0.2', '30.0.3', '33.0.1', '34.0.0', '35.0.0']] [Default is 35.0.0]: 30.0.3

Do you wish to build TensorFlow with iOS support? [y/N]: n
No iOS support will be enabled for TensorFlow.

Preconfigured Bazel build configs. You can use any of the below by adding "--config=<>" to your build command. See .bazelrc for more details.
--config=mkl # Build with MKL support.
--config=mkl_aarch64 # Build with oneDNN and Compute Library for the Arm Architecture (ACL).
--config=monolithic # Config for mostly static monolithic build.
--config=numa # Build with NUMA support.
--config=dynamic_kernels # (Experimental) Build kernels into separate shared objects.
--config=v1 # Build with TensorFlow 1 API instead of TF 2 API.
Preconfigured Bazel build configs to DISABLE default on features:
--config=nogcp # Disable GCP support.
--config=nonccl # Disable NVIDIA NCCL support.
Configuration finished
(fastapi) ➜ LiteRT git:(main) ✗ open ./
(fastapi) ➜ LiteRT git:(main) ✗ git submodule update --init --recursive
Submodule 'third_party/tensorflow' (https://github.com/tensorflow/tensorflow) registered for path 'third_party/tensorflow'
Cloning into '/Users/wangchengyun/IdeProjects/sit/LiteRT/third_party/tensorflow'...
Submodule path 'third_party/tensorflow': checked out '2adc36c677a558b93a454705059baac2b0cdf5a3'
(fastapi) ➜ LiteRT git:(main) ✗ bazel clean
WARNING: Ignoring JAVA_HOME, because it must point to a JDK, not a JRE.
INFO: Reading 'startup' options from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --windows_enable_symlinks
INFO: Options provided by the client:
Inherited 'common' options: --isatty=1 --terminal_columns=139
INFO: Reading rc options for 'clean' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc:
Inherited 'common' options: --experimental_repo_remote_exec --experimental_repo_remote_exec
INFO: Reading rc options for 'clean' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc:
Inherited 'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --features=-force_no_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility
INFO: Reading rc options for 'clean' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.tf_configure.bazelrc:
Inherited 'build' options: --action_env PYTHON_BIN_PATH=/opt/miniconda3/envs/fastapi/bin/python3 --action_env PYTHON_LIB_PATH=/opt/miniconda3/envs/fastapi/lib/python3.11/site-packages --python_path=/opt/miniconda3/envs/fastapi/bin/python3 --action_env ANDROID_NDK_HOME=/Users/wangchengyun/Library/Android/sdk/ndk/25.1.8937393 --action_env ANDROID_NDK_VERSION=25 --action_env ANDROID_NDK_API_LEVEL=21 --action_env ANDROID_BUILD_TOOLS_VERSION=30.0.3 --action_env ANDROID_SDK_API_LEVEL=30 --action_env ANDROID_SDK_HOME=/Users/wangchengyun/Library/Android/sdk
INFO: Reading rc options for 'clean' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc:
Inherited 'build' options: --config=v2 --config=litert_prefixes
INFO: Found applicable config definition build:short_logs in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --output_filter=DONT_MATCH_ANYTHING
INFO: Found applicable config definition build:v2 in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1
INFO: Found applicable config definition build:litert_prefixes in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --define=litert_prefixes=external/org_tensorflow/
INFO: Found applicable config definition build:macos in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --apple_platform_type=macos --copt=-DGRPC_BAZEL_BUILD --features=archive_param_file --copt=-w --define=PREFIX=/usr --define=LIBDIR=$(PREFIX)/lib --define=INCLUDEDIR=$(PREFIX)/include --define=PROTOBUF_INCLUDE_PATH=$(PREFIX)/include --cxxopt=-std=c++17 --host_cxxopt=-std=c++17
INFO: Starting clean (this may take a while). Consider using --async if the clean takes more than several minutes.
(fastapi) ➜ LiteRT git:(main) ✗ bazel build -c opt
--cxxopt=--std=c++17
--config=android_arm64
//tflite/c:tensorflowlite_c
--verbose_failures
WARNING: Ignoring JAVA_HOME, because it must point to a JDK, not a JRE.
INFO: Reading 'startup' options from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --windows_enable_symlinks
INFO: Options provided by the client:
Inherited 'common' options: --isatty=1 --terminal_columns=139
INFO: Reading rc options for 'build' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc:
Inherited 'common' options: --experimental_repo_remote_exec --experimental_repo_remote_exec
INFO: Reading rc options for 'build' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc:
'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --features=-force_no_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility
INFO: Reading rc options for 'build' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.tf_configure.bazelrc:
'build' options: --action_env PYTHON_BIN_PATH=/opt/miniconda3/envs/fastapi/bin/python3 --action_env PYTHON_LIB_PATH=/opt/miniconda3/envs/fastapi/lib/python3.11/site-packages --python_path=/opt/miniconda3/envs/fastapi/bin/python3 --action_env ANDROID_NDK_HOME=/Users/wangchengyun/Library/Android/sdk/ndk/25.1.8937393 --action_env ANDROID_NDK_VERSION=25 --action_env ANDROID_NDK_API_LEVEL=21 --action_env ANDROID_BUILD_TOOLS_VERSION=30.0.3 --action_env ANDROID_SDK_API_LEVEL=30 --action_env ANDROID_SDK_HOME=/Users/wangchengyun/Library/Android/sdk
INFO: Reading rc options for 'build' from /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc:
'build' options: --config=v2 --config=litert_prefixes
INFO: Found applicable config definition build:short_logs in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --output_filter=DONT_MATCH_ANYTHING
INFO: Found applicable config definition build:v2 in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1
INFO: Found applicable config definition build:litert_prefixes in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --define=litert_prefixes=external/org_tensorflow/
INFO: Found applicable config definition build:android_arm64 in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --config=android --cpu=arm64-v8a --fat_apk_cpu=arm64-v8a
INFO: Found applicable config definition build:android in file /Users/wangchengyun/IdeProjects/sit/LiteRT/.bazelrc: --crosstool_top=//external:android/crosstool --host_crosstool_top=@bazel_tools//tools/cpp:toolchain --dynamic_mode=off --define=xnn_enable_avxvnniint8=false --noenable_platform_specific_config --copt=-w --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --define=with_xla_support=false
DEBUG: /private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/local_tsl/third_party/py/python_repo.bzl:98:14:
HERMETIC_PYTHON_VERSION variable was not set correctly, using default version.
Python 3.11 will be used.
To select Python version, either set HERMETIC_PYTHON_VERSION env variable in
your shell:
export HERMETIC_PYTHON_VERSION=3.12
OR pass it as an argument to bazel command directly or inside your .bazelrc
file:
--repo_env=HERMETIC_PYTHON_VERSION=3.12
DEBUG: /private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/local_tsl/third_party/py/python_repo.bzl:109:10: Using hermetic Python 3.11
INFO: Repository pthreadpool instantiated at:
/Users/wangchengyun/IdeProjects/sit/LiteRT/WORKSPACE:68:14: in
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/tensorflow/workspace2.bzl:946:21: in workspace
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/tensorflow/workspace2.bzl:174:20: in _tf_repositories
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/third_party/repo.bzl:136:21: in tf_http_archive
Repository rule _tf_http_archive defined at:
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/third_party/repo.bzl:89:35: in
INFO: Repository cpuinfo instantiated at:
/Users/wangchengyun/IdeProjects/sit/LiteRT/WORKSPACE:68:14: in
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/tensorflow/workspace2.bzl:946:21: in workspace
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/tensorflow/workspace2.bzl:181:20: in _tf_repositories
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/third_party/repo.bzl:136:21: in tf_http_archive
Repository rule _tf_http_archive defined at:
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/third_party/repo.bzl:89:35: in
INFO: Repository XNNPACK instantiated at:
/Users/wangchengyun/IdeProjects/sit/LiteRT/WORKSPACE:68:14: in
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/tensorflow/workspace2.bzl:946:21: in workspace
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/tensorflow/workspace2.bzl:151:20: in _tf_repositories
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/third_party/repo.bzl:136:21: in tf_http_archive
Repository rule _tf_http_archive defined at:
/private/var/tmp/_bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/org_tensorflow/third_party/repo.bzl:89:35: in
INFO: repository @pthreadpool' used the following cache hits instead of downloading the corresponding file.

• Hash 'a4cf06de57bfdf8d7b537c61f1c3071bce74e57524fe053e0bbd2332feca7f95' for https://storage.googleapis.com/mirror.tensorflow.org/github.com/Maratyszcza/pthreadpool/archive/4fe0e1e183925bf8cfa6aae24237e724a96479b8.zip
  If the definition of 'repository @pthreadpool' was updated, verify that the hashes were also updated.
  ERROR: /Users/wangchengyun/IdeProjects/sit/LiteRT/tflite/kernels/BUILD:800:11: no such package '@local_xla//xla/tsl/lib/random': BUILD file not found in directory 'xla/tsl/lib/random' of external repository @local_xla. Add a BUILD file to a directory to mark it as a package. and referenced by '//tflite/kernels:builtin_op_kernels'
  ERROR: Analysis of target '//tflite/c:tensorflowlite_c' failed; build aborted:
  INFO: Elapsed time: 1.774s
  INFO: 0 processes.
  FAILED: Build did NOT complete successfully (104 packages loaded, 2138 targets configured)
  currently loading: @org_tensorflow//tensorflow/compiler/mlir/lite
  Fetching repository @local_config_xcode; Fetching the default Xcode version
  Fetching repository @llvm-project; starting
  Fetching repository @com_google_absl; starting
  Fetching repository @Ruy; starting
  Fetching ...bazel_wangchengyun/e0602d788eedcef614d2e4eb4154b0e9/external/ruy; Extracting 3286a34cc8de6149ac6844107dfdffac91531e72.zip
  Fetching ...yun/e0602d788eedcef614d2e4eb4154b0e9/external/com_google_absl; Extracting fb3621f4f897824c0dbe0615fa94543df6192f30.tar.gz
  Fetching repository @fft2d; starting
  Fetching repository @local_config_cc; Restarting. ... (10 fetches)

[gabrielpopa]

add

local_repository(
    name = "local_xla",
    path = "third_party/tensorflow/third_party/xla",
)

to WORKSPACE file

[yg-dickson]

add

local_repository(
    name = "local_xla",
    path = "third_party/tensorflow/third_party/xla",
)

to WORKSPACE file

I added it to the configure file, but the same error message appears. Did you build it successfully?