You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
On v0.4.4, I cannot reproduce the build of gpu_delegate_plugin_android.aar. I want to rebuild the library locally with the support for arm64 page size 16kb, added in this TF commit (betwen 2.16.1 and 2.17)
I could not find instructions to build this library, are there any?
I tried to guess how to build it:
I followed the instructions here to setup the docker container to compile, then I updated the tensorflow commit to point to 2.16.1, and finally I ran bazel build -c opt --config android_arm64 //tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android.aar in this dev env
I tried with both bazel 5.1.1 and bazel 6.5 without success, the logs are below.
I don't think that the errors matter too much because I think that my approach might be wrong - How can I build this app? What is the proper build environment? Tensorflow-lite as a dockerfile to create the build env, could we have one for tensorflow-lite-support?
TF 2.16.1, BAZEL 6.5
bazel build -c opt --config android_arm64 //tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android.aar
Starting local Bazel server and connecting to it...
WARNING: Option 'java_toolchain' is deprecated
WARNING: Option 'host_java_toolchain' is deprecated
INFO: Options provided by the client:
Inherited 'common' options: --isatty=1 --terminal_columns=168
INFO: Reading rc options for 'build' from /host_dir/.bazelrc:
Inherited 'common' options: --experimental_repo_remote_exec
INFO: Reading rc options for 'build' from /etc/bazel.bazelrc:
'build' options: --action_env=DOCKER_CACHEBUSTER=1723940018419409631 --host_action_env=DOCKER_HOST_CACHEBUSTER=1723940018483040100
INFO: Reading rc options for 'build' from /host_dir/.bazelrc:
'build' options: --java_toolchain=//third_party/toolchains/java:tf_java_toolchain --host_java_toolchain=//third_party/toolchains/java:tf_java_toolchain --action_env ANDROID_NDK_HOME --action_env ANDROID_NDK_API_LEVEL --action_env ANDROID_BUILD_TOOLS_VERSION --action_env ANDROID_SDK_API_LEVEL --action_env ANDROID_SDK_HOME --define framework_shared_object=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=local -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --enable_platform_specific_config --config=short_logs --config=v2
INFO: Found applicable config definition build:short_logs in file /host_dir/.bazelrc: --output_filter=DONT_MATCH_ANYTHING
INFO: Found applicable config definition build:v2 in file /host_dir/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1
INFO: Found applicable config definition build:android_arm64 in file /host_dir/.bazelrc: --config=android --cpu=arm64-v8a --fat_apk_cpu=arm64-v8a
INFO: Found applicable config definition build:android in file /host_dir/.bazelrc: --copt=-w --crosstool_top=//external:android/crosstool --host_crosstool_top=@bazel_tools//tools/cpp:toolchain --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --noenable_platform_specific_config
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/tensorflow/tools/toolchains/python/python_repo.bzl:32:14:
TF_PYTHON_VERSION environment variable was not set correctly; using Python 3.11.
To set Python version, run:
export TF_PYTHON_VERSION=3.11
DEBUG: Rule 'robolectric' indicated that a canonical reproducible form can be obtained by modifying arguments sha256 = "97f169d39f19412bdd07fd6c274dcda0a7c8f623f7f00aa5a3b94994fc6f0ec4"
DEBUG: Repository robolectric instantiated at:
/host_dir/WORKSPACE:55:26: in <toplevel>
/host_dir/workspace8.bzl:78:17: in tflite_support_workspace8
Repository rule http_archive defined at:
/root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/bazel_tools/tools/build_defs/repo/http.bzl:353:31: in <toplevel>
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_google_absl' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'flatbuffers' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'icu' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_googlesource_code_re2' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'six_archive' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'absl_py' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_google_protobuf' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_google_googletest' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_github_gflags_gflags' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'zlib' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'fft2d' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'snappy' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'build_bazel_rules_apple' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'build_bazel_rules_swift' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'build_bazel_apple_support' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'pybind11' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'pybind11_protobuf' because it already exists.
ERROR: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_executable.bzl:804:12: name 'RunEnvironmentInfo' is not defined
ERROR: /host_dir/tensorflow_lite_support/acceleration/configuration/BUILD:124:16: While resolving toolchains for target //tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android: invalid registered toolchain '@local_config_python//:py_toolchain': error loading package '@local_config_python//': at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/python/defs.bzl:9:5: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/entry_points/py_console_script_binary.bzl:77:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/py_console_script_binary.bzl:19:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/py_binary.bzl:19:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_binary_macro_bazel.bzl:17:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_binary_rule_bazel.bzl:19:5: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_executable_bazel.bzl:30:5: compilation of module 'python/private/common/py_executable.bzl' failed
ERROR: Analysis of target '//tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android.aar' failed; build aborted:
INFO: Elapsed time: 28.770s
INFO: 0 processes.
FAILED: Build did NOT complete successfully (40 packages loaded, 9 targets configured)
currently loading: @local_execution_config_python// ... (2 packages)
Fetching @local_jdk; fetching
Fetching @go_sdk; fetching
TF 2.16.1, BAZEL 5.1.1
bazel build //tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android.aar
WARNING: Option 'java_toolchain' is deprecated
WARNING: Option 'host_java_toolchain' is deprecated
INFO: Options provided by the client:
Inherited 'common' options: --isatty=1 --terminal_columns=168
INFO: Reading rc options for 'build' from /host_dir/.bazelrc:
Inherited 'common' options: --experimental_repo_remote_exec
INFO: Reading rc options for 'build' from /etc/bazel.bazelrc:
'build' options: --action_env=DOCKER_CACHEBUSTER=1723940018419409631 --host_action_env=DOCKER_HOST_CACHEBUSTER=1723940018483040100
INFO: Reading rc options for 'build' from /host_dir/.bazelrc:
'build' options: --java_toolchain=//third_party/toolchains/java:tf_java_toolchain --host_java_toolchain=//third_party/toolchains/java:tf_java_toolchain --action_env ANDROID_NDK_HOME --action_env ANDROID_NDK_API_LEVEL --action_env ANDROID_BUILD_TOOLS_VERSION --action_env ANDROID_SDK_API_LEVEL --action_env ANDROID_SDK_HOME --define framework_shared_object=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=local -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --enable_platform_specific_config --config=short_logs --config=v2
INFO: Found applicable config definition build:short_logs in file /host_dir/.bazelrc: --output_filter=DONT_MATCH_ANYTHING
INFO: Found applicable config definition build:v2 in file /host_dir/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1
INFO: Found applicable config definition build:linux in file /host_dir/.bazelrc: --copt=-w --cxxopt=-std=c++17 --host_cxxopt=-std=c++17
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/tensorflow/tools/toolchains/python/python_repo.bzl:32:14:
TF_PYTHON_VERSION environment variable was not set correctly; using Python 3.11.
To set Python version, run:
export TF_PYTHON_VERSION=3.11
DEBUG: Rule 'robolectric' indicated that a canonical reproducible form can be obtained by modifying arguments sha256 = "97f169d39f19412bdd07fd6c274dcda0a7c8f623f7f00aa5a3b94994fc6f0ec4"
DEBUG: Repository robolectric instantiated at:
/host_dir/WORKSPACE:55:26: in <toplevel>
/host_dir/workspace8.bzl:78:17: in tflite_support_workspace8
Repository rule http_archive defined at:
/root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/bazel_tools/tools/build_defs/repo/http.bzl:353:31: in <toplevel>
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_google_absl' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'flatbuffers' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'icu' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_googlesource_code_re2' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'six_archive' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'absl_py' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_google_protobuf' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_google_googletest' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'com_github_gflags_gflags' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'zlib' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'fft2d' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'snappy' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'build_bazel_rules_apple' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'build_bazel_rules_swift' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'build_bazel_apple_support' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'pybind11' because it already exists.
DEBUG: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/org_tensorflow/third_party/repo.bzl:132:14:
Warning: skipping import of repository 'pybind11_protobuf' because it already exists.
INFO: Build options --cpu, --crosstool_top, and --fat_apk_cpu have changed, discarding analysis cache.
ERROR: /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_executable.bzl:804:12: name 'RunEnvironmentInfo' is not defined
ERROR: /host_dir/tensorflow_lite_support/acceleration/configuration/BUILD:124:16: While resolving toolchains for target //tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android: invalid registered toolchain '@local_execution_config_python//:py_toolchain': error loading package '@local_execution_config_python//': at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/python/defs.bzl:4:5: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/config_settings/transition.bzl:21:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/py_test.bzl:19:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_test_macro_bazel.bzl:17:6: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_test_rule_bazel.bzl:20:5: at /root/.cache/bazel/_bazel_root/faa585346465e7978a465c0d1bdb5e9e/external/rules_python/python/private/common/py_executable_bazel.bzl:30:5: compilation of module 'python/private/common/py_executable.bzl' failed
ERROR: Analysis of target '//tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android.aar' failed; build aborted:
INFO: Elapsed time: 0.233s
INFO: 0 processes.
FAILED: Build did NOT complete successfully (0 packages loaded, 9 targets configured)
currently loading: @local_config_python// ... (3 packages)
Fetching @go_sdk; fetching
Fetching @local_jdk; fetching
The text was updated successfully, but these errors were encountered:
On v0.4.4, I cannot reproduce the build of gpu_delegate_plugin_android.aar. I want to rebuild the library locally with the support for arm64 page size 16kb, added in this TF commit (betwen 2.16.1 and 2.17)
I could not find instructions to build this library, are there any?
I tried to guess how to build it:
I followed the instructions here to setup the docker container to compile, then I updated the tensorflow commit to point to 2.16.1, and finally I ran
bazel build -c opt --config android_arm64 //tensorflow_lite_support/acceleration/configuration:gpu_delegate_plugin_android.aar
in this dev envI tried with both bazel 5.1.1 and bazel 6.5 without success, the logs are below.
I don't think that the errors matter too much because I think that my approach might be wrong - How can I build this app? What is the proper build environment? Tensorflow-lite as a dockerfile to create the build env, could we have one for tensorflow-lite-support?
TF 2.16.1, BAZEL 6.5
TF 2.16.1, BAZEL 5.1.1
The text was updated successfully, but these errors were encountered: