tensorflow: STM32F7-Disco Hello World example fails to build

System information

  • OS Platform and Distribution (e.g., Linux Ubuntu 16.04): NAME=“Ubuntu” VERSION=“18.04.3 LTS (Bionic Beaver)” ID=ubuntu ID_LIKE=debian PRETTY_NAME=“Ubuntu 18.04.3 LTS” VERSION_ID=“18.04”

  • TensorFlow installed from (source or binary): Source

  • TensorFlow version: Latest

  • Python version: 2.7.15+

  • Installed using virtualenv? pip? conda?: No, git

  • Bazel version (if compiling from source):

  • GCC/Compiler version (if compiling from source): gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0

  • CUDA/cuDNN version:

  • GPU model and memory: Genuine Intel® CPU T2400 @ 1.83GHz / 2Gb mem

Describe the problem I am attempting to build the Hello World example for the STM32F7-Disco from the following link but it is failing to build as see below. Example link: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/lite/experimental/micro/examples/hello_world

Build Error

:~/developement/tensor-lite-mcu/tensorflow$ make -f tensorflow/lite/experimental/micro/tools/make/Makefile TARGET=mbed TAGS="CMSIS disco_f746ng" generate_hello_world_mbed_project
tensorflow/lite/experimental/micro/tools/make/download_and_extract.sh "https://storage.googleapis.com/mirror.tensorflow.org/github.com/google/flatbuffers/archive/v1.11.0.tar.gz" "02c64880acb89dbd57eebacfd67200d8" tensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers 
downloading https://storage.googleapis.com/mirror.tensorflow.org/github.com/google/flatbuffers/archive/v1.11.0.tar.gz
tensorflow/lite/experimental/micro/tools/make/download_and_extract.sh "https://github.com/ARM-software/CMSIS_5/archive/01c7adb7685da540be9297b5a93e6640ea3333ce.zip" "3dec53cc74f1d5d79036952137be5d5e" tensorflow/lite/experimental/micro/tools/make/downloads/cmsis 
downloading https://github.com/ARM-software/CMSIS_5/archive/01c7adb7685da540be9297b5a93e6640ea3333ce.zip
tensorflow/lite/experimental/micro/tools/make/download_and_extract.sh "https://github.com/AmbiqMicro/TFLiteMicro_CustCMSIS/archive/8f63966c5692e6a3a83956efd2e4aed77c4c9949.zip" "4fb327201034ee0a820b72de1e807d27" tensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext 
downloading https://github.com/AmbiqMicro/TFLiteMicro_CustCMSIS/archive/8f63966c5692e6a3a83956efd2e4aed77c4c9949.zip
tensorflow/lite/experimental/micro/tools/make/download_and_extract.sh "https://github.com/mborgerding/kissfft/archive/v130.zip" "438ba1fef5783cc5f5f201395cc477ca" tensorflow/lite/experimental/micro/tools/make/downloads/kissfft patch_kissfft
downloading https://github.com/mborgerding/kissfft/archive/v130.zip
Finished patching kissfft
tensorflow/lite/experimental/micro/tools/make/download_and_extract.sh "https://storage.googleapis.com/download.tensorflow.org/data/tf_lite_micro_person_data_grayscale_2019_11_07.zip" "e6430de25aa92bcb807d07278a1b5b90" tensorflow/lite/experimental/micro/tools/make/downloads/person_model_grayscale 
downloading https://storage.googleapis.com/download.tensorflow.org/data/tf_lite_micro_person_data_grayscale_2019_11_07.zip
make: *** No rule to make target 'tensorflow/lite/experimental/micro/tools/make/gen/mbed_cortex-m4/prj/hello_world/mbed/third_party/gemmlowp/fixedpoint/fixedpoint.h', needed by 'generate_hello_world_mbed_project'.  Stop.

About this issue

  • Original URL
  • State: closed
  • Created 5 years ago
  • Comments: 20 (3 by maintainers)

Most upvoted comments

I found the fix for the ‘arm_mult_q15’ in the following issue: https://github.com/tensorflow/tensorflow/issues/32037

This requires editing the methods in the “./mbed-os/cmsis/TARGET_CORTEX_M/arm_math.h” file.

Also, after performing a complete update/upgrade and dist-upgrade, I was able to compile the example in that config as well. That is after making the changes in the “arm_math.h” file.