As the title suggests, I am attempting to cross compile a Pi Zero 2W compatible LiteRT following the official guide. I am getting a Cmake error regarding flatc. Doe this mean I need to cross compile flatbuffers first then point the binary path using -DTFLITE_HOST_TOOLS_DIR=<flatc_dir_path> first?
$ cmake -DCMAKE_C_COMPILER=${ARMCC_PREFIX}gcc -DCMAKE_CXX_COMPILER=${ARMCC_PREFIX}g++ -DCMAKE_C_FLAGS="${ARMCC_FLAGS}" -DCMAKE_CXX_FLAGS="${ARMCC_FLAGS}" -DCMAKE_VERBOSE_MAKEFILE:BOOL=ON -DCMAKE_SYSTEM_NAME=Linux -DCMAKE_SYSTEM_PROCESSOR=aarch64 ./tensorflow/tensorflow/lite/
-- Setting build type to Release, for debug builds use'-DCMAKE_BUILD_TYPE=Debug'.
CMake Error at CMakeLists.txt:83 (message):
When cross-compiling, some tools need to be available to run on the host
(current required tools: flatc). Please specify where those binaries can
be found by using -DTFLITE_HOST_TOOLS_DIR=<flatc_dir_path>.