search

lhelontra / tensorflow-on-arm

TensorFlow for Arm
MIT License
1.05k stars 274 forks source link

Illegal instruction (core dumped) #78

Closed lelopez-io closed 4 years ago

lelopez-io commented 4 years ago

@lhelontra, thank you for this tool and the available releases.

I've been able to successfully run your wheels on my raspberry pi but now I need to run it on a beagle bone black. I've tried running the same pi wheel on it but it fails to actually run my script. The error I'm given is 'Illegal Instruction (core dumped)'

I tried cross-compiling using docker and the provided beagle_black.conf file but that fails. I've attached the error messages I get below.

I was wondering if you have any info on how I would go about either fixing this issue or if you could perhaps provide a wheel built without AVX instructions.

ERROR: /root/tensorflow-on-arm/build_tensorflow/sources/tensorflow/tensorflow/lite/kernels/internal/BUILD:320:1: C++ compilation of rule '//tensorflow/lite/kernels/internal:quantization_util' failed (Exit 1): arm-linux-gnueabihf-gcc failed: error executing command
  (cd /root/.cache/bazel/_bazel_root/ce71fd092aa8d5457c1f0e68f2d49c52/execroot/org_tensorflow && \
  exec env - \
    PATH=/root/tensorflow-on-arm/build_tensorflow/sources//bin/:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin \
    PWD=/proc/self/cwd \
    PYTHON_BIN_PATH=/usr/bin/python3.7 \
    PYTHON_LIB_PATH=/usr/local/lib/python3.7/dist-packages \
    TF_CONFIGURE_IOS=0 \
    TF_DOWNLOAD_CLANG=0 \
    TF_NEED_CUDA=0 \
    TF_NEED_OPENCL_SYCL=0 \
    TF_NEED_ROCM=0 \
  /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/arm-linux-gnueabihf-gcc -U_FORTIFY_SOURCE '-D_FORTIFY_SOURCE=1' -fstack-protector -g0 -O2 -DNDEBUG -ffunction-sections -fdata-sections '-std=c++11' -isystem /usr/include -isystem /usr/include/arm-linux-gnueabihf -MD -MF bazel-out/armeabi-opt/bin/tensorflow/lite/kernels/internal/_objs/quantization_util/quantization_util.pic.d '-frandom-seed=bazel-out/armeabi-opt/bin/tensorflow/lite/kernels/internal/_objs/quantization_util/quantization_util.pic.o' -fPIC -iquote . -iquote bazel-out/armeabi-opt/genfiles -iquote bazel-out/armeabi-opt/bin '-march=armv7-a' '-mtune=cortex-a8' '-mfpu=neon' '-mfloat-abi=hard' -O3 '-std=c++11' '-DS_IREAD=S_IRUSR' '-DS_IWRITE=S_IWUSR' -U__GCC_HAVE_SYNC_COMPARE_AND_SWAP_1 -U__GCC_HAVE_SYNC_COMPARE_AND_SWAP_2 -U__GCC_HAVE_SYNC_COMPARE_AND_SWAP_8 -funsafe-math-optimizations -ftree-vectorize -fomit-frame-pointer -DRASPBERRY_PI -Wno-builtin-macro-redefined '-D__DATE__="redacted"' '-D__TIMESTAMP__="redacted"' '-D__TIME__="redacted"' -no-canonical-prefixes -fno-canonical-system-headers -c tensorflow/lite/kernels/internal/quantization_util.cc -o bazel-out/armeabi-opt/bin/tensorflow/lite/kernels/internal/_objs/quantization_util/quantization_util.pic.o)
Execution platform: @bazel_tools//platforms:host_platform
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/math.h:48:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cmath:45,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:17:
/usr/include/arm-linux-gnueabihf/bits/mathdef.h:19:3: error: #error "Never use <bits/mathdef.h> directly; include <complex.h> instead"
 # error "Never use <bits/mathdef.h> directly; include <complex.h> instead"
   ^~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:95:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:101:5: error: 'div_t' does not name a type
   } div_t;
     ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:112:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:121:5: error: 'lldiv_t' does not name a type
   } lldiv_t;
     ^~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:123:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:142:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:152:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:159:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:167:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:178:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:190:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:206:1: error: '__BEGIN_NAMESPACE_C99' does not name a type
 __BEGIN_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:217:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:235:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/xlocale.h:39:14: error: expected constructor, destructor, or type conversion before ';' token
 } *__locale_t;
              ^
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/xlocale.h:42:9: error: '__locale_t' does not name a type
 typedef __locale_t locale_t;
         ^~~~~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:241:6: error: '__locale_t' has not been declared
      __locale_t __loc) __THROW __nonnull ((1, 4));
      ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:241:32: error: nonnull argument references non-pointer operand (argument 2, operand 4)
      __locale_t __loc) __THROW __nonnull ((1, 4));
                                ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:245:21: error: '__locale_t' has not been declared
         int __base, __locale_t __loc)
                     ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:246:14: error: nonnull argument references non-pointer operand (argument 2, operand 4)
      __THROW __nonnull ((1, 4));
              ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:251:5: error: '__locale_t' has not been declared
     __locale_t __loc)
     ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:252:14: error: nonnull argument references non-pointer operand (argument 2, operand 4)
      __THROW __nonnull ((1, 4));
              ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:257:20: error: '__locale_t' has not been declared
        int __base, __locale_t __loc)
                    ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:258:14: error: nonnull argument references non-pointer operand (argument 2, operand 4)
      __THROW __nonnull ((1, 4));
              ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:261:32: error: '__locale_t' has not been declared
    char **__restrict __endptr, __locale_t __loc)
                                ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:262:14: error: nonnull argument references non-pointer operand (argument 2, operand 3)
      __THROW __nonnull ((1, 3));
              ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:265:38: error: '__locale_t' has not been declared
          char **__restrict __endptr, __locale_t __loc)
                                      ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:266:14: error: nonnull argument references non-pointer operand (argument 2, operand 3)
      __THROW __nonnull ((1, 3));
              ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:270:10: error: '__locale_t' has not been declared
          __locale_t __loc)
          ^~~~~~~~~~
In file included from /usr/include/features.h:424:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/os_defines.h:39,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/arm-linux-gnueabihf/./bits/c++config.h:507,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/utility:68,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:60,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:271:14: error: nonnull argument references non-pointer operand (argument 2, operand 3)
      __THROW __nonnull ((1, 3));
              ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:276:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h: In function 'long int atol(const char*)':
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:285:44: error: 'strtol' was not declared in this scope
   return strtol (__nptr, (char **) NULL, 10);
                                            ^
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h: At global scope:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:287:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:296:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:372:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:377:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:464:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:470:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:484:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:513:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:530:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:539:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:551:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:558:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:565:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:711:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:717:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:751:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:776:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:784:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:792:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:799:1: error: '__END_NAMESPACE_C99' does not name a type
 __END_NAMESPACE_C99
 ^~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:859:1: error: '__BEGIN_NAMESPACE_STD' does not name a type
 __BEGIN_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:879:1: error: '__END_NAMESPACE_STD' does not name a type
 __END_NAMESPACE_STD
 ^~~~~~~~~~~~~~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../arm-linux-gnueabihf/libc/usr/include/stdlib.h:954:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:75,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/usr/include/arm-linux-gnueabihf/bits/stdlib-float.h: In function 'double atof(const char*)':
/usr/include/arm-linux-gnueabihf/bits/stdlib-float.h:27:40: error: 'strtod' was not declared in this scope
   return strtod (__nptr, (char **) NULL);
                                        ^
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:59:0,
                 from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib: At global scope:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:124:11: error: '::div_t' has not been declared
   using ::div_t;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:127:11: error: '::abort' has not been declared
   using ::abort;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:140:11: error: '::div' has not been declared
   using ::div;
           ^~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:141:11: error: '::exit' has not been declared
   using ::exit;
           ^~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:143:11: error: '::getenv' has not been declared
   using ::getenv;
           ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:146:11: error: '::malloc' has not been declared
   using ::malloc;
           ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:148:11: error: '::mblen' has not been declared
   using ::mblen;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:158:11: error: '::rand' has not been declared
   using ::rand;
           ^~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:159:11: error: '::realloc' has not been declared
   using ::realloc;
           ^~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:161:11: error: '::strtod' has not been declared
   using ::strtod;
           ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:162:11: error: '::strtol' has not been declared
   using ::strtol;
           ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:164:11: error: '::system' has not been declared
   using ::system;
           ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:220:11: error: '::lldiv_t' has not been declared
   using ::lldiv_t;
           ^~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:226:11: error: '::_Exit' has not been declared
   using ::_Exit;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:230:11: error: '::llabs' has not been declared
   using ::llabs;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:232:10: error: 'lldiv_t' does not name a type
   inline lldiv_t
          ^~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:236:11: error: '::lldiv' has not been declared
   using ::lldiv;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:247:11: error: '::atoll' has not been declared
   using ::atoll;
           ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:248:11: error: '::strtoll' has not been declared
   using ::strtoll;
           ^~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:251:11: error: '::strtof' has not been declared
   using ::strtof;
           ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:260:22: error: '__gnu_cxx::lldiv_t' has not been declared
   using ::__gnu_cxx::lldiv_t;
                      ^~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:262:22: error: '__gnu_cxx::_Exit' has not been declared
   using ::__gnu_cxx::_Exit;
                      ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:264:22: error: '__gnu_cxx::llabs' has not been declared
   using ::__gnu_cxx::llabs;
                      ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:265:22: error: '__gnu_cxx::div' has not been declared
   using ::__gnu_cxx::div;
                      ^~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:266:22: error: '__gnu_cxx::lldiv' has not been declared
   using ::__gnu_cxx::lldiv;
                      ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:268:22: error: '__gnu_cxx::atoll' has not been declared
   using ::__gnu_cxx::atoll;
                      ^~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:269:22: error: '__gnu_cxx::strtof' has not been declared
   using ::__gnu_cxx::strtof;
                      ^~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cstdlib:270:22: error: '__gnu_cxx::strtoll' has not been declared
   using ::__gnu_cxx::strtoll;
                      ^~~~~~~
In file included from /root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/algorithm:62:0,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:16:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h: In function 'void std::random_shuffle(_RAIter, _RAIter)':
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/bits/stl_algo.h:4451:8: error: 'rand' is not a member of 'std'
      + std::rand() % ((__i - __first) + 1);
        ^~~
In file included from tensorflow/lite/kernels/internal/quantization_util.cc:17:0:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cmath: At global scope:
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cmath:1101:11: error: '::double_t' has not been declared
   using ::double_t;
           ^~~~~~~~
/root/tensorflow-on-arm/build_tensorflow/sources/toolchain/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/bin/../lib/gcc/arm-linux-gnueabihf/6.3.1/../../../../arm-linux-gnueabihf/include/c++/6.3.1/cmath:1102:11: error: '::float_t' has not been declared
   using ::float_t;
           ^~~~~~~
In file included from ./tensorflow/lite/kernels/internal/compatibility.h:20:0,
                 from tensorflow/lite/kernels/internal/quantization_util.cc:20:
./tensorflow/lite/kernels/internal/types.h: In constructor 'tflite::RuntimeShape::RuntimeShape(int, const tflite::RuntimeShape&, int)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: 'abort' was not declared in this scope
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:64:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_GE(x, y) ((x) >= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/types.h:285:5: note: in expansion of macro 'TFLITE_CHECK_GE'
     TFLITE_CHECK_GE(new_shape_size, shape.DimensionsCount());
     ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/internal/types.h: In function 'tflite::Dims<4> tflite::ToRuntimeDims(const tflite::RuntimeShape&)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: 'abort' was not declared in this scope
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:72:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_LE(x, y) ((x) <= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/types.h:306:3: note: in expansion of macro 'TFLITE_CHECK_LE'
   TFLITE_CHECK_LE(dimensions_count, 4);
   ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h: In function 'tflite::QuantizationParams tflite::ChooseQuantizationParams(double, double, bool)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: there are no arguments to 'abort' that depend on a template parameter, so a declaration of 'abort' must be available [-fpermissive]
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:72:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_LE(x, y) ((x) <= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:39:3: note: in expansion of macro 'TFLITE_CHECK_LE'
   TFLITE_CHECK_LE(rmin, 0.);
   ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/op_macros.h:49:28: note: (if you use '-fpermissive', G++ will accept your code, but allowing the use of an undeclared name is deprecated)
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:72:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_LE(x, y) ((x) <= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:39:3: note: in expansion of macro 'TFLITE_CHECK_LE'
   TFLITE_CHECK_LE(rmin, 0.);
   ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/op_macros.h:49:28: error: there are no arguments to 'abort' that depend on a template parameter, so a declaration of 'abort' must be available [-fpermissive]
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:64:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_GE(x, y) ((x) >= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:40:3: note: in expansion of macro 'TFLITE_CHECK_GE'
   TFLITE_CHECK_GE(rmax, 0.);
   ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/op_macros.h:49:28: error: there are no arguments to 'abort' that depend on a template parameter, so a declaration of 'abort' must be available [-fpermissive]
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:56:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_EQ(x, y) ((x) == (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:43:5: note: in expansion of macro 'TFLITE_CHECK_EQ'
     TFLITE_CHECK_EQ(rmin, 0.);
     ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/op_macros.h:49:28: error: there are no arguments to 'abort' that depend on a template parameter, so a declaration of 'abort' must be available [-fpermissive]
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:56:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_EQ(x, y) ((x) == (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:44:5: note: in expansion of macro 'TFLITE_CHECK_EQ'
     TFLITE_CHECK_EQ(rmax, 0.);
     ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/op_macros.h:49:28: error: there are no arguments to 'abort' that depend on a template parameter, so a declaration of 'abort' must be available [-fpermissive]
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:64:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_GE(x, y) ((x) >= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:91:3: note: in expansion of macro 'TFLITE_CHECK_GE'
   TFLITE_CHECK_GE(nudged_zero_point, qmin);
   ^~~~~~~~~~~~~~~
./tensorflow/lite/kernels/op_macros.h:49:28: error: there are no arguments to 'abort' that depend on a template parameter, so a declaration of 'abort' must be available [-fpermissive]
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:72:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_LE(x, y) ((x) <= (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
./tensorflow/lite/kernels/internal/quantization_util.h:92:3: note: in expansion of macro 'TFLITE_CHECK_LE'
   TFLITE_CHECK_LE(nudged_zero_point, qmax);
   ^~~~~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc: In function 'void tflite::QuantizeMultiplier(double, int32_t*, int*)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: 'abort' was not declared in this scope
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:52:57: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK(condition) (condition) ? (void)0 : TFLITE_ABORT
                                                         ^~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc:69:3: note: in expansion of macro 'TFLITE_CHECK'
   TFLITE_CHECK(q_fixed <= (1ll << 31));
   ^~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc: In function 'void tflite::QuantizeMultiplierGreaterThanOne(double, int32_t*, int*)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: 'abort' was not declared in this scope
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:68:55: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_GT(x, y) ((x) > (y)) ? (void)0 : TFLITE_ABORT
                                                       ^~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc:81:3: note: in expansion of macro 'TFLITE_CHECK_GT'
   TFLITE_CHECK_GT(double_multiplier, 1.);
   ^~~~~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc: In function 'void tflite::QuantizeMultiplierSmallerThanOneExp(double, int32_t*, int*)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: 'abort' was not declared in this scope
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:76:55: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_LT(x, y) ((x) < (y)) ? (void)0 : TFLITE_ABORT
                                                       ^~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc:89:3: note: in expansion of macro 'TFLITE_CHECK_LT'
   TFLITE_CHECK_LT(double_multiplier, 1.);
   ^~~~~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc: In function 'int64_t tflite::IntegerFrExp(double, int*)':
./tensorflow/lite/kernels/op_macros.h:49:28: error: 'abort' was not declared in this scope
 #define TFLITE_ABORT abort()
                            ^
./tensorflow/lite/kernels/internal/compatibility.h:56:56: note: in expansion of macro 'TFLITE_ABORT'
 #define TFLITE_CHECK_EQ(x, y) ((x) == (y)) ? (void)0 : TFLITE_ABORT
                                                        ^~~~~~~~~~~~
tensorflow/lite/kernels/internal/quantization_util.cc:99:3: note: in expansion of macro 'TFLITE_CHECK_EQ'
   TFLITE_CHECK_EQ(8, sizeof(double));
   ^~~~~~~~~~~~~~~
Target //tensorflow/tools/pip_package:build_pip_package failed to build
INFO: Elapsed time: 758.513s, Critical Path: 161.90s
INFO: 1048 processes: 1048 local.
FAILED: Build did NOT complete successfully
lhelontra commented 4 years ago

Hi, downgrade the gcc version, replace in your config file:

CROSSTOOL_URL="https://releases.linaro.org/components/toolchain/binaries/6.3-2017.05/arm-linux-gnueabihf/gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf.tar.xz"
CROSSTOOL_DIR="gcc-linaro-6.3.1-2017.05-x86_64_arm-linux-gnueabihf/"

To

CROSSTOOL_URL="https://releases.linaro.org/components/toolchain/binaries/5.5-2017.10/arm-linux-gnueabihf/gcc-linaro-5.5.0-2017.10-x86_64_arm-linux-gnueabihf.tar.xz"
CROSSTOOL_DIR="gcc-linaro-5.5.0-2017.10-x86_64_arm-linux-gnueabihf"
lelopez-io commented 4 years ago

Thank you, I'll have to try this

JproChen commented 4 years ago

Hi, I have the same problem when using pynq-board, with arm, I installed the 1.12 arm7.whl, it can successfully import, but when running tf functions will have the error illegal instruction too, Is there any solution? thanks a lot.

lhelontra commented 4 years ago

It should probably be different instructions contained in armv7 (for rpi) whl. Best way is edit rpi.conf and replace for correct options for your board.