Onnx-tensorrt: make -j8 popping out the error

Created on 11 Jul 2018  路  12Comments  路  Source: onnx/onnx-tensorrt

I was able to complete the first three steps:

mkdir build
cd build
cmake .. -DTENSORRT_ROOT=/opt/tensorrt

in the fourth step

make -j8

I get the following error:

Error generating file
/home/anish-fujitsu/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o

CMakeFiles/nvonnxparser_plugin.dir/build.make:886: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o' failed
make[2]: * [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o] Error 1
CMakeFiles/Makefile2:180: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/all' failed
make[1]: [CMakeFiles/nvonnxparser_plugin.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: ** [all] Error 2

picture anish9

Most helpful comment

_we have found the solution for the above error_
make sure you are updated to the latest CUDA and nvcc versions and not only that ensure the right nvcc is invoked.
_to ensure perform the following steps:_

from the terminal

nano /home/username/.bashrc
or
nano /home/$USER/.bashrc

Inside there add the following:

export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}$ 
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

Files we used on the fixing of the following error:

  1. libcudnn7-dev_7.1.4.18-1+cuda9.2_amd64.deb
  2. libcudnn7-doc_7.1.4.18-1cuda9.2_amd64.deb
  3. nv-tensorrt-repo-ubuntu1604-cuda9.2-ga-trt4.0.1.6-20180612_1-1_amd64.deb
  4. cuda-repo-ubuntu1604_9.2.148-1_amd64.deb
  5. libcudnn7_7.1.4.18-1+cuda9.2_amd64.deb
picture anish9 on 13 Jul 2018
👍2

All 12 comments

Could you do make VERBOSE=1 and paste the detailed error message that it outputs?

yinghai picture yinghai on 11 Jul 2018

5 errors detected in the compilation of "/tmp/tmpxft_000016ac_00000000-7_Split.cpp1.ii".
-- Removing /home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
/usr/bin/cmake -E remove /home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
CMake Error at nvonnxparser_plugin_generated_Split.cu.o.cmake:266 (message):
Error generating file
/home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o

CMakeFiles/nvonnxparser_plugin.dir/build.make:1769: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_Split.cu.o' failed
make[2]: * [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_Split.cu.o] Error 1
make[2]: Waiting for unfinished jobs....
/usr/include/string.h: In function 'void __mempcpy_inline(void, const void, size_t)':
/usr/include/string.h:652:42: error: 'memcpy' was not declared in this scope
return (char *) memcpy (__dest, __src, __n) + __n;
^
-- Removing /home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_ResizeNearest.cu.o
/usr/bin/cmake -E remove /home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_ResizeNearest.cu.o
CMake Error at nvonnxparser_plugin_generated_ResizeNearest.cu.o.cmake:266 (message):
Error generating file
/home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_ResizeNearest.cu.o

CMakeFiles/nvonnxparser_plugin.dir/build.make:1140: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_ResizeNearest.cu.o' failed
make[2]: * [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_ResizeNearest.cu.o] Error 1
/usr/include/string.h: In function 'void* __mempcpy_inline(void, const void, size_t)':
/usr/include/string.h:652:42: error: 'memcpy' was not declared in this scope
return (char *) memcpy (__dest, __src, __n) + __n;
^
-- Removing /home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
/usr/bin/cmake -E remove /home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
CMake Error at nvonnxparser_plugin_generated_FancyActivation.cu.o.cmake:266 (message):
Error generating file
/home/anish-macbook/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o

CMakeFiles/nvonnxparser_plugin.dir/build.make:886: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o' failed
make[2]: * [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o] Error 1
make[2]: Leaving directory '/home/anish-macbook/onnx-tensorrt/build'
CMakeFiles/Makefile2:180: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/all' failed
make[1]: [CMakeFiles/nvonnxparser_plugin.dir/all] Error 2
make[1]: Leaving directory '/home/anish-macbook/onnx-tensorrt/build'
Makefile:149: recipe for target 'all' failed
make: ** [all] Error 2

anish9 picture anish9 on 11 Jul 2018

hi yinghai, can you figure out what could be the reason???

anish9 picture anish9 on 11 Jul 2018

@yinghai Is it okay to install it with Tensorrt-4? coz I am getting the same error:

CMakeFiles/nvonnxparser_plugin.dir/build.make:77: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_Split.cu.o' failed
make[2]: *** [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_Split.cu.o] Error 1
make[2]: *** Waiting for unfinished jobs....
[ 27%] Linking CXX static library libonnx_proto.a
[ 27%] Built target onnx_proto
/usr/include/string.h: In function 'void* __mempcpy_inline(void*, const void*, size_t)':
/usr/include/string.h:652:42: error: 'memcpy' was not declared in this scope
   return (char *) memcpy (__dest, __src, __n) + __n;
                                          ^
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_ResizeNearest.cu.o
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_ResizeNearest.cu.o
CMake Error at nvonnxparser_plugin_generated_ResizeNearest.cu.o.cmake:266 (message):
  Error generating file
  /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_ResizeNearest.cu.o


CMakeFiles/nvonnxparser_plugin.dir/build.make:70: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_ResizeNearest.cu.o' failed
make[2]: *** [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_ResizeNearest.cu.o] Error 1
/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a type specifier

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: attributes are not permitted in a function definition

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a "{"

/usr/include/device_functions.hpp(258): error: identifier "predicate" is undefined in device code

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a type specifier

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: attributes are not permitted in a function definition

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: function "__launch_bounds__" has already been defined

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a "{"

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: identifier "__val_paramf" is undefined in device code

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_task.hpp(240): error: a value of type "int" cannot be assigned to an entity of type "_ZN6thrust6system4cuda6detail5bulk_6detail9cuda_taskINS3_14parallel_groupINS3_16concurrent_groupINS3_5agentILm1EEELm0EEELm0EEENS4_7closureINS2_17for_each_n_detail15for_each_kernelENS_5tupleINS4_6cursorILj0EEENS_12zip_iteratorINSF_IPKfPfNS_9null_typeESM_SM_SM_SM_SM_SM_SM_EEEENS_6detail16wrapped_functionINSP_23unary_transform_functorIZN21FancyActivationPlugin9doEnqueueIfEEiiPKPKvPPvSY_P11CUstream_stEUnvdl1_M21FancyActivationPluginFiiPKPKvPPvPvP11CUstream_stEN21FancyActivationPlugin9doEnqueueIfEE1_fEEvEEjSM_SM_SM_SM_SM_SM_EEEEE9grid_typeE"

/usr/include/thrust/system/cuda/detail/bulk/malloc.hpp(462): internal error: assertion failed at: "../../edg/EDG_4.9/src/c_gen_be.c", line 1621


10 errors and 1 catastrophic error detected in the compilation of "/tmp/tmpxft_00003734_00000000-10_FancyActivation.cpp3.i".
Compilation aborted.
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
CMake Error at nvonnxparser_plugin_generated_FancyActivation.cu.o.cmake:266 (message):
  Error generating file
  /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o


CMakeFiles/nvonnxparser_plugin.dir/build.make:63: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o' failed
make[2]: *** [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o] Error 1
CMakeFiles/Makefile2:180: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/all' failed
make[1]: *** [CMakeFiles/nvonnxparser_plugin.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: *** [all] Error 2
gr8Adakron picture gr8Adakron on 11 Jul 2018

@anish9 Installing with TensorRT-4.0 is OK as I just tried. Regarding your compilation error, I suspect your CUDA version is too old. We require at least CUDA 8.0. Here is a similar issue that has been reported: https://github.com/BVLC/caffe/issues/4046

yinghai picture yinghai on 11 Jul 2018

@yinghai we checked our CUDA version and its 9.0

anish9 picture anish9 on 11 Jul 2018

Try adding -D_FORCE_INLINES to NVCCFLAGS as suggested in BVLC/caffe#4046?

yinghai picture yinghai on 11 Jul 2018

I guess your -DTENSORRT_ROOT path is incorrect. In my case, the path is /usr/src/tensorrt.

Faldict picture Faldict on 12 Jul 2018

@yinghai Where is NVCCFLAG?

The error still persists, I couldn't find the files which are mentioned in[ this caffe issue] (https://github.com/BVLC/caffe/issues/4046) | as this issue is 2years old file doesn't match.
This is my computer configuration:

nvcc --version

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2015 NVIDIA Corporation
Built on Tue_Aug_11_14:27:32_CDT_2015
Cuda compilation tools, release 7.5, V7.5.17

And other installation configurations
CUDA-9.0
Tensorrt=4.0
Python=3.6
Ubuntu=16.04

Successful steps:

mkdir build
cd build
cmake .. -DTENSORRT_ROOT=/usr/src/tensorrt

The error:

make -j8

-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
-- Generating dependency file: /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.NVCC-depend
-- Generating dependency file: /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.NVCC-depend
/usr/bin/nvcc -M -D__CUDACC__ /home/afzal/cuda-gpu-setup/onnx-tensorrt/Split.cu -o /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.NVCC-depend -ccbin /usr/bin/c++ -m64 -DONNX_NAMESPACE=onnx2trt_onnx -Xcompiler ,\"-Wall\",\"-O3\",\"-DNDEBUG\" -D_FORCE_INLINES -std=c++11 -Xcompiler -fPIC --expt-extended-lambda -DNVCC -I/usr/include -I/usr/include/x86_64-linux-gnu -I/usr/include -I/usr/include
/usr/bin/nvcc -M -D__CUDACC__ /home/afzal/cuda-gpu-setup/onnx-tensorrt/FancyActivation.cu -o /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.NVCC-depend -ccbin /usr/bin/c++ -m64 -DONNX_NAMESPACE=onnx2trt_onnx -Xcompiler ,\"-Wall\",\"-O3\",\"-DNDEBUG\" -D_FORCE_INLINES -std=c++11 -Xcompiler -fPIC --expt-extended-lambda -DNVCC -I/usr/include -I/usr/include/x86_64-linux-gnu -I/usr/include -I/usr/include
-- Generating temporary cmake readable file: /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend.tmp
/usr/bin/cmake -D input_file:FILEPATH=/home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.NVCC-depend -D output_file:FILEPATH=/home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend.tmp -P /usr/share/cmake-3.5/Modules/FindCUDA/make2cmake.cmake
-- Copy if different /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend.tmp to /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend
/usr/bin/cmake -E copy_if_different /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend.tmp /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend.tmp and /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.NVCC-depend
-- Generating temporary cmake readable file: /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend.tmp
/usr/bin/cmake -D input_file:FILEPATH=/home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.NVCC-depend -D output_file:FILEPATH=/home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend.tmp -P /usr/share/cmake-3.5/Modules/FindCUDA/make2cmake.cmake
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.depend.tmp /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_Split.cu.o.NVCC-depend
-- Generating /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
/usr/bin/nvcc /home/afzal/cuda-gpu-setup/onnx-tensorrt/Split.cu -c -o /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o -ccbin /usr/bin/c++ -m64 -DONNX_NAMESPACE=onnx2trt_onnx -Xcompiler ,\"-Wall\",\"-O3\",\"-DNDEBUG\" -D_FORCE_INLINES -std=c++11 -Xcompiler -fPIC --expt-extended-lambda -DNVCC -I/usr/include -I/usr/include/x86_64-linux-gnu -I/usr/include -I/usr/include
-- Copy if different /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend.tmp to /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend
/usr/bin/cmake -E copy_if_different /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend.tmp /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend.tmp and /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.NVCC-depend
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.depend.tmp /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//nvonnxparser_plugin_generated_FancyActivation.cu.o.NVCC-depend
-- Generating /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
/usr/bin/nvcc /home/afzal/cuda-gpu-setup/onnx-tensorrt/FancyActivation.cu -c -o /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o -ccbin /usr/bin/c++ -m64 -DONNX_NAMESPACE=onnx2trt_onnx -Xcompiler ,\"-Wall\",\"-O3\",\"-DNDEBUG\" -D_FORCE_INLINES -std=c++11 -Xcompiler -fPIC --expt-extended-lambda -DNVCC -I/usr/include -I/usr/include/x86_64-linux-gnu -I/usr/include -I/usr/include
/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a type specifier

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: attributes are not permitted in a function definition

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a "{"

/usr/include/device_functions.hpp(258): error: identifier "predicate" is undefined in device code

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a type specifier

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: attributes are not permitted in a function definition

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: function "__launch_bounds__" has already been defined

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a "{"

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: identifier "__val_paramf" is undefined in device code

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_task.hpp(240): error: a value of type "int" cannot be assigned to an entity of type "_ZN6thrust6system4cuda6detail5bulk_6detail9cuda_taskINS3_14parallel_groupINS3_16concurrent_groupINS3_5agentILm1EEELm0EEELm0EEENS4_7closureINS2_17for_each_n_detail15for_each_kernelENS_5tupleINS4_6cursorILj0EEENS_12zip_iteratorINSF_INS_6detail15normal_iteratorINS_10device_ptrIPfEEEENS_7pointerISM_NS2_3tagENS_11use_defaultESR_EENS_9null_typeEST_ST_ST_ST_ST_ST_ST_EEEENSJ_16wrapped_functionINSJ_23unary_transform_functorINS_8identityISM_EEEEvEEjST_ST_ST_ST_ST_ST_EEEEE9grid_typeE"

/usr/include/thrust/system/cuda/detail/bulk/malloc.hpp(462): internal error: assertion failed at: "../../edg/EDG_4.9/src/c_gen_be.c", line 1621


10 errors and 1 catastrophic error detected in the compilation of "/tmp/tmpxft_000069e9_00000000-10_Split.cpp3.i".
Compilation aborted.
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o
CMake Error at nvonnxparser_plugin_generated_Split.cu.o.cmake:266 (message):
  Error generating file
  /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_Split.cu.o


CMakeFiles/nvonnxparser_plugin.dir/build.make:1769: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_Split.cu.o' failed
make[2]: *** [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_Split.cu.o] Error 1
make[2]: *** Waiting for unfinished jobs....
/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a type specifier

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: attributes are not permitted in a function definition

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a "{"

/usr/include/device_functions.hpp(258): error: identifier "predicate" is undefined in device code

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a type specifier

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: attributes are not permitted in a function definition

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: function "__launch_bounds__" has already been defined

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: expected a "{"

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_launcher/triple_chevron_launcher.hpp(63): error: identifier "__val_paramf" is undefined in device code

/usr/include/thrust/system/cuda/detail/bulk/detail/cuda_task.hpp(240): error: a value of type "int" cannot be assigned to an entity of type "_ZN6thrust6system4cuda6detail5bulk_6detail9cuda_taskINS3_14parallel_groupINS3_16concurrent_groupINS3_5agentILm1EEELm0EEELm0EEENS4_7closureINS2_17for_each_n_detail15for_each_kernelENS_5tupleINS4_6cursorILj0EEENS_12zip_iteratorINSF_IPKfPfNS_9null_typeESM_SM_SM_SM_SM_SM_SM_EEEENS_6detail16wrapped_functionINSP_23unary_transform_functorIZN21FancyActivationPlugin9doEnqueueIfEEiiPKPKvPPvSY_P11CUstream_stEUnvdl1_M21FancyActivationPluginFiiPKPKvPPvPvP11CUstream_stEN21FancyActivationPlugin9doEnqueueIfEE1_fEEvEEjSM_SM_SM_SM_SM_SM_EEEEE9grid_typeE"

/usr/include/thrust/system/cuda/detail/bulk/malloc.hpp(462): internal error: assertion failed at: "../../edg/EDG_4.9/src/c_gen_be.c", line 1621


10 errors and 1 catastrophic error detected in the compilation of "/tmp/tmpxft_000069f8_00000000-10_FancyActivation.cpp3.i".
Compilation aborted.
-- Removing /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
/usr/bin/cmake -E remove /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o
CMake Error at nvonnxparser_plugin_generated_FancyActivation.cu.o.cmake:266 (message):
  Error generating file
  /home/afzal/cuda-gpu-setup/onnx-tensorrt/build/CMakeFiles/nvonnxparser_plugin.dir//./nvonnxparser_plugin_generated_FancyActivation.cu.o


CMakeFiles/nvonnxparser_plugin.dir/build.make:886: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o' failed
make[2]: *** [CMakeFiles/nvonnxparser_plugin.dir/nvonnxparser_plugin_generated_FancyActivation.cu.o] Error 1
CMakeFiles/Makefile2:180: recipe for target 'CMakeFiles/nvonnxparser_plugin.dir/all' failed
make[1]: *** [CMakeFiles/nvonnxparser_plugin.dir/all] Error 2
Makefile:149: recipe for target 'all' failed
make: *** [all] Error 2

the Tree structure of my build repository, _half compiled_:

|-- CMakeCache.txt
|-- CMakeFiles
|   |-- 3.5.1
|   |   |-- CMakeCCompiler.cmake
|   |   |-- CMakeCXXCompiler.cmake
|   |   |-- CMakeDetermineCompilerABI_C.bin
|   |   |-- CMakeDetermineCompilerABI_CXX.bin
|   |   |-- CMakeSystem.cmake
|   |   |-- CompilerIdC
|   |   |   |-- a.out
|   |   |   `-- CMakeCCompilerId.c
|   |   `-- CompilerIdCXX
|   |       |-- a.out
|   |       `-- CMakeCXXCompilerId.cpp
|   |-- cmake.check_cache
|   |-- CMakeDirectoryInformation.cmake
|   |-- CMakeError.log
|   |-- CMakeOutput.log
|   |-- CMakeRuleHashes.txt
|   |-- CMakeTmp
|   |-- feature_tests.bin
|   |-- feature_tests.c
|   |-- feature_tests.cxx
|   |-- Makefile2
|   |-- Makefile.cmake
|   |-- nvonnxparser.dir
|   |   |-- build.make
|   |   |-- cmake_clean.cmake
|   |   |-- DependInfo.cmake
|   |   |-- depend.make
|   |   |-- flags.make
|   |   |-- link.txt
|   |   `-- progress.make
|   |-- nvonnxparser_plugin.dir
|   |   |-- build.make
|   |   |-- cmake_clean.cmake
|   |   |-- cmake_clean_target.cmake
|   |   |-- DependInfo.cmake
|   |   |-- depend.make
|   |   |-- flags.make
|   |   |-- link.txt
|   |   |-- nvonnxparser_plugin_generated_FancyActivation.cu.o.cmake
|   |   |-- nvonnxparser_plugin_generated_FancyActivation.cu.o.depend
|   |   |-- nvonnxparser_plugin_generated_ResizeNearest.cu.o.cmake
|   |   |-- nvonnxparser_plugin_generated_ResizeNearest.cu.o.depend
|   |   |-- nvonnxparser_plugin_generated_Split.cu.o.cmake
|   |   |-- nvonnxparser_plugin_generated_Split.cu.o.depend
|   |   `-- progress.make
|   |-- nvonnxparser_runtime.dir
|   |   |-- build.make
|   |   |-- cmake_clean.cmake
|   |   |-- DependInfo.cmake
|   |   |-- depend.make
|   |   |-- flags.make
|   |   |-- link.txt
|   |   `-- progress.make
|   |-- nvonnxparser_runtime_static.dir
|   |   |-- build.make
|   |   |-- cmake_clean.cmake
|   |   |-- cmake_clean_target.cmake
|   |   |-- DependInfo.cmake
|   |   |-- depend.make
|   |   |-- flags.make
|   |   |-- link.txt
|   |   `-- progress.make
|   |-- nvonnxparser_static.dir
|   |   |-- build.make
|   |   |-- cmake_clean.cmake
|   |   |-- cmake_clean_target.cmake
|   |   |-- DependInfo.cmake
|   |   |-- depend.make
|   |   |-- flags.make
|   |   |-- link.txt
|   |   `-- progress.make
|   |-- onnx2trt.dir
|   |   |-- build.make
|   |   |-- cmake_clean.cmake
|   |   |-- DependInfo.cmake
|   |   |-- depend.make
|   |   |-- flags.make
|   |   |-- link.txt
|   |   `-- progress.make
|   |-- progress.marks
|   `-- TargetDirectories.txt
|-- cmake_install.cmake
|-- CPackConfig.cmake
|-- CPackSourceConfig.cmake
|-- Makefile
`-- third_party
    `-- onnx
        |-- CMakeFiles
        |   |-- CMakeDirectoryInformation.cmake
        |   |-- onnx.dir
        |   |   |-- build.make
        |   |   |-- cmake_clean.cmake
        |   |   |-- cmake_clean_target.cmake
        |   |   |-- DependInfo.cmake
        |   |   |-- depend.make
        |   |   |-- flags.make
        |   |   |-- link.txt
        |   |   |-- onnx
        |   |   |   |-- common
        |   |   |   |-- defs
        |   |   |   |   |-- controlflow
        |   |   |   |   |-- experiments
        |   |   |   |   |-- generator
        |   |   |   |   |-- logical
        |   |   |   |   |-- math
        |   |   |   |   |-- nn
        |   |   |   |   |-- reduction
        |   |   |   |   |-- rnn
        |   |   |   |   |-- tensor
        |   |   |   |   `-- traditionalml
        |   |   |   `-- optimizer
        |   |   `-- progress.make
        |   |-- onnx_proto.dir
        |   |   |-- build.make
        |   |   |-- cmake_clean.cmake
        |   |   |-- cmake_clean_target.cmake
        |   |   |-- DependInfo.cmake
        |   |   |-- depend.make
        |   |   |-- flags.make
        |   |   |-- link.txt
        |   |   |-- onnx
        |   |   `-- progress.make
        |   `-- progress.marks
        |-- cmake_install.cmake
        |-- Makefile
        `-- onnx

Any help? will be appreciated.

gr8Adakron picture gr8Adakron on 12 Jul 2018

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2015 NVIDIA Corporation
Built on Tue_Aug_11_14:27:32_CDT_2015
Cuda compilation tools, release 7.5, V7.5.17

You nvcc is too old. Are you sure your CUDA 9.0 is being invoked?

yinghai picture yinghai on 12 Jul 2018
👍2

_we have found the solution for the above error_
make sure you are updated to the latest CUDA and nvcc versions and not only that ensure the right nvcc is invoked.
_to ensure perform the following steps:_

from the terminal

nano /home/username/.bashrc
or
nano /home/$USER/.bashrc

Inside there add the following:

export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}$ 
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

Files we used on the fixing of the following error:

  1. libcudnn7-dev_7.1.4.18-1+cuda9.2_amd64.deb
  2. libcudnn7-doc_7.1.4.18-1cuda9.2_amd64.deb
  3. nv-tensorrt-repo-ubuntu1604-cuda9.2-ga-trt4.0.1.6-20180612_1-1_amd64.deb
  4. cuda-repo-ubuntu1604_9.2.148-1_amd64.deb
  5. libcudnn7_7.1.4.18-1+cuda9.2_amd64.deb
anish9 picture anish9 on 13 Jul 2018
👍2

Fixes #23

anish9 picture anish9 on 13 Jul 2018
Was this page helpful?
0 / 5 - 0 ratings

Related issues

zhanghongruiupup picture zhanghongruiupup  路  8Comments
wangzilu picture wangzilu  路  8Comments
austinmw picture austinmw  路  6Comments
aif2017 picture aif2017  路  3Comments
felixplum picture felixplum  路  7Comments