I’m trying to write some CUDA code, so I installed
$ sudo apt install nvidia-cuda-toolkit
I tried compiling the tutorial https://devblogs.nvidia.com/even-easier-introduction-cuda/, but got this:
$ nvcc add.cu -o add_cuda
ERROR: No supported gcc/g++ host compiler found, but clang-4.0 is available.
Use 'nvcc -ccbin clang-4.0' to use that instead.
With -v:
$ nvcc add.cu -o add_cuda -v
#$ _SPACE_=
#$ _CUDART_=cudart
#$ _HERE_=/usr/lib/nvidia-cuda-toolkit/bin
#$ _THERE_=/usr/lib/nvidia-cuda-toolkit/bin
#$ _TARGET_SIZE_=
#$ _TARGET_DIR_=
#$ _TARGET_SIZE_=64
#$ NVVMIR_LIBRARY_DIR=/usr/lib/nvidia-cuda-toolkit/libdevice
#$ PATH=/usr/lib/nvidia-cuda-toolkit/bin:/home/matt/.nvm/versions/node/v10.2.1/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin
#$ LIBRARIES= -L/usr/lib/x86_64-linux-gnu/stubs
#$ gcc -std=c++14 -D__CUDA_ARCH__=300 -E -x c++ -DCUDA_DOUBLE_MATH_FUNCTIONS -D__CUDACC__ -D__NVCC__ -D"__CUDACC_VER_BUILD__=85" -D"__CUDACC_VER_MINOR__=1" -D"__CUDACC_VER_MAJOR__=9" -include "cuda_runtime.h" -m64 "add.cu" > "/tmp/tmpxft_00001642_00000000-8_add.cpp1.ii"
ERROR: No supported gcc/g++ host compiler found, but clang-4.0 is available.
Use 'nvcc -ccbin clang-4.0' to use that instead.
# --error 0x1 --
Now, according to https://docs.nvidia.com/cuda/cuda-installation-guide-linux/#system-requirements, I need at least gcc 7.2.0 or clang 5.0. I seem to meet this:
$ gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/7/lto-wrapper
OFFLOAD_TARGET_NAMES=nvptx-none
OFFLOAD_TARGET_DEFAULT=1
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu 7.3.0-16ubuntu3' --with-bugurl=file:///usr/share/doc/gcc-7/README.Bugs --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --prefix=/usr --with-gcc-major-version-only --with-as=/usr/bin/x86_64-linux-gnu-as --with-ld=/usr/bin/x86_64-linux-gnu-ld --program-suffix=-7 --program-prefix=x86_64-linux-gnu- --enable-shared --enable-linker-build-id --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --libdir=/usr/lib --enable-nls --with-sysroot=/ --enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes --with-default-libstdcxx-abi=new --enable-gnu-unique-object --disable-vtable-verify --enable-libmpx --enable-plugin --enable-default-pie --with-system-zlib --with-target-system-zlib --enable-objc-gc=auto --enable-multiarch --disable-werror --with-arch-32=i686 --with-abi=m64 --with-multilib-list=m32,m64,mx32 --enable-multilib --with-tune=generic --enable-offload-targets=nvptx-none --without-cuda-driver --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu
Thread model: posix
gcc version 7.3.0 (Ubuntu 7.3.0-16ubuntu3)
Moreover, using clang caused a linker error:
$ nvcc add.cu -o add_cuda -ccbin clang-4.0
/tmp/tmpxft_000046a5_00000000-10_add.o: In function `main':
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text+0x19b): undefined reference to `std::cout'
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text+0x1ae): undefined reference to `std::basic_ostream<char, std::char_traits<char> >& std::operator<< <std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&, char const*)'
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text+0x1bb): undefined reference to `std::ostream::operator<<(float)'
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text+0x1c1): undefined reference to `std::basic_ostream<char, std::char_traits<char> >& std::endl<char, std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&)'
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text+0x1cd): undefined reference to `std::ostream::operator<<(std::ostream& (*)(std::ostream&))'
/tmp/tmpxft_000046a5_00000000-10_add.o: In function `__cxx_global_var_init':
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text.startup+0x13): undefined reference to `std::ios_base::Init::Init()'
/tmp/tmpxft_000046a5_00000000-5_add.cudafe1.cpp:(.text.startup+0x19): undefined reference to `std::ios_base::Init::~Init()'
clang: error: linker command failed with exit code 1 (use -v to see invocation)
More info:
$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Fri_Nov__3_21:07:56_CDT_2017
Cuda compilation tools, release 9.1, V9.1.85
Is gcc 7.3.0 bugged? Why is nvcc looking for the wrong clang?