CODEWITHSUNDEEP
c++cudacmakenvcc
KingKraut
KingKraut

Reputation: 1

nvcc fatal : Value 'sm_13' is not defined for option 'gpu-architecture'

I am writing, because I have a problem with the installation of lama and therein the compilation of CUDA. I followed the tutorial on this page http://libama.sourceforge.net/doc/d5/de9/page_linuxinstallation.html and the command "make" gives me the following error:

[Kraut@Wzmbx001 build44]$ make
Scanning dependencies of target log4lama
[ 0%] Building CXX object logging/CMakeFiles/log4lama.dir/GenLogger.cpp.o
[ 1%] Building CXX object logging/CMakeFiles/log4lama.dir/GenLoggerCreator.cpp.o
[ 1%] Building CXX object logging/CMakeFiles/log4lama.dir/Level.cpp.o
[ 2%] Building CXX object logging/CMakeFiles/log4lama.dir/Logger.cpp.o
[ 2%] Building CXX object logging/CMakeFiles/log4lama.dir/LoggerProvider.cpp.o
[ 3%] Building CXX object logging/CMakeFiles/log4lama.dir/SourceLocation.cpp.o
Linking CXX shared library liblog4lama.so
[ 3%] Built target log4lama
[ 3%] Building NVCC (Device) object lama/cuda/CMakeFiles/cuda_compile.dir//./cuda_compile_generated_CUDACOOUtils.cu.o
nvcc fatal : Value 'sm_13' is not defined for option 'gpu-architecture'
CMake Error at cuda_compile_generated_CUDACOOUtils.cu.o.cmake:202 (message):
Error generating
/opt/OpenFOAM/lama/build44/lama/cuda/CMakeFiles/cuda_compile.dir//./cuda_compile_generated_CUDACOOUtils.cu.o


make[2]: *** [lama/cuda/CMakeFiles/cuda_compile.dir/./cuda_compile_generated_CUDACOOUtils.cu.o] Error 1
make[1]: *** [lama/cuda/CMakeFiles/amacuda.dir/all] Error 2
make: *** [all] Error 2

I looked for solutions to this problem and found these three answers, which seem to be closest to my problem:

https://devtalk.nvidia.com/default/topic/762051/jetson-tk1/compile-issues/?offset=4#4786865 How can i tell PyCUDA which GPU to use? http://sourceforge.net/p/viennacl/mailman/message/34316211/

In these threads it is mentioned, that one should change or comment out the "-arch=sm_13"-option in a "Makefile" or a file "~.cmake". However I have no idea where to find this file, where I can modify this option. Can anybody help and guide me to where this file is located?

The problem appears to be quite common, since this option refers to the gpu-architecture supported by CUDA. I have CUDA-7.5 on my system, so from searching for a solution to this issue I understand, that sm_13 is not supported anymore. So I hope changing this will solve this problem - Just need to know where!

Upvotes: 0

Views: 4094

Answers (1)

talonmies
talonmies

Reputation: 72350

It would appear that the supported architectures are defined here in CompilerFlags.cmake :

### choosing the right compute capability
### we just start from version 1.3 ( 1.0 - 1.2 is not supported )
LIST ( APPEND CC_CHOICES "13" "20" "21" "30" "35" )
set ( CUDA_COMPUTE_CAPABILITY "13" CACHE STRING "CUDA compute capability (supported up from 13)" )
    set ( CACHE CUDA_COMPUTE_CAPABILITY PROPERTY STRINGS ${CC_CHOICES} )
checkValue( ${CUDA_COMPUTE_CAPABILITY} "${CC_CHOICES}" )
    mark_as_advanced ( CUDA_COMPUTE_CAPABILITY )

modify those as required

Upvotes: 2

Related Questions