I'm trying to compile cp2k version 7.1 using INTEL MKL, without CUDA First I entered the tools/toolchain folder and executed: ./install_cp2k_toolchain.sh --with-mkl=system --with-elpa=no --enable-cuda=no which terminates fine, the final part of the message is: Done! Now copy: cp /home/bruno.focassio/codes/cp2k-7.1/tools/toolchain/install/arch/* to the cp2k/arch/ directory To use the installed tools and libraries and cp2k version compiled with it you will first need to execute at the prompt: source /home/bruno.focassio/codes/cp2k-7.1/tools/toolchain/install/setup To build CP2K you should change directory: cd cp2k/ make -j 32 ARCH=local VERSION="sopt sdbg ssmp popt pdbg psmp" arch files for GPU enabled CUDA versions are named "local_cuda.*" arch files for valgrind versions are named "local_valgrind.*" arch files for coverage versions are named "local_coverage.*" Note that these pre-built arch files are for the GNU compiler, users have to adapt them for other compilers. It is possible to use the provided CP2K arch files as guidance. I copied the files as instructed and when executing the make command: make ARCH=local VERSION="sopt ssmp popt psmp" It returns an error in the following step: nvcc -c -I'/home/bruno.focassio/codes/cp2k-7.1/exts/dbcsr/src' /home/bruno.focassio/codes/cp2k-7.1/exts/dbcsr/src/acc/cuda/dbcsr_cuda_nvtx_cu.cu What am I missing ? Thanks in advance