Problem Compiling CP2K

[Darkasleif]

I am trying to install CP2K with CUDA (12.3) on my windows 11 machine.
I followed the instructions on https://www.cp2k.org/howto:compile_on_windows for a parallel CP2K binary including building the toolchain
./install_cp2k_toolchain.sh --install-all --with-gcc=system --with-mpich=system --with-sirius=no
After that I did as explained at https://www.cp2k.org/howto:compile_with_cuda (adding the 3 lines to the ARCH file)
When I did this ld could not locate lcudar, lcublas, lcufft, and lnvrtc.
I then added:
sudo ln -s /usr/local/cuda/lib64/libcudart.so /usr/lib/libcudart.so
and similarly for the other 3 libraries. That took care of that problem but then I received a long list of issues that I do not understand, see attached log file.

Could you please help me proceed further. I am new at installing those big codes and understanding and fixing these problems are very new to me
Thank you
=cp2k.log

[fstein93]

Sorry for the late reply. Your log file is incomplete. There is no error message given.
Why do you want to use CP2K on a Windows machine with CUDA (presumably a Desktop/Notebook with a consumer GPU)? We do not recommend it because CP2K relies mostly on double-precision operations which are not much (if at all) faster on consumer GPUs than on CPUs. Here, it is better to stay with the latter or switch to a cluster with more performant GPUs (if available).