error encountered while running VASP in GPU

[martin.schlipf]

Now that you use MKL for FFTW does this resolve your issues with the compilation of VASP? Originally, you used the fftw that you got from apt install.

[bhargabkakati]

Dear sir, my issue (the TODO error) still persists while trying to run VASP. It would be really helpful if you shed some light on how to solve the issue. Thank you.

[martin.schlipf]

Can you check ldd /path/to/vasp/executable e.g. ldd bin/vasp_std in the root directory of VASP and check whether FFTW is now linked to something that states intel and mkl like you wrote in your last post. I suspect it still points to libfftw in /usr/local/lib.

[bhargabkakati]

Hello,
I am still stuck with the error. Can you please help?
Thank you.

[martin.schlipf]

Look, you need to provide me with a bit more information about what you tried and what error messages you got. If you just tell me it does not work, I cannot suggest what else you can try.

[bhargabkakati]

Dear sir,
My apologies. I somehow did not receive your last two replies, hence I kept posting about the error not getting fixed. "ldd /path/to/executable" gave me this:

linux-vdso.so.1 (0x00007ffef7db9000)
libqdmod.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/extras/qd/lib/libqdmod.so.0 (0x0000740c2c400000)
libqd.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/extras/qd/lib/libqd.so.0 (0x0000740c2c000000)
liblapack_lp64.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/liblapack_lp64.so.0 (0x0000740c2b200000)
libblas_lp64.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libblas_lp64.so.0 (0x0000740c29200000)
libfftw3.so.3 => /lib/x86_64-linux-gnu/libfftw3.so.3 (0x0000740c28e00000)
libfftw3_omp.so.3 => /lib/x86_64-linux-gnu/libfftw3_omp.so.3 (0x0000740c2c780000)
libmpi_usempif08.so.40 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libmpi_usempif08.so.40 (0x0000740c28a00000)
libmpi_usempi_ignore_tkr.so.40 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libmpi_usempi_ignore_tkr.so.40 (0x0000740c28600000)
libmpi_mpifh.so.40 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libmpi_mpifh.so.40 (0x0000740c28200000)
libmpi.so.40 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libmpi.so.40 (0x0000740c27c00000)
libscalapack_lp64.so.2 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libscalapack_lp64.so.2 (0x0000740c27400000)
libnvhpcwrapcufft.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libnvhpcwrapcufft.so (0x0000740c27000000)
libcufft.so.11 => /usr/local/cuda-12.3/lib64/libcufft.so.11 (0x0000740c1c200000)
libcusolver.so.11 => /usr/local/cuda-12.3/lib64/libcusolver.so.11 (0x0000740c15000000)
libcudaforwrapnccl.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudaforwrapnccl.so (0x0000740c14c00000)
libnccl.so.2 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/nccl/lib/libnccl.so.2 (0x0000740c03c00000)
libcublas.so.12 => /usr/local/cuda-12.3/lib64/libcublas.so.12 (0x0000740bfd400000)
libcublasLt.so.12 => /usr/local/cuda-12.3/lib64/libcublasLt.so.12 (0x0000740bda400000)
libcudaforwrapblas.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudaforwrapblas.so (0x0000740bda000000)
libcudaforwrapblas117.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudaforwrapblas117.so (0x0000740bd9c00000)
libcudart.so.12 => /usr/local/cuda-12.3/lib64/libcudart.so.12 (0x0000740bd9800000)
libcudafor_120.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudafor_120.so (0x0000740bd3800000)
libcudafor.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudafor.so (0x0000740bd3400000)
libacchost.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libacchost.so (0x0000740bd3000000)
libaccdevaux.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libaccdevaux.so (0x0000740bd2c00000)
libacccuda.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libacccuda.so (0x0000740bd2800000)
libcudadevice.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudadevice.so (0x0000740bd2400000)
libcudafor2.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libcudafor2.so (0x0000740bd2000000)
libnvf.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libnvf.so (0x0000740bd1800000)
libnvhpcatm.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libnvhpcatm.so (0x0000740bd1400000)
libstdc++.so.6 => /lib/x86_64-linux-gnu/libstdc++.so.6 (0x0000740bd1000000)
libnvomp.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libnvomp.so (0x0000740bcfe00000)
libnvcpumath.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libnvcpumath.so (0x0000740bcf800000)
libnvc.so => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/compilers/lib/libnvc.so (0x0000740bcf400000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x0000740bcf000000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x0000740c2c756000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x0000740c2c66f000)
libatomic.so.1 => /lib/x86_64-linux-gnu/libatomic.so.1 (0x0000740c2c665000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x0000740c2c65e000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x0000740c2c659000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x0000740c2c654000)
libgomp.so.1 => /lib/x86_64-linux-gnu/libgomp.so.1 (0x0000740c2c3b6000)
libopen-rte.so.40 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libopen-rte.so.40 (0x0000740bcec00000)
libopen-pal.so.40 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libopen-pal.so.40 (0x0000740bce600000)
libucp.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libucp.so.0 (0x0000740bce200000)
libuct.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libuct.so.0 (0x0000740bcde00000)
libucs.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libucs.so.0 (0x0000740bcda00000)
libnuma.so.1 => /lib/x86_64-linux-gnu/libnuma.so.1 (0x0000740c2c643000)
libucm.so.0 => /opt/nvidia/hpc_sdk/Linux_x86_64/24.3/comm_libs/12.3/openmpi4/openmpi-4.1.5/lib/libucm.so.0 (0x0000740bcd600000)
libutil.so.1 => /lib/x86_64-linux-gnu/libutil.so.1 (0x0000740c2c63e000)
libz.so.1 => /lib/x86_64-linux-gnu/libz.so.1 (0x0000740c2c39a000)
/lib64/ld-linux-x86-64.so.2 (0x0000740c2c7a2000)
libnvJitLink.so.12 => /usr/local/cuda-12.3/lib64/libnvJitLink.so.12 (0x0000740bca000000)
libcusparse.so.12 => /usr/local/cuda-12.3/lib64/libcusparse.so.12 (0x0000740bba000000)

[martin.schlipf]

Hello sir,
I did "nvfortran example2.f90 -I /opt/intel/oneapi/mkl/2024.0/include -L /opt/intel/oneapi/mkl/2024.0/include/fftw -lfftw3" with the code you've given and got "a.out" without any error.

libfftw3.so.3 => /lib/x86_64-linux-gnu/libfftw3.so.3 (0x0000740c28e00000)
libfftw3_omp.so.3 => /lib/x86_64-linux-gnu/libfftw3_omp.so.3 (0x0000740c2c780000)

There is something wrong here, you got the example to compile using the MKL fftw but the VASP executable is still linked to the gnu versions. Please check if you updated the makefile.include to use the MKL versions as well. You may need to recompile the code (i.e. `make veryclean`) to make sure that the FFT link to the correct libraries.