[gmx-users] No CUDA-capable device is detected

Jiri Kraus jkraus at nvidia.com
Fri Feb 10 16:16:55 CET 2017

Hi Li,

What does nvidia-smi return on the machine where you try to run GROMACS? Also is CUDA_VISIBLE_DEVICES set? I.e. what does 



Even if there are GPUs in a system they could be masked out by setting CUDA_VISIBLE_DEVICES to an empty string. Sometimes batch system do this if jobs that did not request GPU resources are executed compute nodes with GPUs. The latter might be also enforced using cgroups. 

Hope this helps


> Message: 4
> Date: Fri, 10 Feb 2017 09:57:19 -0500
> From: "Li, Shi" <sli259 at g.uky.edu>
> To: gromacs.org_gmx-users at maillist.sys.kth.se
> Subject: [gmx-users] No CUDA-capable device is detected
> Message-ID:
> 	<CAC+hMcDBrYOJk0UM57TvLoMp4grT0G0aaYhSgfDhgGK4=7QtQA@
> mail.gmail.com>
> Content-Type: text/plain; charset=UTF-8
> Dear GMX users,
> I have a problem with the Gromacs jobs run on GPU node. I have the output
> in the log file attached below. Anyone know if the CUDA and GROMACS are
> compiled correctly according to this output?
> Is so, why does the job cannot run on GPU? Thank you very much!
>   gmx_mpi mdrun -deffnm em
> GROMACS version:    2016.1
> Precision:          single
> Memory model:       64 bit
> MPI library:        MPI
> OpenMP support:     enabled (GMX_OPENMP_MAX_THREADS = 32)
> GPU support:        CUDA
> SIMD instructions:  AVX_256
> FFT library:        fftw-3.3.5-sse2-avx-avx_128_fma
> RDTSCP usage:       enabled
> TNG support:        enabled
> Hwloc support:      disabled
> Tracing support:    disabled
> Built on:           Tue Jan 17 15:14:56 EST 2017
> Built by:           sli259 at compute-gpu-0-2.local [CMAKE]
> Build OS/arch:      Linux 2.6.32-642.11.1.el6.x86_64 x86_64
> Build CPU vendor:   Intel
> Build CPU brand:    Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz
> Build CPU family:   6   Model: 63   Stepping: 2
> Build CPU features: aes apic avx avx2 clfsh cmov cx8 cx16 f16c fma htt lahf
> mmx msr nonstop_tsc pcid pclmuldq pdcm pdpe1gb popcnt pse rdrnd rdtscp
> sse2
> sse3 sse4.1 sse4.2 ssse3 tdt x2apic
> C compiler:         /home/sli259/program/gromacs/openmpi/bin/mpicc GNU
> 4.9.3
> C compiler flags:    -mavx     -O3 -DNDEBUG -funroll-all-loops
> -fexcess-precision=fast
> C++ compiler:       /home/sli259/program/gromacs/openmpi/bin/mpic++
> 4.9.3
> C++ compiler flags:  -mavx    -std=c++0x   -O3 -DNDEBUG -funroll-all-loops
> -fexcess-precision=fast
> CUDA compiler:      /home/sli259/program/gromacs/cuda/bin/nvcc nvcc:
> (R) Cuda compiler driver;Copyright (c) 2005-2016 NVIDIA Corporation;Built on
> Sun_Sep__4_22:14:01_CDT_2016;Cuda compilation tools, release 8.0, V8.0.44
> CUDA compiler flags:-gencode;arch=compute_20,code=sm_20;-
> gencode;arch=compute_30,code=sm_30;-
> gencode;arch=compute_35,code=sm_35;-
> gencode;arch=compute_37,code=sm_37;-
> gencode;arch=compute_50,code=sm_50;-
> gencode;arch=compute_52,code=sm_52;-
> gencode;arch=compute_60,code=sm_60;-
> gencode;arch=compute_61,code=sm_61;-
> gencode;arch=compute_60,code=compute_60;-
> gencode;arch=compute_61,code=compute_61;-use_fast_math;;;-
> Xcompiler;,-mavx,,,,,,;-Xcompiler;-O3,-DNDEBUG,-funroll-all-loops,-fexcess-
> precision=fast,,;
> CUDA driver:        8.0
> CUDA runtime:       0.0
> NOTE: Error occurred during GPU detection:
>       no CUDA-capable device is detected
>       Can not use GPU acceleration, will fall back to CPU kernels.
> Running on 1 node with total 12 cores, 24 logical cores, 0 compatible GPUs
> Hardware detected on host compute-gpu-0-2.local (the node of MPI rank 0):

NVIDIA GmbH, Wuerselen, Germany, Amtsgericht Aachen, HRB 8361
Managing Director: Karen Theresa Burns

This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.

More information about the gromacs.org_gmx-users mailing list