[gmx-users] GPU problem with running gromacs.2018

Szilárd Páll pall.szilard at gmail.com
Fri Feb 16 20:06:10 CET 2018 

Hi,

Indeed, this is an issue introduced in the 2018 release which aborts
mdrun if an error is encountered during GPU detection; My first guess
is that one of your GPUs is in exclusive or prohibited mode. You can
use mdrun by skipping that GPU, i.e. setting the CUDA_VISIBLE_DEVICES
environment variable.

BTW, what you shared is  not your full log! It would have been (and
will be in the future) helpful if you actually shared the _full_ log
file so we get all information not just what you find useful to paste
in your reply; in this particular case seeing the hardware detection
report would have helped a bit.

Cheers,
--
Szilárd


On Fri, Feb 16, 2018 at 1:55 PM, Osmany Guirola Cruz
<osmany.guirola at biocomp.cigb.edu.cu> wrote:
> This is the output with gromacs 2018 and cuda 9
> -------------------------------------------------------Program:     gmx
> mdrun, version 2018Source file: src/gromacs/gpu_utils/gpu_utils.cu
> (line 735)Function:    void findGpus(gmx_gpu_info_t*)
> Assertion failed:Condition: cudaSuccess == cudaPeekAtLastError()Should
> be cudaSuccess
> For more information and tips for troubleshooting, please check the
> GROMACSwebsite at http://www.gromacs.org/Documentation/Errors----------
> -----------------------------------------------------
>
> and this is the log
> Log file opened on Fri Feb 16 07:49:19 2018Host: draco  pid:
> 20855  rank ID: 0  number of ranks:  1                       :-)
> GROMACS - gmx mdrun, 2018 (-:
>                             GROMACS is written by:     Emile
> Apol      Rossen Apostolov  Herman J.C. Berendsen    Par
> Bjelkmar    Aldert van Buuren   Rudi van Drunen     Anton
> Feenstra    Gerrit Groenhof   Christoph Junghans   Anca
> Hamuraru    Vincent Hindriksen Dimitrios Karkoulis    Peter
> Kasson        Jiri Kraus      Carsten Kutzner      Per Larsson
> Justin A. Lemkul    Viveca Lindahl    Magnus Lundborg   Pieter
> Meulenhoff    Erik Marklund      Teemu Murtola       Szilard
> Pall       Sander Pronk      Roland Schulz     Alexey
> Shvetsov     Michael Shirts     Alfons Sijbers     Peter
> Tieleman    Teemu Virolainen  Christian Wennberg    Maarten
> Wolf                              and the project leaders:        Mark
> Abraham, Berk Hess, Erik Lindahl, and David van der Spoel
> Copyright (c) 1991-2000, University of Groningen, The
> Netherlands.Copyright (c) 2001-2017, The GROMACS development team
> atUppsala University, Stockholm University andthe Royal Institute of
> Technology, Sweden.check out http://www.gromacs.org for more
> information.
> GROMACS is free software; you can redistribute it and/or modify itunder
> the terms of the GNU Lesser General Public Licenseas published by the
> Free Software Foundation; either version 2.1of the License, or (at your
> option) any later version.
> GROMACS:      gmx mdrun, version
> 2018Executable:   /usr/local/gromacs/bin/gmxData
> prefix:  /usr/local/gromacsWorking
> dir:  /home/osmany/Downloads/RASCommand line:  gmx mdrun -deffnm md
> GROMACS version:    2018Precision:          singleMemory
> model:       64 bitMPI library:        thread_mpiOpenMP
> support:     enabled (GMX_OPENMP_MAX_THREADS = 64)GPU
> support:        CUDASIMD instructions:  SSE4.1FFT library:        fftw-
> 3.3.7-sse2-avxRDTSCP usage:       disabledTNG
> support:        enabledHwloc support:      hwloc-1.11.6Tracing
> support:    disabledBuilt on:           2018-02-14 14:05:11Built
> by:           osmany at draco [CMAKE]Build OS/arch:      Linux 4.14.0-3-
> amd64 x86_64Build CPU vendor:   IntelBuild CPU brand:    Intel(R)
> Core(TM)2 Quad CPU    Q9650  @ 3.00GHzBuild CPU family:   6   Model:
> 23   Stepping: 10Build CPU features: apic clfsh cmov cx8 cx16 intel
> lahf mmx msr pdcm pse sse2 sse3 sse4.1 ssse3C
> compiler:         /usr/bin/gcc-6 GNU 6.4.0C compiler flags:    -
> msse4.1     -O3 -DNDEBUG -funroll-all-loops -fexcess-
> precision=fast  C++ compiler:       /usr/bin/g++-6 GNU 6.4.0C++
> compiler flags:  -msse4.1    -std=c++11   -O3 -DNDEBUG -funroll-all-
> loops -fexcess-precision=fast  CUDA compiler:      /usr/lib/nvidia-
> cuda-toolkit/bin/nvcc nvcc: NVIDIA (R) Cuda compiler driver;Copyright
> (c) 2005-2017 NVIDIA Corporation;Built on
> Fri_Sep__1_21:08:03_CDT_2017;Cuda compilation tools, release 9.0,
> V9.0.176CUDA compiler flags:-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_70,code=sm_70;-
> gencode;arch=compute_70,code=compute_70;-use_fast_math;-
> D_FORCE_INLINES;; ;-msse4.1;-std=c++11;-O3;-DNDEBUG;-funroll-all-
> loops;-fexcess-precision=fast;CUDA driver:        9.0CUDA
> runtime:       9.0
> ----------------------------------------------------------------------
>
>
> Osmany
>
>
>
>
>
> -----Original Message-----From: Szilárd Páll <pall.szilard at gmail.com>Re
> ply-to: gmx-users at gromacs.orgTo: Discussion list for GROMACS users <gmx
> -users at gromacs.org>Subject: Re: [gmx-users] GPU problem with running
> gromacs.2018Date: Thu, 15 Feb 2018 18:23:05 +0100
> Please provide a full log file output.--Szilárd
>
> On Thu, Feb 15, 2018 at 6:11 PM, Osmany Guirola Cruz<osmany.guirola at bio
> comp.cigb.edu.cu> wrote:
>> Hi
>>
>> I am having problems running mdrun command compiled with GPU
>> support(cuda 9.0).
>> here is  the output of the mdrun command....
>>
>> ........
>> Using 1 MPI thread
>> Using 4 OpenMP threads
>>
>> 1 compatible GPU is present, with ID 0
>> 1 GPU auto-selected for this run.
>> Mapping of GPU ID to the 1 PP rank in this node: 0
>>
>>
>> WARNING: Just caught a previously occurred CUDA error (no kernel
>> image
>> is available for execution on the device), will try to continue.
>>
>>
>> -------------------------------------------------------
>> Program:     gmx mdrun, version 2016.4
>> Source file: src/gromacs/mdlib/nbnxn_cuda/nbnxn_cuda.cu (line 740)
>>
>> Fatal error:
>> cudaFuncSetCacheConfig failed: invalid device function
>>
>> For more information and tips for troubleshooting, please check the
>> GROMACS
>> website at http://www.gromacs.org/Documentation/Errors
>> ...............
>>
>>
>> Thanks.
>> Osmany
>>
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at http://www.gromacs.org/Support/Mailing
>> _Lists/GMX-Users_List before posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.
> --
> Gromacs Users mailing list
>
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.

More information about the gromacs.org_gmx-users mailing list