[gmx-users] [gmx-developers] GTX 680 non detected on OS X

Uliano Guerrini uliano.guerrini at unimi.it
Tue May 31 14:59:13 CEST 2016


uliano at MacPro:~$ ls /Developer/NVIDIA/CUDA-7.5/lib
libcublas.7.5.dylib         libcusolver_static.a
libcublas.dylib             libcusparse.7.5.dylib
libcublas_device.a          libcusparse.dylib
libcublas_static.a          libcusparse_static.a
libcudadevrt.a              libnppc.7.5.dylib
libcudart.7.5.dylib         libnppc.dylib
libcudart.dylib             libnppc_static.a
libcudart_static.a          libnppi.7.5.dylib
libcufft.7.5.dylib          libnppi.dylib
libcufft.dylib              libnppi_static.a
libcufft_static.a           libnpps.7.5.dylib
libcufftw.7.5.dylib         libnpps.dylib
libcufftw.dylib             libnpps_static.a
libcufftw_static.a          libnvToolsExt.1.dylib
libcuinj.7.5.dylib          libnvToolsExt.dylib
libcuinj.dylib              libnvblas.7.5.dylib
libculibos.a                libnvblas.dylib
libcurand.7.5.dylib         libnvrtc-builtins.7.5.dylib
libcurand.dylib             libnvrtc-builtins.dylib
libcurand_static.a          libnvrtc.7.5.dylib
libcusolver.7.5.dylib       libnvrtc.dylib
libcusolver.dylib           stubs

I compiled also the git current version with the same result


> On 31 May 2016, at 14:44, Mark Abraham <mark.j.abraham at gmail.com> wrote:
> 
> OK, but if all you have is static libraries, then that is all that CMake
> can do, which is why I asked whether you have shared libraries where CMake
> can find them?
> 
> Mark
> 
> On Tue, May 31, 2016 at 2:39 PM Uliano Guerrini <uliano.guerrini at unimi.it>
> wrote:
> 
>> Hi Mark,
>> 
>> the patch was active, in the sense that I pasted the 3 lines if (APPLE)…
>> in between of the corresponding location (not same lines) of the relevant
>> file of gramacs 5.1.2.
>> 
>> I recompiled now without any difference (libcudart always static).
>> 
>> uliano
>> 
>>> On 31 May 2016, at 11:31, Mark Abraham <mark.j.abraham at gmail.com> wrote:
>>> 
>>> Hi,
>>> 
>>> The point of the patch that I linked was to specifically disable static
>>> linking in what sounds like your case. What linking resulted from it? Do
>>> you even have shared libraries installed? What does NVIDIA do/recommend
>> on
>>> Mac?
>>> 
>>> Mark
>>> 
>>> On Tue, 31 May 2016 11:25 Uliano Guerrini <uliano.guerrini at unimi.it>
>> wrote:
>>> 
>>>> 
>>>>> 
>>>>> The output you pasted into that email clearly shows that libcudart is
>>>>> *statically* linked, so it's normal that it does not show up in the
>> otool
>>>>> output (you can list the symbols in your binary with some OS X tool
>>>> similar
>>>>> to nm if you want to verify).
>>>>> 
>>>>> I have *no Idea* of the mechanism that is able to load libcudart
>> without
>>>>>> having it listed by otool -L
>>>>>> 
>>>>> 
>>>>> The mechanism is called binary loader (if you linked statically) ;)
>>>> 
>>>> actually it seems that libcudart *is* statically linked to libgromacs
>>>> 
>>>> configuring with:
>>>> 
>>>> uliano at MacPro:~/local/src/gromacs_build$ cmake ../gromacs-5.1.2
>>>> -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++ -DGMX_GPU=on
>>>> -DCMAKE_PREFIX_PATH="/opt/local;/Developer/NVIDIA/CUDA-7.5/lib"
>>>> -DCUDA_TOOLKIT_ROOT_DIR=/Developer/NVIDIA/CUDA-7.5
>>>> -DCMAKE_INSTALL_PREFIX=/opt/gromacs
>> -DCUDA_HOST_COMPILER=/usr/bin/clang++
>>>> 
>>>> I get:
>>>> 
>>>> uliano at MacPro:~$ nm  /opt/gromacs/lib/libgromacs.1.dylib | grep cudart
>> |
>>>> head
>>>> 000000000090cd70 t __ZN6cudart10configData11addArgumentEPKvmm
>>>> 000000000090cd30 t __ZN6cudart10configData5resetE4dim3S1_mP11CUstream_st
>>>> 000000000090cc90 t __ZN6cudart10configDataC1Ev
>>>> 000000000090ccd0 t __ZN6cudart10configDataC2Ev
>>>> 000000000090cd10 t __ZN6cudart10configDataD1Ev
>>>> 000000000090cd20 t __ZN6cudart10configDataD2Ev
>>>> 000000000092c160 t __ZN6cudart10cuosCallocEmm
>>>> 000000000092a3e0 t __ZN6cudart10cuosGetEnvEPKcPcm
>>>> 000000000092c140 t __ZN6cudart10cuosMallocEm
>>>> 000000000092a3d0 t __ZN6cudart10cuosSetEnvEPKcS1_
>>>> 
>>>> but still:
>>>> 
>>>> NOTE: Error occurred during GPU detection:
>>>>     CUDA driver version is insufficient for CUDA runtime version
>>>>     Can not use GPU acceleration, will fall back to CPU kernels.
>>>> 
>>>> so, maybe this is not a linking problem
>>>> 
>>>> uliano
>>>> 
>>>> --
>>>> 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.
>> 
>> 
>> 
>> --
>> 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