Re: compilation of NAMD 2.12 fails

From: David Hardy (dhardy_at_ks.uiuc.edu)
Date: Fri Apr 28 2017 - 10:15:04 CDT

Dear Vlad,

I would try adding -std=c++11 flag to the compiler option list to see if the compiler will catch the C++ 11 conventions as advertised.

If this doesn't work, is it possible for you to upgrade your GCC installation?

It appears that the use of "auto" typing is pervasive through the code. You might want to verify that changing the first instance is in fact solving your compile issue, i.e.,

in ComputeQM.C line 664, change "auto" to "void *"

and then see if recompiling knocks out the first two errors.

Best regards,
Dave

> On Apr 28, 2017, at 9:14 AM, Vlad Cojocaru <vlad.cojocaru_at_mpi-muenster.mpg.de> wrote:
>
> Dear Joshua, Dear Dave,
>
> Thanks for your suggestions ... Removing "gfortran" from the charm compilation does not change anything... Also, mpicxx points to g++ 4.8.5. I also tried to compile the new 2.13 nightly build code and I got completely different errors, so I guess I will first try to get 2.12 working ...
>
> Best
> Vlad
>
> On 04/27/2017 07:36 PM, David Hardy wrote:
>> Dear Vlad,
>>
>> Have you verified that mpicxx is in fact wrapping gcc 4.8.5? Looks like "mpicxx -showme" will tell you.
>>
>> Compilation appears to be failing on the auto-typed variables, but according to
>> https://gcc.gnu.org/projects/cxx-status.html <https://urldefense.proofpoint.com/v2/url?u=https-3A__gcc.gnu.org_projects_cxx-2Dstatus.html&d=DwMD-g&c=8hUWFZcy2Z-Za5rBPlktOQ&r=cCnbnDbBv_O0m2oP3VzZwpeeKpMkzLjk2HDqfuC9GM4&m=MAnLjZOTrJLBX7kgYTl7DVKKTdzOHV5D7AC4uEk_X_Q&s=PCDjVvLzXPZMq78kJu8GgNps7J9Qn0258WljU2i1O0A&e=>
>> it should be supported for gcc 4.4 and above.
>>
>> You would reduce some warnings by explicitly setting -std=c++11 compiler option.
>>
>> Best regards,
>> Dave
>>
>>
>>> On Apr 27, 2017, at 11:37 AM, Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov <mailto:Joshua.Vermaas_at_nrel.gov>> wrote:
>>>
>>> Hi Vlad,
>>>
>>> The first thing I would try is to recompile charm 6.7.1 without
>>> specifying a fortran compiler. This was a few years ago, but I remember
>>> that causing some issues that were similarly hard to diagnose when
>>> putting NAMD together, which has no fortran components.
>>>
>>> -Josh
>>>
>>> On 04/27/2017 09:40 AM, Vlad Cojocaru wrote:
>>>> Dear all,
>>>>
>>>> Compiling NAMD 2.12 on a linux x86_64 linux platform with gcc 4.8.5,
>>>> openmpi 2.1.0, charm 6.7.1, fftw3 fails with the attached error message.
>>>> The problematic source file appears to be ComputeQM.C ....
>>>>
>>>> Any advices on how to fix this would be highly appreciated.
>>>>
>>>> Best wishes
>>>> Vlad
>>>>
>>>
>>
>
> --
> Vlad Cojocaru, Ph.D., Project Group Leader
> Department of Cell and Developmental Biology
> Max Planck Institute for Molecular Biomedicine
> Röntgenstrasse 20, 48149 Münster, Germany
> Tel: +49-251-70365-324; Fax: +49-251-70365-399
> Email: vlad.cojocaru[at]mpi-muenster.mpg.de
> http://www.mpi-muenster.mpg.de/43241/cojocaru <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mpi-2Dmuenster.mpg.de_43241_cojocaru&d=DwMD-g&c=8hUWFZcy2Z-Za5rBPlktOQ&r=cCnbnDbBv_O0m2oP3VzZwpeeKpMkzLjk2HDqfuC9GM4&m=MAnLjZOTrJLBX7kgYTl7DVKKTdzOHV5D7AC4uEk_X_Q&s=J1WnjhuF6iTTtoKMRQlPm62agPLtq1Pt1YxyGCRy3OQ&e=>

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:21:14 CST