[mvapich-discuss] Compiling Mvapich2 with PGI Compilers

[Thompson, Matt (GSFC-610.1)[SCIENCE SYSTEMS AND APPLICATIONS INC]]

Mark, et al,

I just tried a plain boring ./configure using PGI 14.6 on an RHEL 6.5 
box here (so roughly equivalent to CentOS 6.4) and it worked here.

Well, I had to pass --with-device=ch3:nemesis (just a desktop machine, 
no Infiniband) and undefine F90 in my environment, but it worked.

Looking at the config file, one thing you might want to see is if CPP is 
defined in your environment. In my PGI modules, I always specifically do 
not set CPP because it seems to often cause issues. For example, from 
your config.log:

> configure:6160: checking how to run the C preprocessor
> configure:6230: result: /app/navapps/pgi/14/linux86-64/14.6/bin/pgprepro
> configure:6250: /app/navapps/pgi/14/linux86-64/14.6/bin/pgprepro   conftest.c
> configure:6250: $? = 0

It's using pgprepro. On my configure, it detected pgcc -E. You might 
also try CPP=cpp as well.

If not that...maybe LD_LIBRARY_PATH isn't pointing to both PGI's lib and 
libso directory?

Matt


On 09/04/2014 03:25 PM, Mr. Mark L. Dotson (Contractor) wrote:
> I tried using 2.0 and received the same error. Config.log is attached.
>
>
>
> Mark
>
>
> On 09/04/14 15:12, Jonathan Perkins wrote:
>> Thanks for providing the config.log.  I inspected it but its not clear
>> to me why its failing.  It seems that my initial guess is not in play
>> here.  Can you try mvapich2-2.0 and see if it fails configure as well?
>>
>> http://mvapich.cse.ohio-state.edu/download/mvapich/mv2/mvapich2-2.0.tar.gz
>>
>>
>> If it does fail, please also send that config.log.
>>
>> On Thu, Sep 04, 2014 at 02:44:02PM -0400, Mr. Mark L. Dotson
>> (Contractor) wrote:
>>> The config.log is attached.
>>>
>>> Thanks,
>>>
>>> Mark
>>>
>>>
>>> On 09/04/14 14:36, Jonathan Perkins wrote:
>>>> Hi Mark.  Can you send us your config.log?
>>>>
>>>> I think there is some library that cannot be loaded at runtime but is
>>>> only manifesting itself when attempting the datatype checks.
>>>>
>>>> On Thu, Sep 04, 2014 at 12:41:16PM -0400, Mr. Mark L. Dotson
>>>> (Contractor) wrote:
>>>>> I am trying to compile mvapich2-2.0b on CentOS 6.4 with PGI compilers
>>>>> 14.6 with a simple ./configure. These are the pertinent messages
>>>>> for the
>>>>> errors I am receiving:
>>>>>
>>>>> checking for size of Fortran type integer... configure: WARNING:
>>>>> Unable
>>>>> to compile the C routine for finding the size of a integer
>>>>>
>>>>> checking for size of Fortran type real... configure: WARNING:
>>>>> Unable to
>>>>> compile the C routine for finding the size of a real
>>>>>
>>>>> checking for size of Fortran type double precision... configure:
>>>>> WARNING:
>>>>> Unable to compile the C routine for finding the size of a double
>>>>> precision
>>>>>
>>>>> checking whether integer*1 is supported... yes
>>>>> checking whether integer*2 is supported... yes
>>>>> checking whether integer*4 is supported... yes
>>>>> checking whether integer*8 is supported... yes
>>>>> checking whether integer*16 is supported... yes
>>>>> checking whether real*4 is supported... yes
>>>>> checking whether real*8 is supported... yes
>>>>> checking whether real*16 is supported... yes
>>>>> configure: error: Unable to configure with Fortran support because
>>>>> configure could not determine the size of a Fortran INTEGER. Consider
>>>>> setting CROSS_F77_SIZEOF_INTEGER to the length in bytes of a Fortran
>>>>> INTEGER
>>>>>
>>>>> I've tried setting the suggested variable to 4 and 8, to no avail.
>>>>> Would
>>>>> anyone know what I can do to fix this?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Mark
>>>>>
>>>>> --
>>>>> Mark Dotson
>>>>> Systems Administrator
>>>>> Lockheed-Martin
>>>>> dotsonml.ctr at afrl.hpc.mil
>>>>> _______________________________________________
>>>>> mvapich-discuss mailing list
>>>>> mvapich-discuss at cse.ohio-state.edu
>>>>> http://mailman.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
>>>>
>>>
>>>
>>> --
>>> Mark Dotson
>>> Systems Administrator
>>> Lockheed-Martin
>>> dotsonml.ctr at afrl.hpc.mil
>>
>
>
>
>
> _______________________________________________
> mvapich-discuss mailing list
> mvapich-discuss at cse.ohio-state.edu
> http://mailman.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
>


-- 
Matt Thompson          SSAI, Sr Software Test Engr
NASA GSFC, Global Modeling and Assimilation Office
Code 610.1, 8800 Greenbelt Rd, Greenbelt, MD 20771
Phone: 301-614-6712              Fax: 301-614-6246