[mvapich-discuss] compile warnings

[Jonathan Perkins]

Thanks for the note Sandeep.  This warning is probably harmless.
However this is an older version of MVAPICH2 which is no longer
maintained.  I suggest upgrading to 1.9 or the latest 2.0 candidate
(2.0b) when you get a chance so that we can provide help if you
encounter any problems.

On Mon, Feb 3, 2014 at 11:38 AM, Sarangi, Sandeep
<sandeep_sarangi at unc.edu> wrote:
> Hello,
>
> I get these warnings when compiling some hello word MPI code using mvapich2-1.7 (with Intel):
>
> ld: Warning: alignment 16 of symbol `mpifcmb4_' in /usr/mpi/intel/mvapich2-1.7/lib/libmpich.a(setbot.o) is smaller than 32 in /tmp/ifort8tVxj2.o
> ld: Warning: alignment 16 of symbol `mpifcmb6_' in /usr/mpi/intel/mvapich2-1.7/lib/libmpich.a(setbot.o) is smaller than 32 in /tmp/ifort8tVxj2.o
> ld: Warning: alignment 16 of symbol `mpifcmb8_' in /usr/mpi/intel/mvapich2-1.7/lib/libmpich.a(setbot.o) is smaller than 32 in /tmp/ifort8tVxj2.o
>
> It appears that others have reported similar warnings:
>
> http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/2011-August/003509.html
>
> I just want to verify- are the above warnings trivial and can they be safely ignored? I'm using a central build on my University's cluster and would prefer not to ask the Sys Admin to rebuild these libraries (using the recommendation in the above email thread).
>
> Thanks,
> Sandeep
>
> _______________________________________________
> mvapich-discuss mailing list
> mvapich-discuss at cse.ohio-state.edu
> http://mailman.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
>
>



-- 
Jonathan Perkins
http://www.cse.ohio-state.edu/~perkinjo