[mvapich-discuss] mvapich2-1.6rc1 problem (fwd)

[Jonathan Perkins]

Thanks for the investigation Bryan.  There may be something funky
going on during the configure or make steps.  If you build in the
following manner...

./configure 2>&1 | tee configure.log
make 2>&1 | tee make.log
make install 2>&1 | tee make-install.log

You can send these log files as well as config.log.  Please send this
directly to me with a cc to mvapich-core at cse.ohio-state.edu.

After we determine the cause of your issue we'll post the conclusions
back onto this list.

On Tue, Dec 7, 2010 at 12:32 PM, Bryan Putnam <bfp at purdue.edu> wrote:
> On Mon, 6 Dec 2010, Bryan Putnam wrote:
>
>> On Mon, 6 Dec 2010, Jonathan Perkins wrote:
>>
>> > Can you give the output of `mpiname -a' with the mvapich2-1.5 version
>> > as well?  I'm wondering about the --with-pm selection.  It seems that
>> > you should not be setting this to remshell if you're trying to use
>> > mpirun_rsh.
>>
>> Turns out that removing the "--with-pm..." did not fix the problem. If I
>> build mvapich2-1.6rc1 with PGI compilers, everything appears OK. However
>> if I build with Intel or gcc compilers, C programs will not run, although
>> Fortran and C++ programs seem to be running OK.
>>
>> Note that I don't see this problem with mvapich2-1.5, and have not seen
>> the problem with any versions of mpich2 up to the current mpich2-1.3.1.
>
> I just now rebuilt mvapich2-1.5 as well as mvapich2-1.5.1 which I hadn't
> looked at closely before. A C code built with mvapich2-1.5 works fine, but
> I'm seeing the same problems with mvapich2-1.5.1 that I see with
> mvapich2-1.6rc1. So, it appears that this problem may have started with
> mvapich-1.5.1.
>
> Thanks,
> Bryan
>
>



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