[mvapich-discuss] mvapich2-x for stock rhel6

Jeff Hammond jeff.science at gmail.com
Mon Feb 24 17:57:46 EST 2014


You probably want to write devel at mpich.org about the ADIO hook issue
in hopes of getting feedback from the devs that wr[io]te that code.

It would be nice if MVAPICH2-X was open-source but clearly there is
some motivation to not do this since it hasn't been done yet.

Best,

Jeff

On Mon, Feb 24, 2014 at 3:33 PM, Martin Pokorny <mpokorny at nrao.edu> wrote:
> On 02/21/2014 06:09 PM, Jonathan Perkins wrote:
>>
>> Thanks for your note Martin.  I was able to reproduce this problem and
>> created a new package that should resolve it.  Can you try the
>> following and let us know if it works?
>>
>>
>> http://mvapich.cse.ohio-state.edu/download/mvapich2x/mvapich2-x-2.0b-1.rhel6.stock.tar.gz
>
>
> The new package works for me. Thanks, Jonathan.
>
>> On Fri, Feb 21, 2014 at 6:27 PM, Martin Pokorny<mpokorny at nrao.edu>  wrote:
>>>
>>> Also, is there any chance that a source package for mvapich2-x will be
>>> available soon?
>
>
> As for this question about source availability...I'm working on an ADIO
> implementation for a file system, which uses both MPI and OpenSHMEM APIs.
> AFAICT, there aren't enough hooks in ADIO to allow me to insert a new ADIO
> "module" implementation without rebuilding the MPI library. Do you have any
> suggestions for how I might proceed in these circumstances?
>
>
> --
> Martin Pokorny
> Software Engineer - New Mexico Systems Group lead
> National Radio Astronomy Observatory - New Mexico Operations
>
> _______________________________________________
> mvapich-discuss mailing list
> mvapich-discuss at cse.ohio-state.edu
> http://mailman.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss



-- 
Jeff Hammond
jeff.science at gmail.com



More information about the mvapich-discuss mailing list