Fabrice --
I have committed a fix to our development master; it is pending moving over to
the v1.10 and v2.x release branches (see
https://github.com/open-mpi/ompi-release/pull/610 and
https://github.com/open-mpi/ompi-release/pull/611, respectively). Once the fix
is in the release branches, it will show up in nightly Open MPI snapshot
tarballs:
http://www.open-mpi.org/nightly/v1.10/
http://www.open-mpi.org/nightly/v2.x/
Check out
https://github.com/open-mpi/ompi/commit/3698621df0bffd3711be2f2d81afd8e3ae2947e6
if you care about the details of the fix, and/or want to just apply that fix
directly to your v1.10 tree (be warned, however, if you apply that patch,
you'll need to run the GNU Autotools, which can be a little tricky if you
haven't previously done the setup described in the HACKING file -- it may be
simpler to wait for the fix to show up in the nightly tarballs, which you can
build exactly like you build the real official release tarballs).
Thanks for reporting the issue.
> On Sep 24, 2015, at 4:55 PM, Jeff Squyres (jsquyres) <jsquy...@cisco.com>
> wrote:
>
> Intel apparently changed something in their 2016 compiler (compared to the
> 2015 compiler); the Open MPI configure script decided to use a different
> pragma. Per the issue I opened up on Github, I need to look at the configure
> script and see what's going wrong.
>
>
>> On Sep 24, 2015, at 4:51 PM, Fabrice Roy <fabrice....@obspm.fr> wrote:
>>
>> Hi,
>>
>> I have made some other tests. I don't know if it can help you but here is
>> what I observed.
>>
>> Using the array contructor [] solves the problem for a scalar, as someone
>> wrote on the Intel forum.
>> The same code with tok declared as an integer and
>> call mpi_bcast([tok],1,mpi_integer,0,mpi_comm_world,ierr)
>> works fine.
>>
>> But I get the same compilation error (no matching specific subroutine) if
>> tok is a 2d array:
>> integer, dimension(2:2) :: tok
>> call mpi_bcast(tok,1,mpi_integer,0,mpi_comm_world,ierr)
>> does not compile.
>>
>> In this case, I can also solve the problem with the array constructor but I
>> don't understand why I have to use this.
>> And if I try to send only a part of my 2d array, it doesn't work.
>> call mpi_bcast([tok(1,2)],1,mpi_integer,0,mpi_comm_world,ierr)
>> compiles but I don't get the good result.
>>
>> Thanks for your help,
>>
>> Fabrice
>>
>>
>> Le 24/09/2015 16:32, Jeff Squyres (jsquyres) a écrit :
>>> Yes -- typo -- it's not a problem with mpi_f08, it's a problem with the mpi
>>> module using the "ignore TKR" implementation.
>>>
>>> See https://github.com/open-mpi/ompi/issues/937.
>>>
>>>
>>>> On Sep 24, 2015, at 4:30 PM, Gilles Gouaillardet
>>>> <gilles.gouaillar...@gmail.com> wrote:
>>>>
>>>> Jeff,
>>>>
>>>> I am not sure whether you made a typo or not ...
>>>>
>>>> the issue only occuex with f90 bindings (aka use mpi)
>>>> f08 bindings (aka use mpi_f08) works fine
>>>>
>>>> Cheers,
>>>>
>>>> Gilles
>>>>
>>>> On Thursday, September 24, 2015, Jeff Squyres (jsquyres)
>>>> <jsquy...@cisco.com> wrote:
>>>> I looked into the MPI_BCAST problem -- I think we (Open MPI) have a
>>>> problem with the mpi_f08 bindings and the Intel 2016 compilers.
>>>>
>>>> It looks like configure is choosing to generate a different pragma for
>>>> Intel 2016 vs. Intel 2015 compilers, and that's causing a problem.
>>>>
>>>> Let me look into this a little more...
>>>>
>>>>
>>>>
>>>>> On Sep 24, 2015, at 11:09 AM, Fabrice Roy <fabrice....@obspm.fr> wrote:
>>>>>
>>>>> Hello,
>>>>>
>>>>> Thanks for the quick answer.
>>>>> I think I cannot use mpi_f08 in my code because I am also using parallel
>>>>> HDF5 which does not seem to be compatible with the Fortran 2008 module.
>>>>> I will ask Intel what they think about this problem.
>>>>> Thanks,
>>>>>
>>>>> Fabrice
>>>>>
>>>>>
>>>>> Le 24/09/2015 02:18, Gilles Gouaillardet a écrit :
>>>>>> Fabrice,
>>>>>>
>>>>>> i do not fully understand the root cause of this error, and you might
>>>>>> want to ask Intel folks to comment on that.
>>>>>>
>>>>>> that being said, and since this compiler does support fortran 2008, i
>>>>>> strongly encourage you to
>>>>>> use mpi_f08
>>>>>> instead of
>>>>>> use mpi
>>>>>>
>>>>>> a happy feature/side effect is that your program compiles and runs just
>>>>>> fine if you use mpi_f08 module (!)
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> Gilles
>>>>>>
>>>>>>
>>>>>> On 9/24/2015 1:00 AM, Fabrice Roy wrote:
>>>>>>> program testmpi
>>>>>>> use mpi
>>>>>>> implicit none
>>>>>>>
>>>>>>> integer :: pid
>>>>>>> integer :: ierr
>>>>>>> integer :: tok
>>>>>>>
>>>>>>> call mpi_init(ierr)
>>>>>>> call mpi_comm_rank(mpi_comm_world, pid,ierr)
>>>>>>> if(pid==0) then
>>>>>>> tok = 1
>>>>>>> else
>>>>>>> tok = 0
>>>>>>> end if
>>>>>>> call mpi_bcast(tok,1,mpi_integer,0,mpi_comm_world,ierr)
>>>>>>> call mpi_finalize(ierr)
>>>>>>> end program testmpi
>>>>>> _______________________________________________
>>>>>> users mailing list
>>>>>> us...@open-mpi.org
>>>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>>>> Link to this post:
>>>>>> http://www.open-mpi.org/community/lists/users/2015/09/27657.php
>>>>> --
>>>>> Fabrice Roy
>>>>> Ingénieur en calcul scientifique
>>>>> LUTH - CNRS / Observatoire de Paris
>>>>> 5 place Jules Janssen
>>>>> 92190 Meudon
>>>>> Tel. : 01 45 07 71 20
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> users mailing list
>>>>> us...@open-mpi.org
>>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>>> Link to this post:
>>>>> http://www.open-mpi.org/community/lists/users/2015/09/27660.php
>>>>
>>>> --
>>>> Jeff Squyres
>>>> jsquy...@cisco.com
>>>> For corporate legal information go to:
>>>> http://www.cisco.com/web/about/doing_business/legal/cri/
>>>>
>>>> _______________________________________________
>>>> users mailing list
>>>> us...@open-mpi.org
>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>> Link to this post:
>>>> http://www.open-mpi.org/community/lists/users/2015/09/27663.php
>>>> _______________________________________________
>>>> users mailing list
>>>> us...@open-mpi.org
>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>> Link to this post:
>>>> http://www.open-mpi.org/community/lists/users/2015/09/27664.php
>>>
>>
>> --
>> Fabrice Roy
>> Ingénieur en calcul scientifique
>> LUTH - CNRS / Observatoire de Paris
>> 5 place Jules Janssen
>> 92190 Meudon
>> Tel. : 01 45 07 71 20
>>
>>
>> _______________________________________________
>> users mailing list
>> us...@open-mpi.org
>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
>> Link to this post:
>> http://www.open-mpi.org/community/lists/users/2015/09/27667.php
>
>
> --
> Jeff Squyres
> jsquy...@cisco.com
> For corporate legal information go to:
> http://www.cisco.com/web/about/doing_business/legal/cri/
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
