Re: [OMPI users] MPI_DATATYPE_NULL and MPI_AlltoallW

Hi Gilles, Thank you for the follow-up. I appreciate the discussion and am glad that you will put this on your agenda. Jim On Wed, Jan 13, 2016 at 5:28 PM, Gilles Gouaillardet wrote: > Jim, > > your initial question was > > i think that this is a bug in open-mpi - would

Re: [OMPI users] MPI_DATATYPE_NULL and MPI_AlltoallW

Jim, your initial question was i think that this is a bug in open-mpi - would you agree? and so far, the answer is we disagree, this is not an OpenMPI bug, this is the MPI 3.1 standard. and your last question was Can you make any argument in support of not allowing it (other that that's

Re: [OMPI users] RMA operations with java buffers

Hi Marko, You can probably find examples of what you'd like to do on github: https://github.com/open-mpi/ompi-java-test There are numerous MPI-2 RMA examples in the one-sided subdirectory. If you've never used github before, jus click on the download as zip button in the upper right hand

[OMPI users] RMA operations with java buffers

Hello,   I work with the java open mpi version and I want to send byte arrays with the mpi get function. The window provides a large buffer containing the array values and a single call of get should provide the process with a small part of that buffer but not necessarily starting at position 0

Re: [OMPI users] MPI_DATATYPE_NULL and MPI_AlltoallW

Bill Gropp's statement on the Forum list is clear: null handles cannot be used unless explicitly permitted. Unfortunately, there is no exception for MPI_DATATYPE_NULL when count=0. Hopefully, we will add one in MPI-4. While your usage model is perfectly reasonable to me and something that I

Re: [OMPI users] MPI_DATATYPE_NULL and MPI_AlltoallW

It seems to me that when there is a question of interpretation of the standard one should ask the consequences of each potential interpretation. It just makes sense that MPI_DATATYPE_NULL should be allowed when the count is 0, otherwise you need to insert some random datatype just to fill the

[OMPI users] MPI_Type_free and non-blocking operations

Thomas, thanks for the report, at first glance, libnbc (the default module that implements non blocking collective) does not retain/release datatypes, that is why you ran into this kind of trouble. I quickly checked the code, and it seems this kind of mechanism is also missing for

[OMPI users] Call for Papers: 4th Alchemy Workshop on Manycore programming

Please accept our apologies if you receive multiple copies of this CfP. *** * ALCHEMY Workshop 2016 * Architecture, Languages, Compilation and Hardware support for Emerging ManYcore systems * * Held in

[OMPI users] MPI_Type_free and non-blocking operations

Dear friends of Open MPI, I am currently facing a problem in connection with MPI_Ibcast and MPI_Type_free. I've been able to isolate the problem in a minimalistic test program which I attached. Maybe some of you can tell me what I am doing wrong or confirm that this might be a bug in Open

Re: [OMPI users] MPI_DATATYPE_NULL and MPI_AlltoallW

Thanks Jeff, i found it at http://lists.mpi-forum.org/mpi-forum/2016/01/3152.php i'd like to re-iterate what i wrote earlier about example 4.23 MPI_DATATYPE_NULL is used as a recv type on non root tasks, and per the mpi 3.1 standard, recv type is "significant only at root" in the case of

Re: [OMPI users] MPI_DATATYPE_NULL and MPI_AlltoallW

There's a thread about this on the MPI Forum mailing list already ;-) Jeff On Tuesday, January 12, 2016, Gilles Gouaillardet wrote: > Jim, > > if i understand correctly, George point is that OpenMPI is currently > correct with respect to the MPI standard : >