[Nek5000-users] Nek5000-users Digest, Vol 92, Issue 2

[nek5000-users at lists.mcs.anl.gov]

Dear Sudhakar,

Thank you for bringing this to our attention.  We're looking into it.

The issue is that mpi_irecv is not posted prior to the mpi_send.
This situation yields platform-dependent behavior.  (We normally would write
mpi_irecv(X) followed by mpi_send(Y) to X.

We're currently working with the neknek group on a rewrite that
will fix the problem.  In the meantime, if you are targeting an HPC system, 
you might find that there is no issue on that system and in fact be able to
proceed with your development.  (We've had many people using neknek in
the past on fairly large systems without running into the problem---but
it is technically not correct as written and we're going to resolve it.)

Paul 

________________________________________
From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov]
Sent: Thursday, October 06, 2016 4:41 AM
To: nek5000-users at lists.mcs.anl.gov
Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 92, Issue 2

Hi all,

Sorry for the confusion in the first post. Any 3D Neknek simulation is
stuck endlessly in the following line

call mpi_send(jsend,len,mpi_byte,id,100+nid, intercomm,ierr)

within "intpts_locate" subroutine in multimesh.f.

Could anyone please share if you have experienced similar issues, and
how can this be resolved?
Thank you very much,

Best regards,
Sudhakar


Quoting nek5000-users-request at lists.mcs.anl.gov:

> Send Nek5000-users mailing list submissions to
>       nek5000-users at lists.mcs.anl.gov
>
> To subscribe or unsubscribe via the World Wide Web, visit
>       https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
> or, via email, send a message with subject or body 'help' to
>       nek5000-users-request at lists.mcs.anl.gov
>
> You can reach the person managing the list at
>       nek5000-users-owner at lists.mcs.anl.gov
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Nek5000-users digest..."
>
>
> Today's Topics:
>
>    1. Issue in 3D NEKNEK simulations (nek5000-users at lists.mcs.anl.gov)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 5 Oct 2016 15:03:33 +0200
> From: nek5000-users at lists.mcs.anl.gov
> To: nek5000-users at lists.mcs.anl.gov
> Subject: [Nek5000-users] Issue in 3D NEKNEK simulations
> Message-ID:
>       <mailman.4026.1475672736.3602.nek5000-users at lists.mcs.anl.gov>
> Content-Type: text/plain; charset="utf-8"; Format="flowed"
>
> Hi,
>
> I am coupling the Navier-Stokes and Darcy equations using the NekNek
> framework. Everything is working fine in 2D. However, any 3D example in
> neknek is stuck endlessly in the following line
>
> callmpi_send
> <http://www.mcs.anl.gov/research/projects/nek5/docs/html/mpi__dummy_8f.html#af98c20c7332635506f852083f7ee504a>(jsend,len,mpi_byte,id
> <http://www.mcs.anl.gov/research/projects/nek5/docs/html/findpts__test_8c.html#a37e2ea357c25dfa93b154d03c233cd23>,100+nid,
> intercomm,ierr)
>
> within the "intpts_locate" subroutine in multimesh.f. When I modify the
> eddy_neknek test case given in the example folder, the behavior is the
> same, so the issue is not specific to my examples. I already checked
> that the problem is not the memory or the length of the data processed
> in MPI_send. Did anyone experience the same before?
>
> Thank you very much,
>
> Best regards,
>
> Sudhakar
>
> The Royal institute of Technology (KTH)
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20161005/e1f98db4/attachment-0001.html>
>
> ------------------------------
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>
>
> End of Nek5000-users Digest, Vol 92, Issue 2
> ********************************************


_______________________________________________
Nek5000-users mailing list
Nek5000-users at lists.mcs.anl.gov
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users